Hindawi

Advances in Materials Science and Engineering

Volume 2024, Article ID 9808340, 1 page
https://doi.org/10.1155/2024/9808340

Retraction
Retracted: Research on the Injection Mold Design and Molding
Process Parameter Optimization of a Car Door Inner Panel

Advances in Materials Science and Engineering

Received 8 January 2024; Accepted 8 January 2024; Published 9 January 2024

Copyright © 2024 Advances in Materials Science and Engineering. Tis is an open access article distributed under the Creative
Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the
original work is properly cited.

Tis article has been retracted by Hindawi following an Te corresponding author, as the representative of all
investigation undertaken by the publisher [1]. Tis in- authors, has been given the opportunity to register their
vestigation has uncovered evidence of one or more of the agreement or disagreement to this retraction. We have kept
following indicators of systematic manipulation of the a record of any response received.
publication process:
(1) Discrepancies in scope
References
(2) Discrepancies in the description of the research [1] K. Yang, Y. Wang, and G. Wang, “Research on the Injection
reported Mold Design and Molding Process Parameter Optimization of
a Car Door Inner Panel,” Advances in Materials Science and
(3) Discrepancies between the availability of data and Engineering, vol. 2022, Article ID 7280643, 18 pages, 2022.
the research described
(4) Inappropriate citations
(5) Incoherent, meaningless and/or irrelevant content
included in the article
(6) Manipulated or compromised peer review
Te presence of these indicators undermines our con-
fdence in the integrity of the article’s content and we cannot,
therefore, vouch for its reliability. Please note that this notice
is intended solely to alert readers that the content of this
article is unreliable. We have not investigated whether au-
thors were aware of or involved in the systematic manip-
ulation of the publication process.
Wiley and Hindawi regrets that the usual quality checks
did not identify these issues before publication and have
since put additional measures in place to safeguard research
integrity.
We wish to credit our own Research Integrity and Re-
search Publishing teams and anonymous and named ex-
ternal researchers and research integrity experts for
contributing to this investigation.
Hindawi
Advances in Materials Science and Engineering
Volume 2022, Article ID 7280643, 18 pages
https://doi.org/10.1155/2022/7280643

Research Article
Research on the Injection Mold Design and Molding Process
Parameter Optimization of a Car Door Inner Panel

E D
Kefan Yang , Youmin Wang, and Guoqing Wang

Correspondence should be addressed to Kefan Yang; 2200120138@stu.ahpu.edu.cn

C T
College of Mechanical Engineering, Anhui Polytechnic University, Wuhu, Anhui 241000, China

Received 24 April 2022; Revised 6 July 2022; Accepted 26 July 2022; Published 31 August 2022

Academic Editor: Achraf Ghorbal

A
Copyright © 2022 Kefan Yang et al. Tis is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Numerical simulation of the injection molding of the inner panel of the car door and the optimization of its process parameters
were completed. Te inner panel and its injection mold were designed by UG software and simulated by the flling and cooling

R
module of the Moldfow software. Diferent gate schemes were selected to compare quality indicators such as flling time, air
pockets, and weld lines to obtain the optimal number of gates. We established the objective function model with gate position as
the independent variable, and used a multi-population genetic algorithm to solve the optimal position of the function model to get
the best gate position. Te mold temperature, melt temperature, cooling time, holding pressure, and holding time were selected as
the infuencing factors, and volume shrinkage and warpage deformation were selected as the evaluation indicators to design and

T
complete the orthogonal test. Te test data were simulated by Moldfow, and the optimal combination of process parameters was
determined by range and variance analysis. Te BP neural network model related to the molding process parameters, volume
shrinkage, and warpage deformation was built, and the trained network model was optimized with the ant colony algorithm. Te
optimal parameter combination was: mold temperature 76°C, melt temperature 205°C, cooling time 23.8s, holding pressure
54.7Mpa, and holding time 22.1s. Te simulation results showed that volume shrinkage was 13.32% and warpage deformation was

E
4.315 mm. Te design of an injection mold for the car door inner panel was completed and its molding process parameters
were optimized.

R
1. Introduction
In the automobile manufacturing industry [1], replacing
traditional steel components with plastic parts has become a
signifcant trend. As an advanced processing method, in-
jection molding production is widely used because of its
advantages of short production cycle and high efciency.
However, there are also some quality defects in forming and
manufacturing, such as volume shrinkage and warpage
caused by uneven cooling, the insufcient fow of molten
plastic leading to fusion marks, cavitation, and fying edges.
Tese defects are mainly related to mold design, molding
process parameter settings, material selection, and other
factors. As a result, injection molding simulation is widely
used in the production of automotive components.
Fan [2] proposed to build a BP neural network rela-
tionship between process parameters and warpage and
optimize the BP neural network based on a genetic algorithm
to reduce warpage deformation. Christoph et al. [3] in-
vestigated the deformation of injection-molded plastic parts,
verifed the quality of plastic parts based on laminated sheet-
like parts, and found that crystalline conformal materials are
prone to more signifcant warpage and shrinkage during the
injection process. Marton Huszar et al. [4] reduced the
deformation by selecting the optimal injection material and
gate position; the research found that diferent injection
materials produce diferent warpage, in which PP and PS
produce the maximum and minimum deformation. Lu [5]
pointed out that for the warpage deformation direction of
injection-molded products, the magnitude of warpage de-
formation of stable warpage is proportional to volume
shrinkage, and non-stable warpage is due to the bending of
the product itself. Heidari Behzad shiroud et al. [6] reduced
the values of warpage and contraction to 0.287222 mm and
2 Advances in Materials Science and Engineering

E D
T
(a) (b)

Figure 1: Tree-dimensional model of the car door inner plate. (a) Front view. (b) Reverse side view.

6 7 8

C
9
5

4
10

A
793.37

11
12

R
1
13

14

T
15
962.89
Figure 2: Die assembly drawing of the car door inner plate.
Figure 4: Cavity diagram.

