applied

sciences
Article
Research on the Fatigue Life Prediction for a New Modified
Asphalt Mixture of a Support Vector Machine Based on Particle
Swarm Optimization
Yu Sun 1,2 , Dongpo He 1, * and Jun Li 2

1 School of Civil Engineering, Northeast Forestry University, Harbin 150040, China; donglin_2016@163.com
2 Department of Municipal and Environmental Engineering, Heilongjiang Institute of Construction Technology,
Harbin 150000, China; jinbaobei_2008@163.com
* Correspondence: Hdp@nefu.edu.cn

Abstract: SMC (Styreneic Methyl Copolymers) is a novel normal temperature asphalt modifier
with superior performance. It has the advantages of a low construction temperature, good road
performance, good energy savings and an emission reduction effect, and can improve the performance
of an asphalt mixture. The fatigue performance of an asphalt mixture is one of the important technical
parameters in the structural design of asphalt pavements. The fatigue performance of an asphalt
mixture under specific traffic and environmental conditions has an important guiding significance
and normative function for the design, construction, and maintenance of asphalt pavement. In this
paper, the mixture of an SMC normal-temperature-modified asphalt and styrene–butadiene styrene
block copolymer (SBS)-modified asphalt (SMCSBS) compound-modified asphalt was investigated,





 and an SMCSBS composite modified asphalt mixture with a different SMC content was prepared.
A semi-circular bending fatigue test (SCB) was conducted to analyze and compare the fatigue
Citation: Sun, Y.; He, D.; Li, J.
properties of the modified asphalt mixture. On this basis, this paper proposes a fatigue life prediction
Research on the Fatigue Life
Prediction for a New Modified
model of an SMCSBS composite modified asphalt mixture based on a particle swarm optimization
Asphalt Mixture of a Support Vector support vector machine (PSO-SVM). SMC content (SMC accounts for the mass percentage of SMCSBS
Machine Based on Particle Swarm composite modified asphalt)/%, asphalt aggregate ratio, stress ratio and loading frequency/Hz were
Optimization. Appl. Sci. 2021, 11, used as training data to establish the prediction model, and RMSE and R2 were used to evaluate the
11867. https://doi.org/10.3390/ performance of the model. Experimental results show that the prediction results of the PSO-SVM
app112411867 method are more accurate than the experimental observation data and can effectively improve
the prediction accuracy of the model. Compared with the M50 model tree (M50 ), artificial neural
Academic Editor: Nikos D. Lagaros network (ANN), and support vector machine (SVM) method, the PSO-SVM method can achieve
better prediction performance and a better prediction effect.
Received: 10 November 2021
Accepted: 12 December 2021
Keywords: SMCSBS composite modified asphalt mixture; the fatigue life; support vector machine (SVM)
Published: 14 December 2021

Publisher’s Note: MDPI stays neutral

with regard to jurisdictional claims in
1. Introduction
published maps and institutional affil-
iations. With increasing traffic volume and heavy load traffic, a large number of reflection
cracks appear in a traditional pavement structure under fatigue load, and the service life
of a road is greatly shortened [1]. Fatigue failure is one of the main forms of current
asphalt pavement damage. It is very important to study the fatigue performance of asphalt
pavement under specific traffic and environmental conditions [2]. In order to ensure good
Copyright: © 2021 by the authors.
Licensee MDPI, Basel, Switzerland.
serviceability and durability of asphalt pavement, it is of great significance to study the
This article is an open access article
fatigue performance of asphalt mixtures and establish an appropriate fatigue prediction
distributed under the terms and model [3–6].
conditions of the Creative Commons Hveem is the first scholar to study the fatigue failure of asphalt pavement based on
Attribution (CC BY) license (https:// elasticity. Hveem’s research shows that there is a correlation between pavement cracks,
creativecommons.org/licenses/by/ fatigue damage types, and pavement deformation caused by wheel pressure [7]. Santucci
4.0/). and Schmidt obtained the result of a low fatigue life of an asphalt mixture with high void

Appl. Sci. 2021, 11, 11867. https://doi.org/10.3390/app112411867 https://www.mdpi.com/journal/applsci

