Materials Science Forum Submitted: 2019-01-09

ISSN: 1662-9752, Vol. 953, pp 53-58 Revised: 2019-01-11

doi:10.4028/www.scientific.net/MSF.953.53 Accepted: 2019-01-11
© 2019 Trans Tech Publications Ltd, Switzerland Online: 2019-05-06

Finite Element Modelling and Multi-Objective Optimization of Composite

Submarine Pressure Hull Subjected to Hydrostatic Pressure
Elsayed Fathallah*
Department of Civil Engineering, Military Technical College, Cairo, Egypt
Saidhabib2000@mtc.edu.eg

Keywords: Multi-objective optimization; buckling load; Tsai-Wu; failure criteria; fiber orientations

Abstract. Excellent mechanical behavior and low density of composite materials make them
candidates to replace metals for many underwater applications. This paper presents a comprehensive
study about the multi-objective optimization of composite pressure hull subjected to hydrostatic
pressure to minimize the weight of the pressure hull and maximize the buckling load capacity
according to the design requirements. Two models were constructed, one model constructed from
Carbon/Epoxy composite (USN-150), the other model is metallic pressure hull constructed from
HY100. The analysis and the optimization process were completely performed using ANSYS
Parametric Design Language (APDL). Tsai-Wu failure criterion was incorporated in the optimization
process. The results obtained emphasize that, the submarine constructed from Carbon/Epoxy
composite (USN-150) is better than the submarine constructed from HY100. Finally, an optimized
model with an optimum pattern of fiber orientations was presented. Hopefully, the results may
provide a valuable insight for the future of designing composite underwater vehicles.

Introduction
The pressure hulls are one of the most main structures of the underwater vehicles which provide high
load capacity for electronic systems and buoyancy [1, 2]. The optimization and buckling behavior
have been attracted some recent attention [2-3]. Zhu et al. [4], studied the buckling behaviors of a
spherical shells with an opening. The critical load with an opening was 4.4% -8% lower than that of
the complete spherical shell. Ross and Little [5], predicted the buckling pressure of a carbon fiber
vessel subjected to an external pressure. It was higher than the experimental by 20%, because the FE
results assume perfect circular geometry. Smith et al. [6], studied an integrated approach to design
and analyze structural modelling of an underwater vehicles. Sekulski [7], presented the multi-
objective optimization of high speed catamaran structure incorporating the strength criteria. John et
al. 8], investigated the corrosion effect on stability and strength of a pressure hull. The results
illustrated that, there is a reduction in the overall collapse and the yield pressures with values 20%
and 40%, respectively. Metin and Tolga [9], studied the buckling of the composite plates considering
the linearly varying in plane loads and different plate theories. Zhang et al. [10], investigated the egg-
shaped pressure hulls to improve the poor hydrodynamics of the spherical pressure hull, the low
buckling resistance and the difficult interior arrangement. Han et al. [11], studied the buckling and
post-buckling behavior of composite cylindrical shells under external pressure. In this study, the
multi-objective optimization of a composite and metal pressure hull shell under external hydrostatic
pressure are presented in order to obtain minimum mass (the weight / displacement) ratio and
maximum buckling pressure. The modelling and the multi-objective optimization are performed
using APDL.

Buckling in Composite Pressure Hull.

To ensure the stability, the critical buckling strength ( N cr ) of the composite pressure hull must
exceed the actual load ( N act ). The buckling load strength factor ( λ ), is introduced to identify the
buckling of the pressure hulls and defined as:

All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans
Tech Publications Ltd, www.scientific.net. (#508715224, Chalmers University of Technology, Göteborg, Sweden-06/08/19,03:19:50)
54 Materials Science and Industrial Applications

N cr
λ= (1)
N act
Where: N act and N cr denote the actual and the critical buckling load, respectively. Buckling will be
occurring when N cr less than N act ( λ < 1 ), which N cr is defined as [12]:
 C11 C12 C13 
  C C22 C23 
   21
 R  C31 C32 C33  (2)
N cr = 
  n 2 + 0.5  mπ R  
2
  C11 C12 
    