E
optimal parameter combination. Saad [8] proposed an
optimized injection molding process parameters under
product defect constraints. Tey used MATLAB to build a
function to solve the optimization problem and fnally

R
obtain the optimal combination of process parameters.
Ramesh et al. [9] used matrix design and parameter group
788.06

algorithms to investigate the variety combination of pa-

954.04
Figure 3: Core diagram.
13.6613% using analysis of variance through the simulation
analysis of the essential components of artifcial bone joint.
Sateesh et al. [7] took the water meter top cover as the
research object, optimized the injection molding parameters
using the grey correlation analysis method, and obtained the
rameters for the automotive instrumentation injection
molding process and verifed this using processing
experiments.
Tis study takes the car door inner panel as the research
object; the inner panel and its injection mold are designed by
UG software and simulated by the flling and cooling module
of the Moldfow software. Diferent gate schemes are selected
to compare the forming quality index and to obtain the
optimal number of gates. We established the objective
function model with gate position as the independent
variable, and used a multi-population genetic algorithm to
solve the optimal position of the function model to get the
best gate position. Te mold temperature, melt temperature,
cooling time, holding pressure, and holding time are selected
as the infuencing factors, and volume shrinkage and
warpage deformation are selected as the evaluation
Advances in Materials Science and Engineering
Gating compatibility
=1.000
Best
Worst
(a)
Gating compatibility
=1.000
Best
R A
T
Worst
(c)
E
Figure 5: Matching cloud diagram of gates. (a) Matching cloud diagram of single gate. (b) Matching cloud diagram of two gates.
(c) Matching cloud diagram of four gates.
R
indicators to design and complete the orthogonal test. Te
test data are simulated by Moldfow, and the optimal
combination of process parameters is determined by range
and variance analysis. Te BP neural network model related
to the molding process parameters, volume shrinkage, and
warpage deformation is built, and the ant colony algorithm
is used to optimize the optimal process parameters. Te
design of an injection mold for the car door inner panel is
completed and its molding process parameters are
optimized.
2. Structure and Mold Design of the Car Door
Inner Plate
Te structural design of the car door inner panel is closely
linked to its structural strength and stifness. At the be-
ginning of the parts design, the choice of material needs to be
verifed, and its thickness needs to be calculated and verifed.
Once the structural design is complete, the corresponding
3
Gating compatibility
=1.000
D
Best
Worst
T E
C
(b)
injection mold design needs to be carried out according to
the door structure.
2.1. Structural Design of the Car Door Inner Panel. Plastic
parts commonly used in automobiles are mainly modifed
plastics. PP plastic is one of the most widely used plastics, a
comprehensive performance thermoplastic. Its heat resis-
tance and rigidity strength meet the requirements of plastic
instead of steel [10]. Tis paper uses modifed polypropylene
as the injection plastic combined with the use requirements
and strength check of the car door inner plate. Te modifed
PP composite (PP-LGF-30) has the advantage of high ma-
terial stifness and impact strength [11]. It can be remelted
for remanufacturing after scrapping without wasting re-
sources, and the material properties remain good through
remanufacturing [12].
Te three-dimensional car door inner panel model is
modeled, as shown in Figure 1. Its surface area is 49698
4 Advances in Materials Science and Engineering

Filling time Filling time

=6.443 [s] =4.930 [s]

D
[s] [s]
6.443 4.930

4.832 3.697

E
3.222 2.465

1.611
1.232

T
0.0000
0.0000

(a) (b)
Filling time
=3.783 [s]

A
[s]
3.783

2.837

1.892

0.9458
C
R
0.0000

(c)

T
Figure 6: Filling time of three gate location schemes. (a) Single gate. (b) Two gates. (c) Four gates.

mm2 , the projected area is 462648 mm2 , and the thickness parts after pressure maintenance and cooling, completing a

E
is usually taken as 1.5mm–3 mm. When using modifed PP
material as the molding material of the car door inner plate,
it is necessary to ensure that the structural strength of the
door inner plate is not lower than that of the original steel
parts. Too thick plastic parts will lead to a waste of re-
product manufacturing process. Te assembly drawing of
the die for the car door inner panel is shown in Figure 2.
1- Support plate; 2- Moving die base plate; 3- Cooling
system; 4- Fixed template; 5- Fixing bolt; 6- Cavity; 7- Gate
sleeve; 8- Locating ring; 9- Fixed die base plate; 10- Guide

R
sources, and too thin plastic parts will reduce the strength
of the door inner plate and afect the safety of the regular
use of the car [11]. Terefore, the thickness of the door
inner plate is 3 mm.
2.2. Design of Injection Mold for the Car Door Inner Panel.
Te injection mold design is an essential link in the pro-
duction process of car door inner plates. Te molding
process of the door inner plate is based on the mold design,
and the design of the mold structure directly afects the
quality of the plastic parts. Te design rationality of the mold
structure is closely related to the durability, service life,
production, and processing mode of the mold itself and the
molding process of plastic parts. Te injection mold is
mainly composed of a moving mold and a fxed mold. Te
screw rod of the injection molding machine pressurizes the
molten plastic into the mold cavity and takes out the plastic
post; 11- Push rod fxing plate; 12- Push plate; 13- Moving
die base plate; 14- Reset rod; 15- Push rod.
As shown in Figures 3 and 4, combined with the
structural characteristics of the car door inner plate, the
parting surface is selected. Te axial and radial dimensions of
the design core are 788.06 mm and 954.04 mm, respectively,
and the cavity is 793.37 mm and 962.89 mm, respectively.
Trough calculation, it is obtained that the core pulling
distance of the injection mold for the car door inner plate is
190 mm, the inclination angle is 20° , the diameter is 20 mm,
the length of the inclined guide column is 555 mm, and the
minimum mold opening stroke is 522 mm.
3. Research on the Optimal Gate Scheme of
Injection Mold for the Car Door Inner Plate
Uneven flling and shrinkage deformation of plastic parts
often occur in the process of injection molding, resulting in
Advances in Materials Science and Engineering 5

Clamping force : XY garph

Clamping force : XY garph 600.0
1000.0

D
500.0
800.0
400.0

600.0

tonne
E
300.0
tonne

400.0
200.0

200.0

T
100.0

0.0000
0.0000
0.0000 5.000 10.00 15.00 20.00 25.00 30.00 35.00 40.00 0.0000 5.000 10.00 15.00 20.00 25.00 30.00 35.00
Time [s] Time [s]

C
(a) (b)
Clamping force : XY garph
400.0

350.0
300.0

A
250.0
tonne

200.0

150.0
100.0

R
50.00
0.0000
0.0000 5.000 10.00 15.00 20.00 25.00 30.00 35.00
Time [s]

T
(c)

Figure 7: Clamping force of three gate location schemes. (a) Single gate. (b) Two gates. (c) Four gates.

