ArticlePDF Available

Mechanical Strength Enhancement of 3D Printed Acrylonitrile Butadiene Styrene Polymer Components Using Neural Network Optimization Algorithm

Authors:

Abstract and Figures

Fused filament fabrication (FFF), a portable, clean, low cost and flexible 3D printing technique, finds enormous applications in different sectors. The process has the ability to create ready to use tailor-made products within a few hours, and acrylonitrile butadiene styrene (ABS) is extensively employed in FFF due to high impact resistance and toughness. However, this technology has certain inherent process limitations, such as poor mechanical strength and surface finish, which can be improved by optimizing the process parameters. As the results of optimization studies primarily depend upon the efficiency of the mathematical tools, in this work, an attempt is made to investigate a novel optimization tool. This paper illustrates an optimization study of process parameters of FFF using neural network algorithm (NNA) based optimization to determine the tensile strength, flexural strength and impact strength of ABS parts. The study also compares the efficacy of NNA over conventional optimization tools. The advanced optimization successfully optimizes the process parameters of FFF and predicts maximum mechanical properties at the suggested parameter settings.
Content may be subject to copyright.
polymers
Article
Mechanical Strength Enhancement of 3D Printed
Acrylonitrile Butadiene Styrene Polymer Components
Using Neural Network Optimization Algorithm
Jasgurpreet Singh Chohan 1, Nitin Mittal 1, Raman Kumar 1, Sandeep Singh 1,
Shubham Sharma 2, Jujhar Singh 2, Kalagadda Venkateswara Rao 3, Mozammel Mia 4, * ,
Danil Yurievich Pimenov 5and Shashi Prakash Dwivedi 6
1University Centre for Research and Development, Chandigarh University, Mohali-140413, India;
jaskhera@gmail.com (J.S.C.); mittal.nitin84@gmail.com (N.M.); ramankakkar@gmail.com (R.K.);
drsandeep1786@gmail.com (S.S.)
2Department of Mechanical Engineering, IKG Punjab Technical University, Kapurthala-144603, India;
shubham543sharma@gmail.com or shubhamsharmacsirclri@gmail.com (S.S.);
jujharsing2085@gmail.com (J.S.)
3Center for Nanoscience and Technology, Institute of Science and Technology, Jawaharlal Nehru
Technological University Hyderabad, Telangana State 500085, India; kalagadda2003@gmail.com
4Department of Mechanical Engineering, Imperial College London, Exhibition Rd., London SW7 2AZ, UK
5Department of Automated Mechanical Engineering, South Ural State University, Lenin Prosp. 76,
454080 Chelyabinsk, Russia; danil_u@rambler.ru
6Department of Mechanical Engineering, G.L. Bajaj Institute of Technology and Management,
Greater Noida 201308, India; spdglb@gmail.com
*Correspondence: m.mia19@imperial.ac.uk
Received: 2 September 2020; Accepted: 26 September 2020; Published: 30 September 2020


Abstract:
Fused filament fabrication (FFF), a portable, clean, low cost and flexible 3D printing
technique, finds enormous applications in dierent sectors. The process has the ability to create
ready to use tailor-made products within a few hours, and acrylonitrile butadiene styrene (ABS) is
extensively employed in FFF due to high impact resistance and toughness. However, this technology
has certain inherent process limitations, such as poor mechanical strength and surface finish, which can
be improved by optimizing the process parameters. As the results of optimization studies primarily
depend upon the eciency of the mathematical tools, in this work, an attempt is made to investigate
a novel optimization tool. This paper illustrates an optimization study of process parameters of
FFF using neural network algorithm (NNA) based optimization to determine the tensile strength,
flexural strength and impact strength of ABS parts. The study also compares the ecacy of NNA
over conventional optimization tools. The advanced optimization successfully optimizes the process
parameters of FFF and predicts maximum mechanical properties at the suggested parameter settings.
Keywords:
optimization; neural network algorithm; fused filament fabrication; mechanical
strength; simulation
1. Introduction
Globalization has intensified the market scenario of manufacturing units due to rapidly changing
customer demands, competition among peers and requirements for high quality and reliable products.