  L   
 C21 C22 
Where: L is the length, R is the radius, m and n are the number of buckle half waves in the axial and
circumferential direction, respectively. They C, A, B and D are expressed as in [13].
Composite Failure Criteria. The successful design requires an efficient and safe use of materials.
Therefore, theories are needed to develop and compare the state of the stresses and the strains in the
material.
Tsai-Wu Failure Criteria. Tsai-Wu failure criterion is the most generalized criterion, it distinguishes
between the compressive and tensile strength and used for the failure prediction in a ply. The criterion
can be expressed as 14]:
 1 1  1 1 σ 112 σ 222 τ 122
= σ 11 
FI −  + σ 22  −  − − − 2 (3)
 Xt Xc   Yt Yc  X t × X c Yt × Yc S
Where: Xt, Yt, Xc, Yc, σ 11 , σ 22 and τ 21 are aforementioned. The failure will be occurring when the
calculated stresses are reaching the ultimate stresses and the IF reaches or exceeds the value 1.
Von Mises yielding criteria. The von Mises yielding criteria is employed here to assess the capability
of both the metal and the core materials to withstand the yielding failure. The failure index IF is
defined as follows.
σ 2 − σ 11σ 22 + σ 222
(4)
11
IF =
σ0
Where: σ 0 denote the material allowable yielding strength, σ 11 and σ 22 are the in-plane principal
stresses.

Submarines Pressure Hulls and Material Property Requirements

Preferably pressure hull structure should have minimal bouncy factor through achieving the hull
density to be close to the sea water as possible. Fig. 1 shows the different forms of a submarine
pressure hull. Strong lightweight materials for the pressure hulls are the most urgent requirement in
the development of a technology for deep diving vehicles. Strength and lightweight are not the only
requirements. Also, toughness, formability, weldability, resistance to corrosion, fracture toughness,
low cycle fatigue strength, reliability and cost are importance. The material properties and the strength
parameters are given in Table 1.
Finite element modeling and simulation. The pressure hulls used in this study presented in Fig. 2,
with total length (Ltot), major diameter (Dmajor), minor diameter (Dminor), intersecting angel (θ)
and radius (R). The pressure hull modelled using (Shell 93) in case of metallic hull (HY100). For
composite pressure hull the structure modelled using (Shell 99) with stacking sequence. The ring
beams and stringers are modelled using BEAM189. The boundary conditions [(-θ /θ ) 4 ]s applied here
as in [15]. To predict the failure index, Tsai-Wu failure criteria is used. Fig. 3 indicates the mesh for
the overall model.
 Materials Science Forum Vol. 953 55

Water Pressure hull Pressure Hatch

hull Casing

Atmospheric Pressure
Atmospheric
Water
Pressure
Water
Casing

a) Cylinder/cone/dome pressure hull. b) The spherical pressure hull.

c) Ring-stiffened circular cylinder, blocked by end caps.

Fig. 1 Possible forms of submerged vehicles.
Long beam
Total Thickness Ring beam

(b2xh2)
(b1xh1) R
θo (b2xh2)

Dminor

Ltot Dmajor

Fig. 2 Multiple intersecting cross elliptical submersible pressure hulls.

General optimization concept. The most design problems are often involving constrained
minimization. One example of such constrained minimization problem is minimizing the weight of a
structure under constraints such as stress and deflection. In general optimization statement, the
objective is to minimize a single function which known as objective function and defined as “ f ”,
where “ f ” is a function of a set of independent variables “x”.
f = f (x) (5)

The structure optimization model can be expressed as follows:

minimize f (x1 , x 2 ,......., x n ), (x1 , x 2 ,......., x n )T ∈ Ω (6)

Where: x = ( x1 , x 2 ,......., x n ) is a column vector of n real-valued design variables, Ω is the feasible

T

region or feasible set. For unconstrained problems, the feasible region is the entire space. These design
variables subject to upper and lower limit and can be defined as:
x L ≤ x i ≤ x u , (i =
1,.......n) (7)

Where: “n” is the number of design variables, “xL” and “xu” are the lower and the upper limit
constraints, respectively. The minimization function is subject to another set of variables, known as
56 Materials Science and Industrial Applications

the state which contains the design and are dependent to the design variables. These state variables,
“g”, “h” and “w” could be bounded by upper and lower limits:
g L ≤ g i ( x ) ≤ g u , (i =
1,.......m)
h L ≤ h i ( x ) ≤ h u , (i =
1,.......m) (8)
w L ≤ w i ( x ) ≤ w u , (i =
1,.......m)

Where: “m” is the number of state variables, “u” and “L” are the upper and the lower limit,
respectively. In this study, the first order optimization method is used.
Optimization Statement. The multi-objective optimization problem can be stated mathematically as:
Total hull weight
F(X): Minimize B.F = (9)
The fluid displaced by the body volume.
N cr (10)
λ=
And Maximize buckling capacity:
N act

