materials
Article
Experimental Study on Mechanical Properties of Concrete at
Super-Early Age
Qiuwei Yang 1,2 , Yun Sun 1,2, * and Xi Peng 1,2
Citation: Yang, Q.; Sun, Y.; Peng, X.
Experimental Study on Mechanical
Properties of Concrete at Super-Early
Age. Materials 2022, 15, 7582.
https://doi.org/10.3390/ma15217582
Academic Editor: Magdalena Janus
Received: 5 October 2022
Accepted: 25 October 2022
Published: 28 October 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Materials 2022, 15, 7582. https://doi.org/10.3390/ma15217582
1 School of Civil and Transportation Engineering, Ningbo University of Technology, Ningbo 315211, China
2 Engineering Research Center of Industrial Construction in Civil Engineering of Zhejiang, Ningbo University
of Technology, Ningbo 315211, China
* Correspondence: sunyun@nbut.edu.cn
Abstract: Few studies have reported the cohesion and friction angle of concrete at a super early
age. However, these two mechanical parameters are necessary to study the influence of engineering
vibration on super-early-age concrete. In view of this, the mechanical properties of the super-early
age-concrete are investigated in this work by direct shear testing. Firstly, the shear strength of the
super-early-age concrete is measured by the direct shear experiment under different normal pressures
at different times. Secondly, the cohesion and friction angle of the super early age concrete are
calculated according to the Mohr–Coulomb criterion of failure. To overcome the great discreteness
and randomness in the measured data, a new robust regression analysis algorithm is presented to
replace the traditional regression analysis method to obtain more reliable and reasonable mechanical
parameters. According to the experimental and theoretical analysis results, it is found that the friction
angles of the super early age concrete are located in the interval of [50◦ , 70◦ ]. The cohesion of the
concrete is about 78.7 kPa at the initial setting state and about 190.9 kPa at the final setting state,
respectively. It has been shown that the cohesion of the concrete at a super-early age tends to increase
rapidly with time. The method and test results of this work can be used as a reference for relevant
engineering practice. Specifically, the proposed regression method can be extended to the data
analysis of other mechanical parameters of concrete, as well as other brittle materials such as rock.
The test results of early concrete cohesion and friction angle can be used to analyze the adverse effects
of vibration on newly cast concrete members in pile driving and blasting engineering.
Keywords: concrete; super-early age; robust regression analysis; cohesion; friction angle
1. Introduction
In the construction projects of civil engineering, great attention is paid to the physical
and mechanical properties of the concrete at an early age [1–3]. The mechanical properties
of concrete after curing can be predicted by investigating the development law of the
mechanical properties of concrete at an early age. On the flip side, the curing conditions of
concrete can also be improved by the change rule of the mechanical properties of concrete
at an early age. In some engineering practice, it is often necessary to discuss the impact
of vibration on the mechanical properties of early-age concrete. To this end, much of the
literature in recent years has reported the physical and mechanical properties of concrete
at an early age, such as compressive strength, tensile strength, temperature, moisture,
shrinkage, creep, impact resistance, etc.
To determine early-age strength, Swaddiwudhipong et al. [4] carried out the direct
tension experiment to investigate the tensile strain capacity of the concrete with uniaxial
tension at an early age. The experimental results indicated that the concrete tensile strain
capacity is a relatively independent parameter. Yoshitake et al. [5] carried out the direct
tension experiment to determine the tensile Young’s moduli of six concrete and mortar
mixtures. It is found that the tensile modulus is about 1.0–1.3 times the compressive
https://www.mdpi.com/journal/materials
Materials 2022, 15, 7582 2 of 10