Appl. Sci. 2021, 11, 11867 2 of 14

content through a fatigue test of controlling strain mode [8]. Zhesheng Ge used the grey
relational analysis method to study the influence of load interval time, loading frequency,
test temperature, void fraction, asphalt penetration, and asphalt content on the fatigue
performance of a mixture, and found that the order of influence degree from large to small
is as follows: Load interval time, test temperature, asphalt type, gradation type, asphalt
content, and loading frequency [9]. Hongzhou Zhu evaluated the influence of asphalt
property, aggregate gradation, and mixture volume on fatigue performance by using a grey
correlation. The research showed that the mixture saturation, the mixture void fraction,
the rubber powder mass ratio, and the asphalt film thickness have a great influence on
fatigue performance [10]. Jiangmiao Yu and Xiaoning Zhang studied the influence of the
environment on the fatigue performance of a mixture. The study showed that, within a
certain test temperature range, the fatigue life increased with the increase of temperature.
The fatigue life of the asphalt mixture specimen decreased after long-term aging [11].
Styreneic Methyl Copolymers (SMC) are a kind of normal temperature asphalt modi-
fier, with a methyl styrene block copolymer extracted from waste plastics, waste rubber,
and other methyl styrene polymer materials as the main raw material, and epoxy resin,
an epoxy resin curing agent, and other additives in a certain proportion with the polymer
solution. After the epoxy modification of the SMC, it has good compatibility with the
asphalt material, and then the asphalt material is liquid at a normal temperature and can
be melted or dispersed in the asphalt to change the construction and workability of an
asphalt binder at a normal temperature [12]. The mixture of an SMC normal-temperature-
modified asphalt and styrene–butadiene styrene block copolymer (SBS)-modified asphalt
(SMCSBS) composite modified asphalt mixture, compared with traditional hot mix asphalt
mixtures, not only has the material composition structure, strength structure, and basic
characteristics [13], but also has the advantages of an SMC normal temperature-modified
asphalt with a low construction temperature, good road performance, good energy sav-
ings and an emission reduction effect [14]. Qu Hongwei [15] proposed to apply an SMC
normal temperature-modified asphalt in a highway covering construction and elaborated
the specific construction points, pointing out that an SMC normal temperature-modified
asphalt has the advantages of energy savings and environmental protection, convenient
construction, and significant economic benefits. Wang Jianjun [16] used SBS and SMC
to prepare a compound-modified constant temperature warm-mix asphalt and applied
it to an AC-13 (the mixture ratio of asphalt mixture is asphalt concrete (AC-13)) asphalt
mixture. Through laboratory tests, the high-temperature performance, low-temperature
performance, and moisture susceptibility of the mixture were evaluated, and a performance
analysis was conducted for the road material. At present, the research on an SMC normal
temperature asphalt modifier is still in the exploratory stage [17], and there are few studies
on SMCSBS-modified asphalt mixtures. This paper tries to study the fatigue performance
of an SMCSBS-modified asphalt mixture.
At present, there are many kinds of fatigue life prediction models for asphalt mixtures.
As early as the 1960s, Jones, D. et al. [18] established a classic fatigue life prediction model
based on strain control. Zhang, B et al. [19] used double load envelopes and twin cohesive
models to predict composite fatigue delamination.
Artificial intelligence has a wide range of applications in various fields. Compared
with classical statistical methods, artificial intelligence models have better a prediction
and recognition ability. In order to save testing time, meet equipment requirements, and
cost and complexity, various prediction models, such as Bayesian, artificial neural network
(ANN) [20], and support vector machine regression (SVM), have been introduced. Yan et al.
proposed a Bayesian damage identification method based on the analysis probability model
of scattering coefficient estimation and ultra-fast wave scattering simulation scheme [21].
Tan et al. pointed out that the error between the prediction results of the low-temperature
flexural strain of asphalt and the actual measurement results based on the BP (back propa-
gation) neural network model of the MATLAB platform is within the accuracy range of the
engineering requirements [22]. As a powerful new learning machine, the support vector
Appl. Sci. 2021, 11, 11867 3 of 14

machine (SVM), based on statistical learning theory, has attracted extensive attention from
its initial application in pattern recognition, to its application in regression estimation [23].
For example, Pal investigates classification based on support vector machines to predict
the liquefaction potential from the actual standard penetration test and cone penetration
test field data [24]. Maalouf, Khoury and Trafalis investigated the application of SVM in
the resilient modulus (MR ) prediction of hot mix asphalt samples [25]. In a study by Soltani
et al., an Support Vector Machine Firefly Algorithm (SVM-FFA) was used to predict the
stiffness of polyethylene terephthalate (PET)-modified asphalt mixtures [26]. Compared
with the artificial neural network, genetic algorithm, and the support vector machine, this
method is effective in stiffness prediction.
To sum up, in this paper, the fatigue performance of an asphalt mixture is studied from
the following aspects: SMC content (SMC accounts for the mass percentage of SMCSBS
composite modified asphalt)/%, asphalt stone ratio, stress ratio, and load frequency/Hz.
So far, the research of AI models in predicting the fatigue life of asphalt mixtures is very
limited. Therefore, the main goal of this study is to use an artificial intelligence method to
predict the fatigue life of an SMCSBS asphalt mixture with different dosages and find the
best prediction model.

2. Objectives
In this paper, the main objectives of the work presented are as follows:
(1) Disperse an SMC modifier into the SBS asphalt to prepare an SMCSBS-modified
asphalt with different SMC contents, including 8%, 10%, and 12%, and then prepare
the SMCSBS-modified asphalt mixture for a semi-circular bending test and semi-
circular bending fatigue test (SCB).
(2) Establish the M50 model tree (M5), artificial neural network (ANN), support vector
machine (SVM), and PSO-SVM models to find the best prediction model for the
fatigue life of different SMCSBS asphalt mixtures.

3. Materials
Under the condition of the same amount of SBS (4%), different amounts of SMC
normal temperature-modified asphalt were added. According to the content of SMC,
an asphalt modifier is 10%–12% of the asphalt quality (DBJ 64/T060-2016) [27], and the
literature [28] on the properties of SMCSBS composite modified asphalts, 8%, 10%, and
12% of SMC content are selected to prepare a composite modified asphalt. The SMC
normal temperature-modified asphalt was added into the SBS-modified asphalt according
to the above content, the heating temperature was 110–120 ◦ C, and the mixing time was
1–2 h (DBJ 64/T060-2016). Referring to the SMC Normal Temperature Modified Asphalt
and Asphalt Mixture Design and Construction Technical Guide by the Scientific Research
Institute of the Ministry of Transport (2014) testing method, the main technical indexes of
normal temperature modified asphalts with different SMC contents were obtained through
testing. The test results are shown in Table 1.

Table 1. Process indexes of modified asphalts with different content.

25 ◦ C Penetration/0.1 mm Softening Point ◦ C 5 ◦ C Ductility

SMC Content/%
Average Standard Deviation Average Standard Deviation Average Standard Deviation
8 191 0.211 74 0.050 89.7 0.531
10 228 0.325 56.5 0.400 >100 0
12 248 0.561 55 0.636 >100 0

One dense-graded asphalt mixture (AC-16) was used for evaluation in this research.
AC-16 mixtures are commonly used in China for the surface course of expressway bitumi-
nous concrete pavements. According to the JTG F40-2004 standard [29], the recommended
mid-value gradation for this mixture was selected. The aggregate gradation and mix design
are shown in Table 2.
Appl. Sci. 2021, 11, 11867 4 of 14

Table 2. Aggregate gradation and mix design.

Passing Rate (by Mass)

Sieve Size(mm)
Up Limits Bottom Limits Composite Gradation
19 100 100 100.0
16.0 100 90 99.6
13.2 92 76 84.2
9.5 80 60 76.7
4.75 62 34 52.0
2.36 48 20 34.6
1.18 36 13 25.1
0.60 26 9 15.3
0.30 18 7 10.1
0.15 14 5 5.2
0.075 8 4 2.4
Binder PG64-34 SMCSBS modified asphalt
Apparent density (g/cm3 ) 2.766

The fatigue performance of SBSSMC modified asphalt mixtures are studied by select-
ing 4.3%, 4.8%, and 5.3% of the asphalt aggregate ratio as specimens.