E
defective products and waste of resources. Tese problems Figure 6 shows that the flling time is 6.443s for a single
are due to the unreasonable design of the mold flling system gate, 4.930s for two gates, and 3.783s for four gates. Te
and cooling system. shortest flling time is achieved when the number of gates is
4.

R
3.1. Study on the Optimal Number of Gates. Before the mold
fow analysis, the location of the gates and the number of
gates need to be determined [13], and the position of the gate
will also change with the number of gates. In this paper, 1, 2,
and 4 gates are mainly selected. Te matching cloud diagram
of the three gating solutions is shown in Figure 5, where the
closer to the red area, the worse the matching; the more
immediate to the blue area, the better the matching [14].
Te injection molding flling process is simulated and
analyzed based on Moldfow software. Te results are pre-
sented in a simulation cloud diagram, mainly including
process parameters such as flling time, clamping force,
cavitation, and welding lines.
3.1.2. Clamping Force. In the process of melt fow, it will
produce external torque to the die. If the clamping force is
small, it will lead to the separation of the concave-convex die
[15]. Terefore, we need to set a larger clamping force than
the torque generated by the melt. Moldfow is used to
simulate and analyze the clamping force, and the results are
shown in Figure 7.
As shown in Figure 7, the maximum clamping force of
single gate is 808.2 t, that of two gates is 563.3 t, and that of
four gates is 350.7 t. Te three clamping forces meet the
maximum injection pressure of the selected injection
molding machine.

3.1.1. Filling Time. Filling time is an essential key index for 3.1.3. Welding Line. Tere may be multiple gates in the
the simulation analysis of most injection-molded parts. Te injection molding process, which will produce welding lines,
simulation cloud diagram of flling time of three gate resulting in the reduction of the surface quality of plastic
schemes is shown in Figure 6. parts. It is essential to minimize the number of welding lines
6 Advances in Materials Science and Engineering

Welding line Welding line

=135.0 [°C] =135.0 [°C]

[°C] [°C]
135.0 135.0

101.4

67.71

34.06
101.3

67.58

E D
T
33.87

0.4132
0.1601

C
(a) (b)
Welding line
=135.0 [°C]

A
[°C]
135.0

101.4

R
67.86

34.28

T
0.7113

E
(c)

Figure 8: Welding line of three gate location schemes. (a) Single gate. (b) Two gates. (c) Four gates.
when designing the gates. Moldfow is used to analyze the reduce the quality and structural stability of plastic parts and

R
welding lines. Te results are shown in Figure 8.
As shown in Figure 8, when the number of gates in-
creases, the number of welding lines also increases. Te rise
in welding lines may lead to the decline of the quality of
plastic parts and afect structural stability. Terefore, we
should try to reduce the number of welding lines.
3.1.4. Cavitation. During the flling process of the plastic
part, the fow rate of the melt, the cooling time, and the
structural diferences of the plastic part lead to the creation
of cavitation in certain specifc locations of the plastic part,
and the number and position of cavitation have a signifcant
impact on the quality of plastic parts. Moldfow carries out
the simulation analysis of cavitation, and the results are
shown in Figure 9.
It can be seen from Figure 9 that diferent gate positions
and quantities will lead to diferent cavitation, which will
increase warpage deformation and shrinkage deformation of
plastic parts. Terefore, we need to select the location and
number of gates that produce less cavitation for injection
molding [16].
Te analysis results of diferent gate schemes are listed in
Table 1 for comparative analysis to select the best gate
number and location.
It can be seen from Table 1 that the flling time of single
gate is the longest, and the clamping force is the largest,
which will lead to excessive injection time and reduce
production efciency. Te flling time and clamping force of
the four gates are the smallest, but the excessive number of
fusion lines and cavitation have a signifcant impact on the
product quality. Tis paper selects the two-gate scheme for
injection molding of the car door inner plate. Te positions
of the two gates are verifed by flling fow simulation, and
the positions are shown in Figure 10. Te initial coordinates
are selected in N1 (20968) and N2 (14276).
Advances in Materials Science and Engineering 7

Cavitation Cavitation

E D
(a)
Cavitation

T
(b)

C
R A (c)

Figure 9: Cavitation of three gate location schemes. (a) Single gate. (b) Two gates. (c) Four gates.

T
Table 1: Analysis results of three diferent intersection schemes.
Number of gates Filling time/s Clamping force/t Welding line Cavitation
1 6.443 808.2 5 2

E
2 4.930 563.3 7 2
4 3.783 350.7 9 4

Te selection of diferent gate positions has a signifcant

R
infuence on the maximum pressure at the injection port, the
temperature diference between the surface of the molded
part, the overheating unit, and the overpressure unit during
N1
the flling process [17]. Terefore, the above four main
infuencing factors need to be weighted by coefcients to
N2 fnally determine the objective function model of gate po-
sition, as shown in equation:

min F(X) � α ∗ Pin + β ∗ Nnovp + c ∗ Nfth + λ ∗ Td ,

Figure 10: Initial gate location diagram of two gate schemes. T
(1)
Td ≤ 20, X ∈ Ω, α + β + c + λ � 1, X � 􏼈X1 , X2 , X3 . . . , Xi 􏼉 ,

3.2. Study on Optimal Gate Position. Te two-gate scheme is
initially selected for the injection molding simulation of the
car door inner panel by comparing the results of the flling
fow simulation with single, two, and four gates. Further,
optimization of the position is carried out on the basis of the
two-gate scheme.
where F(X)—Objective function model; Xi —Coordinates
of gate position; Ω—Feasible area; Pin —Maximum inlet
pressure, Mpa; Nnovp —Percentage of overvoltage units;
Nfth —Percentage of overheating unit; Td —Temperature
diference, K; α, β, c, λ—Weights for Pin , Nnovp , Nfth , and
Td respectively.
8 Advances in Materials Science and Engineering

Best :-6.34564 Mean :-6.34562

0

-1

D
-2

Fitness value
-3

E
-4

-5

T
-6

-7
0 100 200 300 400 500 600 700 800 900 1000
Generation

C
Best fitness
Mean fitness

Figure 11: Algorithm iteration graph.