The demand for low cost and customized products has led to a paradigm shift from traditional
manufacturing techniques to additive manufacturing technologies by industries [
1
]. Additive
manufacturing is a collection of non-conventional techniques that follow the principle of layer
Polymers 2020,12, 2250; doi:10.3390/polym12102250 www.mdpi.com/journal/polymers
Polymers 2020,12, 2250 2 of 18
by layer manufacturing in contrast to traditional subtractive and joining techniques [
2
]. The last
decade witnessed a marginal increase in the development and market share of various additive
manufacturing techniques due to their implementation in medical, aerospace, automobile, military
and ornamental industries [
3
]. Additive manufacturing enables direct fabrication of a product from
three-dimensional computer-aided design (CAD) data through successive layer stacking of suitable
material. These techniques are also popularly referred to as 3D printing, rapid prototyping, solid
freeform fabrication, e-manufacturing and digital fabrication. There are more than thirty dierent
additive manufacturing techniques available, which dier on fabrication principle, materials used,
accuracy, bed size, part strength and applicability; however, fused filament fabrication (FFF) is the
most recognized [4].
FFF has acquired considerable attention from researchers, material scientists, innovators and
medical practitioners due to the unique advantages of this technique, such as material flexibility,
minimum environmental degradation, low cost, portability and higher accuracy [
5
]. FFF uses CAD
data as input, which is transferred to slicing software for tool path generation. These tool paths are
coded instructions that control the three-dimensional motion of the nozzle head and build platforms to
create the product within a few hours. The raw material used is a thin wire of thermoplastic material
with a lower melting point and zero toxicity. This wire is pushed by rollers into the nozzle head, where
heaters convert it into semi-molten filament. Finally, the material is deposited by a nozzle moving in
X and Y direction on the build platform, as shown in Figure 1. As one layer is deposited, the build
platform moves downwards (in Z direction), and the next layer of material is deposited. The process is
repeated until the desired part is achieved. Sometimes, another filament of washable material is used
to support the overhanging part, which is easily washed away after fabrication [6].
Polymers 2020, 12, x FOR PEER REVIEW 2 of 19
layer by layer manufacturing in contrast to traditional subtractive and joining techniques [2]. The last
decade witnessed a marginal increase in the development and market share of various additive
manufacturing techniques due to their implementation in medical, aerospace, automobile, military
and ornamental industries [3]. Additive manufacturing enables direct fabrication of a product from
three-dimensional computer-aided design (CAD) data through successive layer stacking of suitable
material. These techniques are also popularly referred to as 3D printing, rapid prototyping, solid
freeform fabrication, e-manufacturing and digital fabrication. There are more than thirty different
additive manufacturing techniques available, which differ on fabrication principle, materials used,
accuracy, bed size, part strength and applicability; however, fused filament fabrication (FFF) is the
most recognized [4].
FFF has acquired considerable attention from researchers, material scientists, innovators and
medical practitioners due to the unique advantages of this technique, such as material flexibility,
minimum environmental degradation, low cost, portability and higher accuracy [5]. FFF uses CAD
data as input, which is transferred to slicing software for tool path generation. These tool paths are
coded instructions that control the three-dimensional motion of the nozzle head and build platforms
to create the product within a few hours. The raw material used is a thin wire of thermoplastic
material with a lower melting point and zero toxicity. This wire is pushed by rollers into the nozzle
head, where heaters convert it into semi-molten filament. Finally, the material is deposited by a
nozzle moving in X and Y direction on the build platform, as shown in Figure 1. As one layer is
deposited, the build platform moves downwards (in Z direction), and the next layer of material is
deposited. The process is repeated until the desired part is achieved. Sometimes, another filament of
washable material is used to support the overhanging part, which is easily washed away after
fabrication [6].
Figure 1. Schematic of the fused filament fabrication (FFF) process used for 3D printing of
thermoplastics.
FFF supplies customized products with the minimum lead time and manufacturing cost, but the
mechanical strength of the part is always a matter of interest for researchers as considerable variation
in mechanical properties is experienced due to variation in design [7]. Moreover, issues related to the
lower mechanical strength of FFF parts may hinder the usability of these products for certain
applications. Thus, there is always a requirement for intelligent optimization tools for the prediction
and maximization of the mechanical strength of FFF parts. There are several input parameters of FFF
technology, which have a significant impact on the tensile strength, compressive strength, flexural
and impact strength of FFF parts [5]. Although numerous studies have been conducted for
optimization of process parameters of FFF, recent studies have focused on the development of
Figure 1.
Schematic of the fused filament fabrication (FFF) process used for 3D printing of thermoplastics.