4. Methods
4.1. Specimen Forming and Semi-Circular Bending (SCB) Fatigue Test
At present, the fatigue test methods of asphalt mixtures mainly include the four-point
bending fatigue test, direct tensile fatigue test, indirect tensile fatigue test, and semi-circular
bending (SCB) fatigue test. Hasan et al. compared and analyzed the difference between the
four-point bending fatigue and semi-circular bending (SCB) test by studying the fatigue
performance of coarse and fine asphalt mixtures [30]. Both mixtures have similar fatigue
properties (crack resistance). Dong et al. found that, compared with the indirect tensile
fatigue test, the SCB and beam fatigue test had a faster failure rate and were more sensitive
to stress ratio [31]. Considering this comprehensively, this paper chose the semi-circular
bending fatigue test.
The semi-circular bending fatigue test evolved from the semi-circular bending test.
The test uses a semi-circular specimen, and the load is applied by a round rod in the middle
of its top. At the same time, there are two supporting round rods at the bottom of the
semi-circular specimen, and the distance between the supporting points is determined to
be 0.8 times the diameter of the specimen, as shown in Figure 1. A cylindrical specimen
of ϕ100 × 63.5 was made by a rotary compaction method for the asphalt mixture fatigue
test, which was directly cut into two parts along the diameter and used as a5semi-circular
Appl. Sci. 2021, 11, x FOR PEER REVIEW of 14
bending fatigue test specimen.

(a) (b)

Figure Implementation
1. Implementation
Figure 1. of of
SCBSCB testing:
testing: (a) preparation
(a) preparation of specimen;
of specimen; (b)for
(b) setup setup
SCB for
test.SCB test.

4.2. Experimental Scheme Design

The UTM-30 universal testing machine, equipped with a temperature control room,
was used for the SCB strength and fatigue test at 15 °C. The fatigue test conducted was a
stress-controlled loading method with four different stress ratios of 0.2, 0.3, 0.4, and 0.5.
Appl. Sci. 2021, 11, 11867 5 of 14

4.2. Experimental Scheme Design

The UTM-30 universal testing machine, equipped with a temperature control room,
was used for the SCB strength and fatigue test at 15 ◦ C. The fatigue test conducted was a
stress-controlled loading method with four different stress ratios of 0.2, 0.3, 0.4, and 0.5.
The loading frequencies of the fatigue test were 10 Hz and 15 Hz, respectively. The fatigue
test scheme of the SMCSBS composite modified asphalt mixtures is shown in Table 3.

Table 3. Fatigue test scheme of SMCSBS composite modified asphalt mixtures.

Parameter Levels
SMC content 8%, 10%, 12%
Asphalt aggregate ratio 4.3%, 4.8%, 5.3%
Stress ratio 0.2, 0.3, 0.4, 0.5
Loading frequency 10 Hz, 15 Hz
Test temperature 15 ◦ C

In order to determine the control amplitude of the semi-sine wave load, the crack
resistance of the semi-circular bending test was tested, referring to research results, such as
Molenaar [32]. The calculation formula is shown as Equation (1).

4.8P
σm = (1)
dt
where, σm is at the bottom of the specimen tensile strength value, MPa; P is the specimen of
peak load, N; d is the thickness of the specimen, mm; and t is the specimen diameter, mm.
In the stress control mode, the relationship between the fatigue life and control stress
in the fatigue test is the fatigue equation. Referring to the research results of Liao et al. [33]
on the anti-fatigue performance of asphalt mixtures, the double logarithmic regression
equation can generally be used as the fatigue equation of the mixture, such as Equation (2).

lgNf = k − nlgσ (2)

where, Nf is the fatigue life; σ is the control stress, MPa; and k, n is the regression coefficient
of fatigue equation.

4.3. Test Results

4.3.1. Analysis of the Influence of SMC Content on Fatigue Performance
For the sake of comparison, the loading frequency 10 Hz, fatigue life of different SMC
content, and different asphalt aggregate ratio were selected, as shown in Figure 2. It can
be seen from Figure 2a that under the same stress ratio conditions, the fatigue life of the
mixture with a 10% SMCSBS asphalt aggregate ratio of 4.3 was the largest, followed by
the fatigue life of the mixture with an 8% SMCSBS asphalt aggregate ratio of 4.8, which
is consistent with Sun et al.’s research on the properties of SMCSBS composite modified
asphalts [28]. As shown in Figure 2b, under the same stress amplitude, it is obvious that
the mixture with the 10% SMCSBS asphalt aggregate ratio of 4.3 had the maximum fatigue
life, the mixture with the 12% asphalt aggregate ratio of 4.3 had the minimum fatigue
life, and the others were close. The results show that the effect of SMC content on fatigue
performance is that the fatigue life of the mixture was the largest when the SMC content
was 10%.
 which is consistent with Sun et al.’s research on the properties of SMCSBS composite
modified asphalts [28]. As shown in Figure 2b, under the same stress amplitude, it is ob-
vious that the mixture with the 10% SMCSBS asphalt aggregate ratio of 4.3 had the
maximum fatigue life, the mixture with the 12% asphalt aggregate ratio of 4.3 had the
Appl. Sci. 2021, 11, 11867
minimum fatigue life, and the others were close. The results show that the effect of 6SMC
of 14
content on fatigue performance is that the fatigue life of the mixture was the largest
when the SMC content was 10%.