Material strength Constraints:

g1 : FI ( i ) − 1 ≤ 0, i = 1, 2, 3, ......n;

g2 :
σ
−1 ≤ 0
(11)
σy

Where: 𝐹𝐹𝐹𝐹 is the failure index of the angle-ply laminated for each ith layer and (n) is the number of
layers; σ and σ y are the actual stress and yield strength, respectively;
Side Constraints: The radii of pressure hull ((Dmin) and (Dmax)), the thickness (t) of individual ply
thicknesses, the fibre orientation angle (α) of individual ply, the operating depth (H) and the
intersecting angle (θ).
DiL ≤ Di ≤ DiU , i =
max, min;
t L ≤ ti ≤ t U , i = 1, 2, 3, ..., n;
α L ≤ α i ≤ α U     
,  i = 1, 2, 3, ..., n; (12)
H L ≤ H ≤ HU ;
θL ≤θ ≤θU 
Where: 𝐷𝐷𝑖𝑖 , 𝐷𝐷𝑖𝑖𝐿𝐿 and 𝐷𝐷𝑖𝑖𝑈𝑈 denote the ith cross pressure hull diameters and the upper and lower limits,
respectively; 𝑡𝑡𝑖𝑖 , 𝑡𝑡 𝐿𝐿 and 𝑡𝑡 𝑈𝑈 denote the ith ply thickness and the lower and upper bounds, respectively;
𝛼𝛼𝑖𝑖 , 𝛼𝛼 L and 𝛼𝛼 U denote the ith orientation angle of each ply and the upper and lower bounds,
respectively; 𝐻𝐻, 𝐻𝐻 𝑈𝑈 and 𝐻𝐻 𝐿𝐿 denote the operating depth for the cross-elliptical pressure hull and the
upper and lower limits, respectively; 𝜃𝜃, 𝜃𝜃 L and 𝜃𝜃 U denote the ith intersecting angle and the lower
and upper bounds, respectively.

Fig. 3 Finite element modeling of pressure hulls.

 Materials Science Forum Vol. 953 57

Table 1 Strengths of unidirectional composites and material properties of the sandwich

components.
Material Material and strength properties
E11= 131GPa, E22= 10.8GPa, , E33= 10.8GPa, G12= 5.65GPa,
Carbon/epoxy composite (USN-150) G23= 5.65GPa, υ12=0.28, υ23=0.0.59, Xt=2000MPa,
Xc=1400Mpa, Yt=61MPa, Yc=130MPa,S=70 MPa, ρ= 1540kg/m3
E= 210GPa, υ=0.3, ρ=7890 kg/m3, Compressive yield strength
HY100
(Fy)= 690 Mpa, Ultimate strength( Fu)= 739.5Mpa.

The Optimization Results. For the metallic multiple intersecting cross elliptical pressure hull are
summarized in Table 2. These results will be used as a reference for the comparison with the
composite hulls. From Table 2, the Multi-objective function is 0.07373 with buoyancy factor (B.F)
equals to 0.39 and buckling strength factor (λ) equals to 5.2899. The maximum von Mises stress
(σMAX) equals to 765 MPa and the maximum deflection value (δMAX) equals to 2.3297mm. The
total hull weight equals to 778.589kg.
Table 2 The results of the optimal design for metallic pressure hull. (Steel HY100).
Maximum Maximum Von mises
2.3297mm 0.76557×109 Pa
deflection(δMAX) stress(σMAX)
Buckling strength factor(λ) 5.2899 Dmajor 2.0m
Dminor 1.6836m Buoyancy factor (B.F) 0.39005
Total weight 778.5896kg MOF 7.373×10-2
For the same loadings, the optimization procedures were performed on the composite pressure hull.
Using (Carbon/Epoxy composite (USN-150)). The optimization results are presented in Table 3.The
optimization results obtained in case of (Carbon/Epoxy composite (USN-150)) indicates that the
MOF is 0.0487with (IR) = 33.9 %. The B.F equals to 0.1976 with improvement ratio (IR) = 49.33%.
The total hull weight equals to 394.574 kg with improvement ratio (IR) = 49.322 % .While, the
buckling load capacity (λ) in this case is (4.05) which is smaller than the metallic one with percent =
23.39058 %. From the above results, the pressure hull constructed from (Carbon/Epoxy composite
(USN-150)) has better MOF. Moreover, (Carbon/Epoxy composite (USN-150)) has minimum weight
and minimum B.F.