FFF supplies customized products with the minimum lead time and manufacturing cost, but the
mechanical strength of the part is always a matter of interest for researchers as considerable variation
in mechanical properties is experienced due to variation in design [
7
]. Moreover, issues related to
the lower mechanical strength of FFF parts may hinder the usability of these products for certain
applications. Thus, there is always a requirement for intelligent optimization tools for the prediction
and maximization of the mechanical strength of FFF parts. There are several input parameters
of FFF technology, which have a significant impact on the tensile strength, compressive strength,
flexural and impact strength of FFF parts [
5
]. Although numerous studies have been conducted
for optimization of process parameters of FFF, recent studies have focused on the development of
Polymers 2020,12, 2250 3 of 18
conventional mathematical tools and algorithms which can optimize and forecast the mechanical
strength of FFF parts.
Hambali et al. [
2
] tested the deformation behavior of acrylonitrile butadiene styrene (ABS) material
parts made by FFF technology. The parts were fabricated at dierent orientations, and it was found
that the maximum strength occurred at a 45
angle. The study was validated by finite element
analysis, which predicted the mechanical strength with 95% accuracy. Other studies also indicated
the significant impact of orientation angle [
3
] and infill density [
8
] on the mechanical strength of FFF
parts. Ahn et al. [
7
] investigated the anisotropic behavior of FFF parts and found that the air gap and
raster orientation significantly aected the tensile strength of FFF parts. The cross raster structure
(0
/90
) yielded the maximum tensile strength due to its directional stability, and it was suggested
that a negative air gap be maintained for higher strength and stiness. In addition to conventional
optimization studies, researchers have used mathematical models and algorithms for the analysis of
process parameters. Multi-criteria genetic algorithm was used for the optimization and prediction
of the surface finish and manufacturing time of FFF parts [
4
] and aluminum-based metal matrix
composites [
9
]. The teaching–learning based optimization tool was utilized by Rao and Rai [
1
] to
enhance the sliding wear resistance and strength of FFF parts; the experimental data was compared
with genetic algorithm and advanced quantum-behaved particle swarm optimization algorithm, which
found strong correlation between experimental and computational data. Rayegani and Onwubolu [
5
]
developed a relationship between dierent factors, such as part orientation, raster angle, raster width
and air gap, with tensile strength of FFF parts using dierential evolution techniques along with the
group method of data handling (GMDH). It was deduced that deposition and raster angle defined
the FFF part strength. Moreover, the fine-tuning of input parameters was performed using advanced
tools to achieve maximum mechanical strength. Natarajan et al. [
10
] also implemented a modified
teaching–learning based optimization method (NSMTLBO), referred to as non-dominated sorting type,
to acquire maximum finish and minimum material removal rate.
In another study, a hybrid version of particle swarm technique was applied to FFF input parameters,
such as layer thickness, deposition angle, material type and infill strategy, and for improving the
surface finish, hardness, flexural strength and tensile strength in combination with bacterial foraging
optimization [
11
]. The tool yielded promising results as higher strength (more than 7%) was achieved
using the hybrid optimization technique as compared to conventional algorithms. Recent studies
have reported the impact of orientation angle and infill percentage in mechanical properties of ABS
samples [
12
]. Moreover, in the case of polylactic acid material, the anisotropy of samples had a
significant impact on tensile strength as compared to the orientation angle [
13
]. In addition to
optimization tools, some advanced post-processing techniques, such as annealing and ultrasound
treatment, have been recommended for strength enhancement [14].
Zhang et al. [
15
] found that residual stress and porosity were directly proportional to printing
speed during the fabrication of virgin and composite ABS structures. Moreover, the raster angle of
±
45
manifested lesser shrinkage and porosity, which further influenced dimensional accuracy and
mechanical strength, respectively. Similarly, Rankouhi et al. [
16
] found a major impact of layer thickness
and raster orientation on the mechanical strength of FFF parts, with a reduction in layer thickness
and 0
raster angle strengthening the ABS structures. Another study reported the maximum impact
strength of the ABS part at 0
orientation angle [
17
]. Christiyan et al. [
18
] evaluated the mechanical
strength of magnesium silicate reinforced ABS composites and found maximum tensile and flexural
strength at minimum values of layer thickness and printing speed. The increase in infill density also
enhanced the tensile behavior of ABS test samples [
19
]. The flexural and impact strength of ABS
samples under standard conditions was maximum with thinner layer thickness and 0
deposition
angle [20,21].