A
N3

N1

R
N2
N4

T
(a) (b)

Figure 12: Gate position comparison diagram. (a) Initial gate position. (b) Optimized gate position.

E
Filling time Moldfow, and the optimum node position is then recal-
=4.840 [s] culated in conjunction with the optimization function.
[s] However, the above function model has too many con-
4.840
straints and needs to be simplifed to facilitate solution.

R
Terefore, the following theorem is introduced [18].
3.630 If Ψ is a positive variable, G � 􏼈gK (X), K � 1, 2 . . . q􏼉,
and j is the set, then:
2.420 q
1
E(G) � · ln 􏽘 exp 􏼂Ψ · gK (X)􏼃. (2)
Ψ K�1
1.210
Equation 2 is called PCE function, which is a constraint
function with parameters as variables. Combining Equations
0.0000 (1) (2) yields the constrained optimization function
equation:

Figure 13: Optimized flling time diagram. ⎪

⎧ min F(X),
⎪
⎪
⎪
⎨
(3)
⎪
⎪
⎪ 1 ⎨ q
⎧ ⎬
⎫
⎪
⎩ s · t · gΨ (X) � · ln ⎩ 􏽘 exp 􏼂Ψ · gK (X)􏼃⎭ ≤ 0.
Te function value of the gate position optimization Ψ K�1
function needs to be calculated based on the Moldfow flling
analysis module. Te coordinates of the mesh nodes and the Te exact penalty function transforms equation 3 into a
values of the various parameters are obtained through sequential unconstrained optimization function equation:
Advances in Materials Science and Engineering 9

Clamping force : XY garph the optimal solution is in the jth population, in equation 6, A
600.0 is the feasible area of the optimal solution, and B is the
uncertain region of the optimal solution. Using equation 6,
500.0

D
the probability that the optimal solution falls in population j,
Pj , can be expressed as:
400.0
exp 􏽨r · Gj (X)􏽩
tonne

300.0 Pj � , (7)
􏽐M exp 􏽨r · Gj (X)􏽩

E
j�1
200.0
where: r � α − 1/α; α is the weight coefcient, taken as 0.6; Rj
100.0 is the spatial contraction factor for each cluster; and
Rj � 1 − Pj .
0.0000

T
0.0000 5.000 10.00 15.00 20.00 25.00 30.00 35.00 When the optimal solution of the objective function is in
Time [s] the jth population, Pj � 1, Pi � 0, i ≠ j, and 􏽐M ∗
j�1 Pj G(X ) �
∗
G(X ); when optimization begins, Pj � 1/m, j � 1, 2 . . . M,
Figure 14: Optimized clamping force diagram.
and M takes the maximum value; when the optimization is
completed, the uncertainty of the optimal solution is re-

C
Table 2: Comparison of data before and after gate position duced to zero, i.e., minB � 0; in summary, the optimal so-
optimization. lution X∗ of equation 7 is the solution of the original
State Filling time/s Clamping force/t objective function.
Let the space at the beginning be C0 (0), contraction
Before optimization 4.930 563.3
After optimization 4.840 551.0 factor be Rj , j � 1, 2 . . . M, and M be the number. Te

A
contraction space is shown in equation:
Gj (K + 1) � Rj Cj (K). (8)
⎨
⎧ m ⎬
⎫
α
min φΨ (X) � F(X) + · ln⎩ 1 + 􏽘 exp 􏼂Ψ · gK (X)􏼃⎭ . Te upper and lower limits of the variables are shown in
Ψ K�1
equation:

R
(4)
⎪
⎧
⎪ Cj (K + 1)
Te exact penalty function method automatically adjusts ⎪
⎪ Dij (K + 1) � max 􏼨􏼢Xij (K) − 􏼣, Dij (0) 􏼩,
⎪
⎪ 2
the constraints in the objective function based on the ⎪
⎨
tightness principle, such that all the constraints in the ob- ⎪
⎪

T
⎪
⎪ Cj (K + 1)
jective function are within a reasonable range. In equation 4, ⎪
⎪
⎪
⎩ Dij (K + 1) � min 􏼨􏼢Xij (K) + 􏼣, Dij (0) 􏼩,
α is the penalty factor, and the function solution is in the 2
feasible area when α takes the value [103 , 105 ].
(9)
For a multi-population genetic algorithm, the constraint

E
function needs to be transformed into an unconstrained where Xij (k) is the value of the jth group and the ith design
function, and the unconstrained objective function is shown variable when iterating to the K-th generation. Te iteration
in equation: is completed when the space is reduced to the specifed
max G(X) � C − φΨ (X), (5) accuracy.
In the fnite element model, the nodes are considered as

R
where G(X) is the adaptation value function after uncon- each gate position to be marked, the gate position is regarded
strained processing, and C is taken to be a large enough as NP , the number of nodes is N, and the number of gates to
positive number to ensure that the value of G(X) is positive. be optimized is n as shown in equation:
Using the multi-objective coefcient weighting method N(N − 1) · · · [N − (n − 1)]
combined with equation 5, the evolutionary model is con- NP � . (10)
1×2×3×···×n
structed as equation:
M
Te melt temperature used for the injection molding
⎪
⎧
⎪ process is 230°C, and the mold temperature is 70°C. Material
⎪
⎪ min A � − 􏽘 Pj G(X),
⎪
⎪ parameters refer to the relevant data for the injection
⎪
⎪ j�1
⎪
⎪ molding material polypropylene PP-LGF-30. Te meshed
⎪
⎪ M
⎨ parts have 117601 triangular elements and 58817 node el-
⎪ min B � − 􏽘 Pj ln Pj , (6)
⎪
⎪ j�1
ements. Set the α, β, c and λ of the objective function to 0.3,
⎪
⎪ 0.15, 0.4, and 0.15 respectively, the number of populations is
⎪
⎪
⎪
⎪ M
taken as 10, the size is 20, the hybrid probability is 0.2, the
⎪
⎪ s · t · 􏽘 Pj � 1; Pj ∈ [0, 1],
⎪
⎩ variation probability is 0.03, and the initial gate positions are
j�1
N1 (20968) and N2 (14276). After starting the algorithm to
where G(X) is the transformed unconstrained ftness ob- fnd the optimum, the objective function of the gate posi-
jective function, and Pj (1, 2, . . ., M) is the probability that tions starts to converge after 900 iterations, as shown in
10 Advances in Materials Science and Engineering

Table 3: Table of horizontal factor settings.

