Ebook Solutions For Biots Poroelastic Theory in Key Engineering Fields Theory and Applications PDF Full Chapter PDF

Solutions for Biot's Poroelastic Theory
in Key Engineering Fields. Theory and

Applications - eBook PDF
Visit to download the full and correct content document:
https://ebooksecure.com/download/solutions-for-biots-poroelastic-theory-in-key-engin
eering-fields-theory-and-applications-ebook-pdf/
SOLUTIONS FOR
BIOT’S POROELASTIC
THEORY IN KEY
ENGINEERING FIELDS
SOLUTIONS FOR
BIOT’S POROELASTIC
THEORY IN KEY
ENGINEERING FIELDS
Theory and Applications
YUANQIANG CAI
HONGLEI SUN
Elsevier
Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
© 2017 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means, electronic
or mechanical, including photocopying, recording, or any information storage and retrieval system,
without permission in writing from the publisher. Details on how to seek permission, further information
about the Publisher’s permissions policies and our arrangements with organizations such as the
Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website:
www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by the Publisher
(other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and experience broaden
our understanding, changes in research methods, professional practices, or medical treatment may
become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in evaluating and
using any information, methods, compounds, or experiments described herein. In using such information
or methods they should be mindful of their own safety and the safety of others, including parties for
whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume
any liability for any injury and/or damage to persons or property as a matter of products liability, negligence
or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in
the material herein.
Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
ISBN: 978-0-12-812649-3
For information on all Elsevier publications visit our website

at https://www.elsevier.com/books-and-journals
Publisher: Glyn Jones

Acquisition Editor: Simon Tian
Editorial Project Manager: Katie Chan
Production Project Manager: Anusha Sambamoorthy
Cover Designer: Victoria Pearson
Typeset by SPi Global, India
Introduction
All structures made by human beings have to be placed on or in the soil. At the very
beginning, these soil foundations are only subjected to static loads, which are loads that
build up gradually over time or with negligible dynamic effects, also known as monotonic
loads. In the 1930s, due to the rapid development of machine manufacturing and trans-
portation industries, the dynamic interaction between structural foundations and the
underlying soil behavior under the action of cyclic loads started to receive considerable
attention in a number of engineering fields. Cyclic loads are loads which exhibit a degree
of regularity both in their magnitude and frequency. Stress reversals, rate effects and
dynamic effects are the important factors that distinguish cyclic loads from static loads.
Practically speaking, no real cyclic loads exist in nature; however, many kinds of loads
can be simplified into cyclic loads for the convenience of study, analysis and design. For
example, the operation of a reciprocating or a rotary machine typically produces a cyclic
load. The passing of a long train can be considered a cyclic load. Even cars running on a
road on the same line can be simplified as a cyclic load. Cyclic loads acting on the struc-
tures and soil can produce elastic waves in the ground which will act on the surrounding
foundations and soil. These actions can cause environmental and safety problems. As a
result, it is very important to take a deep look into this area to advance the knowledge
regarding the theory of vibrations, the principles of wave propagation, and numerical
methods in finding appropriate solutions for problems of practical interest.
In the past few decades, many studies have been carried out on soil-structure inter-
actions under cyclic loads. Most of them have treated the soil as an elastic or viscoelastic
medium. However, there is underground water in what is considered soil medium, such
that the soil is actually a two-phase medium. Biot [1] pioneered the development of
an elastodynamic theory for a fluid-filled elastic porous medium. Since its publication,
Biot’s theory has had wide applications in the geotechnical professions for analyzing wave
propagation characteristics under cyclic loads. The aim of this book is to provide a tutorial
and a state-of-the-art compilation of the advances in the applications of Biot’s theory.
vii
CHAPTER 1
Basic Equations and Governing

Equations
This chapter introduces the basic equations, the governing equations and the boundary
conditions of Biot’s theory and its transformed forms, which are essential for solving
engineering problems in a fully saturated poroelastic medium.
We use two kinds of coordinate systems in this book, a Cartesian coordinate system
and a cylindrical coordinate system. The relationships between these coordinate systems
are given in Table 1.1.1, in which angle θ (0 θ 2π) is measured from the positive
direction of the x-axis to that of the y-axis. In these coordinates, we use (ux, uy, uz),
(ur, uθ, uz) to denote the displacements at a point in the solid in Cartesian and cylindrical
coordinate systems, respectively. The footnotes demonstrate along which direction the dis-
placement occurs. We can also use them in matrix form as {u} ¼ [ux, uy, uz]T or [ur, uθ, uz]T,
respectively, where the superscript T stands for transpose.
We write the stresses and strains in the Cartesian and cylindrical systems,
respectively, as
2 1 1 3
ε γ γ
2 3 6 xx 2 xy 2 zx 7
σ xx τxy τzx 6 7
6 1 7
4 τxy σ yy τyz 5 and 6 1 γ εyy γ 7
6 2 xy 2 yz 7
τzx τyz σ zz 6 7
4 5
1 1
γ γ yz εzz
2 zx 2
2 1 1 3
εrr γ rθ γ zr
2 3 6 2 2 7
σ rr τrθ τzr 6 7
6 1 7
4 τrθ σ θθ τθz 5 and 6 1 γ εθθ γ 7
6 2 rθ 2 θz 7
τzr τθz σ zz 6 7
4 5
1 1
γ γ θz εzz
2 zr 2
where σ ij is the stress tensor, τij is the shear stress, γ ij is the shear strain, and εij is the
strain tensor.
Solutions for Biot’s Poroelastic Theory in Key Engineering Fields © 2017 Elsevier Inc.
http://dx.doi.org/10.1016/B978-0-12-812649-3.00001-0 All rights reserved. 1
2 Solutions for Biot's Poroelastic Theory in Key Engineering Fields
Table 1.1.1 Direction cosines between coordinate axes in Cartesian and cylindrical coordinates
x y z
r cos θ sin θ 0
θ sin θ cos θ 0
z 0 0 1
The transformation rule between the stresses and strains in different coordinates is
σ p0 q0 ¼ lp0 i lq0 j σ ij (1.1.1)
εp0 q0 ¼ lp0 i lq0 j εij (1.1.2)
where σ ij denotes the stresses in Cartesian coordinates, σ p0 q0 stands for the stresses in a new
Cartesian coordinate system after rotation, and lp0 i and lq0 j represent the direction cosines
between two coordinate axes.
Combining Eq. (1.1.1) with Table 1.1.1, the relationships between the stresses in
cylindrical coordinates and Cartesian coordinates can be easily derived:
σ r ¼ σ x cos 2 θ + σ y sin 2 θ + 2τxy sin θ cos θ
σ α ¼ σ x sin 2 θ + σ y cos 2 θ 2τxy sin θ cos θ

τrα ¼ σ y σ x sin θ cos θ + τxy cos 2 θ sin 2 θ
(1.1.3)
τzr ¼ τzx cos θ + τyz sin θ
ταz ¼ τzx sin θ + τyz cos θ
σz ¼ σz
1.1. BASIC EQUATIONS

The basic equations for a linear elastic medium are geometric equations (strain-displacement
relations), equations of motion and constitutive equations (stress-strain relations).
1.1.1 Geometric Equations

The geometric equations in Cartesian coordinates are given as:
@ux @uy @uz
εx ¼ , γ yz ¼ +
@x @z @y
@uy @uz @ux
εy ¼ , γ zx ¼ + (1.1.4)
@y @x @z
@uz @ux @uy
εz ¼ , γ xy ¼ +
@z @y @x
In cylindrical coordinates, they become
Basic Equations and Governing Equations 3
@ur @uθ @uz

εr ¼ , γ θz ¼ +
@r @z r@θ
1 @uθ ur @uz @ur
εθ ¼ + , γ zr ¼ + (1.1.5)
r @θ r @r @z
@uz 1 @ur @uθ uθ
εz ¼ , γ rα ¼ +
@z r @θ @r r
The tensor form of the geometric equations in Cartesian coordinates can be written as
1
εij ¼ ui, j + uj, i (1.1.6)
2
with 2εij ¼ γ ij when i 6¼ j.
The matrix form of the equations is given as
ε ¼ E T ðrÞu (1.1.7)
in which u ¼ [ux, uy, uz]T and E(r) is an operator matrix defined by

2 3
@ @ @
6 @x 0 0 0 @z @y 7
6 7
6 7
6 @ @ @ 7
E ðrÞ ¼ 6
6 0 @y 0 @z 0 @x 7
7 (1.1.8)
6 7
6 7
4 @ @ @ 5
0 0 0
@z @y @x
1.1.2 Equations of Motion

When the governing equations of a dynamic problem are set at the equilibrium position,
the body force is omitted in the equations. In Cartesian coordinates, we have the equa-
tions of motion as
@σ x @τyx @τzx @ 2 ux
+ + ¼ρ 2
@x @y @z @t
@σ y @τxy @τzy @ 2 uy
+ + ¼ρ 2 (1.1.9)
@y @x @z @t
@σ z @τxz @τyz @ 2 uz
+ + ¼ρ 2
@z @x @y @t
where ρ is the density of the material.
Using Eq. (1.1.3), in cylindrical coordinates, the equations of motion become
@σ r 1 @τrθ @τzr σ r σ θ @ 2 ur
+ + + ¼ρ 2
@r r @θ @z r @t
@τrθ 1 @σ θ @τθz 2τrθ @ 2 uθ
+ + + ¼ρ 2 (1.1.10)
@r r @θ @z r @t
@τzr 1 @τθz @σ z τzr @ 2 uz
+ + + ¼ρ 2
@r r @θ @z r @t
The tensor form of the equations of motion in Cartesian coordinates can be written as
σ ij, j ¼ ρu€i , ði ¼ x,y,z or r, θ, zÞ (1.1.11)
where the dots above the symbols denote partial differentiation with respect to time t.
1.1.3 Constitutive Equations

The tensor form of Hooke’s law in Cartesian coordinates can be written as
σ ij ¼ cijkl εkl , ði, j, k, l ¼ x, y, z or r, θ, zÞ (1.1.12)
where cijkl are components of a fourth-rank tensor including 81 components. Since the
stress vectors are symmetric, the exchange of the indices i and j does not alter the result.
Noting that the strain vectors are symmetric as well, the same process can be done to the
indices k and l; and then we have the relationships
cijkl ¼ cjikl and cijkl ¼ cijlk
In addition, since we are considering the adiabatic process, we still have the following
relationship:
cijkl ¼ cklij
Thus, among the 81 components of cijkl, the maximum number of independent ones is
21. For a homogeneous medium, the number of independent components goes down to
2, which are the Lame constant λ and μ, or Young’s modulus E and Poisson’s ratio ν. And
the tensor form of Hooke’s law can be simplified as
σ ij ¼ 2μεij + λδij εkk (1.1.13)

1 when i ¼ j
where δij ¼
0 when i 6¼ j
Expanding Eq. (1.1.13), we have

σ x ¼ λe + 2μεx
σ y ¼ λe + 2μεy
σ z ¼ λe + 2μεz
(1.1.14)
τxy ¼ μγ xy ¼ 2μεxy
τyz ¼ μγ yz ¼ 2μεyz
τzx ¼ μγ zx ¼ 2μεzx
νE
where λ and μ are Lame’s constant with the relationship λ ¼ .
ð1 + νÞð1 2νÞ
e ¼ εx + εy + εz , which is called the volume strain or the matrix dilation.
The corresponding matrix form of the equations is given as
σ ¼ Cε (1.1.15)
where σ and ε are vectors of stress and engineering strain, respectively. In Cartesian coor-
dinates, they become
h iT
σ ¼ σ x , σ y , σ z , γ xy , γ yz , γ zx
h iT
ε ¼ εx , εy , εz , γ xy , γ yz , γ zx
And C should be a nonsingular and reversible matrix, which can be written as

2 3
λ + 2μ λ λ 0 0 0
6 λ λ + 2μ λ 0 0 07
6 7
6 7
6 λ λ λ + 2μ 0 0 0 7
C¼6 6 7 (1.1.16)
6 0 0 0 μ 0 077
6 7
4 0 0 0 0 μ 05
0 0 0 0 0 μ
1.2. GOVERNING EQUATIONS OF A FULLY SATURATED

POROELASTIC MEDIUM
This section introduces the governing equations for a fully saturated poroelastic medium.
1.2.1 Governing Equations in Cartesian Coordinates

As for modeling the dynamic responses of a saturated porous medium, Biot [3] was the
first to give the theory that presents three kinds of coupling (viscous, inertial, and
mechanical) between the porous solid skeleton and pore fluid, and demonstrated that
two kinds of longitudinal waves (P1 and P2 waves) and one kind of rational wave (the
S wave) exist in the saturated porous medium. The existence of the P2 wave always dis-
tinguishes the two-phase medium from the single-phase one. The basic variables of Biot’s
theory are always the solid skeleton displacement (u) and the average displacement of
pore fluid relative to the solid skeleton (w).
Considering the concept of effective stress of the saturated mixture, the relationship
between effective stress, total stress and pore pressure can be expressed as
σ 0ij ¼ σ ij + αδij p (1.1.17)
where σ ij0 is the effective stress tensor, σ ij is the total stress tensor, δij is the Kronecker delta,
and α is the Biot constant that depends on the geometry of material voids. For the most
part, in soil mechanics problems, α 1 can be assumed. The relationship between total
stress and effective stress becomes
σ 0ij ¼ σ ij + δij p (1.1.18)
which corresponds to the classical effective stress definition by Terzaghi. Thus the tensor
form of the equations of motion for a fully saturated poroelastic medium becomes (omit-
ting the body force):
σ ij, j ¼ ρu€i + ρf w€i (1.1.19)
where the dots above the symbols denote partial differentiation with respect to time t;
thus u€i is the acceleration of the solid part, wi is the fluid displacement relative to
the solid part, and w€i is the fluid acceleration relative to the solid part. For fully saturated
porous media (no air trapped inside), density is equal to ρ ¼ nρf + ð1 nÞρs , where n is
the porosity, and ρs and ρf are the soil particle and water densities, respectively.
For the pore fluid, the equation of momentum balance can be expressed as
p, i ¼ bw_ i ρf u€i mw€i (1.1.20)
where the parameter b ¼ ρfg/kD, where kD is the Darcy permeability of the soil medium
and g is the gravity; p is the pore water pressure.
According to the classical effective stress definition by Terzaghi in Eq. (1.1.17), the
constitutive Eq. (1.1.13) becomes
σ ij ¼ 2μεij + λδij εkk αδij p (1.1.21)
The final equation is the mass conservation of the fluid flow, which is expressed by
p_ ¼ αM e_ + M ς_ (1.1.22)
where
ς ¼ wi, i (1.1.23)
Substituting Eqs. (1.1.21)–(1.1.23) into Eqs. (1.1.19), (1.1.20), a u-w formulation can be
obtained as follows:

μui, jj + λ + α2 M + μ uj, ji + αMwj, ji ¼ ρu€i + ρf w€i (1.1.24)
αMuj, ji + Mwj, ji ¼ ρf u€i + mw€i + bw_ i (1.1.25)
Zienkiewicz et al. [2] proposed a simplified u-p formulation in the context of finite ele-
ment analysis of the liquefaction of saturated sand soil. By neglecting the second time
derivatives of the relative fluid displacement from the original Biot [3] u-w formulation,
the u-p formulation is deduced for reducing the primary variables in the sense of finite
element calculation; there are 3 + 3 and 3 + 1 nodal variables in three-dimensional analysis
for the u-w and u-p formulation, respectively. Moreover, the solid displacement (u) and
the pore fluid pressure ( p) are always the main concern.
By simply neglecting the pore fluid relative acceleration terms, i.e., w€i in Eqs. (1.1.19),
(1.1.20), it is easy to eliminate w_ i using Eqs. (1.1.20), (1.1.22) leaving the ui and p as pri-
mary variables. This simplified formulation is economical and convenient in numerical
analysis, for there are only 3 + 1 nodal variables in the three-dimensional analysis. The
equation set becomes
σ ij, j ¼ ρu€i (1.1.26)

ρf 1
u_i, i + u€i p, i ¼ p=M
_ (1.1.27)
b b ,i
1.2.2 Governing Equations in Cylindrical Coordinates