14000 14000
8%, asphalt stone ratio4.3 8%, asphalt stone ratio4.3
12000 8%, asphalt stone ratio4.8 12000 8%, asphalt stone ratio4.8
8%, asphalt stone ratio5.3 8%, asphalt stone ratio5.3
10000 10000 10%, asphalt stone ratio4.3
10%, asphalt stone ratio4.3

The fatigue life

The fatigue life

10%, asphalt stone ratio4.8

10%, asphalt stone ratio4.8 8000
8000 10%, asphalt stone ratio5.3
10%, asphalt stone ratio5.3
6000 12%, asphalt stone ratio4.3
6000 12%, asphalt stone ratio4.3
12%, asphalt stone ratio4.8
12%, asphalt stone ratio4.8
4000
4000 12%, asphalt stone ratio5.3
12%, asphalt stone ratio5.3
2000
2000
0
0
0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
0.2 0.3 0.4 0.5 0.6
stress amplitude(Mpa)
stress ratio
(a) (b)
Figure
Figure 2.
2. Values
Values of
of different SMC content
different SMC content in
in the
the SCB
SCB fatigue
fatigue test:
test: (a)
(a) stress
stress ratio;
ratio; (b)
(b) stress
stress amplitude.
amplitude.

4.3.2. Analysis
Analysis of
of the
the Influence
Influence of Loading Frequencies on Fatigue Performance
From the influence of different SMC contents on fatigue performance, it can be seen
that the fatigue life of the mixture of 10% SMCSBS was the largest. Therefore, only under
performance of
this condition, the fatigue performance of different
different asphalt aggregate ratios and different
loading frequencies
loading frequencieswas
wasanalyzed,
analyzed,asasshown
shown inin Figure
Figure 3. can
3. It It can be seen
be seen fromfrom Figure
Figure 3 that3
thatmatter
no no matter
underunder the same
the same stressstress
ratioratio condition
condition or under
or under the same
the same stressstress amplitude
amplitude con-
condition,
dition,
Appl. Sci. 2021, 11, x FOR PEER REVIEW the greater
the greater the loading
the loading frequency,
frequency, the greater
the greater the fatigue
the fatigue life. life. The fatigue
The fatigue life life
was
7 of 14
was the largest when the asphalt aggregate ratio was 4.3 and the loading
the largest when the asphalt aggregate ratio was 4.3 and the loading frequency was 15 frequency was
15 Hz.
Hz.

30000 asphalt stone ratio4.3，10Hz 30000 asphalt stone ratio4.3，10Hz

asphalt stone ratio4.3，15Hz asphalt stone ratio4.3，15Hz
25000 25000 asphalt stone ratio4.8，10Hz
asphalt stone ratio4.8，10Hz
The fatigue life
The fatigue life

20000 asphalt stone ratio4.8，15Hz 20000 asphalt stone ratio4.8，15Hz

asphalt stone ratio5.3，10Hz asphalt stone ratio5.3，10Hz
15000 asphalt stone ratio5.3，15Hz 15000 asphalt stone ratio5.3，15Hz

10000 10000

5000 5000

0 0
0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.6 0.8 1.0 1.2 1.4 1.6 1.8

stress ratio stress amplitude(Mpa)

(a) (b)
Figure
Figure3.3.Values
Valuesof
ofdifferent
different loading
loading frequencies
frequencies in
in the
the SCB
SCB fatigue
fatigue test:
test: (a)
(a) stress
stress ratio;
ratio; (b)
(b) stress
stress amplitude.
amplitude.

4.3.3.
4.3.3.Fatigue
FatigueEquation
Equation
Under
Underdifferent
differentloading
loadingfrequencies
frequencies(10(10Hz,
Hz,15
15 Hz),
Hz), aa mixture
mixture withwith anan SMC
SMC content
content
of10%
of 10%and
and asphalt
asphalt aggregate
aggregate ratio
ratio of
of 4.3
4.3 was
was taken
taken toto establish
establish thethe logarithmic
logarithmic equation
equation
offatigue
of fatiguelife
lifeand
andstress
stressratio/stress
ratio/stress amplitude,
amplitude, as as shown
shown in Figure
in Figure 4. The
4. The correlation
correlation co-
coefficients
efficients of of
allall fitting
fitting equationswere
equations were>0.96.
>0.96.The
Thefatigue
fatiguelife
lifeofofthe
thespecimens
specimens decreased
decreased
with the
with the increase of of stress
stressratio/stress
ratio/stress amplitude.
amplitude.At Atthethesame
sametime,time,the more
the more SMCSMCcontent
con-
was was
tent not better; when
not better; the content
when of SMC
the content was more
of SMC was morethan 10%,
than the10%,fatigue life was
the fatigue lifelower
was
with more
lower with SMC
more content.
SMC content.

0.3
10%-4.3-10
-0.3 10%-4.3-10
10%-4.3-15
0.2 10%-4.3-15
10%-4.3-10 The fatigue life
-0.4 10%-4.3-10 The fatigue life
10%-4.3-15 The fatigue life
0.1 10%-4.3-15 The fatigue life
logσ

logσa

-0.5 y=1.00239-0.41175x
y=1.5248-0.41175x
R2=0.98962 0.0
y=0.98141-0.37033x R2=0.98962
-0.6 y=1.50382-0.37033x
R2=0.96899 -0.1
R2=0.96899
 Under different loading frequencies (10 Hz, 15 Hz), a mixture with an SMC content
of 10% and asphalt aggregate ratio of 4.3 was taken to establish the logarithmic equation
of fatigue life and stress ratio/stress amplitude, as shown in Figure 4. The correlation co-
efficients of all fitting equations were >0.96. The fatigue life of the specimens decreased
with the increase of stress ratio/stress amplitude. At the same time, the more SMC con-
Appl. Sci. 2021, 11, 11867 7 of 14
tent was not better; when the content of SMC was more than 10%, the fatigue life was
lower with more SMC content.