Table 3 The results of the optimal design for pressure hull without core.
(Carbon/epoxy composite (USN)).
Buckling strength
FTWSR FTWSR 4.05636769
factor (λ)
FI_1 0.741309805 FI_12 0.821134469 Layer thickness (t) 1.2946(mm)
FI_2 0.722204168 FI_13 0.862284155 B.F 0.197617464
FI_3 0.719470819 FI_14 0.870768505 Total weight 394.574729 kg
FI_4 0.701161177 FI_15 0.915660726 θ 40o
FI_5 0.698515548 FI_16 0.964664573 α 49o
FI_6 0.689078750 Dmajor 2.0 m h1 75.4 mm
FI_7 0.705410043 Dminor 1.6836m b1 56 mm
FI_8 0.707480479 Operating depth (H) 500m h2 50mm
Maximum deflection 4.65067861
FI_9 0.698181393 b2 5mm
(δMAX) (mm)
FI_10 0.772656298 R 0.50 m h3 50mm
FI_11 0.810096248 MOF 4.87178×10-2 b3 5mm
58 Materials Science and Industrial Applications

Conclusions
The methodology for the multi-objective optimization of multiple intersecting cross elliptical
pressure hull is presented and performed using APDL. From the results it is observed that, the pressure
hull constructed from (Carbon/Epoxy composite (USN-150)) has minimum MOF than the pressure
hull constructed from HY100. (Carbon/Epoxy composite (USN-150)) has minimum weight,
minimum B.F. and minimum buckling load than pressure hull constructed from HY100. A higher
strength material provides a better buoyancy to increase the permitted payload as in (Carbon/Epoxy
composite (USN-150)). The fiber orientation is the most influential design parameter. Moreover, it
has a great effect upon the Tsai-Wu failure and the buckling strength factor. All of these results from
this study will be helpful in the future of development and designing composite underwater vehicles.

References
[1] J. Zhang, M. Zhang, W. Tang, W. Wang, and M. Wang, "Buckling of spherical shells subjected
to external pressure: A comparison of experimental and theoretical data," Thin-Walled
Structures, vol. 111, pp. 58-64, 2017/02/01/ 2017.
[2] M. Zhang, W. Tang, F. Wang, J. Zhang, W. Cui, and Y. Chen, "Buckling of bi-segment
spherical shells under hydrostatic external pressure," Thin-Walled Structures, vol. 120, pp. 1-8,
2017/11/01/ 2017.
[3] E. Fathallah, H. Qi, L. Tong, and M. Helal, "Multi-Objective Optimization of Composite
Elliptical Submersible Pressure Hull for Minimize the Buoyancy Factor and Maximize
Buckling Load Capacity," Applied Mechanics and Materials, vol. 578, pp. 75-82, 2014.
[4] Y. Zhu, Q. Ma, J. Zhang, W. Tang, and Y. Dai, "Opening reinforcement design and buckling
of spherical shell subjected to external pressure," International Journal of Pressure Vessels and
Piping, vol. 158, pp. 29-36, 2017/12/01/ 2017.
[5] C. T. F. Ross and A. P. F. Little, "The Buckling of a Corrugated Carbon Fibre Cylinder Under
External Hydrostatic Pressure," Ocean Engineering vol. 28, pp. 1247-1264, 2001.
[6] M. J. Smith, T. Macadam, and J. R. MacKay, "Integrated modelling, design and analysis of
submarine structures," Ships and Offshore Structures, vol. 10, pp. 349-366, 2015.
[7] Z. Sekulski, "Multi-objective topology and size optimization of high-speed vehicle-passenger
catamaran structure by genetic algorithm," Marine Structures, vol. 23, pp. 405-433, 2010.
[8] J. R. MacKay, M. J. Smith, F. van Keulen, T. N. Bosman, and N. G. Pegg, "Experimental
investigation of the strength and stability of submarine pressure hulls with and without artificial
corrosion damage," Marine Structures, vol. 23, pp. 339-359, 2010/07/01/ 2010.
[9] M. Aydogdu and T. Aksencer, "Buckling of cross-ply composite plates with linearly varying
In-plane loads," Composite Structures, vol. 183, pp. 221-231, 2018/01/01/ 2018.
[10] J. Zhang, M. Wang, W. Wang, W. Tang, and Y. Zhu, "Investigation on egg-shaped pressure
hulls," Marine Structures, vol. 52, pp. 50-66, 2017/03/01/ 2017.
[11] J. Y. Han, H. Y. Jung, J. R. Cho, J. H. Choi, and W. B. Bae, "Buckling analysis and test of
composite shells under hydrostatic pressure," Journal of Materials Processing Technology, vol.
201, pp. 742-745, 2008/05/26/ 2008.
[12] J. R. Vinson and R. L, The Behavior Of Structures Composed Of Composite Materials:
Springer, Dordrecht, 2008.
[13] A. K. Kaw, Mechanics of Composite Materials, Second Edition ed.: CRC Press Taylor &
Francis Group, 2006.
[14] P. D. Soden, M. J. Hinton, and A. S. Kaddour, "A comparison of the predictive capabilities of
current failure theories for composite laminates," Composites Science and Technology, vol. 58,
pp. 1225-1254, 1998/07/01/ 1998.
[15] J. Arora, Introduction to Optimum Design vol. 4th Edition: Elsevier 2016.