In cylindrical coordinates, the expanded form of Eqs. (1.1.24), (1.1.25) become

1 2 @uθ @e @ς
μ r ur 2 ur 2
2
+ λ + α2 M + μ αM ¼ ρu€r + ρf w€r
r r @r @r @r

1 2 @u r @e @ς
μ r2 uθ 2 uθ 2 + λ + α2 M + μ αM ¼ ρu€θ + ρf w€θ
r r @θ r@θ r@θ
@e @ς
μr2 uz + λ + α2 M + μ αM ¼ ρu€z + ρf w€z
@z @z (1.1.28)
@e @ς
αM M ¼ ρf u€r + mw€r + bw_ r
@r @r
@e @ς
αM M ¼ ρf u€θ + mw€θ + bw_ θ
r@θ r@θ
@e @ς
αM M ¼ ρf u€z + mw€z + bw_ z
@z @z
where e and ς are the matrix dilation and the fluid dilation relative to the solid, respec-
tively, which are expressed in cylindrical coordinates as

@ur ur @uθ @uz @wr wr @wθ @wz
e¼ + + + , ς¼ + + +
@r r r@θ @z @r r r@θ @z
and r2 denotes the Laplacian operator which is given by
@2 1 @ 1 @2 @2
r2 ¼ + + +
@r 2 r @r r 2 @θ 2
@z2
1.3. BOUNDARY CONDITIONS

In order to obtain a meaningful solution to a specific engineering problem, we need the
external stimuli, called boundary conditions, to solve the governing equations of the fully
saturated poroelastic medium.
The problem is in the area Ω with boundary Γ. For the total momentum balance on
the part of the boundary Γ tN we specify the total traction T i (or in terms of the total stress,
σ ijnj with nj being the jth component of the normal at the boundary) while for Γ sD, the
displacement ū, is given.
For the fluid phase, again the boundary is divided into two parts, Γ pN on which the
values of p are specified and Γ wD where the normal outflow wi is prescribed (for instance, a
zero value for the normal outward velocity on an impermeable boundary).
Summarizing, for the overall assembly, we can thus write
Compulsive boundary conditions

ui ¼ ui on Γ ¼ Γ sD
(1.1.29)
wi ¼ w i on Γ ¼ Γ wD
Natural boundary conditions

σ ij nj ¼ T i on Γ ¼ Γ tN
p (1.1.30)
pni ¼ pn on Γ ¼ Γ N
The relationships between the boundaries
8 p
< Γ ¼ ΓD [ ΓD [ ΓN [ ΓN
> s w t
Γ sD \ Γ tN ¼ ∅ (1.1.31)
>
: w p
ΓD \ ΓN ¼ ∅
CHAPTER 2
Solutions for Saturated Soil Under

Moving Loads and Engineering
Applications
This chapter introduces solutions for saturated ground governed by Biot’s theory to a
moving rectangular pressure. Furthermore, the applications of Biot’s theory to the inves-
tigation of dynamic responses of highway roads and railway tracks subjected to a moving
traffic load are presented in Sections 2.1 and 2.2, respectively.
2.1. GENERAL SOLUTIONS OF SATURATED GROUND UNDER

A MOVING LOAD
The governing equations are given as Eqs. (1.1.24), (1.1.25). The constitutive relation-
ships are given as Eqs. (1.1.21), (1.1.22).
2.1.1 Solutions for Saturated Ground Generated by a Moving

Rectangular Pressure
Fourier transforms with respect to time t are defined as
ð +∞
~f ðx, y, z, ωÞ ¼ f ðx, y, z, t Þeiωt dt (2.1.1)
∞
ð∞
1 ~f ðx, y, z, ωÞeiωt dω
f ðx, y, z, tÞ ¼ (2.1.2)
2π ∞
Fourier transform pairs with respect to x and y are defined as

ð∞ ð∞
~f ðξ, η, z, a0 Þ ¼ ~f ðx, y, z, a0 Þeiðξx + ηyÞ dxdy (2.1.3)
∞ ∞
ð∞ ð∞
~f ðx, y, z, a0 Þ ¼ 1 ~f ðξ, η, z, a0 Þeiðξx + ηyÞ dξdη (2.1.4)
4π 2 ∞ ∞
By using the Fourier transform given in Eqs. (2.1.1), (1.1.24), (1.1.25), (1.1.21), and
(1.1.22) are transformed from partial differential equations to ordinary differential equa-
tions. The governing equations in the x direction are given as
@ θ~ @ ζ~
μr2 u~x + λ + α2 M + μ αM ¼ ω2 ρ~ux ω2 ρf w~x (2.1.5)
@x @x
@ θ~ @ ζ~
αM M ¼ ω2 ρf u~x ω2 mw~x + iωbw~x (2.1.6)
@x @x
@~ p @ ζ~ @ θ~
¼ M αM (2.1.7)
@x @x @x
Substituting Eq. (2.1.7) into Eq. (2.1.5) yields
@ θ~ @~
p
μr2 u~x + ðλ + μÞ
α f ¼ ω2 ρ~
ux ω2 ρf w~x (2.1.8)
@x @x
Substituting Eq. (2.1.7) into Eq. (2.1.6) gives
@~pf

¼ ω2 ρf u~x ω2 mw~x + iωbw
~x (2.1.9)
@x
and then the following equation can be obtained from Eq. (2.1.9):
@~pf
ω2 ρf u~x
w~x ¼ @x (2.1.10)
iωb ω2 m
Then substituting Eq. (2.1.10) into Eq. (2.1.9) yields
@ θ~ @~
p
μr2 u~x + ðλ + μÞ + ω2 ðρ ϑρf Þ~
ux ðα ϑÞ f ¼ 0 (2.1.11)
@x @x
ω2 ρf
where ϑ ¼ 2 :
ω m iωb
Similarly, the following equations in the y and z direction can be obtained:
@ θ~ @~
p
μr2 u~y + ðλ + μÞ + ω2 ðρ ϑρf Þ~
uy ðα ϑÞ f ¼ 0 (2.1.12)
@y @y
@~
pf
¼ ω2 ρf u~y ω2 mw~y + iωbw~y (2.1.13)
@y
@ θ~ @~
p
μr2 u~z + ðλ + μÞ uz ðα ϑÞ f ¼ 0
+ ω2 ðρ ϑρf Þ~ (2.1.14)
@z @z
@~ pf
¼ ω2 ρ~
uz ω2 mw~z + iωbw~z (2.1.15)
@z
Taking the derivative of Eqs. (2.1.9), (2.1.13), (2.1.15) with respect to x, y, z, respec-
tively, and adding them together yields
r2 p~f ¼ ω2 ρf θ~ + ω2 mζ~ iωbζ~ (2.1.16)
Solutions for Saturated Soil Under Moving Loads 11
Eq. (1.1.22) in the transformed domain can be expressed as

p~f ¼ M ζ~ αM θ~ (2.1.17)
ω2 ρf ω2 ρf ðα ϑÞ ~
r2 p~f + p~f + θ¼0 (2.1.18)
ϑM ϑ
By applying the operation r2 on Eq. (2.1.18), the following equation can be obtained:
a20 ρf 2 a2 ρ ðα ϑÞ 2 ~
r4 p~f + r p~f + 0 f r θ¼0 (2.1.19)
ϑM ϑ
With the aid of Eq. (2.1.18), the following equation can then be obtained:
Mr2 p~f + ðω2 m iωbÞ~
pf
θ~ ¼ 2 (2.1.20)
ω ρf M αM ðω2 m iω2 bÞ
Taking the derivative of Eqs. (2.1.11), (2.1.12), (2.1.14) with respect to x, y, z, respec-
tively, and adding them together, the following equations can be obtained:
ðλ + 2μÞr2 θ~ + ω2 ðρ ϑρf Þθ~ ðα ϑÞr2 p~f ¼ 0 (2.1.21)
and thus
ðα ϑÞr2 p~f ω2 ðρ ϑρf Þθ~
r2 θ~ ¼ (2.1.22)
λ + 2μ
By substituting Eqs. (2.1.20), (2.1.22) into Eq. (2.1.19), the following equation can be
obtained:
r4 p~f + β1 r2 p~f + β2 p~f ¼ 0 (2.1.23)

where
ðω2 m iωbÞðλ + α2 M + 2μÞ + ω2 Mρ 2αω2 Mρf
β1 ¼ (2.1.24)
ðλ + 2μÞM
ρω2 ðω2 m iωbÞ ω4 ρ2f
β2 ¼ (2.1.25)
ðλ + 2μÞM
Applying Fourier transforms with respect to x and y, Eq. (2.1.23) is transformed from a
partial differential equation to an ordinary differential equation:
d4 p~f 2 ~f 4
2 d p

4
+ β 1 2ξ2
2η 2
+ ξ + η4 + 2ξ2 η2 β1 ξ2 β1 η2 + β2 p~f ¼ 0 (2.1.26)
dz dz
By solving Eq. (2.1.26), the expression p~f can be obtained:
p~f ¼ Aeγ1 z + Beγ2 z (2.1.27)

At z ! ∞, the dynamic responses become zero; therefore Reðγ i Þ > 0,i ¼ 1,2, where
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
γ 1 ¼ ξ + η L1 , γ 2 ¼ ξ2 + η2 L22
2 2 2 (2.1.28)
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
β1 + β21 4β2 β1 β21 4β2
L12 ¼ , L22 ¼ (2.1.29)
2 2
By using the Fourier transforms given in Eq. (2.1.3) in Eq. (2.1.17), the following equa-
tion can be obtained:

d2 ω2 ρf ω2 ρf ðα ϑÞ ~
ξ η + 2 p~f +
2 2
p~f + θ¼0 (2.1.30)
dz ϑM ϑ

ω2 ρf ðα ϑÞ ~ ω2 ρf γ 1 z ω2 ρ f
θ ¼ γ1 ξ η +
2 2 2
Ae + γ2 ξ η +
2 2 2
Beγ 2 z
ϑ ϑM ϑM
(2.1.31)
and then
θ~ ¼ Aχ 1 eγ 1 z + Bχ 2 eγ2 z (2.1.32)

where
ϑML12 ω2 ρf ϑML22 ω2 ρf
χ1 ¼ , χ ¼ (2.1.33)
ω2 ρf M ðα ϑÞ 2
ω2 ρf M ðα ϑÞ
Applying Fourier transforms to Eq. (2.1.14) with respect to x and y:

d2 @ θ~ @~
p
uz ðα ϑÞ f ¼ 0
μ ξ2 η2 + 2 u~z + ðλ + μÞ + a20 ðρ ϑρf Þ~ (2.1.34)
dz @z @z
By substituting Eqs. (2.1.27), (2.1.32) into Eq. (2.1.34),
d2 u~z 2
μ 2
+ S μξ 2
μη 2
u~z ðλ + μÞðγ 1 Aχ 1 eγ1 z + γ 2 Bχ 2 eγ2 z Þ
dz
+ðα ϑÞðγ 1 Aeγ1 z + γ 2 Beγ 2 z Þ ¼ 0 (2.1.35)
where
S2 ¼ ω2 ðρ ϑρf Þ (2.1.36)
The general solution for Eq. (2.1.35) is u~z ¼ Aa1 γ 1 eγ1 z + Ba2 γ 2 eγ2 z and the particular
solution is u~z ¼ Ceγ3 z ; then the solution of u~z can be expressed as
u~z ¼ Aa1 γ 1 eγ1 z + Ba2 γ 2 eγ2 z + Ceγ 3 z (2.1.37)

where
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
.
S2
γ 3 ¼ ξ + η μ, Reðγ 3 Þ > 0
2 2 (2.1.38)
χ 1 ðλ + μÞ α + ϑ χ ðλ + μÞ α + ϑ
a1 ¼ , a2 ¼ 2 2 (2.1.39)
S μL1
2 2 S μL22
By applying Fourier transforms with respect to x and y, Eq. (2.1.12) can be expressed as
d2 u~y
μ 2 + S2 μξ2 μη2 u~y + iη½ðλ + μÞðAχ 1 eγ 1 z + Bχ 2 eγ2 z Þ
dz
ðα ϑÞðAeγ1 z + Beγ 2 z Þ ¼ 0 (2.1.40)
and thus
u~y ¼ iηðAa1 eγ 1 z + Ba2 eγ2 z Þ + iDeγ3 z (2.1.41)
Applying Fourier transforms with respect to x and y into θ ¼ ui, i , the following equation
can be obtained:
@~
uz
θ~ ¼ iξ~
ux + iη~
uy + (2.1.42)
@z
By substituting Eqs. (2.1.37), (2.1.41) into Eq. (2.1.42), the solutions for u~x can be
obtained as
i
u~x ¼ χ 1 + a1 ξ2 L12 Aeγ1 z + χ 2 + a2 ξ2 L22 Beγ 2 z + ðCγ 3 + DηÞeγ 3 z
ξ
(2.1.43)
The constitutive equation in saturated soil can be expressed as
@uz
σ z ¼ λθ + 2μ αpf (2.1.44)
@z

@ux @uz
τxz ¼ μ + (2.1.45)
@z @x

@uy @uz
τyz ¼ μ + (2.1.46)
@z @y
The expressions in the transformed domain can be expressed as:
@~uz
σ z ¼ λθ~ + 2μ
e α~
p (2.1.47)
@z f
@~
ux
eτxz ¼ μ u
+ iξμ~ (2.1.48)
@z z
@~
uy
eτyz ¼ μ u
+ iημ~ (2.1.49)
@z z
Substituting Eqs. (2.1.27), (2.1.32), (2.1.37), (2.1.41), (2.1.43) into Eqs. (2.1.47)–(2.1.49)
yields
σ z ¼ g3 Aeγ1 z + g4 Beγ 2 z 2μγ 3 Ceγ 3 z

e (2.1.50)
μi
eτxz ¼ g1 γ 1 Aeγ 1 z + g2 γ 2 Beγ 2 z + Cγ 23 + Cξ2 + ηγ 3 D eγ 3 z (2.1.51)
ξ
eτyz ¼ μi½2ηa1 γ 1 Aeγ 1 z + 2ηa2 γ 2 Beγ 2 z + ðCη Dγ 3 Þeγ 3 z (2.1.52)
where

g1 ¼ χ 1 + a1 2ξ2 L12 (2.1.53)

g2 ¼ χ 2 + a2 2ξ2 L22 (2.1.54)
g3 ¼ χ 1 λ 2μa1 γ 21 α (2.1.55)
g4 ¼ χ 2 λ 2μa2 γ 22 α (2.1.56)
A uniformly distributed rectangular pressure is applied to the surface of the

poroelastic half-space. Thus, the boundary conditions on the surface of the half-space
(z ¼ 0 m) are given as: τxz ¼ 0kPa,τyz ¼ 0kPa, pf ¼ 0kPa, and the vertical stress is
expressed as
P jx Vt j a, jyj b
σ z ðx, y, tÞ ¼ (2.1.57)
0 jx Vt j > a, jyj > b
The vertical stress in the transformed domain can be expressed as

8 8πP
>
> sin ξa sin ηb δðξV + a0 Þ ξη 6¼ 0
>
> ξη
>
>
>
>
>
>
< 8πP a sin ηb δða0 Þ ξ ¼ 0,η 6¼ 0
e
σ z ðξ, η, a0 Þ ¼ η (2.1.58)
>
>
>
> 8πP
>
> b sin ξa δðξV + a0 Þ ξ 6¼ 0,η ¼ 0
>
> ξ
>
>
:
8πPab δða0 Þ ξ ¼ 0,η ¼ 0
By substituting the boundary condition into Eqs. (2.1.27), (2.1.50)–(2.1.52), and intro-
ducing the auxiliary spatial coordinate xt ¼ x Vt, the solution in the time domain is
obtained:
ð +∞ ð +∞ ð +∞
1
uz ðxt , y, zÞ ¼ e
σ z ðξ, η, ωÞϕðξ, η, z, ωÞeiξx eiηy eia0 t dξdηdω
ð2π Þ3 ∞ ∞ ∞
ð +∞ ð +∞
1
¼ e
σ z ðξ, ηÞϕðξ,η, z, ξV Þeiξxt eiηy dξdη (2.1.59)
ð2π Þ2 ∞ ∞
where
1 2 2
ϕðξ, η, z, α0 Þ ¼ γ 3 + ξ + η2 ðγ 2 a2 eγ2 z γ 1 a1 eγ 1 z Þ + ðγ 1 g5 γ 2 g6 Þeγ3 z
Δ0
(2.1.60)
2 2
Δ0 ¼ ðg4 g3 Þ γ 3 + ξ + η 2μγ 3 ðγ 1 g5 γ 2 g6 Þ
2
(2.1.61)
2.2. APPLICATION IN HIGHWAY ENGINEERING