0.3
10%-4.3-10
-0.3 10%-4.3-10
10%-4.3-15
0.2 10%-4.3-15
10%-4.3-10 The fatigue life
-0.4 10%-4.3-10 The fatigue life
10%-4.3-15 The fatigue life
0.1 10%-4.3-15 The fatigue life
logσ

logσa
-0.5 y=1.00239-0.41175x
y=1.5248-0.41175x
R2=0.98962 0.0
y=0.98141-0.37033x R2=0.98962
-0.6 y=1.50382-0.37033x
R2=0.96899 -0.1
R2=0.96899
-0.7
-0.2
3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8
3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8
lgNf lgNf

0.3
10%-4.3-10 10%-4.3-10
-0.3
10%-4.3-15 0.2 10%-4.3-15
---- 10%-4.3-10 The fatigue life ---- 10%-4.3-10 The fatigue life
-0.4 ---- 10%-4.3-15 The fatigue life
---- 10%-4.3-15 The fatigue life 0.1

logσa
logσ

-0.5 y=1.00239-0.41175x y=1.5248-0.41175x

0.0
R2=0.98962 R2=0.98962
y=0.98141-0.37033x y=1.50382-0.37033x
-0.6 -0.1
R2=0.96899 R2=0.96899
-0.7 -0.2

3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8

lgNf lgNf

(a) (b)
Figure
Figure 4. 4. Comparisonofof
Comparison fatigueequations
fatigue equationsofof10%
10%SMC
SMCand
andasphalt
asphaltaggregate
aggregateratio
ratio4.3
4.3inin the
the SCB
SCB fatigue
fatigue test:(a)(a)stress
test: stress
ratio; (b) stress amplitude.
ratio; (b) stress amplitude.

5. Predictive Simulation and Analysis

5.1. Particle Swarm Optimization (PSO)
Particle swarm optimization (PSO), introduced by Eberhart and Kennedy [34,35], is
a group intelligence algorithm that has been successful in both continuous and discrete
optimization. The feasible solution of each optimization problem is abstracted into a
“particle” without volume and mass in the search space. The PSO algorithm generates
the initial population and the population of each particle motion in the feasible solution
space, decides its own speed position and orientation of each particle movement following
the current optimum particles, and obtains an optimal solution. In the PSO algorithm,
if it selects particles as the random initial population, the position of each particle in the
dimensional space can be expressed as the solution of the optimization problem; the degree
of its performance depends on the fitness value determined by the objective function of
the optimization problem and, through constant iteration, adjusts its direction change and
speed size, finally finding the global optimal solution. The velocity of the particle is as
follows [36]:

vi (t + 1) = w·vi (t) + c1 r1 (pbesti (t) + xi (t)) + c2 r2 (gbest(t) + xi (t)), (3)

xi (t + 1) = xi (t) + vi (t + 1), (4)

In Equations (3) and (4), i = 1, 2, · · · , m; j = 1, 2, · · · , n; t is the number of iterations; c1 ,
c2 is the learning factor, which measures the individual learning ability of pbesti (t) and the
global learning ability of gbest(t), respectively. r1 , r2 is the random number transformed
within the range of [0,1], xi (t) is the position information of the particle i at time t, vi (t) is
the velocity information of the particle i at time t, pbesti (t) is the current optimal position
of the particle i (namely, the optimal solution), gbest(t) is the best location found by the
whole population (namely, the global optimal solution), and w is the inertia factor, which
Appl. Sci. 2021, 11, 11867 8 of 14

is calculated according to Equation (5) and used to control the influence of the last velocity
on the current state.
Haxgen − Ggeneration
w = (w1 − w2 )· (5)
Haxgen
where, w1 , w2 is the weight factor; Haxgen is the maximum number of iterations; and
Ggeneration is the number of current iterations.
Equations (3) and (4) are used to update the velocity and position of each particle. In
order to limit the particle velocity and reduce some poor results, the contraction coefficient
is applied to the PSO algorithm [37]. In the PSO algorithm with the contraction coefficient,
the update velocity is changed as follows:


vi 0 (t + 1) = x vi (t) + c1 r1 pbesti (t) + xi (t) + c2 r2 gbest(t) + xi (t) ,


(6)

where, the contraction factor x is as follows:

2
x =  p  , φ = c1 + c2 > 4 (7)
2 − φ − φ2 − 4φ


5.2. Support Vector Machine (SVM)

A support vector machine (SVM) is a machine learning method based on statistical
theory. It is a classification regression method [38]. The principle of optimal classification is
a statistical concept for the classification methods of support vector machines. Through
Vapnik [38], the value of the smallest generalization error in the separable class is selected.
It comes from the principle of structural risk minimization. The different classes separate
a plane so that the boundaries between them are at maximum. The boundary is the sum
of the distances from the partition plane to the vertical. For two inseparable classes, a
hyperplane with a maximum edge value and minimum misclassification error is identified.
A predefined constant controls the margin and the size of the error. The design method of
the support vector machine extends to the nonlinear decision surface, and the data projects
into a feature space to generate a classification problem.

5.2.1. SVM Basic Principles

The SVM basic model is defined as a linear classifier, with the largest interval on the
feature space. For a given training sample set (xi , yi ), the interval maximization learning
strategy can be used to solve a convex quadratic programming problem. Then, a Lagrange
function is introduced to solve the dual problem, and finally, the optimal classification
decision function is obtained. The specific process is as follows:
Given the training sample set (xi , yi ), i = 1, . . . , l; x ∈ Rn , y ∈ {+1, −1}, the hyper-
plane is denoted as (w · x) + b = 0. In order for the classification to be correctly classified
and have the classification interval for all samples, it is required to meet the following
constraints:
yi [(w · xi ) + b] ≥ 1 i = 1, 2, · · · , l, (8)
2
The classification interval is calculated as kw k
, so the problem of constructing an
optimal hyperplane is transformed into an optimization problem under constraints:

1 1 T 
minφ(w) = kwk2 = w ·w , (9)
w 2 2
The above problems are transformed into quadratic programming problems under
constraint conditions:
min 12 wT · w + C ∑ni=1 ξi ,
w
s.t. yi wT xi + b ≥ 1 − ξi ,


(10)
ξi ≥ 0, i = 1, 2, · · · , n
Appl. Sci. 2021, 11, 11867 9 of 14

where, ξi is the non-negative relaxation variable introduced to measure the classification

error, w is the normal vector of the hyperplane, and C is the penalty factor of the error
term ∑ni ξi .
To solve the constrained optimization problems, a Lagrange function is introduced:

1 T n n n
L(w, b, λ) = w · w + C ∑i=1 ξi − ∑i=1 λi [yi (wT xi + b) − 1 + ξi ] − ∑i=1 αi ξi , (11)
2
where λi ≥ 0, i = 1, 2, · · · , n. The dual problem of the Equation (10) is:

maxQ(λ) = ∑ni=1 λi − 1
∑ni,j=1 yi yj λi λj xi ·xj ,


2
n (12)
s.t. ∑ yi λi = 0, 0 ≤ λi ≤ C
i=1

Finally, the optimal classification decision function is:

f(x, λ) = sgn ∑SV yi λi xi ·xj + b ,

(13)

For the linear indivisible case, the dual problem is:



maxQ(λ) = ∑ni=1 λi − 21 ∑ni,j=1 yi yj λi λj φ(xi )T · φ xj ,
n (14)
s.t. ∑ yi λi = 0, 0 ≤ λi ≤ C
i=1

where, φ(xi )T · φ xj = K xi , xj is the kernel function.