The previous studies reported the impact of dierent input parameters on the mechanical
properties of ABS parts made by FFF technology. Although smaller layer thickness is uniformly
recommended by previous literature, the recommended settings of the raster angle, orientation angle
Polymers 2020,12, 2250 4 of 18
and air gap are dierent for each study. These issues are due to conventional optimization algorithms
and mathematical modeling tools used for forecasting optimum parameters, which yield conflicting
results. Moreover, limited studies are performed where advanced optimization algorithms are used
for prediction of the best parametric settings for mechanical strength enhancement. In the present
work, the ecacy of neural network algorithm (NNA) is tested to solve the mechanical strength
issues of FFF parts. As the introduction of advanced optimization algorithms have yielded significant
improvements compared to previous studies performed by regular algorithms, investigating the
optimization problems using this optimization and simulation technique is necessary.
In the next section, a detailed methodology of the formulation and implementation of NNA is
elaborated, along with a selection of objective functions and process parameters. The third section
presents a comparative analysis of dierent algorithms and a discussion of the simulation results.
Finally, the fourth section includes confirmatory experiments that are conducted to validate the
predicted results.
2. Materials and Methods
2.1. Neural Network Optimization Algorithm
Sadollah et al. [
22
] developed a metaheuristics based on neural networks (NNs) named neural
network algorithm (NNA). The authors found that a metaheuristic optimization algorithm could be
used to model artificial NNs (ANNs) to solve optimization problems. NNA was developed using the
ANNs framework to solve optimization problems. NNA is a parameter-free optimization algorithm
where users do not require any algorithm parameters to be adjusted.
ANNs iteratively update the ANN’s weights
wij
to minimize the mean square error and map the
input information according to the target information required at its output. ANNs are designed to
produce new alternatives, where the population’s best search agent is considered as the target and all
searching agents seek the solution through algorithm processes.
NNA draws on the concept and structure of ANN. NNA begins within a search space with an
initial population of randomly generated solutions. Each individual in the population is referred to as
a “pattern solution”, which is a vector of 1 ×D, representing the input data of the NNA.
PatternSolutioni=[xi,1,xi,2 ,xi,3,. . . ,xi,D].
To start the NNA optimization algorithm, within the boundaries of the search space, a pattern
solution matrix Xwith size Npop ×Dis generated randomly, given by:
X=
X1
X2
.
.
.
XNpop
=
x1,1 x1,2 x1,D
x2,1 x2,2 x2,D
...
...
...
xNpop,1 xNpop ,2 xNpop,D
(1)
where:
xij =LBj+randUB jLBj,i=1, 2, . . . ,Npop,j=1, 2, . . . ,D
where LB and UB represent the search space’s lower and upper limits.
Similar to ANNs, every pattern solution
Xi
in NNA will have its corresponding
Wi
weight, where:
Wi=hwi,1,wi,2 ,wi,3,. . . ,wi,Npop iT
Polymers 2020,12, 2250 5 of 18
The weights array Wis given by:
W=hW1,W2,. . . ,Wi,. . . ,WNpop i=
w11 wi1. . . wNpop1
w12 wi2. . . wNpop2
...
...
...
w1Npop wiNpop,2 . . . wNpop Npop
(2)
where
W
represents a matrix of a
Npop ×Npop
randomly distributed number in [0, 1], used to generate
new candidate solutions.
The initial weights in NNA are randomly generated and the network changes their weights
according to the network error propagation as the iteration increases. The weight values are limited,
as the overall weight of this pattern solution must not exceed one and the solution for the weight
pattern is defined as:
wij ∈ U(0, 1),i,j=1, 2, 3, . . . ,Npop (3)
Npop
X
j=1
wij =1, i=1, 2, 3, . . . ,Npop (4)
Such weight-value constraints are used to regulate bias movement and to generate new pattern
solutions. NNA will get trapped in the local optimum without these constraints. The fitness
fi
of
each solution is determined using the corresponding pattern solution
Xi
by evaluating the objective
function fobj.
fi=fobj(Xi) = fob j(xi1,xi2,xi3,. . . ,xiD ),i=1, 2, 3, . . . ,Npop (5)
After fitness calculation for all pattern solutions, the pattern solution with the best fitness is
considered as the target solution, with a target location
XTarget
, target fitness
FTarget
and target weight
WTarget
. The new pattern solution is created using the weight summation technique used in ANNs
as follows:
XNew
j(k+1)=
Npop
X
i=1
wij(k).Xi(k),j=1, 2, 3, . . . ,Npop (6)
Xi(k+1)=Xi(k)+XNew
i(k+1),i=1, 2, 3, . . . ,Npop (7)
where kis an iteration index.