With increasing traffic volume and traffic running speeds, the serviceability of highway roads
is increasingly influenced by the traffic. Road traffic loads cause distresses such as cracks and
settlement. Furthermore, due to the low Rayleigh wave velocity in soft soils, vehicles can
possibly approach the Rayleigh wave velocity in the poroelastic half-space. Vibrations
and stresses in the pavement structure and the ground will be amplified in this circumstance,
causing disturbances to residences along the line. It is of great significance to predict the
dynamic responses of pavement-ground systems to the moving load using an accurate model.
2.2.1 Governing Equations for the Pavement Structure

The dynamic response of a pavement on a saturated poroelastic half-space was first inves-
tigated by Cai et al. [4]. The pavement structure was simplified as a thin plate extending to
infinity in the horizontal direction. The model of pavement on the saturated ground is
shown in Fig. 2.2.1.
The Kirchhoff small-deflection thin-plate theory was employed as by Kim and
Roessert [5]. Denoting the reaction of the ground to the plate by F(x, y, t), the governing
equation of the plate is given as follows:
@ 4 w ðx, y, tÞ @ 4 wðx, y, t Þ @ 4 w ðx, y, t Þ @ 2 wðx, y, t Þ
Dp + 2 + + mb ¼ qðx, y, tÞ + F ðx, y, tÞ
@x4 @x2 @y2 @y4 @t 2
(2.2.1)
where w is the vertical displacement of the plate; mb is the mass density of the plate per
unit area; and q(x, y, t) is the traffic load pressure acting on the surface of the road. Dp is the
flexural rigidity of the plate and can be defined as follows:
Eh3
Dp ¼ (2.2.2)
12 1 νp 2
where E, h, and νp are the elastic modulus, thickness, and Poisson’s ratio of the plate,
respectively.
The traffic load is modeled as four rectangular load pressures expressed as:
(
da d1 da + d1 dw d2 dw + d2
qðx, y, tÞ ¼ Q jx Vt j and jyj (2.2.3)
2 2 2 2
0 otherwise
where Q is the magnitude of the load pressure; d1 and d2 are the loaded lengths of a tire
print in the x and y directions, respectively; dw is the distance between the left and right
wheels; and da is the distance between the front and rear wheels. V is the vehicle speed.
The constitutive equations of the thin plate are given as
2
E @ w @2w
σx ¼ + νp 2 (2.2.4)
1 ν2p @x2 @y
h
Plate y x
Observation point
Infinite poroelastic half space
z
Fig. 2.2.1 Theoretical model of the plate-ground system.
2
E @ w @2w
σy ¼ + ν p (2.2.5)
1 ν2p @y2 @x2
2.2.2 Coupling Between the Pavement and Saturated Ground
As the contact surface between the pavement and the poroelastic half-space is assumed to
be smooth and fully permeable, the boundary conditions of the half-space at z ¼ 0 m are
given as follows:
σ z ðx, y, 0, tÞ ¼ F ðx, y, tÞ (2.2.6)
τxz ðx, y, 0, t Þ ¼ 0 (2.2.7)
τyz ðx, y, 0, tÞ ¼ 0 (2.2.8)
pf ðx, y, 0, tÞ ¼ 0 (2.2.9)
uz ðx, y, 0, t Þ ¼ w ðx, y, tÞ (2.2.10)
2.2.3 Solution Methods for the Pavement and Saturated Ground

The solutions of the plate-ground system were obtained based on the Fourier transform.
Applying a triple Fourier transform with respect to x, y, and t, Eq. (2.2.1) yields
h 2 i
Dp ε2 + η2 mb ω2 w~ ¼ ~qðε, η, ωÞ + F~ðε, η, ωÞ (2.2.11)
The traffic load in the transformed domain can be expressed as:

8
> d1 ε d2 η
<
2 e d2a εi e d2w ηi 1 + eda εi 1 + edw ηi δðω + εV Þ εη 6¼ 0
sin sin
~qðε, η, ωÞ ¼ 8πQ 2
>
: εη
8πQd1 d2 δðωÞ ε ¼ 0,η ¼ 0
(2.2.12)
Combined with the boundary conditions Eqs. (2.2.6)–(2.2.10) and the solutions of the
saturated ground presented in Section 2.1, the solutions of the plate-ground system can
be derived as:
uz ðx, y, z, t Þ
ð ð ð ϕðε, η, z, ωÞ
1 ∞ ∞ ∞ h i eiεx eiηy eiωτ dεdηdω
¼ ~
q ð ε, η, ω Þ D ðε 2
2 + η2 Þ m ω2 ϕðε, η, 0, ωÞ + 1
ð2π Þ3 ∞ ∞ ∞ p b
(2.2.13)
where
1 2 2
ϕðε, η, z, ωÞ ¼ γ 3 + ε + η2 ðγ 2 a2 eγ2 z γ 1 a1 eγ1 z Þ + ðγ 1 g5 γ 2 g6 Þeγ3 z (2.2.14)
Δ0
g5 ¼ g1 + 2η2 a1 , g6 ¼ g1 + 2η2 a2 (2.2.15)

Δ0 ¼ ðg4 g3 Þ γ 23 + ε2 + η2 2γ 3 ðγ 1 g5 γ 2 g6 Þ (2.2.16)
With the nontrivial properties of the delta function and introducing auxiliary spatial
coordinate xt ¼ x Vt, Eq. (2.2.13) can be written as
ð ð
1 ∞ ∞ ϕðε, η, z, εc0 Þei½εxt + ηy
uz ðxt , y, zÞ ¼ q ðε, η Þ dεdη
ð2π Þ2 ∞ ∞ Dp ðε2 + η2 Þ2 mb ε2 V 2 ϕðε, η,0, εV Þ + 1
(2.2.17)
2.2.4 Numerical Results and Conclusions
Dimensionless parameters were adopted in the numerical analysis to demonstrate the
general trends in the numerical results. The parameters were nondimensionalized
pffiffiffiffiffiffiffiffi λ M ρ
through x ¼ x1 =a, y ¼ y1 =a, z ¼ z1 =a, τ ¼ ðt=aÞ μ=ρ, λ* ¼ , M* ¼ , ρ* ¼ f ,
μ μ ρ
m ab
m* ¼ , b* ¼ pffiffiffiffiffi, where a is the unit length. The dimensionless parameters for fully
ρ ρμ
saturated ground are presented in Table 2.2.1. The dimensionless parameters of the plate
are presented in Table 2.2.2. The dimensionless parameters of the vehicle can be found in
Table 2.2.3.
The vertical displacements at the surface of the pavement between are studied in
Fig. 2.2.2 considering four different b* values. With the assumption of a constant fluid
viscosity value, then the intrinsic permeability of the soil medium decreases as b increases.
Table 2.2.1 Dimensionless parameters for fully water-saturated poroelastic soil medium
Parameter Value
Shear modulus of solid, μ* 1
Poisson’s ratio, ν 0.35
Solid density, ρs * 1.219
Water density, ρw* 0.671
Porosity, n 0.4
Coefficient of material damping, D 0.05
Ratio between the fluid viscosity and the intrinsic permeability, b* 0.1–100
Bulk modulus of the fluid, Kf * 1.299
Table 2.2.2 Dimensionless parameters of the plate

Parameter Value
Elastic modulus, E* 454.55
Poisson’s ratio, νp 0.15
Plate thickness, h* 0.30
Mass density, mb* 0.47
Damping ratio, D1 0
Table 2.2.3 Dimensionless parameters of tandem-axle loads

Parameter Value
Load pressure, Q* 0.013
Load length, d1* 0.2
Load width, d2* 0.2
Distance between left and right wheels, dw* 1.5
Distance between front and rear wheels, da* 3
In Fig. 2.2.2A for c0 ¼ 0:2, the displacement curves are almost symmetric and have one
peak at the center of the tandem axles. The maximum displacement decreases with
increasing b*. In Fig. 2.2.2B for the high load velocity of c0 ¼ 1:0, the displacements
become much larger than those of c0 ¼ 0:2, and significant ground vibration is observed.
The maximum displacements apparently decrease as b* increases.
The dimensionless maximum vertical displacement against dimensionless load veloc-
ity for different values of flexural rigidity is plotted in Fig. 2.2.3. The displacements are
normalized with respect to the displacement at c0 ¼ 0. The maximum surface displace-
ment response has nearly a static character up to a velocity of c0 ¼ 0:5. However, with a
further increase in load velocity, the displacements increase rapidly and reach a maximum
value near c0 ¼ 0:93 and then decrease rapidly. The critical speed for the poroelastic half-
space is the Rayleigh wave speed, which equals 0.936Vs with a soil Poisson’s ratio of 0.35
according to Richart [6]. Thus, the critical speed of the pavement system is a little smaller
than the Rayleigh wave speed of poroelastic soil. This phenomenon is due to the
dynamic interaction between the plate and poroelastic half-space. It is clearly observed
in Fig. 2.2.3 that the peak values of the normalized maximum displacement curves
decrease as the plate flexural rigidity Dp* increases.
The acceleration responses at the ground surface are investigated in Fig. 2.2.4. For c0 ¼
0:2 in Fig. 2.2.4A, the acceleration curve has one peak at the center of the tandem axle.
0.04
b* = 0.1
0.025 b* = 1
0.03
b* = 0.1 b* = 10
0.020 b* = 1 b* = 100
0.02
b* = 10
u*z(xt,0,0)
u*z(xt,0,0)
0.015 b* = 100
0.01
0.010
0.00
0.005
–0.01
0.000
–30 –20 –10 0 10 20 30 –40 –30 –20 –10 0 10 20 30 40
(A) xt (B) xt
Fig. 2.2.2 Effects of the soil permeability coefficients on the displacement responses. (A) c0 ¼ 0:2;
(B) c0 ¼ 1:0:
2.0
Dp* = 0.05
Normalized max. vert. displacement

1.8
Dp* = 0.5
1.6 Dp* = 5
D * = 50
1.4 p
1.2
1.0
0.8
0.6
0.0 0.2 0.4 0.6 0.8 1.0 1.2
C
0
Fig. 2.2.3 Maximum vertical displacement against dimensionless velocity c0.
The permeability of the soil has little effect on ground acceleration responses. When the
load velocity exceeds the Rayleigh wave speed, as shown in Fig. 2.2.4B, the accelerations
apparently vibrate to a wide extent, and the magnitude of the accelerations is much larger
than that for c0 ¼ 0:2. Thus, as the vehicle speed exceeds the critical speed, the vibrations of
the road increase significantly. The permeability affects the acceleration responses signif-
icantly at this vehicle speed. The acceleration responses decrease as b* increases.
0.015
b* = 0.1
Dimensionless vertical acceleration
Dimensionless vertical acceleration
0.001
0.010 b* = 1
b* = 10
0.000 0.005 b* = 100
b* = 0.1
0.000
–0.001 b* = 1
b* = 10
–0.005
–0.002 b* = 100
–0.010
–0.003
–30 –20 –10 0 10 20 30 –40 –30 –20 –10 0 10 20 30 40
xt xt
(A) C0 = 0.2 (B) C0 = 1.0
Fig. 2.2.4 Acceleration responses at the surface of the ground. (A) c0 ¼ 0.2; (B) c0 ¼ 1.0.
1.0
1.2
0.8 Dp* = 0.05 Dp* = 0.05
0.6 Dp* = 0.5 Dp* = 0.5
0.8
Dp* = 5 Dp* = 5
0.4
s*z(xt,0,0)
Dp* = 50 Dp* = 50
s*z(xt,0,0)
0.4
0.2
0.0 0.0
–0.2
–0.4
–0.4
–0.6 –0.8
–30 –20 –10 0 10 20 30 –40 –20 0 20 40
xt xt
(A) C0 = 0.2 (B) C0 = 1.0
Fig. 2.2.5 The vertical stresses of the plate for different Dp*. (A) c0 ¼ 0.2; (B) c0 ¼ 1.0.
The vertical stresses of the plate generated by the moving traffic were obtained from
the compatibility condition between the plate and the poroelastic half-space. In
Fig. 2.2.5, the vertical stress at the interface between the plate and the half-space is plotted
against xt for different Dp* when c0 ¼ 0:2. Clearly, the magnitude of the stress decreases
rapidly with increasing Dp*. When D∗p ¼ 0:05, the positive peak stresses occurred at the
point at which the load was applied. And negative stress was caused at the center of the
tandem axles because the plate was arched during the load area when the flexural rigidity
of the plate was small. As Dp* increased, the maximum stresses occurred at the center of
the tandem axles, and no negative stress was caused. In Fig. 2.2.5B, the vertical stress for
high load velocity c0 ¼ 1:0 is presented. The vertical stresses of the plate fluctuate signif-
icantly faced with the load and the frequency of fluctuations decreased with increasing
Dp*. Compared with Fig. 2.2.5A, the magnitude of the vertical stress increased slightly
when the load velocity increased to c0 ¼ 1:0. The effect of c0 on the magnitude of the
vertical stress at the contact surface was not pronounced.
2.3. APPLICATIONS IN RAILWAY ENGINEERING

In terms of environmental pollution, the consequences of railway traffic become more
important in view of the fact that the speed of the trains has increased rapidly, and the vibra-
tions and dynamic stresses in the track base have intensified. It will be important to predict
the dynamic responses of the track and ground induced by the moving traffic load.
2.3.1 Governing Equations for the Track System

The vibrations of a saturated poroelastic half-space generated by high-speed trains were
first investigated by Cao et al. [7]. The vehicle-track-ground coupling system is shown in
Fig. 2.3.1. The track model proposed by Picoux and Le Houedec [8] is introduced in this
chapter.
In order to calculate the receptance of the track-ground system, a unit vertical har-
monic load eiΩt was applied on the rail, which was pointing downwards and located at
x ¼ 0 when t ¼ 0, moving along the rails at speed c. The governing equation for a rail
represented by an Euler beam is written as
@ 4 uR ðx, tÞ @ 2 uR
EI + m R + kP ½uR ðx, t Þ uS ðx, tÞ ¼ eiΩt δðx ct Þ (2.3.1)
@x 4 @t 2
in which uR is the vertical displacement of the Euler beam, EI is the bending stiffness of
the rail beam, mR is the mass of the rail per unit length, kP denotes the spring constant of
the rail pads, and uS is the vertical displacement of the sleepers.
It is noted by Knothe and Grassie [9] and Grassie et al. [10] that the effect of the dis-
creteness of the rail supports may be neglected when the dominating frequency content
of the response is not in the vicinity of the so-called pinned-pinned resonance frequency.
This frequency is normally located in the range 700–1000 Hz, depending on the rail
properties and the sleeper distance. The main frequency range of interest for the percep-
tion of ground vibration is about 5–80 Hz; therefore it is suitable to represent the sleepers
by a continuous mass:
@ 2 uS ðx, tÞ
mS + kP ½uS ðx, tÞ uR ðx, t Þ ¼ FS ðx, tÞ (2.3.2)
@t 2
where mS is the mass of the sleeper per unit length and FS is the load between the sleepers
and the ballast.
The ballast was first considered by the Cosserat model (Suiker et al. [11]). At the top
and bottom of the ballast, the system can be written:
Rail
Sleeper
Ballast
x
Poroelastic half-space
z
Fig. 2.3.1 Theoretical model of the vehicle-track-ground coupling system.
mB @ 2 uS ðx, tÞ @ 2 uB ðx, tÞ
2 + + kB ½uS ðx, tÞ uR ðx, tÞ ¼ FS ðx, tÞ (2.3.3)
6 @t 2 @t 2
mB @ 2 uS ðx, tÞ @ 2 uB ðx, tÞ
+ 2 + kB ½uS ðx, t Þ + uB ðx, tÞ ¼ FB ðx, t Þ (2.3.4)
6 @t 2 @t 2
where mB is the mass of the ballast per unit length, kB is the spring constant between bal-
last and sleepers, FB is the ballast load on the soil, and uB is the vertical displacement of the
ballast.
2.3.2 Coupling Between the Track and Saturated Ground