The corresponding optimal classification decision function can be obtained as follows:

f(x, λ) = sgn(∑SV yi λi K xi , xj + b),

(15)

In fact, the whole process is to transform a linear indivisible problem into a simpler
high-dimensional linear separable problem, which depends on the kernel function of SVM,
and is defined as: K xi , xj = φ(xi )T · φ xj .

5.2.2. SVM Parameter Optimization and Kernel Function

For a given sample, SVM penalty factor C, selection of the kernel function and the
determination of kernel parameter δ are key issues affecting SVM learning ability and
generalization ability. The purpose of introducing penalty factor C into SVM is to deal
with fault tolerance at fault points near hyperplane boundaries. The penalty factor of
an appropriate value can improve the system’s robust fault tolerance. On the contrary,
overfitting or underfitting will result in classification problems.
The kernel function is very important to SVM. It makes raw data linearly separable and
avoids the difficulty of a dimension disaster. There are many types of SVM kernel functions.
The commonly used kernel functions include the linear kernel function, the polynomial
kernel function, and the radial basis function (RBF) kernel function, respectively:

K xi , xj = xi T xj ,


(16)

 δ
K xi , xj = xi T xj + 1 , (17)

kxi − xj k2
!


K xi , xj = exp − (18)
2δ2
For a linear kernel, the form of a function is simple. Therefore, it is more suitable
for dealing with the linear separable problems. For the polynomial kernel function, its
function form is more complex, and requires extensive calculation time. For the RBF
Appl. Sci. 2021, 11, 11867 10 of 14

kernel function, it can realize nonlinear mapping, the required number of the RBF kernel
function parameters is smaller, and the numerical difficulty is also small. A large number
of experimental results show that RBF kernel functions have better performance than other
commonly used kernel functions.

5.3. Simulation of the PSO-SVM

5.3.1. Main Steps
Follows the following main steps (as shown in Figure 5):
Step 1: Load the obtained data set and divide it into two parts. The first part includes
70% of the data for training and building AI “black boxes,” with the remaining 30% for
model validation. Input parameters are SMC content (/%), asphalt aggregate ratio, stress
ratio, and loading frequency (/Hz). The output of an AI numerical tool is the fatigue life.
Step 2: Build the model using the training data set. Use particle swarm optimization
to optimize the SVM penalty factor C and kernel function parameter δ to generate the
optimal PSO-SVM prediction model.
Step 3: Validate the model with test data. Various standards, namely, R2 and RMSE,
Appl. Sci. 2021, 11, x FOR PEER REVIEW
are used to validate four developed models in training and testing data sets. 11 of 14
Step 4: Predict fatigue life, and predict results.

THE Predict fatigue Build the Best Verify the

END life model data(30%)

Optimization (PSO)
Trained (SVM)

Initial population

Initiate position and velocities

Update positions and velocities

Training (SVM)
Find the best position (gbest)

SCB experi-
Build the training data(70%) ments

BEGIN

Figure 5. The methodological chart of the present study.

study.

5.3.2. Model Performance Evaluation

In this
thisstudy,
study,subsequent
subsequentstatistical indicators
statistical were were
indicators specified to evaluate
specified the performance
to evaluate the per-
0
of M5 , ANN, SVM, and the PSO-SVM. 2 were
formance of M5′, ANN, SVM, and the The Root Mean
PSO-SVM. The Square ErrorSquare
Root Mean (RMSE)Error
and R (RMSE)
usedRaswere
and 2 fitness functions
used to determine
as fitness functionsthe applicability
to determine of applicability
the each factor. Equations (19) and
of each factor. (20)
Equa-
respectively
tions (19) andrepresent RSME and
(20) respectively R2 equations:
represent RSME and R2 equations:
(1) RMSE

∑
MSE = , (19)
Appl. Sci. 2021, 11, 11867 11 of 14

(1) RMSE r
∑ni=1 Pi1 − Pi2
MSE = , (19)
n
(2) R2
∑ni=1 Pi2 − Pi2 Pi1 − Pi1



2
R = n (20)
∑i=1 Pi2 − Pi2 ∑ni=1 Pi1 − Pi1

where, Pi2 and Pi1 are the fatigue life value predicted by the model and the measured value,
respectively, and n is the total number of test data.

5.3.3. Performance Analysis

There were 72 samples in this study—70% of the training set and 30% of the test set.
In order to verify the advantages of the proposed PSO-SVM method, its performance
Appl. Sci. 2021, 11, x FOR PEER REVIEW 12 of 14
was compared with that of other commonly used intelligent computing-based methods,
0
such as M5 , ANN, and SVM. The predicted fatigue lives are represented in Figure 6 in
the form of scatter plots. Table 4 summarizes performance indices of different methods to
conclusion that thelife.
predict the fatigue PSO-SVM outperformed
The comparison other
presented inmethods in terms
Table 4 leads to theofconclusion
accuracy that
of pre-
the
diction
PSO-SVM of the output.
outperformed other methods in terms of accuracy of prediction of the output.