modulus in the first week of the material. Hasan and Kabir [6] developed a reasonable
polynomial mathematical model to predict the compressive strength of concrete at 28 days,
based on early results. Peng et al. [7] proposed an improved hyperbolic model to predict
the compressive strength of the early-age concrete. Wolfs et al. [8] proposed a numerical
model to analyze the mechanical properties of early 3D printing concrete. They found that
in terms of the compressive and shear strength, Young’s modulus linearly increases with
fresh concrete age. Wan et al. [9] investigated the early-age experimental characterization
of the ultra-high-performance concretes. Based on the tested results, they proposed a
computational model for the aging analysis of the ultra-high-performance concretes. Shen
et al. [10] investigated the influence of internal curing on the early expansion of high-
strength concrete. It was found that the early expansion increased with the increase in
internal curing water. For the early-age temperature and moisture, Shen et al. [11] used the
temperature stress test machine to study the effect of polypropylene plastic fiber length
on the early age properties of the high-strength concrete. Azenha et al. [12] performed the
experimental study to monitor concrete deformations during early ages after casting. It
was found that the temperature sensitivity of the sensor to the curing instant of concrete
and its signal during the pre-curing period is the key. Wu et al. [13] developed a finite
element method based on ANSYS software to compute the early temperature stress field of
concrete. The theoretical solution obtained by their method is very close to the experimental
value. Zhang et al. [14] performed the experimental study to obtain the variation law of the
moisture transfer coefficient for early-age concrete. They found that the moisture diffusion
coefficient at the early age of concrete has a great relationship with the age. Zhang and
Huang [15] proposed a general moisture distribution model for early-age concrete. Their
model can well predict water distribution and its change over time. Hilaire et al. [16]
employed a theoretical model to forecast the massive concrete cracking at the early age.
Using the humidity and deformation sensors, Zhang et al. [17] investigated the change
rule of internal relative humidity and global deformation for the early age concrete after
casting. It was found that the global deformation behaves first as a plastic swelling, and
then shrinking occurs in the initial few hours after casting.
For the early-age shrinkage and creep, Altoubat and Lange [18] found that the seal
curing for the concrete samples cannot avoid the early-age shrinkage, but the wet curing
can effectively release the shrinkage stresses. Di Luzio and Cusatis [19] proposed the
solidification–microprestress–microplane (SMM) model to predict the mechanical prop-
erties of concrete at an early age and beyond. From the experimental data, it was found
that their SMM model can describe the early shrinkage, creep and cracking of the concrete
well. Soliman and Nehdi [20] studied the influences of drying conditions on the early
shrinkage of high-strength concrete. They found that autogenous and drying shrinkage
are dependent phenomena. Using fibre Bragg gratings, Wong et al. [21] found that the
total shrinkage of the reactive powder concrete for seven days is 488 µε, with the early
age shrinkage contributing about 77% of this. Zhao et al. [22] studied the early creep of
fly ash mass concrete using a thermal stress tester. They found that the ultra-high content
fly ash concrete had only medium strength at the early stage, and had poor fresh and
hardened engineering performance. Cortas et al. [23] studied the influence of aggregate
water saturation on early shrinkage development and cracking risk of concrete. It is found
that the development of early performance of concrete depends on the amount of water
added in the mixing process. Liu et al. [24] found that multiaxial compression caused by
early temperature rise in mass concrete can strengthen concrete and reduce tensile cracks.
Farah et al. [25] conducted early bending creep tests on concrete beams and found that the
Young’s modulus and residual strength of partially damaged beams decrease under creep
load. Østergaard et al. [26] found that the concrete will show higher tensile creep strain if it
is loaded at the age of less than or equal to one day. For the early-age impact resistance,
Ansell and Silfwerbrand [27] summarized the experience in actual project construction
and proposed the maximum vibration levels for the early age concrete. Dunham et al. [28]
studied the influence of vibration on concrete strength during initial setting and final
Materials 2022, 15, 7582 3 of 10
Materials 2022, 15, x FOR PEER REVIEW 3 o

setting. Wojtowicz et al. [29,30] discussed the impact range and influence mechanism of
studied the influence of vibration on concrete strength during initial setting and final
vibration generated by pile driving on the newly cast reinforced concrete members.
ting. Wojtowicz et al. [29,30] discussed the impact range and influence mechanism of
However, few studies
bration generatedhave reported the cohesion
by pile driving and friction
on the newly angle ofconcrete
cast reinforced super-early-
members.
age concrete. In pile However,
driving and blasting engineering, these two mechanical
few studies have reported the cohesion and friction parameters
angle of super-ea
are necessary parameters
age concrete. to study
In pilethe influence
driving of vibration
and blasting on super-early-age
engineering, concrete. paramet
these two mechanical
Herein, the termare
super early age refers to the period between initial setting
necessary parameters to study the influence of vibration on super-early-age and final setting concr
of the concrete. Herein,
The cohesion
the term and friction
super earlyangle are important
age refers to the period parameters to determine
between initial setting and final
whether cracks occur
ting ofinthe theconcrete.
super-early-age
The cohesion concrete under external
and friction angle arevibration
importantload such to de
parameters
as blast, vehiclemine
vibration,
whether and so on.
cracks Theinresearch
occur significanceconcrete
the super-early-age of this work
undermainlyexternallies
vibration lo
in two aspects. such
The asfirst is tovehicle
blast, investigate the and
vibration, mechanical
so on. The properties of the super-early-
research significance of this work mai
age concrete byliesdirect
in twoshear testing.
aspects. The isobtained
The first cohesion
to investigate and friction
the mechanical angles of the
properties thesuper-ea
early-age concrete
agecan be used
concrete by to evaluate
direct shear the damage
testing. effect of cohesion
The obtained vibrationand on fresh
frictionconcrete
angles of the ea
members. The second is to propose
age concrete can be used a new regression
to evaluate algorithm
the damage to compute
effect of vibration theoncohesion
fresh concrete me
and friction angle of the
bers. Thesuper-early-age
second is to propose concrete.
a new This new regression
regression algorithmapproach
to compute canthe
alsocohesion a
be employed to friction
analyzeangle
the other
of the mechanical
super-early-agecharacteristics
concrete. Thisof thenew
brittle materials.
regression From can also
approach
employed
the test and theoretical to analyze
results, it is the
shownotherthatmechanical characteristics
the friction angle of the of the brittle materials. From
super-early-age
test
concrete has little and theoretical
relationship withresults,
concrete it is shown
age, thatcohesion
but the the friction angle ofrelated
is closely the super-early-age
to the c
age of concrete. crete has little
Generally, therelationship
cohesion ofwith concrete
concrete age, but
increases thethe
with cohesion
increase is closely related to the
in concrete
of concrete.
age. The framework of theGenerally, the cohesion
study is presented asofthe
concrete increases
following. with 2the
Section increase the
provides in concrete a
The framework
experimental materials and process of theofstudy is presented
the direct shear test as for
thethe
following.
concreteSection 2 provides the exp
at a super-early
mental materials
age. Section 3 describes the mainand processand
principles of the direct shear
formulas of thetest
newforregression
the concrete at a super-early a
algorithm.
Section 3 describes the main principles
Finally, the conclusions of this study are summarized in Section 4. and formulas of the new regression algorith
Finally, the conclusions of this study are summarized in Section 4.
2. Experimental Materials and Process
2. Experimental Materials and Process
As stated before, there are few reports on the experiments of cohesion and friction
angle of the concrete As
at astated before, age.
super-early thereHowever,
are few reports on the
these two experiments
parameters mayofbecohesion
used to and frict
angle of
analyze the influence of the concrete
vibration onatthe
a super-early
concrete atage. However, these
a super-early age. two parameters
To this may be used
end, direct
shear experiments are carried out in this work to obtain the cohesion and friction anglethis
analyze the influence of vibration on the concrete at a super-early age. To of end, dir
shear experiments are carried out in this work to obtain the
the concrete at super-early age. This experiment is implemented by the method similar to cohesion and friction angle
the concrete at super-early age. This experiment is implemented
the soil direct shear test. As seen in Figure 1a, the materials used in this experiment are by the method simila
the soil(natural
cement, fine aggregate direct shear test.sand),
yellow As seen
waterin Figure 1a, the
and coarse materialsTable
aggregate. used 1inpresents
this experiment
cement, fine aggregate (natural
the concrete mix proportion in this experiment. yellow sand), water and coarse aggregate. Table 1 prese
the concrete mix proportion in this experiment.