The dimensionless variables were adopted to obtain the general trends. All the displace-
ments were nondimensionalized with respect to the unit length a. Pore water pressures
and stresses were nondimensionalized with respect to the shear modulus μ. The load
between the rail, sleeper, ballast, and ground was nondimensionalized with respect to
μa2. All variables were then replaced by the corresponding dimensionless quantities,
denoted by a superscript asterisk (*). The dimensionless time is defined as:
pffiffiffiffiffiffiffiffi
τ ¼ ðt=aÞ μ=ρ (2.3.5)
kB kP KV
The following nondimensional parameters are also defined: k∗B ¼ , k∗P ¼ , K∗V ¼ ,
μa μa μa
∗ ∗
M V λ M ρ m ab EI m m
M∗V ¼ 3 , λ* ¼ , M* ¼ , ρ∗f ¼ , m* ¼ , b* ¼ pffiffiffiffiffi, β ¼ 2 , m∗R ¼ 2 , m∗S ¼ S2 ,
f R
ρa μ μ ρ ρ ρμ μa ρa ρa
pffiffiffiffiffiffiffiffiffi
c0 ¼ V =Vs , Vs is the shear wave velocity of the half-space, expressed as Vs ¼ μ=ρs , a is
the unit length, V is the vehicle speed.
The drainage conditions between the track and the poroelastic half-space were assumed
to be permeable. Then the boundary conditions of the half-space are given as follows:
τxz ðx*, y*, 0, τÞ ¼ 0 (2.3.6)
1
σ zz ðx*, y*, 0, τÞ ¼ Π ðy*ÞFB∗ ðx*, τÞ (2.3.7)
2a
τyz ðx*, y*, 0, τÞ ¼ 0 (2.3.8)
pðx*, y*, 0, τÞ ¼ 0 (2.3.9)
uz ðx*, 0, 0, τÞ ¼ uB ðx*, τÞ (2.3.10)
The load distributing function is defined as
1 jyj L*Bal
ΠðyÞ ¼ (2.3.11)
0 jyj > L*Bal
LBal is the half-width of the ballast.
From Eqs. (2.3.6) to (2.3.10), the following equation can be obtained:

ΠðηÞ ∗
u~z ðε, η, z*, ωÞ ¼ ~
∗ F B ðε, ωÞϕðε, η, z*, ωÞ
2LBal
(2.3.12)
where
1 h 2 2 i
ϕðε, η, z*, ΩÞ ¼ γ 3 + ε + η2 γ 2 a2 eγ2 z* γ 1 a1 eγ 1 z* + ðγ 1 g5 γ 2 g6 Þeγ3 z*
Γ
2
gi ¼ Ei + 2ε bi 2 Fi i ¼ 1, 2, g3 ¼ λ*E1 2γ 21 F1 α, g4 ¼ λ*E2 2γ 22 F2 α

g5 ¼ g1 + 2η2 F1 , g6 ¼ g2 + 2η2 F2 , Γ ¼ ðg4 g3 Þ γ 23 + ξ2 + η2 2γ 3 ðγ 1 g5 γ 2 g6 Þ
In Eq. (2.3.12), F~B ðε, ωÞ remains unknown and will be resolved in the next section.
2.3.3 Solutions for the Track-Ground System

Applying the Fourier transform with respect to x* and τ for Eqs. (2.3.1)–(2.3.4), (2.3.7),
the following dimensionless equations can be obtained by eliminating FS*(ξ, ω):
uR ðξ, ωÞ k∗P u~S ðξ, ωÞ ¼ α2 ðξ, ωÞ
α1 ðξ, ωÞ~ (2.3.13)
k∗P u~R ðξ, ωÞ + α3 ðξ, ωÞ~
uS ðξ, ωÞ + α4 ðξ, ωÞ~
uB ðξ, ωÞ ¼ 0 (2.3.14)
∗
α4 ðξ, ωÞ~ uB ðξ, ωÞ ¼ F~B ðξ, ωÞ
uS ðξ, ωÞ + α5 ðξ, ωÞ~ (2.3.15)
∗
u~B ðξ, ωÞ ¼ α6 ðξ, ωÞF~B ðξ, ωÞ (2.3.16)
where
α1 ðξ, ωÞ ¼ δξ4 m∗R ω2 + k∗P , α2 ðξ, ωÞ ¼ 2πδðω + ξc0 ΩÞ,
α3 ðξ, ωÞ ¼ m∗B ω2 =2 + k∗P + k∗B m∗S ω2 , α4 ðξ, ωÞ ¼ m∗B ω2 =6 k∗B
α5 ðξ, ωÞ ¼ m∗B ω2 =3 + k∗B , α6 ðξ, ωÞ ¼ Ψðξ, 0, ωÞ
ð
1 ∞
Ψðξ, 0, ωÞ ¼ ΠðηÞϕðξ, η, 0, ωÞeiηy dy
2π ∞
By using Eqs. (2.3.13)–(2.3.16), F~B ðξ, ωÞ is obtained:
∗ α2 ðξ, ωÞα4 ðξ, ωÞk∗P
F~ B ðξ, ωÞ ¼
α1 ðξ, ωÞα6 ðξ, ωÞα4 2 ðξ, ωÞ α3 ðξ, ωÞα1 ðξ, ωÞ k*P 2 ð1 + α5 ðξ, ωÞα6 ðξ, ωÞÞ
(2.3.17)
Thus, the rail displacement in the Fourier transform domain is finally given by
∗
α2 ðξ, ωÞα4 ðξ, ωÞ2 α2 ðξ, ωÞα3 ðξ, ωÞα5 ðξ, ωÞ α4 ðξ, ωÞF~B ðξ, ωÞk∗p
u~*R ðξ, ωÞ ¼
α1 ðξ, ωÞα4 ðξ, ωÞ2 α1 ðξ, ωÞα3 ðξ, ωÞα5 ðξ, ωÞ + α5 ðξ, ωÞk*p 2
(2.3.18)
The displacement of the rail and ground in the time domain can be expressed as follows
by introducing an auxiliary spatial coordinate x∗t ¼ x* c0 τ, and the time-domain results
are obtained by the FFT algorithm.
ð
1 ∞ *
u*R ðx*, τÞ ¼ u~*R ðξ,Ω* ξc0 Þeiξxt dξ eiΩ*τ (2.3.19)
2π ∞
ð ð
1 ∞ ∞ *
u*z ðx*, y*, z*, τÞ ¼ 2 u~* ðξ, η, Ω* ξc0 Þeiηy* eiξxt dηdξ eiΩ*τ (2.3.20)
4π ∞ ∞ z
Eqs. (2.3.19), (2.3.20) can also be expressed as:
∗ iΩ*τ
u*R ðx*, τÞ ¼ uΩ
R xt e (2.3.21)
∗ iΩ*τ
u*z ðx*, y*, z*, τÞ ¼ uΩ
z xt , y*, z* e (2.3.22)
Eqs. (2.3.21), (2.3.22) denote that, in the auxiliary spatial coordinate, the displacements
of the track-ground system are harmonic and have the same vibration frequency as the
dynamic load.
2.3.4 Coupling of Vehicle-Track-Ground System

The receptance herein denotes the displacement amplitude of the wheelsets due to a unit
vertical harmonic load with an excitation frequency Ω. The vehicle model used in Sheng
and Jones [12] was introduced. As shown in Fig. 2.3.1, the vehicles were represented as
multiple rigid body systems and the vertical dynamics of the vehicles were coupled to the
track-ground model by introducing linear Hertzian contact springs between each wheel-
set and the rails. The differential equation of motion for a single vehicle is given by
€ V ðt Þ + KV ZV ðt Þ ¼ BPðtÞ
MV Z (2.3.23)
where MV and KV denote the mass and stiffness matrices of the vehicle respectively,
ZV(t) denotes the displacement vector, P(t) denotes the wheel-rail force vector, and
B is a matrix of unit and zero elements (see Appendix A).
The roughness-induced dynamic loads between the wheel and rail were harmonic loads
with angular frequency Ω, where Ω ¼ 2πc=λ1 , λ1 is the wavelength of the rail profile, and c
is the vehicle speed. As shown by Sheng and Jones [12], P(t) and ZV(t) can be expressed as:
PðtÞ ¼ P0 ðΩÞeiΩt and ZV ðtÞ ¼ Z0V ðΩÞeiΩt . Then Eq. (2.3.23) can be written as
1
Z0V ðΩÞ ¼ KV Ω2 MV BP0 ðΩÞ (2.3.24)
The receptance between the jth and kth wheelsets within a vehicle is denoted by ΔW jk (W
means wheelset), where j, k ¼ 1, 2, …, N; N is the number of wheelsets of the vehicle.
ΔWjk denotes the displacement amplitude of the jth wheelset due to a unit vertical har-
monic load with an excitation frequency Ω exerted at the kth wheelset. The displacement
vector of the wheelset in the vehicle is expressed as
0 T
Z0W ðΩÞ ¼ ZW1 0
ðΩÞ, ZW2 0
ðΩÞ, ZW3 0
ðΩÞ, …, ZWN ðΩÞ (2.3.25)
and
T
P0 ðΩÞ ¼ P10 ðΩÞ, P20 ðΩÞ, P30 ðΩÞ, …, PN0 ðΩÞ (2.3.26)
is the wheel-rail force vector for a vehicle.
The displacement vector of the wheelsets is part of that for the corresponding vehicle.
Therefore, it can be written as
Z0W ðΩÞ ¼ AZ0V ðΩÞ (2.3.27)

where A ¼ BT (see Appendix A). Thus
1
Z0W ðΩÞ ¼ ΔW P0 ðΩÞ ¼ A KV Ω2 MV BP0 ðΩÞ (2.3.28)
2 W 3
Δ11 ⋯ ΔW 1N
6 7 1
ΔW ¼ 6 7
4 ⋮ ⋯ ⋮ 5 ¼ A KV Ω MV B
2
(2.3.29)
ΔW
N 1 ⋯ ΔNN
W
Eq. (2.3.29) gives the receptance matrix at the wheelsets for a single vehicle. Suppose
there are N1 identical vehicles being considered; the total number of the wheel-rail loads
is M ¼ N1 N . The receptance matrix at the wheelsets for the train, denoted by ΔT (where
T stands for train), is given by
2 W 3
Δ ⋯ 0
6 7
ΔT ¼ diag ΔW , …, ΔW ¼ 6 4 ⋮ ⋮ ⋮ 7
5 (2.3.30)
W
0 ⋯ Δ
The elements of matrix ΔT are denoted by ΔT lk, where k,l ¼ 1, 2, …, M.
Thus, the receptance at the jth wheel-rail contact point due to a unit load at the kth
wheel-rail contact point on the rail is determined by

Ω ∗
ΔRjk ¼ u R ljk (2.3.31)
where
ljk∗ ¼ a∗j a∗k (2.3.32)
is the dimensionless distance between the two contact points.
The complex amplitudes of the displacements at the wheel-rail contact points on the
rails are given by
∗
z0 R ðΩ*Þ ¼ ΔR P0 *ðΩ*Þ (2.3.33)
where
2 3
ΔR
11 Δ12 ⋯ Δ1M
R R
6 R 7
6 Δ21 ΔR
22 ⋯ Δ2M
R
7
R6
Δ ¼6 7 (2.3.34)
7
4 ⋮ ⋮ ⋮ ⋮ 5
ΔR
M1 ΔM2 ⋯ ΔMM
R R
The receptance matrix is nonsymmetrical due to the load motion.
z0 *R ðΩ*Þ ¼ ðz0 *R1 ðΩ*Þ, z0 *R2 ðΩ*Þ, z0 *R3 ðΩ*Þ, …, z0 *RM ðΩ*ÞÞ
T
(2.3.35)
Eq. (2.3.35) represents the displacement vector of the rail at the wheel-rail contact points
observed in the auxiliary spatial coordinate.
The rail irregularities are presented by a sinusoidal profile of amplitude A. The profile
of the rail irregularities is given by
zðxÞ ¼ Aeið2π=λ1 Þx (2.3.36)
where λ1 denotes the wavelength. The process was assumed to be linear, so that a dis-
placement input was generated at the excitation frequency f ¼ c=λ1 . The angular fre-
quency is obtained by Ω ¼ 2πc=λ1 . At time t, the lth wheelset arrives at x ¼ al + ct;
thus the displacement input of the rail profile at the lth wheel-rail contact point is
zl ðt Þ ¼ z0l ðΩÞeiΩt ¼ Aeið2π=λ1 Þðal + ctÞ ¼ Aeið2π=λ1 Þal eiΩt (2.3.37)
The coupling of a wheelset with rail is illustrated in Fig. 2.3.2, where zWl0 (Ω)eiΩt denotes
the displacement of the lth wheelset.
The wheel and rail deform locally according to the Hertz theory under the action of
the contact force. Thus the wheel and rail are coupled by a Hertz spring. Provided that
the dynamic contact force is a small fraction of the axle load, the Hertz spring can be taken
to be linear. The stiffness of the Hertzian contact spring is denoted by kHl. It is also
assumed that the wheelset is always in contact with the rail. Thus
z0Wl ðΩÞ ¼ z0Rl ðΩÞ + z0l ðΩÞ + P 0 ðΩÞ=kHl (2.3.38)
From Eqs. (2.3.21) to (2.3.28), the following relations can be derived:
Z⬘Wl(Ω)
P⬘l(Ω)
KHI
P⬘l(Ω) Z⬘l(Ω)
Fig. 2.3.2 Coupling wheel-rail model.
XM
z0Wl ðΩÞ ¼ ΔTlk Pk0 ðΩÞ (2.3.39)
k¼1
X
M
z0Rl ðΩÞ ¼ ΔR 0
lk Pk ðΩÞ (2.3.40)
k¼1
By nondimensionalizing these previous equations and substituting Eqs. (2.3.39), (2.3.40)

into Eq. (2.3.38) the following equation can be obtained
X
M 0 0 0
ΔTlk + ΔR
lk Pk *ðΩ*Þ + P *l ðΩ*Þ=k*Hl ¼ zl *ðΩ*Þ (2.3.41)
k¼1
The unknown item Pl0 (Ω*) can be obtained by solving Eq. (2.3.41). The displacement of
the ground and the rails at the exciting frequency Ω* are given by superposition.
X
M
u∗R ðx*, τÞ ¼ uΩ 0
R ðxt * a*l ÞPl ðΩ*Þe
iΩ*τ
(2.3.42)
l¼1
X
M
u∗z ðx*, τÞ ¼ uΩ 0
z ðxt * al *, y*,z*ÞPl ðΩ*Þe
iΩ*τ
(2.3.43)
l¼1
2.3.5 Numerical Results

In order to demonstrate the characteristics of the response of the system, the calculations
were carried out for a single-axle vehicle model comprising a suspended mass MC and an
unsprung mass MW. A detailed description of the model is given in Appendix A. Each
mass had one degree of freedom in the vertical direction. The frequency of interest for the
perception of ground vibration was about 5–80 Hz, and for a train speed in the range of
10–120 m/s, the corresponding wavelengths of the corrugated rail lay within the range of
Table 2.3.1 Dimensionless parameters for the single-axle vehicle model