25000 PredictedData 25000 PredictedData

Fit line Fit line
20000 —— Act.=Pred. 20000 —— Act.=Pred.
Predicted value
Predicted value

15000 15000

10000
10000

5000
5000

0
0
0 10000 20000 0 10000 20000

Actual value Actual value

(a) R2 = 0.779 (b) R2 = 0.814
PredictedData 25000 PredictedData
25000 Fit line Fit line
—— Act.=Pred.
20000 —— Act.=Pred.
Predicted value

Predicted value

20000

15000
15000

10000
10000

5000
5000

0
0 10000 20000 0 10000 20000

Actual value Actual value

(c) R2 = 0.861 (d) R2 = 0.903
Figure
Figure 6.
6. Fatigue
Fatigue life
life prediction
prediction by
by (a) M50 , (b)
(a) M5′, (b) ANN,
ANN, (c)
(c) SVM,
SVM, and
and (d)
(d) PSO-SVM.
PSO-SVM.

Table 4. Summary of prediction capability for testing parts using M5′, ANN, SVM, and the
PSO-SVM.

Method RMSE (mm/day) R2

M5′ 4035.33 0.779
ANN 3988.36 0.814
Appl. Sci. 2021, 11, 11867 12 of 14

Table 4. Summary of prediction capability for testing parts using M50 , ANN, SVM, and the PSO-SVM.

Method RMSE (mm/day) R2

M50 4035.33 0.779
ANN 3988.36 0.814
SVM 3563.26 0.861
PSO-SVM 2253.71 0.903

6. Conclusions
The following conclusions can be drawn based on this investigation:
(1) Under the same load frequency (10 Hz), the fatigue life of different SMC contents is
shown as follows: under the same stress ratio/stress amplitude, the fatigue life of a
mixture reaches the maximum when the effect of SMC content on fatigue performance
is 10%. This also shows that the more SMC content is not better; when the content of
SMC is more than 10%, the fatigue life is lower with more SMC content.
(2) Taking the fatigue life of the mixture of 10% SMCSBS as an example, the fatigue
performance of different asphalt aggregate ratios and different loading frequencies is
shown as follows: whether under the same stress ratio/stress amplitude, the greater
the loading frequency, the greater the fatigue life. The fatigue life is the largest when
the asphalt aggregate ratio is 4.3 and the loading frequency is 15 Hz.
(3) Four artificial intelligence models M50 , ANN, SVM, and PSO-SVM were used to
predict the fatigue life of an SMCSBS asphalt mixture, with input parameters (SMC
content/%, asphalt aggregate ratio, stress ratio, and load frequency/Hz) and out-
put parameters (fatigue life). The model was validated using standards, such as
RMSE and R2 . The results show that M50 (RMSE = 4035.33 and R2 = 0.779), ANN
(RMSE = 3988.36 and R2 = 0.814), and SVM (RMSE = 3563.26 and R2 = 0.861), but the
PSO-SVM (RMSE = 2253.71 and R2 = 0.903) with particle swarm optimization showed
the most stable algorithm compared with other methods. Particle swarm optimization
(PSO) support vector machine (SVM) ha good performance in predicting the fatigue
life of an SMCSBS asphalt mixture.

Author Contributions: Conceptualization, Y.S. and D.H.; methodology, D.H.; software, Y.S.; vali-
dation, Y.S. and J.L.; investigation, J.L.; writing—original draft preparation, Y.S.; writing—review
and editing, Y.S., D.H. and J.L. All authors have read and agreed to the published version of the
manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data sharing is not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Li, K.; Huang, M.; Zhong, H.; Li, B. Comprehensive evaluation of fatigue performance of modified asphalt mixtures in different
fatigue tests. Appl. Sci. 2019, 9, 1850. [CrossRef]
2. Shafabakhsh, G.; Mirabdolazimi, S.M.; Sadeghnejad, M. Evaluation the effect of nano-TiO2 on the rutting and fatigue behavior of
asphalt mixtures. Constr. Build. Mater. 2014, 54, 566–571. [CrossRef]
3. Ma, H.Y.; Chen, H.X.; Yang, P.W.; Xing, M.L.; Wang, K.; Kuang, D.L. Prediction of fatigue life of asphalt slurry based on improved
S-VECD model. J. Chang. Univ. Nat. Sci. Ed. 2019, 4, 35–43.
4. Ren, R.B.; Geng, L.T.; Chen, M.M.; Yan, F.L. Study on fatigue performance test and prediction model of rubber—Plastic asphalt
mixture. J. Shandong Jianzhu Univ. 2016, 31, 583–587.
5. Xie, C.L.; Zhang, Y.; Geng, H.B. Prediction model of asphalt mixture fatigue performance based on BP neural network.
J. Chongqing Jiaotong Univ. (Nat. Sci. Ed.) 2018, 02, 35–40.
Appl. Sci. 2021, 11, 11867 13 of 14