(a) (b)
Figure 1. (a) Experimental materials; (b) Concrete mixer.
Figure 1. (a) Experimental materials; (b) Concrete mixer.
Table 1. Concrete mix proportion.

Cement: Water: Sand: Coarse Aggregate = 1:0.43:1.22:2.27

Materials 2022, 15, 7582 4 of 10

Table 1. Concrete mix proportion.

Materials 2022, 15, x FOR PEER REVIEW 4 of 10
Materials 2022, 15, x FOR PEER REVIEW Cement: Water: Sand: Coarse Aggregate = 1:0.43:1.22:2.27 4 of 10

Cement Water Fine Aggregate Coarse Aggregate

Cement Water Fine Aggregate Coarse Aggregate3
488.37 kg/m3 210 kg/m3 595.57 kg/m3 1106.06 kg/m
Cement
488.37 kg/m 3 Water
210 kg/m 3 Fine Aggregate
595.57 kg/m 3 Coarse
1106.06Aggregate
kg/m3
488.37 kg/m3 210 kg/m3 595.57 kg/m3 1106.06 kg/m3
At
At8:50
8:50a.m.,
a.m., the above materials
the above materialswere
werepoured
pouredinto intothe
the concrete
concrete mixer
mixer (shown
(shown in
in Fig-
Figure
ure 1b) 1b) to begin mixing. Subsequently, the round concrete specimens, as shown in
At to begin
8:50 a.m.,mixing.
the above Subsequently,
materials werethe round
pouredconcrete
into thespecimens,
concrete mixeras shown
(shown in in
Figure
Fig-
Figure
2, were 2, were
producedproduced for the direct
for the Subsequently, shear
direct shear tests tests at different times. Figure 3 shows the
ure 1b) to begin mixing. theatround
different times.specimens,
concrete Figure 3 shows the two
as shown pieces
in Figure
two pieces of equipment
ofwere
equipment used used in the test: a concrete penetration resistance instrument and
2, produced forinthethe test:
direct a concrete
shear tests atpenetration resistance
different times. Figure 3instrument
shows the two and pieces
direct
direct shear instrument. The concrete penetration resistance instrument is used to measure
shear
of instrument.
equipment usedThein concrete
the test: penetration resistance instrument
a concrete penetration resistance is used to measure
instrument the
and direct
the penetration resistance of the concrete at different times. The setting state of concrete can
penetration resistance of the concrete at different times. The setting
shear instrument. The concrete penetration resistance instrument is used to measure the state of concrete can
be assessed by the measured values of the penetration resistance for the early-age concrete.
be assessed resistance
penetration by the measured values of
of the concrete at the penetration
different times. resistance
The settingfor theofearly-age
state concretecon-can
The direct shear instrument is used to measure the shear strength of the concrete under
crete.
be The direct
assessed by the shear instrument
measured valuesis of
usedthetopenetration
measure the shear strength
resistance for the of the concrete
early-age con-
different normal pressures at different times. Figure 4 presents the experiment process of
underThe
crete. different
directnormal pressures atisdifferent
shear instrument used to times.
measure Figure
the 4shear
presents the experiment
strength of the concretepro-
the direct shear test. Figure 5 presents the common failure modes of the direct shear test for
cess ofdifferent
under the directnormal
shear test. Figureat
pressures 5 presents
differentthe common
times. Figurefailure modes
4 presents theofexperiment
the direct shear
pro-
these samples.
test for
cess these
of the samples.
direct shear test. Figure 5 presents the common failure modes of the direct shear
test for these samples.