After the generation of the new pattern solutions, the weight matrix is also modified using
the equation:
WUpdated
i(k+1)=Wi(k)+2.rand.WTarget (k)Wi(k),,i=1, 2, 3, . . . ,Npop (8)
where the constraints (3) and (4) during the optimization process must be satisfied.
A bias operator is used to better explore the search space by adjusting a fixed range of pattern
solutions created in the new
Xi(k+1)
population as well as using the updated weight matrix
WUpdated
i(k+1)
. This operator helps the algorithm prevent premature convergence by changing the
few individuals in the population to investigate certain positions in the search space that were not
explored yet.
An adjustment factor
βNNA
is used, using the bias operator, to determine the proportion of pattern
solutions to be transformed. The value of
βNNA
is set to 1 initially, which means all the entities in the
Polymers 2020,12, 2250 6 of 18
population are biased. At each iteration, the
βNNA
value will be reduced adaptively using any possible
reduction technique, such as:
βNNA(k+1)=1 k
Max_iteration!,k=1, 2, 3, . . . ,Max_iteration (9)
βNNA(k+1)=βNNA(k).αNNA,k=1, 2, 3, . . . ,Max_iteration (10)
where
αNNA
is a positive number slightly smaller than 1. The flowchart of the NNA process is given in
Figure 2.
Polymers 2020, 12, x FOR PEER REVIEW 6 of 19
𝛽𝑁𝑁𝐴(𝑘+1)=1𝑘
𝑀𝑎𝑥
_
𝑖𝑡𝑒𝑟𝑎𝑡𝑖𝑜𝑛,𝑘=1,2,3,,𝑀𝑎𝑥
_
𝑖𝑡𝑒𝑟𝑎𝑡𝑖𝑜𝑛 (9)
𝛽𝑁𝑁𝐴(𝑘+1)=𝛽𝑁𝑁𝐴(𝑘).𝛼𝑁𝑁𝐴,𝑘=1,2,3,,𝑀𝑎𝑥
_
𝑖𝑡𝑒𝑟𝑎𝑡𝑖𝑜𝑛 (10)
where 𝛼 is a positive number slightly smaller than 1. The flowchart of the NNA process is given
in Figure 2.
Figure 2. Flowchart adopted during implementation neural network algorithm (NNA) (NFE, number
of function evaluations; TF, transfer function).
Reduction in the value of 𝛽𝑁𝑁𝐴 is done to enhance the algorithm's exploitation during the final
iterations. The transfer function operator is used in NNA (unlike ANNs) to generate better-quality
solutions using the following equation:
𝑋
𝑖(𝑘+1)=𝑇𝐹
𝑋
𝑖(𝑘+1)=
𝑋
𝑖(𝑘+1)+2.𝑟𝑎𝑛𝑑.
𝑋
𝑇𝑎𝑟𝑔𝑒𝑡(𝑘)
𝑋
𝑖(𝑘+1)
𝑖=1,2,3,,𝑁 (11)
In NNA, the bias operator has more chances to produce a new pattern solution at early iterations,
meaning that there are more possibilities to find unvisited pattern solutions. As the number of
iterations increases, the probability of applying the bias operator decreases. The transfer function
(𝑇𝐹) operator has more chances of enhancing the exploitation, particularly at the final iterations.
The generation of a new updated solution in NNA does not depend solely on the preceding
value of that solution but also depends on all the mathematically defined population (known as a
dynamic optimization model), as follows:
Figure 2.
Flowchart adopted during implementation neural network algorithm (NNA) (NFE, number
of function evaluations; TF, transfer function).
Reduction in the value of
βNNA
is done to enhance the algorithm’s exploitation during the final
iterations. The transfer function operator is used in NNA (unlike ANNs) to generate better-quality
solutions using the following equation:
X
i(k+1)=TF(Xi(k+1)) =Xi(k+1)+2.rand.XTarget(k)Xi(k+1)i=1, 2, 3, . . . ,Npop (11)
In NNA, the bias operator has more chances to produce a new pattern solution at early iterations,
meaning that there are more possibilities to find unvisited pattern solutions. As the number of iterations
increases, the probability of applying the bias operator decreases. The transfer function
(TF)
operator
has more chances of enhancing the exploitation, particularly at the final iterations.