Factors
Level A B C D E

D
Mold temperature(℃) Melt temperature(℃) Cooling time (s) Holding pressure (Mpa) Holding time (s)
1 60 200 15 20 10
2 70 210 20 30 15
3 80 220 25 40 20

E
4 90 230 30 50 20

Table 4: Experimental design and results.

Factors Indexes

T
Number
A B C D E Volume shrinkage (%) Warpage deformation (mm)
1 60 200 15 20 10 15.27 7.708
2 60 210 20 30 15 15.65 7.087
3 60 220 25 40 20 16.19 6.990

C
4 60 230 30 50 25 16.73 6.966
5 70 200 20 40 25 15.19 5.417
6 70 210 15 50 20 15.63 5.428
7 70 220 30 20 15 16.16 8.244
8 70 230 25 30 10 16.71 8.258
9 80 200 25 50 15 15.14 5.310

A
10 80 210 30 40 10 15.59 6.348
11 80 220 15 30 25 16.13 7.123
12 80 230 20 20 20 16.68 8.578
13 90 200 30 30 20 15.11 5.977
14 90 210 25 20 25 15.53 6.821

R
15 90 220 20 40 10 16.10 7.661
16 90 230 15 50 15 16.65 6.218

Table 5: Analysis of range of volume shrinkage.

T
Factors Level1 Level2 Level3 Level4 Range
Mold temperature 15.96 15.89 15.85 15.81 0.15
Melt temperature 15.07 15.60 16.15 16.69 1.62
Cooling time 15.92 15.84 15.86 15.86 0.06

E
Holding pressure 15.91 15.86 15.73 16.01 0.28
Holding time 15.92 15.87 15.87 15.86 0.06

R
Table 6: Analysis of variance of volume shrinkage.
Factors Level1 Level2 Level3 Level4 Freedom Mean-variance F
Mold temperature 0.3045 0.3796 0.3851 0.3977 3 0.366725 24.81
Melt temperature 0.014025 0.0021 0.00115 0.000925 3 0.00455
Cooling time 0.2709 0.35745 0.4157 0.43245 3 0.369125 85.64
Holding pressure 0.30235 0.412375 0.207625 0.5181 3 0.360112 18.59
Holding time 0.296975 0.370125 0.41105 0.3987 3 0.369212 113.95

Figure 11, where the horizontal coordinates are the number
of iterations, and the vertical coordinates are the adaptation
values.
As shown in Figure 12, by transcoding the binary co-
ordinates after 900 iterations, the two gate positions ob-
tained are N3 (28425) and N4 (46424).
Based on the optimized gate position, the flling fow
module of Moldfow is used to simulate the optimized two
gate parts. Te optimized flling time is 4.840s, as shown in
Figure 13, and the maximum clamping force is 551.0 t, as
shown in Figure 14.
Te flling time and clamping force before and after
optimizing the two gate positions are listed, as shown in
Table 2.
It can be seen from Table 2 that the new gate position is
obtained by establishing the objective optimization function
Advances in Materials Science and Engineering 11

17

16.5

D
16

15.5

E
15

14.5

T
14
Level 1 Level 2 Level 3 Level 4

Mold temperature Holding pressure

Melt temperature Holding time

C
Cooling time

Figure 15: Volume shrinkage range analysis line chart.

A
Table 7: Analysis of range of warpage deformation.
Factors Level1 Level2 Level3 Level4 Range
Mold temperature 7.188 6.837 6.840 6.669 0.519
Melt temperature 6.103 6.421 7.505 7.505 1.402
Cooling time 7.494 7.186 7.055 6.884 0.61

R
Holding pressure 6.133 7.111 6.604 5.981 1.13
Holding time 7.494 6.715 6.743 6.582 0.912

T
Table 8: Analysis of variance of warpage deformation.
Factors Level1 Level2 Level3 Level4 Freedom Mean-variance F
Mold temperature 0.09227 2.00014 1.42092 0.42236 3 0.983923 84.38
Melt temperature 0.92284 0.39873 0.24540 0.91631 3 0.62082

E
Cooling time 1.52024 1.32561 1.13780 0.74157 3 1.181305 174.52
Holding pressure 3.34708 0.65057 0.68519 0.44561 3 1.282113 251.91
Holding time 0.49267 1.17431 1.43605 0.46362 3 0.891663 43.37