Parameter Value
Suspended mass, MC* 12.9
Unsprung mass, MW* 1.173
kS* 0.089
CS1* 0.165
k0 S* 0.1
0.125–24 m. The parameters for the vehicle model are listed in Table 2.3.1 and the
parameters for the track are given in Table 2.3.2, and correspond to those used in Sheng
and Jones [12]. The parameters of the saturated poroelastic half-space are presented in
Table 2.3.3, which are the same as those used in Cai et al. [4]. The parameters were
selected according to Tables 2.3.1–2.3.3 if not denoted in the figures.
The roughness-induced dynamic wheel-rail forces are shown in Fig. 2.3.3 against the
excitation frequency f* (f * ¼ c0 =λ∗1 ). It is shown that the magnitude of the wheel-rail
forces increased as f* increased and reached a maximum value at the critical frequency
around f * ¼ 0:55, then decreased as f* increased further. With a close inspection of
Fig. 2.3.3, one can also see that the magnitude of the dynamic wheel-rail forces have
a local peak at frequency 0.02. This frequency is close to the natural frequency of the
suspended mass on the suspension, as noted in Sheng and Jones [12].
In order to study the attenuation of the ground vibration, the magnitudes of accel-
erations for both axle and dynamic loads are presented against y* in Fig. 2.3.4. In
Fig. 2.3.4A, when c0 ¼ 0:2, the quasistatically induced acceleration decreases rapidly as
y* increases, and is nearly zero at y* ¼ 5. The dynamically induced acceleration is much
larger at y* ¼ 0 and decreases more slowly. The dynamically induced component of
vibration is the dominant one for the acceleration response of the free-field.
Table 2.3.2 Dimensionless parameters for railway track

Parameter Value
Mass of the beam per unit length of track, mR* 0.08
Bending stiffness of the rail beam, β 0.42
Loss factor of the rail 0.01
Rail pad stiffness, kP* 11.67
Rail pad loss factor 0.15
Mass of the sleeper per unit length of the track, mS* 0.328
Mass of the ballast per unit length of the track, mB* 0.84
Ballast stiffness per unit length of the track, kB* 10.5
Loss factor of the ballast 1.0
Contact width of ballast and ground, 2LBal* 2.7
Table 2.3.3 Dimensionless parameters for fully water-saturated poroelastic soil medium
Parameter Value
Lame constant, λ* 2
Water density, ρf* 0.53
Parameter of soil structure, m* 1.5625
Hysteretic damping ratio, D 0.02
Ratio between the fluid viscosity and the intrinsic permeability, b* 10
The parameter for the compressibility of the soil particle, α 0.97
The parameter for the compressibility of the fluid, M* 12
In Fig. 2.3.4A, the acceleration response for the elastic half-space soil medium is also pre-
sented. The acceleration responses predicted by the elastic soil model are larger than those
predicted by the poroelastic soil model for both axle and dynamic loads at y* ¼ 0, and the
acceleration responses attenuate more slowly in the elastic soil medium. The free-field
acceleration response predicted by the elastic soil model is larger than that predicted
by the poroelastic one. In Fig. 2.3.4B, when c0 ¼ 0:6, the quasistatically induced accel-
eration is larger than the dynamically induced acceleration at y* ¼ 0, but the dynamically
induced vibration component dissipates more slowly and gives rise to the majority of the
vibration for the free-field off the track. For the case of an elastic soil medium, it can be
seen that the acceleration response dissipates more slowly in the elastic soil. Using the
elastic soil medium model will overestimate the acceleration response under the free-field
condition for c0 ¼ 0:6. In Fig. 2.3.4C, for c0 ¼ 1:0, the quasistatically induced accelera-
tions are much larger and are the dominant component for the ground vibration at
0.0016
0.0014
Magnitude of the wheel-rail force
0.0012
0.0010
0.0008
0.0006
0.0004
0.0002
0.0000
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Excitation frequency
Fig. 2.3.3 Dynamic wheel-rail force at different excitation frequencies.
0.0025 0.007
Magnitude of ground surface accelerations

0.006
0.0020
Axle load for poroelastic half-space 0.005 Axle load for poroelastic half-space
Axle load for elastic half-space
0.0015 Axle load for elastic half-space
Dynamic load for poroelastic half-space 0.004 Dynamic load for poroelastic half-space
Dynamic load for elastic half-space Dynamic load for elastic half-space
0.0010 0.003
0.002
0.0005
0.001
0.0000 0.000
0 10 20 30 40 0 10 20 30 40
y* y*
(A) c0 = 0.2 (B) c0 = 0.6

0.025
0.020
Axle load for poroelastic half-space
Axle load for elastic half-space
0.015 Dynamic load for poroelastic half-space
Dynamic load for elastic half-space
0.010
0.005
0.000
0 10 20 30 40
y*
(C) c0 = 1.0
Fig. 2.3.4 Ground surface accelerations against the distance. (A) c0 ¼ 0.2; (B) c0 ¼ 0.6; (C) c0 ¼ 1.0.
various y*. The acceleration response predicted by the elastic soil model is apparently
smaller than that predicted by the poroelastic soil model.
Therefore, for vehicle speeds below the Rayleigh wave speed, the free-field
acceleration response is dominated by the dynamic load. But when the vehicle speed
approaches the Rayleigh wave speed of the half-space, the axle load is the dominant
one for the acceleration response of the free-field. Using the elastic soil medium model
will overestimate the acceleration response level of the free-field for the train speed below
Rayleigh wave speed, while it will underestimate the response level for the train speed
above the Rayleigh wave speed.
In Fig. 2.3.5, the train-induced excess pore water pressures are studied. In Fig. 2.3.4A,
when c0 ¼ 0:2, the quasistatically induced pore water pressure reaches a positive peak at
the load point and a negative peak behind the load. The dynamically induced pore water
pressure fluctuates along xt* and the maximum value is larger than that for the quasista-
tically induced one. When the vehicle speed exceeds the critical speed, as shown in
Fig. 2.3.5B, the quasistatically induced pore water pressure becomes much larger and
is the dominant component for the pore water pressure responses.
0.0010
0.006
0.0008
0.005
0.0006 Axle load
0.004
Dynamic load Axle load
0.0004
P*(x*t,0,1)
P*(x*t,0,1)
0.003 Dynamic load
0.0002 0.002
0.0000 0.001
–0.0002 0.000
–0.0004 –0.001
–0.0006 –0.002
–40 –20 0 20 40 –40 –20 0 20 40
x* x*
(A) t (B) t
Fig. 2.3.5 Excess pore water pressure responses.
2.4. ENGINEERING APPLICATION

The Beijing-Tianjin intercity railway line is the earliest constructed high-speed railway
line in China; it connects the two municipalities of Beijing and Tianjing. Construction of
this intercity railway line was started on Jul. 4, 2005 and was completed on Dec. 15, 2007.
The designed operating speed was 350 km/h. In order to ensure ride quality as well as
reduce maintenance costs, the slab track B€ ogl imported from Germany was adopted on
the railway line. The subgrade of the railway lines was mainly composed of soft clay, and
then treated by CFG piles with a diameter of 0.4 m and a length of 28 m. After debug-
ging, the railway started test runs on Jul. 1, 2008. During this period, Zhejiang University
was invited to measure the dynamic responses of the track and embankment. Two
locomotive types, Harmony CRH2 and CRH3, were used during the test runs. The
measured vibration of the track and embankment is presented herein.
The testing site was chosen at Yangcun. The instrumentation of the velocity sensor and
the accelerometer on the cross-section of the railway line is shown in Fig. 2.4.1. V and H in
Fig. 2.4.1 denote the sensors in the vertical and horizontal directions, respectively.
302 V1 Rail
30B
V3
31A V4
17.36 m
V5
H6
V8 V7
Fig. 2.4.1 Cross-section arrangement of the sensors for the vibration test.
1.0
Velocity (cm/s)
0.5
0.0
–0.5
–1.0
1 2 3 4 5 6 7 8
Time (s)
Fig. 2.4.2 Measured velocity at the sleeper (V1) under a train speed of 145.1 km/h maximum value
1.046, minimum value 0.906.
0.06
0.05
0.04
Velocity (cm/s)
0.03
0.02
0.01
0.00
–0.01
–0.02
–0.03
1 2 3 4 5 6 7 8
Time (s)
Fig. 2.4.3 Measured velocity at the embankment (V3) under a train speed of 145.1 km/h maximum
value 0.058, minimum value 0.025.
0.04
0.02
Velocity (cm/s)
0.00
–0.02
–0.04
–0.06
1 2 3 4 5 6 7 8
Time (s)
Fig. 2.4.4 Measured velocity at the free-field (V7) under a train speed of 145.1 km/h maximum value
2.0
1.5
1.0
Velocity (cm/s)
0.5
0.0
–0.5
–1.0
–1.5
2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Time (s)
Fig. 2.4.5 Measured velocity at sleeper (V1) under a train speed of 335.1 km/h maximum value 1.866,
minimum value 1.419.
0.3
0.2
0.1
Velocity (cm/s)
0.0
–0.1
–0.2
–0.3
–0.4
2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
Time (s)
Fig. 2.4.6 Measured velocity at the embankment (V3) under a train speed of 335.1 km/h maximum
value 0.245, minimum value 0.348.
0.02
0.01
Velocity (cm/s)
0.00
–0.01
–0.02
–0.03
–0.04
–0.05
2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
Time (s)
Fig. 2.4.7 Measured velocity at the free-field (V8) under a train speed of 335.1 km/h maximum value
Measured velocity
Calculated velocity
1.0
0.5
Velocity (cm/s)
0.0
–0.5
–1.0
1 2 3 4 5 6 7 8
Time (s)
Fig. 2.4.8 Comparison between the calculated and measured velocity at the sleeper (V1) under a train
speed of 145.1 km/h.
Measured velocity
Calculated velocity
1.0
0.5
Velocity (cm/s)
0.0
–0.5
–1.5
1 2 3 4 5 6 7 8
Time (s)
Fig. 2.4.9 Comparison between the calculated and measured velocity at the sleeper (V1) under a train
speed of 335.1 km/h.
The measured results follow. When the train runs at the speed of 145.1 km/h, the
measured velocity at the sleepers, embankment, and free field are presented in
Figs. 2.4.2–2.4.4. The peak velocity at the sleeper was about 1.0 cm/s, and then
decreased as the distance from the railway centerline increased. When the train speed
increased to 335.1 km/h, the measured velocity at the sleepers, embankment, and free
field are presented in Figs. 2.4.5–2.4.7, and it can be seen that the velocity at the sleeper
was amplified to 1.5 cm/s, then decreased as the distance increased.
In order to validate the correctness of the theoretical model proposed in the previous
section, the comparison between the calculated and measured velocity at the sleeper
(V1) under a train speed of 145.1 km/h and 335.1 km/h is shown in Figs. 2.4.8 and
2.4.9, respectively. It is shown that the calculated and measured results agreed well.
Based on the in situ test results, the following results were obtained:
(1) The velocity of the track was intensified significantly as the train speed increased; the
velocity amplitude of the track at 350 km/h was about two times what it was at
120 km/h.
(2) The vibration decayed rapidly along the transverse direction; the vibration velocity
at the free-field was significantly smaller than that at the sleeper of the track.
CHAPTER 3
Problems for Vibrations of Foundations

This chapter introduces the procedures for solving the problem of a vibrating foundation
in saturated ground using Biot’s theory.
3.1. VERTICAL VIBRATIONS OF A RIGID FOUNDATION EMBEDDED

IN SATURATED SOIL
A simplified analytical method was used to investigate the vertical dynamic response of a
rigid, massive, cylindrical foundation embedded in a poroelastic soil layer. The founda-
tion was subjected to time-harmonic vertical loading and was perfectly bonded to the
surrounding soil in the vertical direction. The behavior of the poroelastic soil is governed
by Biot’s poroelastodynamic theory and its governing equations are solved by means
of the Hankel transform. The soil at the side of the foundation was simplified to be com-
posed of a series of infinitesimally thin independent poroelastic layers, while the soil
underlying the foundation base was modeled as a single-layered type of poroelastic
soil based on rigid bedrock. The contact surface between the foundation base and the
poroelastic soil was smooth and fully permeable. A computer code was developed to
solve this interaction problem. Selected numerical results are presented to show the influ-
ence of nondimensional frequency of excitation, soil layer thickness, poroelastic material
parameters, depth ratio and mass ratio of the foundation on the dynamic response of a
rigid, massive, cylindrical foundation embedded in a poroelastic soil layer.
3.1.1 Governing Equations of Saturated Soil for Vertical Vibrations

The problem considered in this study is shown in Fig. 3.1.1, where a rigid cylindrical
foundation with a radius r0 and a mass m was embedded to a depth h in a poroelastic soil
layer of thickness H. The foundation was subjected to a time-harmonic vertical loading
P(t) ¼ Peiωt acting along its axis of symmetry, where ω is the frequency of motion and
pffiffiffiffiffiffiffi
i ¼ 1: A cylindrical coordinate system with its origin o(r, θ, z) located at the center
of the foundation was defined such that the z-axis was perpendicular to the free surface of
the soil layer, and the motion was axisymmetric with respect to the z-axis. The governing
equations of a poroelastic medium for axisymmetric motion, in the absence of compres-
sion of the solid matrix and pore fluid, are given according to Eqs. (1.1.10), (1.1.20) as:
P (t)
Soil surface
m r0
h
o
r
H
q
z
Rigid bedrock
Fig. 3.1.1 Rigid cylindrical foundation embedded in a poroelastic soil layer.
@σ rr σ rr σ θθ @σ rz @pf
+ + ¼ ρu€r + ρf w€r (3.1.1)
@r r @z @r
@σ rz σ rz @σ zz @pf
+ + ¼ ρu€z + ρf w€z (3.1.2)
@r r @z @z
@pf ρ
¼ ρf u€r + f w€r + bw_ r (3.1.3)
@r n
@pf ρ
¼ ρf u€z + f w€z + bw_ z (3.1.4)
@z n
The constitutive relations for a homogeneous poroelastic material can be expressed
according to Eq. (1.1.21) as
@ur
σ rr ¼ λe + 2μ (3.1.5)
@r
@uz
σ zz ¼ λe + 2μ (3.1.6)
@z
ur
σ θθ ¼ λe + 2μ (3.1.7)
r

@ur @uz
σ rz ¼ μ + (3.1.8)
@z @r
@ u_r u_r @ u_z @ w_ r w_ r @ w_ z
+ + + + + ¼0 (3.1.9)
@r r @z @r r @z
The motion under consideration was assumed to be time-harmonic with the factor eiωt,
which was henceforth suppressed from all expressions for brevity. It was convenient to
use nondimensional forms of length parameters including the coordinates by selecting the
Problems for Vibrations of Foundations 39
radius of the foundation as a unit of length. Stresses and pore pressure can also be nor-
malized into the nondimensional forms, with respect to the shear modulus μ of the bulk
material. All variables were replaced by the nondimensional quantities, but the previous
notations were used for convenience. In addition, the following nondimensional fre-
quency and material properties were introduced:
rffiffiffi
ρ λ ρ br0
a0 ¼ r0 ω, λ* ¼ , ρ* ¼ f , b* ¼ pffiffiffiffiffi , M* ¼ M=μ (3.1.10)
μ μ ρ ρμ
3.1.2 Boundary Conditions for Vertical Vibrations
For the vibration problem of a rigid foundation perfectly bonded to the surrounding soil
in the vertical direction, as shown in Fig. 3.1.1, it was assumed that the contact surface
between the foundation base and the poroelastic soil was smooth and fully permeable.
The boundary conditions on z ¼ 0 and z ¼ H/r0 can then be written in nondimen-
sional quantities as follows:
uz ðr, 0Þ ¼ U0 , 0 r 1 (3.1.11)
σ zz ðr, 0Þ ¼ 0, 1 < r ∞ (3.1.12)
σ rz ðr, 0Þ ¼ 0, 0 r ∞ (3.1.13)
pf ðr, 0Þ ¼ 0, 0 r ∞ (3.1.14)
uz ðr,H=r0 Þ ¼ 0, 0 r 1 (3.1.15)
ur ðr,H=r0 Þ ¼ 0, 0 r 1 (3.1.16)
wz ðr,H=r0 Þ ¼ 0, 0 r 1 (3.1.17)
where U0 ¼ U/r0 with U denoting the vertical displacement of the rigid foundation.
3.1.3 A Simplified Analytical Method for Vibrations of Foundations