Polymers 2020,12, 2250 7 of 18
The generation of a new updated solution in NNA does not depend solely on the preceding value
of that solution but also depends on all the mathematically defined population (known as a dynamic
optimization model), as follows:
Xi(k+1)=f(Xi(k),X(k)),i=1, 2, 3, . . . ,Npop (12)
2.2. Selection of Process Parameters
The materials selected for the present study were ABS commercial-grade P400 (Supplier: Stratasys
Ltd., Eden Prairie, Minnesota, Inc., USA), which was manufactured by the polymerization of styrene
(42–60%) and acrylonitrile (18–35%) in an atmosphere of butadiene (6–30%). The criss-cross bonding
of long-chained butadiene with shorter styrene-acrylonitrile chains yields excellent toughness and
heat resistance. The filament also contained mineral oil, tallow and wax in small concentrations
(0–2%), as reported in the manufacturer’s datasheet [
23
]. The output of the case study performed
by
Panda et al.
[
24
] was considered to identify the optimum parameter settings of the FFF process to
enhance the mechanical properties of the test samples. The five input parameters considered for the
present study and their units were as follows:
x1Layer thickness in mm.
x2Orientation in degrees.
x3Raster angle in degrees.
x4Raster width in mm.
x5Air gap in mm.
The major reason for the selection of the aforementioned process parameters was that previous
studies found a strong correlation of these parameters with the mechanical strength of FFF parts [
7
].
The response parameters used for the present study were tensile strength (TS) in MPa, flexural strength
(FS) in MPa and impact strength (IS) in MJ/m
2
. The primary motivation behind the selection of such
response parameters was that FFF products withstand dierent types of loading conditions when
used for automobile, aerospace and medical applications. Thus, it was obligatory to test each type of
mechanical strength using the advanced optimization tool. These output parameters were taken as
secondary data as reported by Panda et al. [
24
], and were designated as objective functions for the
optimization study.
The objective functions are expressed by Equations (13)–(15) below:
Maximize TS =13.5625 +0.7156x11.3123x2+0.9760x3+0.5183x5+1.1671x2
1
1.3014x2
20.4363x1x3+0.4364x1x40.4364x1x5+0.4364x2x3+0.4898x2x5
0.5389x3x4+0.5389x3x50.5389x4x5
(13)
Maximize FS =29.9178 +0.8719x14.8741x2+2.4251x30.9096x4+1.6626x5
1.7199x1x3+1.7412x1x41.1275x1x5+1.0621x2x5+1.0621x3x5+1.0408x4x5
(14)
Maximize IS =0.401992 +0.034198x1+0.008356x2+0.013673x3+0.021383x2
1
+0.008077x2x4
(15)
The upper and lower limits of these parameters are defined as per machine constraints, and most
commercial FFF printers are capable of fabricating products within the following parameter limits [
24
].
The parameter bounds are expressed by Equations (16)–(20).
0.127 x10.254 (16)
0x230 (17)
0x360 (18)
Polymers 2020,12, 2250 8 of 18
0.4064 x40.5064 (19)
0x50.008 (20)
3. Results and Discussion
NNA was implemented in this work to solve the optimization problem of the FFF process. Because
heuristic algorithms are stochastic optimization methods, to generate meaningful statistical results,
they must be run more than 10 times at minimum.
Each simulation was carried out 30 times for this purpose, with population size varying from 20
to 40, and the maximum amount of iterations ranging from 100 to 500. Grey wolf optimization (GWO),
novel bat algorithm (NBA), dragonfly algorithm (DA), salp swarm algorithm (SSA) and sine cosine
algorithm (SCA) [
25
29
] were selected to check and compare the eciency of NNA. The algorithm
parameter settings used for comparison are shown in Table 1. The value of dierent parameters,
such as the number of particles (NP), dimension size (D), maximum number of iterations (Gmax) and
other algorithm variables (A, f
min
, f
max
, c1, c2, c3, w, s, c, e,
α
,
γ
and
β
) are shown in Table 1. It must be
noted that for each of the algorithms, 20–40 search agents and 100–500 iterations for dierent setups
were used.