Figure 2. Round concrete specimens for direct shear test.

Figure 2. Round concrete specimens for direct shear test.
Figure 2. Round concrete specimens for direct shear test.

(a) (b)
(a)penetration resistance instrument; (b) Direct
Figure 3. (a) Concrete (b) shear instrument.
Figure 3. (a) Concrete penetration resistance instrument; (b) Direct shear instrument.
Figure 3. (a) Concrete penetration resistance instrument; (b) Direct shear instrument.
Materials 2022, 15, 7582 5 of 10
Materials 2022, 15, x FOR PEER REVIEW 5 of 10

Materials 2022, 15, x FOR PEER REVIEW 5 of 10

Figure 4. Direct shear test process.

Figure 4. Direct shear test process.
Figure 4. Direct shear test process.

Figure 5. Direct shear failure modes of the samples.

Figure 5. Direct shear failure modes of the samples.
Figure 5. Direct shear failure modes of the samples.
Table 2 presents the tested values of the penetration resistance for the concrete at
Table 2 presents the tested values of the penetration resistance for the concrete at
Table times.
different 2 presents
Tablethe tested values
3 presents of thevalues
the tested penetration resistance
of the shear for the
strength for concrete at
the concrete
different times. Table 3 presents the tested values of the shear strength for the concrete
different times.
under different Table
differentnormal 3 presents
normalpressures
pressuresthe tested values of the shear strength for the concrete
under atat different
different times.
times.
under different normal pressures at different times.
Table 2.
Table 2.Tested
Testedvalues
valuesofofthe
thepenetration
penetration resistance
resistance of the
of the super-early-age
super-early-age concrete.
concrete.
Table 2. Tested values of the penetration resistance of the super-early-age concrete.
Time
Time 12:11
12:11 a.m. 14:00
a.m. 14:00 p.m. 15:05
p.m. 15:05
p.m.p.m. 16:2016:20
p.m.p.m.
The Time
The penetration
penetration resistance12:113.87
resistance a.m.
3.87 MPa 14:00
MPa 6.24p.m.
6.24
MPa 15:05
MPa 21.8 p.m.
21.8
MPa MPa 31.5 16:20
31.5
MPap.m.
MPa
The penetration resistance 3.87 MPa 6.24 MPa 21.8 MPa 31.5 MPa
Table 3.
Table 3.Tested
Testedvalues
valuesofofthe
theshear
shear strength
strength of of
thethe super-early-age
super-early-age concrete
concrete (kPa).
(kPa).

Tested values of The

Table 3. Time The
the Shear
Shear
shear Strength
Strength
strength τ τsuper-early-age
of the under Different
under Normal
Different Normal
concrete Pressure
(kPa). σ
Pressure σ
Time
σσ = =100 kPa σ σ= 200
100kPa kPa σ σ= 300
kPa
= 200 kPakPa σ =σ400= kPa
= 300 400 kPa
12:11
12:11 a.m.
a.m. The Shear Strength
203.6
203.6 372.9τ under Different
372.9 481.1Normal Pressure
481.1 578.3
σ578.3
Time
12:41 a.m.
12:41 a.m. σ =192.8
100 kPa
192.8 σ=374.7
200 kPa
374.7 σ482.8
= 300 kPa
482.8 756.7
σ = 400 kPa
756.7
13:20
12:11 p.m.
13:20a.m.
p.m. 236.0
236.0
203.6 454.9
454.9
372.9 562.2
562.2
481.1 681.9681.9
578.3
14:00
12:41 p.m.
14:00a.m.
p.m. 151.4
192.8
151.4 414.4
374.7
414.4 351.4
482.8
351.4 924.3
756.7
924.3
14:35p.m.
13:20 p.m. 281.1
236.0 390.1
454.9 618.0
562.2 909.9
681.9
14:35 p.m. 281.1 390.1 618.0 909.9
14:00
15:05p.m.
p.m. 151.4
324.3 414.4
320.7 351.4
500.0 924.3
891.9
15:05 p.m.
14:35 324.3 320.7 500.0 891.9
15:50p.m.
p.m. 281.1
331.5 390.1
341.4 618.0
672.1 909.9
720.7
15:50p.m.
15:05 p.m. 331.5
324.3 341.4
320.7 672.1
500.0 720.7
891.9
16:20 p.m. 264.9 434.2 454.9 677.5
15:50
16:20p.m.
p.m. 331.5
264.9 341.4
434.2 672.1
454.9 720.7
677.5
16:20 p.m. 264.9 434.2 454.9 677.5
The setting state of the concrete can be determined according to Table 2. In general,
The settingresistance
the penetration state of the concrete
value in thecan be determined
initial setting stageaccording
of concretetoisTable
located 2. Inin general,
the
the The
penetration
interval setting
of resistance
state
[3.5, 6.9]. The value invalue
of penetration
the concrete the of
can initial
be setting
determined
concrete stage
at the finalof
according concrete
settingto stage
Tableisis2.
located in the
In general,
generally
interval
the of [3.5, 6.9]. The penetration value of concrete at the final setting
in the interval of [19.6, 30.9]. Thus, one can see from Table 2 that the concrete was in thethe
penetration resistance value in the initial setting stage of concrete stage
is locatedis generally
in
in the setting
interval
initial interval of [19.6,
of [3.5,state
6.9]. atThe 30.9]. Thus,
penetration
12:11 a.m., one
and value
in thecan
of seesetting
fromatTable
concrete
final at2 15:05
the final
state that the
p.m.concrete
setting stage
FromisTable was 3, in the
generally
initial
inone
thecansetting
find that the shear strength increases with the increase in normal pressure for thethe3,
interval state
of at
[19.6, 12:11
30.9]. a.m.,
Thus,and
one in the
can final
see fromsetting
Table state
2 at
that 15:05
the p.m.
concrete Fromwas Table
in
concrete find
one
initialcan
setting that the
state
samples at sheartime.
at each
12:11 strength
a.m., and increases
in thethere
However, with
final setting
is theobvious
no increase in normal
state atregularity
15:05 p.m.inpressure
From
the shear for the
Table 3,
one can find
concrete that theatshear
samples eachstrength increases there
time. However, with the increase
is no obviousin normal pressure
regularity in the forshear
the
concrete samples at each time. However, there is no obvious regularity in the shear strength
Materials 2022, 15, 7582 6 of 10