Equation of Dynamic Equilibrium for the Rigid Foundation
In addition to those previously defined, the following notations were also introduced in
this study: Rb(t) ¼ Rbeiωt is the vertical dynamic reaction at the foundation base; Rs(t) ¼
Rseiωt is the vertical dynamic reaction along the side of the foundation. The equation of
the dynamic equilibrium for the foundation is
mU€ðtÞ ¼ P ðtÞ Rb ðt Þ Rs ðtÞ (3.1.18)
where U(t) ¼ Ue .iωt
Because of the difficulty in obtaining an exact analytical solution for a dynamic

response of embedded foundations, a simplified analytical method proposed by Baranov
[13] was extended and used by Novak et al. [14–16] to study the vertical, coupled hor-
izontal and rocking, torsional and coupled vibrations of a rigid foundation embedded in
an ideal elastic soil. In the Baranov-Novak method, the soil was divided into two inde-
pendent parts and it was assumed that the soil reaction at the base of the foundation was
equal to that of a foundation resting on the surface of the soil, while the lateral soil reac-
tion was evaluated independently. In this paper, several similar assumptions were intro-
duced: (I) the soil at the side of the foundation is composed of a series of infinitesimally
thin independent poroelastic layers. In each infinitesimally thin poroelastic layer, the
influence of the gradient of σ zz and pf in the vertical direction and the influence of radial
displacement on Rs(t) are neglected. (II) The dynamic reaction at the base of the foun-
dation is independent of depth of embedment.
Dynamic Reaction at the Base of the Foundation

Under assumption (II), the dynamic reaction at the base of the foundation can be derived
from a dynamic interaction between a surface foundation and a single-layered poroelastic
soil. The governing partial differential equations of poroelastic soil for axisymmetric
motion can be solved by applying a Hankel integral transform with respect to the radial
coordinate r. The μth-order Hankel integral transform of a function f(r, z) with respect to
r and its inverse transform are defined by Sneddon [17] as
ð∞
μ
f~ ðξ, zÞ ¼ rf ðr, zÞJμ ðξr Þdr (3.1.19)
0
ð∞
f μ ðr, zÞ ¼ ξ f~ðξ, zÞJμ ðξr Þdξ (3.1.20)
0
where Jμ is the Bessel function of the first kind of the μth order, and ξ is the Hankel trans-
form parameter.
The following general solutions can be obtained for a zero-order Hankel transform of
uz, wz, σ zz, and pf, and the first-order Hankel transform of σ rz and ur, respectively:
p~f0 ¼ χ 1 A1 eqz + χ 1 B1 eqz + A2 eξz + B2 eξz (3.1.21)
u~1r ¼ χ 2 A1 eqz + χ 2 B1 eqz + χ 3 A2 eξz + χ 3 B2 eξz + A3 esz + B3 esz (3.1.22)
u~z0 ¼ χ 4 A1 eqz χ 4 B1 eqz + χ 3 A2 eξz χ 3 B2 eξz + χ 5 A3 esz χ 5 B3 esz (3.1.23)

1 qz ξz ξz

w~z0 ¼ κ1 u~z0 qχ A
1 1 e + qχ B
1 1 e qz
ξA 2 e + ξB 2 e (3.1.24)
ρ*a0 2
σ~zz
0
¼ γ 1 A1 eqz + γ 1 B1 eqz 2ξχ 3 A2 eξz 2ξχ 3 B2 eξz 2ξA3 esz 2ξB3 esz (3.1.25)
σ~1rz ¼ γ 2 A1 eqz γ 2 B1 eqz 2ξχ 3 A2 eξz + 2ξχ 3 B2 eξz + γ 3 A3 esz γ 3 B3 esz (3.1.26)
where A1, A2, A3, B1, B2, and B3 are arbitrary functions to be determined from the
boundary and continuity conditions of the given problem. The variables κ 1, q, s, χ i
(i ¼ 1, 2, 3, 4, 5), and γ j (j ¼ 1, 2, 3) appearing in the preceding equations are known
Problems for Vibrations of Foundations 41
functions of the material properties and the frequency of excitation, and are given in
Appendix B.
By applying the Hankel transform to Eqs. (3.1.12)–(3.1.17), together with Eqs. (3.1.21)–
(3.1.26), the boundary conditions can be rewritten in a matrix form:
Qa¼f (3.1.27)

Q11 Q12
where a ¼ fA1 , A2 , A3 , B1 , B2 , B3 g; f ¼ σ~zz ðξ, 0Þ,0,0,0,0,0 ; and Q ¼
T T 0
Q21 Q22
denotes a 6 6 matrix whose elements Q11, Q12, Q21, and Q22 are functions of ξ and
are given in Appendix C.
The solution of Eq. (3.1.27) is
φ∗11 0 φ∗ φ∗
A1 ¼ σ~zz ðξ, 0Þ, A2 ¼ 12 σ~0zz ðξ, 0Þ, A3 ¼ 13 σ~0zz ðξ, 0Þ (3.1.28)
jφj jφj jφj
ψ ∗11 0 ψ∗ ψ∗
B1 ¼ σ~zz ðξ, 0Þ, B2 ¼ 12 σ~0zz ðξ, 0Þ, B3 ¼ 13 σ~0zz ðξ, 0Þ (3.1.29)
jψ j jψ j jψ j
1 1
where φ ¼ Q11 Q12 Q22 Q21 ; ψ ¼ Q12 Q11 Q21 Q22 ; jφj and jψ j denote the determi-
nants of φ and ψ, respectively; φ11 , φ12 , φ13 and ψ ∗11 , ψ ∗12 , ψ ∗13 denote the alge-
∗ ∗ ∗
braic complements of φ11, φ12, φ13 and ψ 11, ψ 12, ψ 13, corresponding to φ and ψ,
respectively.
Substitution of Eqs. (3.1.28), (3.1.29) into Eq. (3.1.23) results in the following
relationship:
u~z0 ðξ, 0Þ ¼ pðξÞ~
σ 0zz ðξ, 0Þ (3.1.30)
χ 4 φ∗11 + χ 3 φ∗12 + χ 5 φ∗13 χ 4 ψ ∗11 + χ 3 ψ ∗12 + χ 5 ψ ∗13

where pðξÞ ¼ :
jφj jψ j
Then substituting Eq. (3.1.30) into Eq. (3.1.11) and expressing σ~0zz ðr, 0Þ in terms of its
Hankel transform, a set of dual integral equations describing the vertical vibration of a
rigid foundation embedded in a poroelastic soil layer are finally derived as follows:
ð∞
U0
ξ1 ½1 + H ðξÞBðξÞJ0 ðξr Þdξ ¼ , 0r 1 (3.1.31)
0 1ν
ð∞
BðξÞJ0 ðξr Þdξ ¼ 0, r > 1 (3.1.32)
0
ξpðξÞ
σ 0zz ðξ, 0Þ, H ðξÞ ¼
where BðξÞ ¼ ξ~ 1 and ν is Poisson’s ratio of the solid matrix.
1ν
It can be proved that lim ξpðξÞ ¼ ð1 νÞ, i:e:, lim H ðξÞ ¼ 0.
ξ!∞ ξ!∞
The dual integral equations can be reduced to a Fredholm integral equation of the
second kind by employing the following integral representation:
Another random document with
no related content on Scribd:
SECONDARY OR REMOTE RESULTS OF
LIVER DISEASE.
In gout: Arrest of oxidation of proteids into urea. Deposits of biurate of lime on

joints, and other disorders. Urinary calculi containing urates, cystine, xanthine,
etc., also from imperfect oxidation of albuminoids. Oxalic acid represents a similar
arrest. Kidney degenerations from irritating urates and oxalates. Fatty kidney from
excessive glycogenesis. Digestive disorders from excess or deficiency of bile or
torpid liver. Nervous disorders, dullness, lameness, vertigo, spasms, irritability
from hepatic inactivity and resulting poisons. Sore throat and bronchitis from
hepatic derangement. Skin eruptions in tardy or imperfect action of the liver.
Treatment: Abundant water, succulent vegetables, ensilage, fresh grains, balanced
ration, in carnivora and omnivora oat meal, buttermilk, clear meat juice, avoid
sweets, gravies, spiced animal food. Dangers for pampered horses, dogs, and old
improved meat producing animals. Open air exercise. Laxatives with alkalies,
salines, mercurous and mercuric chloride, pilocarpin, chlorides, iodides, bromides,
nitro-muriatic acid, ipecacuan, euonymus, bitters.
Among the many secondary results of hepatic disorder, and which

are habitually described as affections of other organs a few may be
mentioned as indicating the wide range of influence exercised by the
liver in disease as well as in health.
Gout as it appears in fowls and omnivora is directly due to the
arrest of the transformation of the albuminoids into urea. Circulating
in the system in the form of the less perfectly oxidized and less
soluble uric acid, it determines deposits of biurate of lime around the
joints, with local inflammations, and disorders of circulation and
innervation, and altered spirit, temper, etc.
Urinary calculi in the same animals, are composed largely of urate
of lime, cystine, xanthine and other nitrogenous products
representing various stages of oxidation short of the final transition
into urea and ammonia. Recognizing the active rôle which the
urinary bacteria fill in this respect we must still acknowledge the
great importance, as causative agents, of an excess in the urine of
these comparatively insoluble products.
The oxalic acid found in certain calculi points in the same
direction, as this acid, both in the body and in the laboratory, is
found to result from the oxidation of uric acid (Wohler, Schenck,
Hutchinson).
Degenerations of the Kidneys are to be largely traced to the same
hepatic source. The uric acid diathesis, and the oxalic acid diathesis,
both the result of imperfect liver function, are among the most
frequent causes of irritation of the kidneys, by which channel they
are eliminated from the body. Hence acute and chronic nephritis, as
well as nephritic calculi result from morbid conditions which have
their starting point in the imperfect function of the liver. Again fatty
degeneration of the kidney is very liable to result from derangement
of the glycogenic function of the liver, the tendency to the formation
of fat and the constant irritation caused by the passage of the sugar
contributing to the tissue degradation. In such cases albuminuria is a
not uncommon accompaniment.
Derangements of the Digestive Organs may be said to be a
necessary result of hepatic disorder. Excessive secretion of bile
stimulates peristalsis and may induce diarrhœa, while diminished
secretion tends to constipation, light colored, fœtid stools, intestinal
fermentations and poisoning by the irritant products. A torpid
hepatic circulation means congestion of the whole portal system,
indigestions, colics, chronic muco-enteritis, intestinal hemorrhages,
hemorrhoids, etc.
Derangements of the Nervous System. In this connection may be
named the lameness of the right shoulder which accompanies certain
disorders of the liver, the extreme dullness and depression that
attends on others, the sluggish pulse that appears in certain types,
the unsteadiness of gait (giddiness) in others, the muscular cramps,
and irritability in still others. These appear to be due in some
instances to the nervous sympathy of one part with another, whilst at
other times they as manifestly depend on the circulation in the blood
of partially oxidized and other morbid products of hepatic disorder
which prove direct poisons to the nervous system.
Derangements of the circulation, like extreme rapidity, or slowness
of the pulse, irregularities in rhythm and intermissions, may be
charged more directly on the nervous affection, though primarily
determined by hepatic disorder.
On the part of the Respiratory Organs, affections of a chronic
type, like sore throat and bronchitis may often be traced to hepatic
torpor or disorder.
Skin Diseases are notoriously liable to come from inactive or
disordered liver, the irritant products circulating in the skin or
sweating out through it, giving rise to more or less irritation. The
result may be a simple pruritus, an urticaria, an eruption of papules,
vesicles or even pustules. In any such cases it is proper to look for
other indications of liver disease,—pale color and offensive odor of
the fæces, muco-enteritis, indigestion, icterus or yellow patches on
the mucous membranes, tenderness on percussion over the asternal
ribs, muscular neuralgia, nervous disorder, the passage of bile,
hæmoglobin, albumen, sugar or other abnormal elements in the
urine, etc.
TREATMENT OF SECONDARY AND
FUNCTIONAL DISEASES OF THE LIVER.
Diet. Many hepatic disorders, and especially those that are
exclusively or mainly functional may be corrected by diet alone.
Prominent among dietary influences is the abundant supply of water.
The succulent grasses of spring and early summer constitute the
ideal diet, hastening and increasing elimination, and lessening the
density of the bile, even to the extent of dissolving biliary calculi and
concretions. Upon dry winter feeding such calculi are common
especially in ruminants, whereas after a month or two at pasture they
are extremely rare. In winter the same good may be arrived at by the
use of ensilage, brewer’s grains, roots, fruits, or even scalded hay or
bran. The two extremes of highly albuminous and highly
carbonaceous or saccharine food are to be avoided or used only in
limited amounts. In the one class are clover, alfalfa, sainfoin,
vetches, cowpea, lespedeza, especially in the form of hay, beans,
peas, cotton seed, gluten-meal, rape and linseed cake. In the other
are wheat, buckwheat, Indian corn, sorghum, sweet-corn and
cornstalks. Some agents like beets which are rich in saccharine
matter may be actually beneficial by reason of their laxative and
cholagogue action. In the carnivora the food should be largely of
simple mush of oatmeal, wheat seconds, or barley meal, skimmilk or
buttermilk. If it is needful to tempt the appetite in a fleshfed animal
this should not be done by rich, fat gravies, highly spiced animal
food, or rich saccharine puddings, but rather by the addition of a
little pure juice of lean meat, or some well skimmed beef tea.
It is as important to regulate the quantity as the quality of the food
as the heavy feeder will over-charge the liver as much by an excess of
otherwise wholesome food, as will the ordinary animal by the
indigestible and unwholesome articles. As a rule the improved
breeds of meat producing animals, have acquired such facility in fat
production that much of the surplus is largely and profitably
disposed of in this way, and in their short lives little obvious evil
comes of the overfeeding, but in cases in which this outlet proves
insufficient, as in horses and dogs that are highly fed on stimulating
or saccharine diet, and which are kept for the natural term of their
lives, with little exercise, the evil tends to reach a point of danger.
Nursing mothers and dairy cows find a measure of safety in the free
flow of milk and the yield of butter, but breeding cows that have been
improved till they have no longer a capacity for milking, but must
have their calves raised on the milk of other and milking strains are
correspondingly liable to suffer.
Exercise in the Open Air. As enforced idleness, on a full diet and in
a warm and moist environment is a main cause of hepatic disorder,
so abundant exercise in the open air and especially in a cool season is
beneficial in a marked degree. Beside the bracing effect on the
digestive organs and the improvement of the general tone of the
system, the action of the muscles in hastening the circulation greatly
favors the removal and elimination of waste matters. Still more
advantageous is the increased activity of the respiration and the
aspiratory power of the chest in at once unloading the portal system
and the liver by hastening the progress of the hepatic blood into the
vena cava and right heart, and in furnishing an abundant supply of
oxygen for the disintegration of the albuminoids and amylaceous
products. Such exercise must of course be adapted to the condition of
the animal and its power of sustaining muscular work, but
judiciously employed, it is one of the most effective agencies in
correcting and improving hepatic disorder or hepatic torpor. Idle
horses, the victims of obstinate habits of constipation, muco-enteric
irritation, indigestion, nervous, urinary or cutaneous disorders will
often be greatly benefited or entirely restored by systematic exercise.
This is one of the great advantages of a run at pasture, as the subject
secures at once the laxative cholagogue diet, an abundant supply of
oxygen, a better tone of the muscular and general system, and a more
perfect disintegration of albuminoids. Sea air with its abundance of
ozone is especially advantageous.
In the carnivora while we cannot send them to grass, much can be
done in the way of systematic exercise, and in the case of city dogs a
change to the country, where they can live out of doors and will be
tempted to constant exercise and play, will go far to correct a faulty
liver.
Laxatives. Cholagogues. When a free action of bowels and liver
cannot be secured by succulent food and exercise, we can fall back on
medicinal laxatives. These are advantageous in various ways. Some
laxatives like podophyllin, aloes, colocynth, rhubarb, senna, jalap,
and taraxacum act directly on the liver in increasing the secretion of
bile. These may be used for a length of time in small doses and in
combination with the alkalies. Other aperients act directly on the
bowel carrying away the excess of bile, the albuminoids and
saccharine matter that would otherwise be absorbed, and by a
secretion from the portal veins, abstracting nitrogenous and
saccharine elements which would otherwise overtax the liver to
transform them. Thus indirectly these also act as cholagogues by
withholding the excess of material on which it has to operate, and by
rousing its functions sympathetically with those of the bowels. Thus
sulphates of magnesia and soda, and tartrates and citrates of the
same bases, given in the morning fasting, dissolved in a large
quantity of warm water and conjoined with sodium chloride,
ammonium chloride, sodium carbonate or other alkaline salts, or
with one or more of the vegetable cholagogues above mentioned,
may be continued for a length of time until the normal functions
have been re-established, and will maintain themselves irrespective
of this stimulus.
Calomel (and even mercuric chloride in small doses), though it is
not experimentally proved to be a direct cholagogue, is one of the
very best correctives of impaired hepatic function. It expels the bile
from the duodenum and bowels generally, thereby preventing its
reabsorption; it proves antiseptic to the ingesta; it eliminates much
of the peptone, saccharine and fatty matter from the intestines and
portal system thus relieving the liver materially; and it is supposed
further to modify the other liver functions by a direct action on the
hepatic cells, and by reducing the cohesion of fibrine, and promoting
the disintegration of albumen. Certain it is that calomel gives most
substantial relief in many torpid and other disorders of the liver and
as it is not in itself an active liver stimulant but has rather a soothing
action on that gland it can be safely resorted to in states of hepatic
irritation in which the more direct cholagogues would prove more or
less hurtful.
In some forms of hepatic disorder where a speedy and abundant
secretion is demanded, pilocarpin may be employed, with great
caution so as not to reduce the strength unduly by the attendant
diaphoresis, diuresis, salivation or diarrhœa.
Alkalies have long been recognized as of great clinical value in
hepatic disorders. Though carbonate of soda decreases the secretion
of bile, (Nasse, Röhrig), yet the alkalies generally appear to promote
oxidation, and to hasten the disintegration of albumen and the
albuminoids. They increase the disintegration of sulphur compounds
materially adding to the sulphates and urea in the urine. They
further tend to increase the hippuric acid, carbonate of soda (2 drs.)
even determining the abundant excretion of this acid in man (Nasse).
It may be concluded that the acknowledged value of alkalies in these
diseases, is largely due to their hastening of the metabolic processes
in albuminoids. Small doses of sodium carbonate further stimulate
the gastric secretion and may thus benefit by rendering the process
of digestion more complete and satisfactory.
Chlorine, Iodine, Bromine and their Salts. These halogens are of
great value in many hepatic disorders. The universal craving for
sodium chloride indicates the need of its elements in the animal
body, and whether this is mainly the supply of chlorine for the
hydrochloric acid of the gastric juice, or to fulfill its uses in favoring
the oxidation and disintegration of the nitrogenous matters in the
blood and tissues, or for other more or less obscure uses, it is well to
recognize and act upon the indication. The various mineral waters
which are held in high esteem in liver affections contain a large
proportion of sodium chloride. As a medicinal agent ammonium
chloride maintains an equally high position. Large doses thrice a day,
so as to induce diaphoresis and diuresis greatly relieve hepatic
congestions. This agent determines a great increase in the urea
eliminated so that it is even more effective in the same direction,
than sodium chloride. Free chlorine is also effective in hepatic torpor
and congestion, and to this in part may be attributed the great value
of nitro-muriatic acid.
Bromide and iodide of potassium have been found to be effective
in reducing hepatic enlargement and thus in conducing to a more
healthy activity of the liver.
Ipecacuanha, Euonymus, etc. These agents are more or less
hepatic stimulants and may be found beneficial as combined with the
laxative or alkaline agents in securing a better functional activity in
cases of torpor or deranged function.
Tonics, Bitters. Tonics are often useful when the health has been
undermined by long continued hepatic disorder. The iron tonics are
as a rule contraindicated as tending to check secretion of bile, unless
they can be given with alkalies. Iron sulphate or chloride, combined
with sodium or potassium carbonate so as to establish a mutual
decomposition will obviate this objection. The vegetable bitters
(gentian, cascarilla, calumba, salicin, serpentaria, aloes, nux vomica)
combined with alkalies are often of great value. Quinia, like opium,
checks secretion and is to be avoided or used with judgment and in
combination with cholagogues.
HÆMOGLOBINÆMIA. AZOTÆMIA.
AZOTURIA. HÆMOGLOBINURIA. TOXÆMIA
FROM IMPERFECT HEPATIC FUNCTION.
Definition. Theories, of hysteria, uræmia, spinal myelitis, myelo-renal