6. Zhu, H.Z.; Fan, S.P.; Lu, Z.T.; Yan, H. Analysis of fatigue life prediction model of aged asphalt based on dissipated energy. Road
Traffic Technol. 2019, 36, 8–13, 30.
7. Hveem, F.N. Pavement Deflections and Fatigue Failures; Highway Research Board Bulletin 114; Transportation Research Board:
Washington, DC, USA, 1955.
8. Santucci, L.E.; Schmidt, R.J. The Effect of Asphalt Properties on the Fatigue Cracking of Asphalt Concrete on the Zaca-Wigmore
Test Project. Proc. Assoc. Asphalt Paving Technol. 1969, 38, 39–64.
9. Ge, Z.S. Grey correlation analysis of influencing factors on fatigue performance of asphalt mixture. J. Transp. Eng. 2002, 2, 8–11.
10. Zhu, H.Z. Analysis of key Influencing Factors of asphalt mixture fatigue performance. J. Southeast Univ. (Nat. Sci. Ed.) 2004, 34,
260–263.
11. Yu, J.M.; Zhang, X.N. Evaluation of environmental factors influencing the fatigue performance of asphalt mixtures. J. Southeast
Univ. 2005, 21, 486–489.
12. Tan, Y.R.; Zhang, H.Y.; Cao, D.W.; Xia, L.; Du, R.J.; Shi, Z.Q.; Dong, R.; Wang, X.H. Study on cohesion and adhesion of
high-viscosity modified asphalt. Intern. J. Trans. Sci. Technol. 2019, 8, 394–402.
13. Jia, X.; Zhang, Z.; Guo, C.; Gao, J. Mixture design of SMC recycled asphalt mixture at normal temperature. In Proceedings of the
10th Annual Conference of Maintenance and Management Branch of China Highway Society, Shenzhen, China, 9 January 2020;
pp. 63–70.
14. Wei, W.; Jin, Y. Road performance analysis of waste plastics +SMC compound modified SMA10 asphalt mixture. Highw. Traffic
Technol. 2019, 35, 43–46, 60.
15. Qu, H.W. Application of SMC Modified Asphalt at Room Temperature in Highway Overlay Engineering. Shanxi Constr. 2016, 42,
166–167.
16. Wang, J.J. Road performance analysis of AC-13 asphalt mixture at room temperature. Urban. Roads Br. Flood Control 2019, 8,
211–213.
17. Shu, P.; Wang, B.; Zhang, Y.; Wang, S. Properties of high RAP SMC recycled asphalt mixture at norma temperature in Panzhihua
area. Transp. Energy Sav. Environ. Prot. 2020, 80, 83–87.
18. Jones, D.; Harvey, J.T.; Monismith, C.L. Reflective Cracking Study: Summary Report; Working Paper Series; Institute of Transportation
Studies: California, CA, USA, 2008.
19. Zhang, B.; Kawashita, L.F.; Hallett, S.R. Composites fatigue delamination prediction using double load envelopes and twin
cohesive models—Sciencedirect. Compos. Part A Appl. Sci. Manuf. 2020, 129, 105711. [CrossRef]
20. Park, H.I.; Kweon, G.C.; Lee, S.R. Prediction of resilient modulus of granular subgrade soils and subbase materials using artificial
neural network. Road Mater. Pavement Des. 2009, 10, 647–665. [CrossRef]
21. Yan, W.J.; Chronopoulos, D.; Papadimitriou, C.; Cantero-Chinchilla, S.; Zhu, G.S. Bayesian inference for damage identification
based on analytical probabilistic model of scattering coefficient estimators and ultrafast wave scattering simulation scheme.
J. Sound Vib. 2020, 468, 115083. [CrossRef]
22. Tan, Y.; Song, X.; Zhou, C.; Gong, W. Research on the application of neural network in the model for solid volume parameters
prediction of asphalt mixture. Highw. Eng. 2008, 60–64. [CrossRef]
23. Vapnik, V. Statistical Learning Theory. Ann. Inst. Stat. Math. 2003, 55, 371–389.
24. Pal, M. Support vector machines-based modelling of seismic liquefaction potential. Int. J. Numer. Anal. Methods Geomech. 2006, 30,
983–996. [CrossRef]
25. Maalouf, M.; Khoury, N.; Trafalis, T.B. Support vector regression to predict asphalt mix performance. Int. J. Numer. Anal. Methods
Geomech. 2008, 32, 1989–1996. [CrossRef]
26. Soltani, M.; Moghaddam, T.B.; Karim, M.R.; Shamshirband, S.; Sudheer, C. Stiffness performance of polyethylene terephthalate
modified asphalt mixtures estimation using support vector machine-firefly algorithm. Measurement 2015, 63, 232–239. [CrossRef]
27. DBJ 64/T060-2016, Technical Specification for Construction of SMC Modified Asphalt Pavement; Standards of Ningxia Hui Autonomous
Region: Yinchuan, China, 2016.
28. Sun, Y.; He, D. High and Low-Temperature Performance Evaluation and Microanalysis of SMCSBS Compound-Modified Asphalt.
Materials 2021, 14, 771. [CrossRef] [PubMed]
29. JTG F40-2004, Technical Specifications for Construction of Highway Asphalt; Pavement, Renmin Communication Press: Beijing, China,
2004.
30. Hasan, M.M.; Ahmad, M.; Hasan, M.A.; Faisal, H.M.; Tarefder, R.A. Laboratory performance evaluation of fine and coarse-graded
asphalt concrete mix. J. Mater. Civ. Eng. 2019, 31, 04019259. [CrossRef]
31. Dong, Q.; Zhao, X.; Chen, X.; Ma, X.; Cui, X. Long-term mechanical properties of in situ semi-rigid base materials. Road Mater.
Pavement Des. 2020, 22, 1692–1707. [CrossRef]
32. Molenaar, A.A.A.; Scarpas, A.; Liu, X.; Erkens, S.M.J.G. Semi-circular bending test, simple but useful. Asph. Paving Technol. Assoc.
Asph. Paving Technol.-Proc. Tech. Sess. 2002, 71, 794–815.
33. Liao, M.C.; Chen, J.S.; Tsou, K.W. Fatigue Characteristics of Bitumen-Filler Mastics and Asphalt Mixtures. J. Mater. Civ. Eng. 2012,
24, 916–923. [CrossRef]
34. Kennedy, J.; Eberhart, R. Particle swarm optimization. In Proceedings of the ICNN’95-International Conference on Neural
Networks, Perth, Australia, 27 November–1 December 1995; Volume 4, pp. 1942–1948.
Appl. Sci. 2021, 11, 11867 14 of 14

35. Shi, Y.; Eberhart, R.C. Empirical study of particle swarm optimization. In Proceedings of the 1999 Congress on Evolutionary
Computation-CEC99 (Cat. No. 99TH8406), Washington, DC, USA, 6–9 July 1999; Volume 3, pp. 1945–1950.
36. Liu, Q.J.; Jing, L.H.; Wang, L.M.; Lin, Q.Z. A Method of Particle Swarm Optimized SVM Hyper-spectral Remote Sensing Image
Classification. IOP Conf. Ser. Earth Environ. Sci. 2014, 17, 012205. [CrossRef]
37. Clerc, M.; Kennedy, J. The Particle Swarm: Explosion, Stability and Convergence in Multi-dimensional Complex space. IEEE
Trans. Evol. Comput. 2002, 20, 1671–1676.
38. Vapnik, V. The Nature of Statistical Learning Theory, 2nd ed.; Springer: New York, NY, USA, 1999; ISBN 978-0-387-98780-4.