data at different times. This reflects the great discreteness and randomness of mechanical
properties of the concrete.

3. Regression Analysis Method

According to the Mohr–Coulomb criterion of failure, the relationship between shear
strength τ and normal stress σ of the super-early-age concrete can be expressed as

τ = c + σ · tan( ϕ) (1)

where c and ϕ are the cohesion and friction angle of the concrete, respectively. Using
Equation (1) and Table 3, the cohesion and friction angle of the super-early-age concrete can
be calculated by the linear regression analysis. Traditionally, the linear regression analysis
is based on the least-square solution. Letting c = x0 and tan( ϕ) = x1 , Equation (1) can be
rewritten as the following linear equation

τ = σ · x1 + x0 (2)

For each time point, the regression coefficients x0 and x1 can be calculated by the
least-square solution. Taking 12:11 a.m. as a case, the linear regression system for this time
point can be obtained by substituting the data in Table 3 into Equation (2) as

{τ } = Π · { x } (3)
 

 203.6
372.9
 
{τ } = , (4)

 481.1
578.3
 
 
100 1
200 1
Π= 300 1,
 (5)
400 1
 
x1
{x} = (6)
x0
Note that the values in Equations (4) and (5) are from Table 3, that and the correspond-
ing units are kPa. In Equation (6), x0 is the cohesion values with a unit of kPa, and x1 is a
pure number without unit. It should be pointed out that Equation (3) is the overdetermined
system of the equation. From Equation (3), the regression coefficients x0 and x1 are often
computed by the least-square estimation as.
−1
{ x }c = (Π T Π) ΠT {τ } (7)

in which the symbol “T” indicates the transposition process of a matrix. For the other
time points, the regression coefficients x0 and x1 can also be computed by this regression
approach based on the least-square estimation. All the calculated results of the regression
coefficients x0 and x1 are shown in Table 4. Note that x0 is the cohesion c. The friction angle
ϕ is also given in Table 4, which is calculated by ϕ = arctan( x1 ).
Materials 2022, 15, 7582 7 of 10

Table 4. The regression coefficients obtained by the traditional regression analysis.

12.11 12:41 13:20 14:00 14:35 15:05 15:50 16:20

Time
a.m. a.m. p.m. p.m. p.m. p.m. p.m. p.m.
x1 1.2323 1.7998 1.4450 2.2557 2.1143 1.8821 1.4983 1.2585
( ϕ = arctan( x1 )) (50.94◦ ) (60.94◦ ) (55.32◦ ) (66.09◦ ) (64.69◦ ) (62.02◦ ) (56.28◦ ) (51.53◦ )
x0
100.9 1.8 122.5 −103.55 21.2 38.7 141.85 143.25
(c = x0 , kPa)

From Table 4, it was found that the friction angles of the super-early-age concrete
are located in the interval of [50◦ , 70◦ ]. However, the cohesion values obtained by the
traditional regression analysis in Table 4 are not very satisfactory. It is obvious that the
calculated cohesions for 12:41 a.m., 14:00 p.m., 14:35 p.m., and 15:05 p.m. are wrong since,
these values are abnormally small. The possible reason for this error is that the concrete
mechanical properties may be very discrete and random in many cases. In view of this,
the least-square regression algorithm cannot be successfully used to deal with those test
parameters of concrete. Therefore, it is significant to develop new regression algorithms to
tackle the discrete data. For this purpose, a regression analysis method based on modified
weight coefficients is proposed to calculate the fitting parameters more accurately. The basic
principle of the proposed method is to reduce the weight of the abnormal data according
to the feedback of residual errors. At first, the residual error vector {v} is derived from
Equation (3) as
{v} = |Π · { x }c − {τ }| (8)
The i-th element vi in {v} is the absolute value of the residual error between the tested
shear strength, with the analytical shear strength obtained by Equation (7). Assuming mv and
sv are the mean and standard deviation of {v}, the relative deviation ri can be defined as

ri = | vi − m v | − s v (9)

From Equation (9), a diagonal matrix whose nonzero elements are weight coefficients
can be defined as:  
w1 0 0 0
 0 ... 0
 