congestion, rheumatic lumbago, myosito-myelo-nephritis, rheumatic chill with
destruction of muscle albuminoids. Yet it occurs in our semi-tropical midsummer
with a temperature of 80 or 90, in spring and autumn, and rarely even in the cold,
damp stable in midwinter in the absence of exercise. Constant conditions: One or
more days absolute rest, preceding steady work, a strongly nitrogenous ration,
continued during the rest, sudden active exertion accelerated breathing and
unloading of peptones and proteids from portal vein and liver into the general
circulation. Sanguineous albuminuria from excess of albuminous food, free
ingestion of water, suppressed milk secretion, forced marches. Transfusion of
blood. Excess of albumen dangerous, excess of red globules not dangerous. The
blood concentration of diuresis or diaphoresis is not dangerous. Continuous
muscle decomposition from work bars the disease. Stable miasm untenable. Poison
may be drawn suddenly from the enormous mass of blood in the liver, spleen and
portal system. The absence of icterus antagonizes the bile theory. Benzoic acid,
unaltered peptones, and glycogen are examples of elements destructive to blood.
Normal destruction of red globules in liver, spleen and bone marrow. Sudden
access of resulting hæmoglobin to the blood. Other products of disintegrated
globules. Poisons from food, and antitoxic action of liver in presence of glycogen.
Carbon dioxide favors solution of red globules. Theories of hæmoglobinæmia in
man. Lesions: Blood black, diffluent, iridescent, has no avidity for oxygen, with
excess of urea and extractives, serum of clot red, globules, small, pale, distorted,
not sticky, extravasations, liver, enlarged, congested, blood gorged, spleen
congested, swollen: Lumbar or gluteal muscles pale, infiltrated, with loss of
striation; bone marrow congested, hemorrhagic; kidneys congested infarcted;
urine dark brown or red, with excess of urea and hæmoglobin. End of spinal cord
has congestion or infiltration. Symptoms: History of high condition, constant
work, high feeding, a day’s rest, then exercise and attack. To full life, follows
flagging, droops, moves one or both hind limbs stiffly, knuckles, drags toes,
crouches, trembles, perspires, breathes rapidly, is tender on back, loins, croup or
thigh, muscles firm, paretic, and drops unable to rise. Urine retained, brown, red
or black, sometimes glairy, later may have casts. Appetite may return. In mild
cases, stiffness, lameness, with or without visible muscular lesions or tremors.
Urine glairy, dense, with excess of urea and nitrogenous products. Recover under
careful feeding and exercise, and relapse under original causes. Progress: May
recover under rest. In bad cases accelerated breathing and recumbency forbid rest
and recovery. Recovery in a few hours or after a week. Urinary casts with renal
epithelium, imply nephritis and grave conditions. In persistent paresis, muscles
waste. Modes of death. Mortality 20 per cent. Diagnosis, by history of onset, etc.
Prevention: When highly fed and hard worked, give daily exercise, with
comparative rest, reduce ration, and give laxative or diuretic. Plenty of water.
Treatment: Rest, sling, diffusible stimulants, bleeding, bromides, water ad libitum,
fomentations, unload liver and portal vein, purgative, eserine, barium chloride,
enemata, diuretics, for remaining paresis, derivatives, strychnia, diet, laxative,
non-stimulating, restore to work gradually.
Definition. An acute auto-poisoning occurring in plethoric horse