R
for the gate position, which shortens the flling time and 8
reduces the clamping force, achieving the optimization efect.
7.5
4. Process Parameter Optimization Based on the
Orthogonal Test Design and the Ant Colony 7
Neural Network
6.5
Warpage deformation and volume shrinkage of the car door
inner panel are used as evaluation indicators, and the infuence
6
of the process parameters on the evaluation indicators is in-
vestigated based on orthogonal tests. Optimization of process
5.5
parameters is based on ant colony neural network algorithms
to obtain the optimal combination of parameters to reduce
5
warpage and volume shrinkage and improve product quality. Level 1 Level 2 Level 3 Level 4
Mold temperature Holding pressure
Melt temperature Holding time
4.1. Orthogonal Experimental Design and Data Analysis of the
Cooling time
Car Door Inner Plate. Tis paper investigates fve molding
process parameters: mold temperature, melt temperature, Figure 16: Warpage deformation range analysis line chart.
12
[%]
15.08
11.50
7.918
4.335
A C
R
0.7518
Figure 17: Volume shrinkage diagram.
T
cooling time, holding pressure, and holding time;
each experimental factor is uniformly selected at four
levels, and the infuence of each process parameter on
warpage deformation and volume shrinkage of the car
door inner panel is investigated through orthogonal tests.
E
Te division of experimental factors and levels is shown in
Table 3.
As shown in Table 4, the orthogonal test principle is
applied to establish the L16 (45 ) orthogonal experiment table,
R
and the simulation analysis is carried out for 16 sets of
process parameter combinations in the table, respectively, to
obtain 16 sets of warpage deformation and volume
shrinkage.
4.1.1. Analysis of Volume Shrinkage. Volume shrinkage of
the car door inner panel is studied by range and variance
methods, respectively, as shown in Tables 5 and 6. Ten use
the data in Table 5 to make a line chart, as shown in
Figure 15.
According to the data in Table 5, the infuence degree of
diferent parameters on volume shrinkage is as follows: melt
temperature > holding pressure > mold temper-
ature > cooling time > holding time. According to the cor-
responding F value of each parameter in Table 6, the analysis
correctness of Table 5 can be verifed. Terefore, combi-
nation A4B1C4D4E3 is the optimal combination for volume
shrinkage.
Advances in Materials Science and Engineering
[mm]
5.268
3.269
E D
T
1.270
-0.7294
-2.729
Figure 18: Warpage deformation diagram.
4.1.2. Analysis of Warpage Deformation. Warpage defor-
mation of the car door inner panel is studied by range and
variance methods, respectively, as shown in Tables 7 and 8.
Ten use the data in Table 5 to make a line chart, as shown in
Figure 16.
According to the data in Table 7, the infuence degree of
diferent parameters on warpage deformation is as follows:
melt temperature > holding pressure > holding time-
> cooling time > mold temperature. According to the cor-
responding F value of each parameter in Table 8, the analysis
correctness of Table 7 can be verifed. Terefore, combi-
nation A2B1C1D1E3 is the optimal combination for
warpage deformation.
Once the two sets of process parameters have been
obtained, it is necessary to combine them. According to the
analysis of variance, the melt temperature and holding
pressure have the most signifcant infuence on the com-
prehensive index; thus, the fnal process of selecting process
parameters selects B1D4; for warping deformation, the in-
fuence of holding time and cooling time is more important,
and thus the deformation of these two parameters is mainly
small, i.e., C1E3 is selected; the mold temperature takes the
median value, and the fnal combination is A4B1C1D4E3;
the mold temperature of 90°C, melt temperature of 200°C,
cooling time of 15s, holding pressure of 30Mpa, holding time
of 22s, volume shrinkage, and warpage deformation are
obtained by simulation analysis with Moldfow, as shown in
Figures 17 and 18.
Te process parameters obtained by range and variance
analysis are simulated. Te optimized volume shrinkage is
15.08%, and warpage deformation is 5.268 mm.
Advances in Materials Science and Engineering 13

Mold temperature

D
Volume shrinkage
Melt temperature

E
Cooling time

T
Holding pressure
Warpage deformation

C
Holding time

Input layer Hidden layer Output layer

A
Figure 19: BP network structure.

Best Validation Performance is 0.049497 at epoch 17

101

R
Mean Squared Error (mse)

100

E T 10-1

10-2
0 5 10 15 20

R
Figure 20: Training efect diagram.
4.2. Establishment of the Ant Colony Neural Network. A
three-layer BP neural network model is built to describe the
relationship between process parameters and volume
shrinkage and warpage deformation. Te input layer takes
fve parameters, such as mold temperature as input neurons,
and the output layer takes volume shrinkage and warpage
deformation as neurons. Te structure of the specifc net-
work model is shown in Figure 19.
Te selection of samples usually follows the principle of
average uniformity, using data derived from orthogonal tests
and injection simulation as learning samples, using groups
labeled 2–5, 7–10, and 12–15 to train the network, and 1, 6,
11, and 16 as validation groups. Te model is trained by
MATLAB programming and MATLAB toolbox, and the
training of the BP neural network is stopped after 23 training
sessions. Te training results are shown in Figure 20, and the
training error is 0.0111.
Train Test
Validation Best
Figure 21: Training curve diagram.
Te training curve is shown in Figure 21, and the ftting
degree is shown in Figure 22. It can be seen from the fgure
that the R-value of the training and test samples is 0.99978,
indicating that the correlation between them is good.
Once the construction of the neural network is com-
pleted, its correctness needs to be verifed by entering or-
thogonal experimental data into the software predictions
and then calculating the results of the actual tests against the
software predictions of shrinkage and deformation to obtain
the errors, as shown in Table 9.
Te ftting degree between the predicted and test results
for volumetric shrinkage and warpage deformation is shown
in Figures 23 and 24. After comparing the predicted value
14 Advances in Materials Science and Engineering

Training: R=0.99978 Validation=0.99875

16 16

D
Output~=0.99*Target+0.041
14 14

Output~=1*Target+-0.1
12 12

E
10 10

8 8

T
6 6
6 8 10 12 14 16 6 8 10 12 14 16
Target Target
Data Data

C
Fit Fit
Y=T Y=T
Test: R=0.98317 All: R=0.9968

16 16

A
Output~=0.98*Target+0.13
Output~=0.92*Target+0.76

14 14

12 12

R
10 10

8 8

T
6 6
6 8 10 12 14 16 6 8 10 12 14 16
Target Target
Data Data

E
Fit Fit
Y=T Y=T

Figure 22: Fitting degree diagram.

R
Table 9: Error between the prediction and test results.
Volume shrinkage Warpage deformation
Number
Prediction result Test result Error Prediction result Test result Error
1 14.98 15.27 1.90% 7.719 7.708 0.14%
6 15.51 15.63 0.77% 5.301 5.428 2.34%
11 15.27 16.13 5.33% 7.317 7.123 2.72%
16 16.84 16.65 1.14% 6.015 6.218 3.26%

with the actual value, it is found that the maximum errors of colony neural network (ACO-BPNN) method [19]. Take the
the predicted value and the actual value of volume shrinkage test data obtained in the above chapters as the test data to test
and warpage deformation are 5.33% and 3.26%, respectively, the authenticity of this optimization method. Te groups
the minimum errors are 0.77% and 0.14%, respectively, and numbered 2–5, 7–10, and 12–15 in the orthogonal test table
the average errors are 2.285% and 2.115%, respectively, are selected for training the ant colony neural network, and
indicating the correctness of the neural network model. 1, 6, 11, and 16 are used as the validation groups for the
Optimization of the built and validated neural network accuracy of the neural network. Te sample is predicted and
model is based on the ant colony algorithm called the ant analyzed by MATLAB, and the volume shrinkage and
Advances in Materials Science and Engineering 15

17

16.5

D
16

15.5

E
15

14.5

T
14
1 2 3 4
Predicted result
Test result

C
Figure 23: Volumetric shrinkage ft comparison.

A
7

R
3

T
0
1 2 3 4
Predicted result
Test result

E
Figure 24: Warpage deformation ft comparison.