0
W= , (10)
0
 0 wi 0  
..
0 0 0 .
( vmin
−1
e vi , i f ri < 0
wi = (11)
0 , i f ri ≥ 0
in which vmin is the minimum value of {v}. Using Equation (10), the solution of the
unknowns { x } can be computed by the new regression method as
−1
{ x }r = ( Π T W 2 Π ) ΠT W 2 {τ } (12)

Using Equation (12), the new solution of the regression coefficients x0 and x1 is listed in
Table 5. From Table 5, it was found that the calculated friction angles are also located in the
interval of [50◦ , 70◦ ]. Comparing Table 5 with Table 4, one can see that the cohesion values
obtained by the proposed regression method are more reasonable and reliable than those
obtained by the traditional regression method. Using the proposed regression method,
only the calculated cohesions of 12:41 a.m. and 14:00 p.m. are determined to be wrong,
since these two values are abnormally small. Other cohesion values basically increase with
the increase in time, and the distribution interval is more reasonable than that in Table 4.
For example, the value 78.7 in Table 5 is more reasonable than the value 100.9 in Table 4
for 12.11 a.m. This is because the concrete is in the initial setting state at 12.11 a.m., and its
 located in the interval of [50°, 70°]. Comparing Tables 5 with 4, one can see that the cohe-
sion values obtained by the proposed regression method are more reasonable and reliable
than those obtained by the traditional regression method. Using the proposed regression
method, only the calculated cohesions of 12:41 a.m. and 14:00 p.m. are determined to be
wrong, since these two values are abnormally small. Other cohesion values basically in-
Materials 2022, 15, 7582 8 of 10
crease with the increase in time, and the distribution interval is more reasonable than that
in Table 4. For example, the value 78.7 in Table 5 is more reasonable than the value 100.9
in Table 4 for 12.11 a.m. This is because the concrete is in the initial setting state at 12.11
cohesion
a.m., andshould not be so
its cohesion large.not
should Forbe
16:20 p.m., the
so large. Forvalue
16:20190.9
p.m.,inthe
Table 5 is190.9
value moreinreasonable
Table 5 is
than the value 143.25 in Table 4. This is because the concrete is in the
more reasonable than the value 143.25 in Table 4. This is because the concrete final setting is
state at
in the
16:20 p.m., and its cohesion should be a lot larger.
final setting state at 16:20 p.m., and its cohesion should be a lot larger.

Table5.5.The
Table Theregression
regressioncoefficients
coefficientsobtained
obtainedby
bythe
theproposed
proposedregression
regressionanalysis.
analysis.

Time
Time 12.11
12.11 a.m. 12:41
a.m. 12:41 a.m. 13:20
a.m. 13:20 p.m. 14:00
p.m. 14:00 p.m. 14:35
p.m. 14:35 p.m. 15:05
p.m. 15:05p.m.
p.m. 15:50
15:50p.m.
p.m. 16:20
16:20p.m.
p.m.
x1x1 1.2490
1.2490 1.8738
1.8738 1.4863
1.4863 2.5820
2.5820 2.096
2.096 1.892
1.892 1.7030
1.7030 1.2165
1.2165
(ϕ = arctan( x1 )) (51.32◦ ) 0 (61.91◦ ) 0 (56.07◦ ) 0 (68.82◦ ) 0 (64.49◦ ) 0 (62.14◦ ) 0 (59.58◦ ) 0 (50.58◦ )0
( ϕ = arctan( x1 ) ) (51.32 ) （61.91 ） （56.07 ） （68.82 ） （64.49 ） （62.14 ） （59.58 ） （50.58 ）
x0
78.7 3.603 87.3667 −105.7337 71.5 135.1 161.2 190.9
(c = x0 x
, kPa)
0
78.7 3.603 87.3667 −105.7337 71.5 135.1 161.2 190.9
( c = x 0 ,kPa)
From the above results, it was found that the friction angle of the concrete has little
relationship
From the with concrete
above age,
results, but the
it was cohesion
found is friction
that the closely related
angle oftothe
theconcrete
age of concrete.
has little
Generally,
relationship with concrete age, but the cohesion is closely related to the ageTable
the cohesion of concrete increases with the increase in age. From 5, the
of concrete.
valid data for
Generally, thethe relationship
cohesion of concrete
of concrete cohesion
increases with the with age are
increase listed
in age. in Table
From Table65,and
the
Figure 6. It is obvious that the cohesion of concrete at a super-early age tends to increase
valid data for the relationship of concrete cohesion with age are listed in Table 6 and Fig-
rapidly
ure 6. Itwith time. that the cohesion of concrete at a super-early age tends to increase
is obvious
rapidly with time.
Table 6. The valid cohesion–age data obtained by the proposed regression analysis.
Table 6. The valid cohesion–age data obtained by the proposed regression analysis.
Length of Time 3.35 h 4.5 h 5.75 h 6.25 h 7h 7.5 h
Lengthc of
Cohesion Time
(kPa) 3.35 h
78.7 4.5 h
87.3667 5.75 h
71.5 6.25 h
135.1 7h
161.2 7.5 h
190.9
Cohesion c (kPa) 78.7 87.3667 71.5 135.1 161.2 190.9

250

200
Cohesion (kPa)

150

100

50

0
3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
Length of time (h)