on being subjected to active exertion after a period of idleness, and
manifested by great nervous excitement and prostration, paresis
commencing with the hind limbs and the passage of hæmoglobin in
the urine.
Nature and Causes. The most varied conclusions as to the nature
of this disease have been put forward by different authors. In
England, Haycock called it hysteria, mistakenly supposing that it was
confined to mares, and Williams attributed it to uræmic poisoning,
conveniently ignoring the fact that the sudden manifestation of the
most extreme symptoms in an animal which just before was in the
highest apparent health and spirits contradicted the conclusion. In
France (Trasbot) and Southern Europe (Csokor) it has been looked
on as a spinal myelitis, a conclusion based on the disturbed
innervation of the posterior extremities in the great majority of
cases, but which is not always sustained by the pathological anatomy
of the cord. In Germany veterinarians have viewed the disease from
widely different standpoints. Haubner calls it myelo-renal-
congestion (Nièren-Rückenmarks): Weinmann, a rheumatic
lumbago; Dieckerhoff defines it as an acute general disease of horses,
manifested by a severe parenchymatous inflammation of the
skeleton muscles, with a bloody infiltration of the bone marrow,
especially of the femur, and with acute nephritis and
hæmoglobinuria. He attributes the attack to exposure to cold. If this
were the real cause the attack would be far more common in very
cold weather when the horse is suddenly exposed to cold drafts
between open doors and windows, than when he is harnessed and
driven so as to generate and diffuse animal heat. Yet attacks in the
stable are virtually unknown, and in almost every instance the onset
occurs during a short drive. Friedberger and Fröhner say that the
epithet rheumatismal may be correctly applied to almost all cases
that we meet in practice. They quote Goring as having produced the
disease experimentally by exposure to cold, and go on to explain that
rest in the stable before the attack causes the extreme sensitiveness
to cold that is generated by a warm environment. The implication of
the lumbar, pelvic and femoral muscles they explain by the
stimulation of the nutritive metamorphosis by the action of cold on
the sensitive nerves of the skin. The effect of this cutaneous irritation
is exaggerated by the heat of the stable to which they have been
previously subjected. The products of the destruction of the
albuminoids of the muscles, pass into the blood as hæmoglobin, and
produce the ulterior phenomena. The muscles of the hind quarters
especially suffer because of their greater exposure and because they
are subjected to the hardest work in propelling the animal machine.
In this connection they quote the experiments of Lassar and
Nassaroff in which sudden exposure to cold determines
parenchymatous degeneration of muscles; also the cases of
paroxysmal or winter hæmoglobinuria in certain susceptible men
whenever they are exposed to an extremely low temperature.
There are serious objections to the acceptance of this as the
essential cause, among which the following may be named:
1st. The disease is not confined to the cold season but occurs also
at midsummer when the outdoor temperature is even higher than it
is in the stable.
2d. In our Northern States it appears to be more common in spring
and autumn or early winter, when the extreme colds have either
already passed, or have not yet set in, but when the abrupt changes
of weather (rain-storms, etc.) are liable to shut up the animal indoors
for a day or more at a time.
3d. The popular names quoted with approval by these authors—
Monday disease, Easter disease, Whitsuntide disease—indicate the
prevalence in Europe also, of the malady in the milder, or more
temperate seasons rather than during the prevalence of extreme
cold.
4th. The fact that the disease rarely or never occurs in the stable,
no matter how cold the season, how open the wooden walls or floor,
nor how strong the draft between doors or windows, shows that the
theory of cold as the sole or main cause must be discarded.
It is not necessary to ignore the action of cold as a concurrent
factor in certain cases, or as a stimulant to reflex vaso-motor paresis,
to muscular metamorphosis and the increase of hæmoglobin in the
blood. It is only necessary that this should be held as subordinate
and non-essential to the final result. Several other factors that are
accorded a subordinate place by these writers, are so constant and so
manifestly essential that they must be allotted a much more
important position in the list of causes.
A period of rest is a constant precursor of an attack. The more
extended the inquiry the more certain we become that a short rest is
a prerequisite to equine hæmoglobinæmia. The horse that is kept at
daily steady work may be said to be practically exempt. Even the
non-professional observer recognizes the fact and names the disease
after the weekly or yearly holiday or rest day which was the occasion
of it. To him it is the Monday morning disease, the disease of the day
following Thanksgiving, Christmas, New Year, or Fourth of July. It is
the disease of wet weather, of heavy snowfalls, of the blizzard, or of
the owner’s absence from home, of any time that entails one or two
days of absolute inactivity in the stall.
But again the affection does not appear in the horse that is
absolutely idle for a length of time. It is the short period of rest in
an interval of otherwise continuous work that determines it.
In short the subject must be in good muscular condition and with a
hearty, vigorous appetite and good digestion. The short unwonted
rest interrupts the disposal of the rich products of a vigorous
digestion, and tends to overload the portal veins, the liver, the blood
and tissues with an excess of proteids. The condition of the animal is
so far one of plethora.
Another feature that bears this out is that the attack comes only in
the animal that is heavily fed on a strongly nitrogenous
ration. It is not the disease of the horse kept on straw, or hay, or
which receives a limited amount only of grain. It does not occur in
the animal which has its grain suspended or materially reduced
during the one or two days of idleness. It does not select the horse
that has had a laxative either in the form of food or medicine. This
last may increase the sensitiveness to cold, but it certainly lessens the
tendency to hæmoglobinæmia. The most rational explanation
appears to be that it affords this protection by interfering with the
thoroughness of digestion and absorption, by securing elimination
from the portal veins and liver, and by reducing the amount of
albuminoids in the blood.
A blood abnormally rich in albuminoids, as it is in the transient
plethora induced by a short period of rest, in the well-conditioned
working horse, without any restriction of his diet, may therefore be
set down as one of the most important factors in producing
hæmoglobinæmia. Nor is this without approximate examples in
human pathology. Von Bamberger has shown that “hæmatogenous
albuminuria” will occur in healthy individuals when there is an
excess of albumen in the blood-plasma, as after a too free use of
albuminous food, or after suppression of the milk secretion
(Landois). A similar result comes from increase of blood pressure, as
after drinking freely, or when, under emotion or violent exertion, the
heart’s action is increased in force and the blood is thrown with
greater impetus into the large renal arteries. Senator has found
albuminous urine to attend and follow, for several days, upon forced
marches made by young recruits. Here the muscular work is added to
the increased blood tension superinduced by the more active
contractions of the heart.
In this connection it is interesting to trace the changes in the blood
after transfusion. The dilatability of the capillaries enables the
system to accommodate itself to a very great increase in the volume
of blood An increase of 83 per cent. may be borne without serious
results, but above this limit there is increasing risk and an increase of
150 per cent. entails immediate danger to life. In the restoration of
the blood to its normal condition, the secretion of water sets in
promptly leaving an excess of albuminoids and blood globules. The
next change is in the albuminoids which in two days are almost
entirely transformed into urea. This leaves the blood abnormally rich
in globules (Panum, Lesser, Worm-Müller), the red globules break
up much more slowly and may still be in excess after the lapse of a
month (Tscherjew).
In this light, temporary plethora cannot of itself be accepted as the
main or essential cause of the disease. It must be admitted to be a
more constant and important factor than the mere exposure to cold,
but of itself it is inadequate to the production of hæmoglobinæmia.
In the absence of exertion the general plethora fails to produce the
specific disease; again, after transfusion a plethora of albumen lasts
for one or two days, but hæmoglobinæmia sets in only in the first few
minutes after the animal starts out from the stable, (never after an
hour or two at work): once more, excess of globules may last for a
month, but with steady work there is no danger of this disease, after
the first mile or two has been traversed, on the first day of the
resumption of labor.
A similar plethora of albuminoids and globules may be induced in
a plethoric animal by a profuse diarrhœa, diuresis or perspiration,
the blood having been robbed of its watery constituents, and
concentrated especially as regards its globules and albuminoids, but
hæmoglobinæmia never occurs as the result of such an artificial
concentration. On the contrary a free secretion by the bowels or
kidneys is of the greatest value in cutting short its progress after it
has set in.
The doctrine of poisoning by hæmoglobin produced by excessive
work and disintegration of the muscles is equally insufficient to
account for an attack. Excess of muscular work and of muscle-
decomposition-products, would not reach its maximum within the
first few minutes after the animal has started from the stable, but,
other things being equal, would increase with the continuance of
work and the accumulation in the blood of a constantly increasing
amount of these products. The sharp line of restriction by which the
attack is limited to the initial period of work, while it is never seen
after hard work continued for hours in succession, rules out this
from the list of essential causes. It may be that the products of
muscular decomposition aggravate the attack, but to set them down
as the cause of the attack is to beg the whole question and to
contradict the truth that continuous and severe muscular work with
its consequent increase of waste products is a direct bar to the
development of the disease. It should be noted in this connection
that the increase in the waste of nitrogenous bodies, as shown by the
increase of urea, is dependent far more on the amount of nitrogenous
matters ingested than on the muscle work or decomposition. In
eleven hours just before ascending the Faulhorn, Fick passed 21.686
grs. of urea per hour; in eight hours ascending the hill, 12.43 grs. per
hour; and in six hours after the ascent he passed 13.39 grs. per hour.
A general survey of the field shows that it is not the simple
increase of any normal waste product in the blood which determines
hæmoglobinæmia, and on the other hand the suddenness and
severity of the attack bears all the marks of a profound poisoning.
The nature of the poison has not yet been definitely ascertained, yet
one or two hypothesis may be hazarded, as furnishing a working
theory, in anticipation of the actual demonstration which may be
expected in the early future.
The action of a stable miasm as claimed by some writers is
contradicted by the fact that the disease does not develop so long as
the animal is left to inhale that miasm, and on leaving the stable, the
life and vigor are usually remarkable.
The morbific agent must be sought in some source from which it
can be supplied with great rapidity under the stimulus of a short but
active exertion. The chylopoietic viscera furnish such a source. The
healthy liver contains one-fourth of the entire mass of the blood. The
torpid congested liver of the vigorous high conditioned horse, after a
short period of idleness, on full, rich feeding, must hold much more
than this normal ratio. The spleen, the natural store-house or safety
valve of the portal veins, is also gorged with this liquid in the high
fed, idle animal. This organ which is always turgescent after meals, is
especially so in the over-fed horse, which for twenty-four hours has
been denied the opportunity of working off by exercise, the
superfluous products of an active digestion and absorption. Then the
whole of the portal veins and the capillaries in which they originate
are surcharged with rich blood which cannot make its way with the
necessary dispatch through the inactive liver.
In this condition there is incomparably more than a quarter of the
entire mass of blood, enriched to the highest degree in proteids,
ready to be discharged through the liver and hepatic veins into the
general circulation. Under the action of the hurried breathing and
circulation, caused by the sudden and active exertion, this whole
mass of rich blood is speedily unloaded on the right heart, the lungs
and the systemic circulation. One can hardly conceive of a more
effective method of inducing a sudden plethora, with an excess of
both globules and albuminoids.
The presence of actual poisons in such blood is not so easily
certified.
The absorption of bile elements and especially of taurocholic acid,
which is a solvent of the red blood globules, and would set free their
globulin might account for the characteristic condition of the blood.
The powerful aspiratory action of the chest, would speedily empty
the whole of the liver blood vessels, and lessening their tension below
that of the biliary radicals would determine an active absorption of
bile or of the more diffusible of the bile elements. A manifest
objection to this view is the absence of an icteric tint in the mucous
membranes of the affected animals. The visible mucosæ are of a
brownish red hue, such as might come from hæmoglobin dissolved
in the blood serum, rather than the yellow tint which might be
expected from bile pigment. The theory of poisoning by bile acids
therefore, would require an explanation of concurrent suppression or
decomposition of the bile pigments.
Other sources, however, offer solvents for hæmoglobin, benzoic
acid, which is derived from a cellulose in the fodders, and forms the
source of hippuric acid, dissolves red globules (Landois). In the over-
fed horse with active digestion, but inactive body and liver, this must
accumulate in the liver, spleen and portal system, and when
suddenly drawn into the blood without time for oxidation in the liver
it will contribute to the condition of hæmoglobinæmia.
Peptones, being very diffusible, are very rapidly absorbed, but they
are not found, in healthy conditions, in the portal vein (Neumeister).
These are manifestly transformed into albumen in the intestinal
mucosa (Salvioli), or taken up by the very numerous leucocytes and
transformed or carried elsewhere (Hoffmeister). But peptones
injected into the blood of the dog render it incoagulable, and in large
quantity are fatal (Landois). An excess of glycogen dissolves the red
globules, and the conditions of heavy feeding and torpid liver, are
calculated to produce this in great excess and to store it in the liver
cells.
Under the extra vigorous aspiratory force of the chest, these highly
diffusible agents, present in great excess, are likely to be drawn on
through the mucosa, into the portal vein, liver, and cava, without an
opportunity for complete transformation by leucocytes or liver cells.
These would tend to rob the blood globules of their normal
physiological vigor, would unfit them for maintaining the healthy
functions of lungs, kidneys, brain or muscle, and would unfit the
globules for successful resistance to solvents and other inimical
influences.
Again it is an important function of the liver, spleen and red bone
marrow to disintegrate worn out or abnormal red globules. These are
taken up by the white blood corpuscles of the hepatic capillaries, by
the cells of the spleen and the bone marrow and are stored up chiefly
in the capillaries of the liver, in the spleen, and in the marrow of
bone. They are transformed, partly into colored and partly into
colorless proteids, and are either deposited in the granular form, or
are dissolved (Landois). Quincke says: “That the normal red blood
globules and other particles suspended in the blood stream are not
taken up in this way, may be due to their being smooth and polished.
As the corpuscles grow older and become more rigid, they, as it were,
are caught by the amœboid cells. As cells containing blood corpuscles
are very rarely found in the general circulation, one may assume that
the occurrence of these cells within the spleen, liver, and marrow of
bone, is favored by the slowness of the circulation in these organs.”
From this chain of normal processes of blood disintegration, we may
reasonably infer, a greatly exaggerated work of blood destruction
when, in connection with an increased density of the plasma, and the
presence in the portal blood of poisonous products of digestion, the
red globules have been altered in density, in outline and in vitality, so
that they become ready victims of the amœboid cells of blood and
tissues. Then the stagnant condition of this altered blood in the
compulsorily idle animal favors the greatest excess of this
destruction and the storing up of an increased quantity of
hæmoglobin and other products, to be poured suddenly into the
general circulation as soon as the movement of the blood is
quickened by exercise.
This destruction of the red blood globules by disintegration
contributes to the formation of numerous decomposition-products,
like succinic, formic, acetic, butyric and lactic acids, inosite, leucin,
xanthine, hypoxanthin, and uric acid, some of which are strongly
toxic. The tendency will be to lower the vitality of the red globules
and thus to render them the easier victims of the leucocytes and of
the liver, spleen and marrow cells. Even the freed hæmoglobin
appears to exert a solvent action on the red blood globules. These
are, of course, most concentrated and effective in the seat of their
production, yet when drawn suddenly in large amount, into the
general circulation, by the vigorous aspiratory action of the chest,
they may prove seriously detrimental to the blood at large.
Again a variety of toxic matters are introduced into the system in
the food and others are developed from the food in the stomach and
intestine. Brieger found in the gastric peptones a potent alkaloid
having the effect of urari, and which in excess would determine
muscular paralysis. The alkaloidal and other poisons produced by
fermentations in the intestines have to be safely disposed of. The
ptomaines, if not too abundant, are arrested or even decomposed in
the liver which thus stands as a guardian, at the outlet of the portal
system, to protect the body at large. But this antitoxic function of the
liver is only exercised in the presence of glycogen (Rogers, Landois),
and forced muscular movement soon removes all glycogen from the
liver of the dog (Landois). Again glycogenesis in the liver is now
believed to be dependent on a ferment produced by the pancreas. If
therefore, the sudden active exercise and the aspiratory action of the
chest freed the liver of its glycogen, and hurried the alkaloidal and
other poisons through its capillaries too rapidly to allow of the
protective action of the liver cells, or if the pancreas as well as the
liver had become torpid and had failed to produce the requisite
amount of glycogen-ferment for the liver, the poisoning of the blood
and system at large would be imminent.
Not to mention the other toxic products which come from
imperfect metamorphosis in the liver, it may be noted that a venous
condition of the blood or an excess of carbon dioxide contributes
greatly to the solubility of the red blood globules. It also tends greatly
to modify the fibrinogenous elements. Thus the blood of a suffocated
animal fails to coagulate or coagulates loosely, and the blood of the
portal vein of a suffocated horse is strongly toxic (Sauson). Now the
conditions attendant on the onset of equine hæmoglobinæmia are
such as to give free scope to both of these inimical influences. The
great mass of blood in the portal vein, spleen and liver is venous
blood strongly charged with carbon dioxide, and by the sudden,
active exertion this is forced rapidly through the liver and lungs
without time for full æration, so that the whole mass of the
circulating blood is speedily reduced below par, and laid specially
open to the action of blood solvents. By the same action the systemic
blood is charged with poisons, direct from the food, and fermenting
ingesta, and from the overworked spleen and liver whose functions
are profoundly impaired, and later from other important organs, the
healthy functional activity of which can no longer be maintained by
the deteriorated blood supplied to them.
Hæmoglobinæmia in dogs has been produced experimentally by
the injection of water into the veins the mere dilution of the plasma
dissolving out the coloring matter from the red globules (Hayem);
also by the inhalation of arseniureted hydrogen (Naunyn and
Stadelman); by the ingestion of toluylendiamine, or phosphorus
(Afanassiew, Stadelman); by snake venom, septicæmia, influenza,
contagious pneumonia, petechial fever, anthrax, etc. These cannot be
looked on as causes of the acute hæmoglobinæmia in the horse, but
they serve as illustrations of changes in the plasma, and poisons in
the blood determining the escape of hæmoglobin from the cells.
Ralfe recognizes two forms of hæmoglobinæmia in man:
1st. That in which the hæmoglobin is simply dissolved out of the
blood globules, the solution taking place chiefly in parts exposed to
cold.
2d. A more severe form in which the dissolution is general and
probably attended by some destruction of red globules in the liver,
spleen and even in the kidneys. The general opinion appears to be
that the attacks are due to some nervous disturbance, which causes
vaso-motor disorder and it is supposed that there is an exaggerated
sensibility of the reflex nervous system. It has been suggested that
peripheral irritation causes irritation of the vaso-motor centre, and
in turn this causes local asphyxia in the part stimulated, under which
conditions the red globules part with their hæmoglobin (Roberts).
Murri holds that the disease depends on an increased irritability of
the vaso-motor reflex centre, and the formation, owing to the
disorder of the blood forming organs, of corpuscles unable to
withstand exposure to cold or carbon dioxide.
While it is not assumed to point out the actual poisons of
hæmoglobinæmia in the horse the above suggestions may offer
valuable hints as to the lines of inquiry that may be followed with the
best hope of reaching definite results.
Lesions. These are especially found in the blood, liver, spleen,
muscles, bone marrow and kidneys. The spinal cord and nerve
trunks are occasionally affected.
The blood is charged with carbon dioxide and is black, tarry,
comparatively incoagulable remaining in the veins and showing an
iridescent reflection. It does not absorb oxygen readily though
exposed to the air, and thus bears a strong general resemblance to
the blood of anthrax. It contains an abnormal proportion of urea and
allied extractive matters which greatly increase its density, and
interfere with the healthy exercise of the different cell organisms and
functions. These are not due to excessive muscular activity as stated
by Friedberger and Fröhner, but are derived mainly from the
abundant products of digestion. When the shed blood coagulates it
forms a soft clot without buff and the expressed serum is reddish
from the presence of hæmoglobin, and of hæmatoidin crystals. The
uncoagulated blood drawn over a sheet of white paper stains it
deeply by reason of the same coloring matters in solution. The red
corpuscles may be paler than natural, some even entirely colorless,
and they are often notched or broken up in various irregular forms.
They have lost the natural tendency of the shed equine blood to stick
together, to collect in rouleaux and precipitate to the bottom of the
vessel, so that no buffy coat is formed, should the blood coagulate.
The white corpuscles are relatively increased. Finally the coloring
matters contained in the plasma are imbibed by the different tissues
and give a brown or reddish tinge to such as are naturally white.
Limited blood extravasations are not uncommon especially in the
more vascular organs like the muscles, liver, spleen and kidneys.
The liver is more or less congested and enlarged, friable, yellow, or
mottled yellow and red and exudes black blood freely when incised.
The bile is thick, viscid and dark green, as in cases of experimental
intravenous injection of hæmoglobin.
The spleen is also swollen and congested with blood, and the pulp
is very high colored from the excess of hæmoglobin and other
products of blood destruction. The muscles of the croup are usually

Ebook Solutions For Biots Poroelastic Theory in Key Engineering Fields Theory and Applications PDF Full Chapter PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ebook Solutions For Biots Poroelastic Theory in Key Engineering Fields Theory and Applications PDF Full Chapter PDF

Uploaded by

Copyright:

Available Formats

Solutions for Biot's Poroelastic Theory

in Key Engineering Fields. Theory and

Library of Congress Cataloging-in-Publication Data

For information on all Elsevier publications visit our website

Publisher: Glyn Jones

Basic Equations and Governing

1.1. BASIC EQUATIONS

1.1.1 Geometric Equations

@ur @uθ @uz

in which u ¼ [ux, uy, uz]T and E(r) is an operator matrix defined by

1.1.2 Equations of Motion

σ ij, j ¼ ρu€i , ði ¼ x,y,z or r, θ, zÞ (1.1.11)

1.1.3 Constitutive Equations

σ ij ¼ cijkl εkl , ði, j, k, l ¼ x, y, z or r, θ, zÞ (1.1.12)

cijkl ¼ cjikl and cijkl ¼ cijlk

σ ij ¼ 2μεij + λδij εkk (1.1.13)

Expanding Eq. (1.1.13), we have

And C should be a nonsingular and reversible matrix, which can be written as

1.2. GOVERNING EQUATIONS OF A FULLY SATURATED

1.2.1 Governing Equations in Cartesian Coordinates

σ 0ij ¼ σ ij + αδij p (1.1.17)

σ 0ij ¼ σ ij + δij p (1.1.18)

σ ij, j ¼ ρu€i + ρf w€i (1.1.19)

p, i ¼ bw_ i ρf u€i mw€i (1.1.20)

σ ij ¼ 2μεij + λδij εkk αδij p (1.1.21)

1.2.2 Governing Equations in Cylindrical Coordinates

1.3. BOUNDARY CONDITIONS

Solutions for Saturated Soil Under

2.1. GENERAL SOLUTIONS OF SATURATED GROUND UNDER

2.1.1 Solutions for Saturated Ground Generated by a Moving

Fourier transform pairs with respect to x and y are defined as

Eq. (1.1.22) in the transformed domain can be expressed as

r4 p~f + β1 r2 p~f + β2 p~f ¼ 0 (2.1.23)

By solving Eq. (2.1.26), the expression p~f can be obtained:

p~f ¼ Aeγ1 z + Beγ2 z (2.1.27)

θ~ ¼ Aχ 1 eγ 1 z + Bχ 2 eγ2 z (2.1.32)

u~z ¼ Aa1 γ 1 eγ1 z + Ba2 γ 2 eγ2 z + Ceγ 3 z (2.1.37)

The expressions in the transformed domain can be expressed as:

σ z ¼ g3 Aeγ1 z + g4 Beγ 2 z 2μγ 3 Ceγ 3 z

A uniformly distributed rectangular pressure is applied to the surface of the

The vertical stress in the transformed domain can be expressed as

2.2. APPLICATION IN HIGHWAY ENGINEERING

2.2.1 Governing Equations for the Pavement Structure

Infinite poroelastic half space

2.2.3 Solution Methods for the Pavement and Saturated Ground

The traffic load in the transformed domain can be expressed as:

g5 ¼ g1 + 2η2 a1 , g6 ¼ g1 + 2η2 a2 (2.2.15)

Table 2.2.2 Dimensionless parameters of the plate

Table 2.2.3 Dimensionless parameters of tandem-axle loads

Normalized max. vert. displacement

Fig. 2.2.3 Maximum vertical displacement against dimensionless velocity c0.

Dimensionless vertical acceleration

(A) C0 = 0.2 (B) C0 = 1.0

(A) C0 = 0.2 (B) C0 = 1.0

2.3. APPLICATIONS IN RAILWAY ENGINEERING

2.3.1 Governing Equations for the Track System

2.3.2 Coupling Between the Track and Saturated Ground

From Eqs. (2.3.6) to (2.3.10), the following equation can be obtained:

2.3.3 Solutions for the Track-Ground System

2.3.4 Coupling of Vehicle-Track-Ground System

Z0W ðΩÞ ¼ AZ0V ðΩÞ (2.3.27)

The receptance matrix is nonsymmetrical due to the load motion.

Fig. 2.3.2 Coupling wheel-rail model.