Table 1. Parameter settings for dierent algorithms.
Algorithm Parameters
NBA
NP =20–40; D=5; G
max
=100–500;
A
=0.5;
r
=0.5;
α
=γ=0.9; fmin =0; fmax =1.5
GWO NP =20–40; D=5; Gmax =100–500; a=[0,1,2]
DA
SSA
SCA
NP =20–40; D=5; Gmax =100–500; w=[0.4–0.9], s=
0.1, a=0.1, c=0.7, f=1, e=1
NP =20–40; D=5; Gmax =100–500; c1, c2, c3 =[0,1]
NP =20–40; D=5; Gmax =100–500; a=2
NNA NP =20–40; D=5; Gmax =100–500; βNNA =[1–0.05];
αNNA =0.99
The optimum solutions obtained by simulated algorithms are given in Tables 24for TS, FS and IS
estimation, respectively. It is evident from these outcomes that the optimum NNA’s fitness values are
the same as that of competitive algorithms for TS, FS and IS fitness functions.
Table 2.
Optimum solutions obtained by optimization algorithms for tensile strength (TS) estimation
for 500 iterations with population size 40.
Process
Parameters Units NBA GWO DA SSA SCA NNA
x1mm 0.127 0.127 0.127 0.127 0.127 0.127
x2degrees
9.557223075 9.556302041 9.555264322 9.554816312 9.418607509 9.557224815
x3degrees 60 60 60 60 60 60
x4mm 0.4064 0.4064 0.4064 0.4064 0.4064 0.4064
x5mm 0.008 0.008 0.008 0.008 0.008 0.008
y
1.605823622 1.605823622 1.605823622 1.605823622 1.605823622 1.605823622
Polymers 2020,12, 2250 9 of 18
Table 3.
Optimum solutions obtained by optimization algorithms for flexural strength (FS) estimation
for 500 iterations with population size 40.
Process
Parameters Units NBA GWO DA SSA SCA NNA
x1mm 0.127 0.127 0.127 0.127 0.127 0.127
x2degrees 0 0 0 0 0 0
x3degrees 60 60 60 60 60 60
x4mm 0.4064 0.4064 0.4064 0.4064 0.4064 0.4064
x5mm 0.008 0.008 0.008 0.008 0.008 0.008
y-
162.6744472 162.6744472 162.6744472 162.6744472 162.6744472 162.6744472
Table 4.
Optimum solutions obtained by optimization algorithms for impact strength (IS) estimation
for 500 iterations with population size 40.
Process
Parameters Units NBA GWO DA SSA SCA NNA
x1mm 0.254 0.254 0.254 0.254 0.254 0.254
x2degrees 30 30 30 30 30 30
x3degrees 60 60 60 60 60 60
x4mm 0.5064 0.5064 0.5064 0.5064 0.5064 0.5064
x5mm
0.007951741 0.007951741 0.007951741 0.007951741 0.007951741 0.007951741
y-
1.605823622 1.605823622 1.605823622 1.605823622 1.605823622 1.605823622
The performance of the simulated algorithms for FFF over 30 independent runs for 500 iterations
and with a population size of 40 is given in Tables 57for TS, FS and IS estimation, respectively.
The results are presented in terms of best, worst, mean and standard deviation values over 30 runs.
Table 5.
The performance of simulated algorithms for TS estimation over 30 independent runs for
500 iterations and population size 40.
Algorithm Worst Best Average Median Std. Dev.
NBA 174.6234301 174.9214992 174.9016279 174.9214992 7.56 ×102
GWO 174.6234298 174.9214992 174.911563 174.9214991 5.44 ×102
DA 174.6234301 174.9214992 174.8800297 174.9214992 9.72 ×102
SSA 174.6761253 174.9214992 174.8844652 174.9214992 7.32 ×102
SCA 174.6177085 174.9214213 174.7818994 174.8788781 1.44 ×101
NNA 174.9214992 174.9214992 174.9214992 174.9214992 1.89 ×1010
Table 6.
The performance of simulated algorithms for FS estimation over 30 independent runs for
500 iterations and population size 40.
Algorithm Worst Best Average Median Std. Dev.