Table 10: Error between the prediction and test results.

R
Volume shrinkage Warpage deformation
Number
Prediction result Test result Error Prediction result Test result Error
1 15.13 15.27 0.92% 7.711 7.708 0.04%
6 15.42 15.63 1.34% 5.412 5.428 1.6%
11 16.24 16.13 0.68% 7.139 7.123 0.22%
16 16.53 16.65 0.72% 6.207 6.218 0.18%

warpage deformation obtained are compared with the test
results to obtain the error between them, as shown in
Table 10.
Te ftting degree between the predicted and test results
for volumetric shrinkage and warpage deformation is shown
in Figures 25 and 26. After comparing the predicted value
with the actual value, it is found that the maximum errors of
the predicted value and the actual value of volume shrinkage
and warpage deformation are 1.34% and 1.6%, respectively,
the minimum errors are 0.68% and 0.04%, respectively, and
the average errors are 1.37% and 1.605%, respectively; thus,
the accuracy of the predicted value of ACO-BPNN model is
verifed.
4.3. Optimization of Molding Process Parameters Based on
ACO-BPNN. Te ant colony algorithm is used to optimize
the established BP neural network model. Te established
neural network model is used as the ftness function, and the
minimum value of volume shrinkage and warpage
16 Advances in Materials Science and Engineering

17 Evolutionary convergence curve of ACO

23

16.5 22

D
21
16

Error of mean square

20
15.5

E
19

15 18

17
14.5

T
16

14 15
1 2 3 4 0 5 10 15 20 25 30 35 40 45
Predicted result Number of evolutionary iterations
Test result Optimum ftness

C
Figure 25: Volumetric shrinkage ft comparison. Figure 27: Iterative diagram of the ant colony algorithm.

9 Comparison of predicted value and real value of BP

neural network and afer AOC optimization
8

A
45
7
40
6

5 35

R
4
Index value

30
3

2 25

T
1
20
0
1 2 3 4
15
Predicted result
Test result

E
10
Figure 26: Warpage deformation ft comparison. 0 5 10 15
Real value
BP predicted value
ACO-BP predicted value
Table 11: Parameter settings of ACO-BPNN.

R
Figure 28: Comparison diagram before and after ACO
Parameter m Q Spn ρ η σ NC max Wmin Wmax optimization.
ACO-BPNN 55 50 65 0.5 0.3 1 1000 − 10 10

[%]
deformation is used as the optimization goal of the ant 13.32
colony algorithm to optimize factors such as die temperature
[20]. Te value range of the fve process parameters is 12.10
regarded as the constraint condition of the function, which is
x1 ∈ [60, 90], x2 ∈ [200, 230], x3 ∈ [15, 30], x4 ∈ [20, 50],
and x5 ∈ [10, 25]. Te ftness value is taken as the function 8.668
value, and the minimum of ftness function is taken as the
iterative optimization goal of the ant colony algorithm. Te
5.236
relevant parameter settings of the ant colony algorithm are
shown in Table 11 [21].
Using MATLAB software to calculate the ant colony 1.803
algorithm, the change of the optimal ftness value of the
Figure 29: Volume shrinkage diagram.
objective function can be obtained, as shown in Figure 27. It
Advances in Materials Science and Engineering 17

[mm]
4.315

D
4.086

2.756

E
1.425

0.0942

T
Figure 30: Warpage deformation diagram.

Table 12: Result comparison table.

C
Process parameters Forming index
Optimization Mold Melt Cooling Holding Holding Volume Warpage
method temperature/°C temperature/°C time/s pressure/Mpa time/s shrinkage/% deformation/mm
Orthogonal test 90 200 15 50 20 15.08 5.268
ACO-BPNN 76 205 23.8 54.7 22.1 13.32 4.315

A
can be seen from the function iteration diagram that con-
vergence is reached about the 15th time. Te comparative

R
analysis of the predicted value and the actual value of the BP
neural network before and after ACO optimization is shown
in Figure 28.
As can be seen from Figure 27, the optimal adaptation
value of ACO-BPNN is minimized after about 15 iterations,
obtained, so as to determine the optimal gate
quantity. On this basis, the objective function
about the gate position is established. Taking the
objective function as the ftness function, the gate
position is optimized based on the multi-pop-
ulation genetic algorithm, so as to obtain the
optimal gate position.

T
corresponding to the following molding process parameters:
mold temperature of 76°C, melt temperature of 205°C,
cooling time of 23.8 s, holding pressure of 54.7 MPa, and
holding time of 22.1s. By inputting the molding process
(3) Warping deformation and volume shrinkage of
the car door inner panel are selected as the evaluation
indexes, and mold temperature, melt temperature,
cooling time, holding pressure, and holding time are

E
parameters into Moldfow for flling fow analysis, volume
shrinkage of the car door inner plate is 13.32% and warpage
deformation is 4.315 mm, as shown in Figure 29 and
Figure 30.
Comparing the orthogonal test optimization results
taken as the infuencing factors to establish the or-
thogonal test. Trough the analysis method of range
and variance, the combination of process parameters
is A4B1C1D4E3, i.e., the mold temperature is 90°C,
the melt temperature is 200°C, the cooling time is