Figure6.6.The
Figure Thechanging
changing trend
trend of the
of the validvalid cohesion–age
cohesion–age data obtained
data obtained by the proposed
by the proposed regressionregression
analysis.
analysis.
4. Conclusions
4. Conclusions
In this work, direct shear experiments are carried out to obtain the cohesion and
frictionInangle
this work,
of thedirect shear
concrete at aexperiments
super-early areage.carried
A newout to obtain
robust the cohesion
regression analysis and fric-
method
istion
proposed
angle oftothe
replace the traditional
concrete regression
at a super-early age. Aanalysis method
new robust in orderanalysis
regression to analysis the
method
measured
is proposeddatatowith great
replace thediscreteness
traditional and randomness.
regression analysisFrom the test
method and numerical
in order anal-
to analysis the
ysis, the following
measured data withmain conclusions
great canand
discreteness be obtained.
randomness.(1) The
Fromexisting
the testregression algorithm
and numerical anal-
using the following
ysis, the least-square
mainestimation is notcan
conclusions suitable for computing
be obtained. some mechanical
(1) The existing parameters
regression algorithm
of concrete, since the measured data are very discrete and random. (2) The proposed new
regression algorithm can successfully overcome data discreteness and randomness, obtain-
ing more reasonable and accurate calculation results. (3) It is found that the friction angles
of the super-early age concrete are located in the interval of [50◦ , 70◦ ]. (4) The cohesion
of the concrete is about 78.7 kPa at the initial setting state and about 190.9 kPa at the final
setting state, respectively. It is obvious that the cohesion of the concrete at a super-early age
tends to increase rapidly with time. The theoretical and experimental results of this work
can be used as references in related engineering practice. Specifically, the new regression
Materials 2022, 15, 7582 9 of 10

method proposed in this paper can be extended to the data analysis of other mechanical
parameters of concrete and can also be used in the data analysis of mechanical parameters
of other brittle materials such as rock. The experimental results on the early cohesion and
friction angles of concrete can be used to analyze the adverse effects of vibration on fresh
concrete members in pile driving and blasting projects. It should be highlighted that this
research is carried out only for one concrete mix. As the results are of practical importance,
these tests can be continued in the future for different types of concrete and for different
time parameters.

Author Contributions: Investigation, Y.S.; Methodology, Q.Y.; Software, X.P. All authors have read
and agreed to the published version of the manuscript.
Funding: This research was funded by the Zhejiang public welfare project (LGF22E080021), Ningbo
natural science foundation (202003N4169), Natural Science Foundation of China (52008215), Zhe-
jiang Province Natural Science Foundation (LQ20E080013), and the project of Ningbo science and
technology innovation 2025 (2019B10076).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The data generated and/or analyzed during the current study are
not publicly available for legal/ethical reasons but are available from the corresponding author on
reasonable request.
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the
collection, analyses, or interpretation of data, or in the decision to publish the results.

References
1. Mihashi, H.; Leite, J.P.B. State-of-the-art report on control of cracking in early age concrete. J. Adv. Concr. Technol. 2004, 2, 141–154.
[CrossRef]
2. Nehdi, M.; Soliman, A.M. Early-age properties of concrete: Overview of fundamental concepts and state-of-the-art research. Proc.
Inst. Civ. Eng.-Constr. Mater. 2011, 164, 57–77. [CrossRef]
3. Maruyama, I.; Lura, P. Properties of early-age concrete relevant to cracking in massive concrete. Cem. Concr. Res. 2019, 123,
105770. [CrossRef]
4. Swaddiwudhipong, S.; Lu, H.-R.; Wee, T.-H. Direct tension test and tensile strain capacity of concrete at early age. Cem. Concr.
Res. 2003, 33, 2077–2084. [CrossRef]
5. Yoshitake, I.; Rajabipour, F.; Mimura, Y.; Scanlon, A. A prediction method of tensile Young’s modulus of concrete at early age.
Adv. Civ. Eng. 2012, 2012, 391214. [CrossRef]
6. Hasan, M.M.; Kabir, A. Prediction of compressive strength of concrete from early age test result. In Proceedings of the 4th Annual
Paper Meet and 1st Civil Engineering Congress, Dhaka, Bangladesh, 22–24 December 2011; pp. 978–984.
7. Peng, X.; Zhuang, Z.; Yang, Q. Predictive Modeling of Compressive Strength for Concrete at Super Early Age. Materials 2022, 15,
4914. [CrossRef]
8. Wolfs, R.J.M.; Bos, F.P.; Salet, T.A.M. Early age mechanical behaviour of 3D printed concrete: Numerical modelling and
experimental testing. Cem. Concr. Res. 2018, 106, 103–116. [CrossRef]
9. Wan, L.; Wendner, R.; Liang, B.; Cusatis, G. Analysis of the behavior of ultra high performance concrete at early age. Cem. Concr.
Compos. 2016, 74, 120–135. [CrossRef]
10. Shen, D.; Wang, X.; Cheng, D.; Zhang, J.; Jiang, G. Effect of internal curing with super absorbent polymers on autogenous
shrinkage of concrete at early age. Constr. Build. Mater. 2016, 106, 512–522. [CrossRef]
11. Shen, D.; Liu, X.; Zeng, X.; Zhao, X.; Jiang, G. Effect of polypropylene plastic fibers length on cracking resistance of high
performance concrete at early age. Constr. Build. Mater. 2020, 244, 117874. [CrossRef]
12. Azenha, M.; Faria, R.; Ferreira, D. Identification of early-age concrete temperatures and strains: Monitoring and numerical
simulation. Cem. Concr. Compos. 2009, 31, 369–378. [CrossRef]
13. Wu, S.; Huang, D.; Lin, F.-B.; Zhao, H.; Wang, P. Estimation of cracking risk of concrete at early age based on thermal stress
analysis. J. Therm. Anal. Calorim. 2011, 105, 171–186. [CrossRef]
14. Zhang, J.; Hou, D.; Gao, Y.; Wei, S. Determination of moisture diffusion coefficient of concrete at early age from interior humidity
measurements. Dry. Technol. 2011, 29, 689–696. [CrossRef]
15. Zhang, J.; Qi, K.; Huang, Y. Calculation of moisture distribution in early-age concrete. J. Eng. Mech. 2009, 135, 871–880. [CrossRef]
16. Hilaire, A.; Benboudjema, F.; Darquennes, A.; Berthaud, Y.; Nahas, G. Modeling basic creep in concrete at early-age under
compressive and tensile loading. Nucl. Eng. Des. 2014, 269, 222–230. [CrossRef]
Materials 2022, 15, 7582 10 of 10