R
with those based on ACO-BPNN, volume shrinkage is
reduced from 15.08% to 13.32% and warpage deformation
is reduced from 5.268 mm to 4.315 mm as shown in
Table 12.
(4) Finally, the BP neural network is established, in
5. Conclusion
(1) UG is used to model the car door inner panel, and
then the injection mold is designed based on the
mold design module of UG. Finally, the selection of
the material and thickness of the car door inner panel
and the design of the relevant parameters of the
injection mold are completed.
(2) Based on the flling and cooling module of
Moldfow, the mold fow of the door inner plate is
analyzed. By selecting diferent gate positions and
quantities for analysis, quality indexes such as
flling time, cavitation, and welding line are
15s, the holding pressure is 30 MPa, the holding time
is 20s, volume shrinkage is 15.08%, and warpage
deformation is 5.268 mm.
which parameters such as melt temperature are taken
as the input layer and warpage deformation and
volume shrinkage are taken as the output variables.
Te prediction results and the test results of warpage
deformation and volume shrinkage are ftted and
compared based on MATLAB to verify the accuracy
of the neural network. Te ant colony algorithm is
used to optimize the established neural network
model. Te ftness value of ACO-BPNN reaches the
minimum after about 15 iterations, with corre-
sponding process parameters being: mold temper-
ature of 76°C, melt temperature of 205°C, cooling
time of 23.8s, holding pressure of 54.7Mpa, and
holding time of 22.1s. Using Moldfow to simulate
18
the flling fow, volume shrinkage is 13.32% and
warpage deformation is 4.315 mm. Te optimization
of injection molding process parameters of car door
inner plates is completed.
Data Availability
Te data used to support the fndings of this study are in-
cluded within the article.
Conflicts of Interest
Te authors declare that they have no conficts of interest.
Authors’ Contributions
Kf Y made substantial contributions to the design, experi-
mental research, data collection, and result analysis; Ym W
made critical changes to important academic content; and
Gq W contributed to the fnal review and fnalization of the
articles to be published.
A
Acknowledgments
Tis article belongs to the major projects of the “Te
University Synergy Innovation Program of Anhui Province
(GXXT-2019-004),” the project of the “Teaching Research
R
Project of Anhui Education Department (2019jyxm0229),”
and the project of the “Science and Technology Planning
Project of Wuhu City (2021YF58).”
References
T
[1] W. Chen, Research on Vehicle Lightweight, Vibration and
Noise Reduction Based on Carbon Fiber Composites, Hunan
University, Hunan, China, 2019.
E
[2] K. Fan, Study on Numerical Simulation and Parameter Op-
timization Method for Warpage of Injection Molded Products,
Shandong University of Science and Technology, Shandong,
China, 2017.
[3] C. Doerfel, R. Decker, M. Heinrich et al., “Polypropylene
based piezo ceramic compounds for micro injection molded
R
sensors,” Key Engineering Materials, vol. 742, pp. 807–814,
2017.
[4] M. Huszar, F. Belblidia, H. M. Davies, C. Arnold, D. Bould,
and J. Sienz, “Sustainable injection moulding: the impact of
materials selection and gate location on part warpage and
injection pressure,” Sustainable Materials and Technologies,
vol. 5, pp. 1–8, 2015.
[5] G. Lu, “Warpage analysis and process optimization of in-
jection molding based on BP neural network [J],” Plastics
industry, vol. 45, no. 02, pp. 58–60, 2017.
[6] B. S. Heidari, S. M. Davachi, A. H. Moghaddam et al.,
“Optimization simulated injection molding process for ul-
trahigh molecular weight polyethylene nanocomposite hip
liner using response surface methodology and simulation of
mechanical behavior,” Journal of the Mechanical Behavior of
Biomedical Materials, vol. 81, pp. 95–105, 2018.
[7] N. Sateesh, S. Devakar Reddy, G. Praveen Kumar, and
R. Subbiah, “Optimization of injection moulding process in
manufacturing the top cap of water meter,” Materials Today
Proceedings, vol. 18, pp. 4556–4565, 2019.
Advances in Materials Science and Engineering
[8] M. S. Mukras Saad, “Experimental-based optimization of
injection molding process parameters for short product cycle
time,” Advances in Polymer Technology, vol. 2020, Article ID
1309209, 15 pages, 2020.
D
[9] S. Ramesh, P. Nirmala, G. Ramkumar et al., “Simulation
process of injection molding and optimization for automobile
instrument parameter in embedded system,” Advances in
Materials Science and Engineering, vol. 2021, Article ID
E
9720297, 10 pages, 2021.
[10] G. Wang and X. Fan, “Study on mathematical model of flling
process of Microcellular Plastic Injection,” Molding [J]. Plastic
technology, vol. 48, no. 08, pp. 48–52, 2020.
[11] W. Sun, “Research progress on preparation technology of ABS
T
and PP composites,” Oil refning and chemical industry,
vol. 31, no. 06, pp. 8–11, 2020.
[12] X. Zhi, Service analysis of plastic pipe bending die core and
Study on die steel selection, Beijing Jiaotong University, Beijing
Chinna, 2019.
C
[13] U. Götze, P. Peças, and F. Richter, “Design for eco-efciency
– a system of indicators and their application to the case of
moulds for injection moulding,” Procedia Manufacturing,
vol. 33, pp. 304–311, 2019.
[14] J. M. Mercado-Colmenero, M. A. Rubio-Paramio,
J. D. J. Marquez-Sevillano, and C. Martin-Doñate, “A new
method for the automated design of cooling systems in in-
jection molds,” Computer-Aided Design, vol. 104, pp. 60–86,
2018.
[15] C. Aversa, M. Barletta, E. Pizzi, M. Puopolo, and S. Vesco,
“Wear resistance of injection moulded PLA-talc engineered
bio-composites: efect of material design, thermal history and
shear stresses during melt processing,” Wear, vol. 390,
pp. 184–197, 2017.
[16] W. Chen, X. Xu, Z. Li, F. Chen, and J. Qiu, “Optimal design of
injection mold gate based on mold fow analysis [J],” Modern
machinery, no. 02, pp. 5–8, 2019.
[17] M. Seyednourani, M. Yildiz, and H. S. Sas, “A two-stage
optimization methodology for gate and vent locations and
distribution media layout for liquid composite molding
process,” Composites Part A: Applied Science and
Manufacturing, vol. 149, Article ID 106522, 2021.
[18] F. Yin, H. Mao, and L. Hua, “A hybrid of back propagation
neural network and genetic algorithm for optimization of
injection molding process parameters,” Materials & Design,
vol. 32, no. 6, pp. 3457–3464, 2011.
[19] D. Di Caprio, A. Ebrahimnejad, H. Alrezaamiri, and
F. J. Santos-Arteaga, “A novel ant colony algorithm for solving
shortest path problems with fuzzy arc weights,” Alexandria
Engineering Journal, vol. 61, no. 5, pp. 3403–3415, 2022.
[20] Y. Mokhtari and D. Rekioua, “High performance of maxi-
mum power point tracking using ant colony algorithm in
wind turbine,” Renewable Energy, vol. 126, pp. 1055–1063,
2018.
[21] Bo Jiang, F. Cai, X. Qin, Bo Wang, G. Jiang, and J. Gao,
“Application progress of ant colony algorithm in polymer
polymerization and processing,” Chemical Production and
Technology, vol. 27, no. 02, pp. 20–26+34+28, 2021.