17. Zhang, J.; Dongwei, H.; Wei, S. Experimental study on the relationship between shrinkage and interior humidity of concrete at
early age. Mag. Concr. Res. 2010, 62, 191–199. [CrossRef]
18. Altoubat, S.A.; Lange, D.A. Creep, shrinkage, and cracking of restrained concrete at early age. ACI Mater. J. 2001, 98, 323–331.
19. Di Luzio, G.; Cusatis, G. Solidification–microprestress–microplane (SMM) theory for concrete at early age: Theory, validation and
application. Int. J. Solids Struct. 2013, 50, 957–975. [CrossRef]
20. Soliman, A.M.; Nehdi, M.L. Effect of drying conditions on autogenous shrinkage in ultra-high performance concrete at early-age.
Mater. Struct. 2011, 44, 879–899. [CrossRef]
21. Wong, A.; Childs, P.A.; Berndt, R.; Macken, T.; Peng, G.; Gowripalan, N. Simultaneous measurement of shrinkage and temperature
of reactive powder concrete at early-age using fibre Bragg grating sensors. Cem. Concr. Compos. 2007, 29, 490–497. [CrossRef]
22. Zhao, Z.; Wang, K.; Lange, D.A.; Zhou, H.; Wang, W.; Zhu, D. Creep and thermal cracking of ultra-high volume fly ash mass
concrete at early age. Cem. Concr. Compos. 2019, 99, 191–202. [CrossRef]
23. Cortas, R.; Rozière, E.; Staquet, S.; Hamami, A.; Loukili, A.; Delplancke-Ogletree, M.-P. Effect of the water saturation of aggregates
on the shrinkage induced cracking risk of concrete at early age. Cem. Concr. Compos. 2014, 50, 1–9. [CrossRef]
24. Liu, G.T.; Gao, H.; Chen, F.Q. Microstudy on creep of concrete at early age under biaxial compression. Cem. Concr. Res. 2002, 32,
1865–1870. [CrossRef]
25. Farah, M.; Grondin, F.; Alam, S.; Loukili, A. Experimental approach to investigate creep-damage bilateral effects in concrete at
early age. Cem. Concr. Compos. 2019, 96, 128–137. [CrossRef]
26. Østergaard, L.; Lange, D.A.; Altoubat, S.A.; Stang, H. Tensile basic creep of early-age concrete under constant load. Cem. Concr.
Res. 2001, 31, 1895–1899. [CrossRef]
27. Ansell, A.; Silfwerbrand, J. The vibration resistance of young and early-age concrete. Struct. Concr. 2003, 4, 125–134. [CrossRef]
28. Dunham, M.R.; Rush, A.S.; Hanson, J.H. Effects of induced vibrations on early age concrete. J. Perform. Constr. Facil. 2007, 21,
179–184. [CrossRef]
29. Wojtowicz, A.; Michalek, J.; Ubysz, A. Range of dynamic impact of geotechnical works on reinforced concrete structures. In Proceed-
ings of the XXII International Scientific Conference “Construction the Formation of Living Environment” (FORM-2019), Tashkent,
Uzbekistan, 18–21 April 2019; Volkov, A., Pustovgar, A., Sultanov, T., Adamtsevich, A., Eds.; EDP Sciences: Les Ulis, France, 2019;
pp. 1–8.
30. Wojtowicz, A.; Ubysz, A.; Michałek, J. How vibrations affect steel-to-concrete adherence. In Proceedings of the VI International
Scientific Conference “Integration, Partnership and Innovation in Construction Science and Education” (IPICSE-2018), Moscow,
Russia, 14–16 November 2018; Volkov, A., Pustovgar, A., Adamtsevich, A., Eds.; EDP Sciences: Les Ulis, France, 2018; Volume 251,
pp. 1–8.