NBA 162.1491001 162.6744472 162.6044009 162.6744472 1.82 ×101
GWO 162.6744472 162.6744472 162.6744472 162.6744472 5.78 ×1015
DA 162.6278193 162.6744472 162.6727806 162.6744472 8.51 ×103
SSA 162.6744472 162.6744472 162.6744472 162.6744472 5.78 ×1014
SCA 162.6744472 162.6744472 162.6744472 162.6744472 5.78 ×1014
NNA 162.6744472 162.6744472 162.6744472 162.6744472 5.78 ×1014
Polymers 2020,12, 2250 10 of 18
Table 7.
The performance of simulated algorithms for IS estimation over 30 independent runs for
500 iterations and population size 40.
Algorithm Worst Best Average Median Std. Dev.
NBA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
GWO 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
DA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
SSA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
SCA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
NNA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
The convergence rate of NNA, as shown in Figure 3, is also better than the competitive algorithms.
The convergence characteristics for TS, FS and IS estimation are shown in Figure 3. It is found that for
TS estimation, GWO, DA, SSA and NBA algorithms achieve optimum value, but overall, NNA has the
capability to identify the best value with the least standard deviation. For FS estimation, GWO, SSA,
SCA and NNA have the potential to reach global optimal solutions with the same standard deviation.
For IS estimation, all algorithms are able to reach a near global optimal solution with the same standard
deviation. Overall, the result demonstrates that NNA’s mean, median and standard deviation values
are much better compared to competitive algorithms, which proves the enhanced exploration and
exploitation capabilities of NNA for process parameter optimization of FFF, especially in TS estimation.
The box plots for TS, FS and IS evaluations are displayed in Figure 4a–c, respectively, for NBA,
GWO, DA, SSA, SCA and NNA. These box plots are an excellent medium to ascertain the fitness values
and compare the eectiveness of dierent algorithms. For TS estimation, it is clear that NNA yields
the best fitness values as mean and median, along with the least standard deviation. As compared
to other rival algorithms, the NNA performs better, and these results can be used for the fabrication
of commercial products using ABS. From the box plots of FS estimation, it is clear that GWO, SSA,
SCA and NNA have similar mean and median fitness values, and their standard deviation values are
better in comparison to NBA and DA. Thus, the overall performance of GWO, SSA, SCA and NNA is
found to be better for FS estimation compared to other algorithms. From the box plots of IS estimation,
it is clear that all optimization algorithms have similar mean and median fitness values, their standard
deviation values are almost the same, and the algorithms reach optimum values.
Further, the performance comparison of metaheuristic algorithms for FFF over 30 independent
runs for 100 iterations and population size 40 are given in Tables A1A3 (see Appendix A) for TS,
FS and IS estimation, respectively. The eect of the reduced total number of iterations clearly shows
that NNA’s mean, median and standard deviation values for TS estimation are much better compared
to competitive algorithms, which proves that NNA is able to find an optimum solution with reduced
iterations for process parameter optimization of FFF. Similarly, for FS estimation, SSA, SCA and NNA
are able to achieve global optimum with equal authority. However, in IS estimation, each algorithm
shows a similar eect to achieve the optimum value.
The performance of optimization algorithms for FFF process parameter estimation was evaluated
over 30 independent runs for 100 iterations with population size 20. The results are given in
Tables A4A6 (see Appendix A) for TS, FS and IS estimation, respectively, which demonstrate that
NNA’s mean, median and standard deviation values are much better compared to competitive
algorithms for TS and FS estimation. This proves the enhanced exploration and exploitation capabilities
of NNA for the process parameter optimization of FFF. However, each algorithm is able to achieve
the optimum value for IS estimation, independent of the population size of variables and the given
number of iterations. The maximum values of TS, FS and IS can be achieved by manufacturing FFF
parts using the process parameters predicted by the NNA algorithm.
Polymers 2020,12, 2250 11 of 18
Polymers 2020, 12, x FOR PEER REVIEW 11 of 19
Figure 3. The convergence graph of simulated algorithm estimations over 30 independent runs for
500 iterations and population size 40 for (a) TS (b) FS and (c) IS.
Figure 3.
The convergence graph of simulated algorithm estimations over 30 independent runs for
500 iterations and population size 40 for (a) TS (b) FS and (c) IS.
Polymers 2020,12, 2250 12 of 18
Polymers 2020, 12, x FOR PEER REVIEW 12 of 19
Figure 4. The box plot of simulated algorithm estimations over 30 independent runs for 500 iterations
and population size 40 for (a) TS (b) FS and (c) IS.
The performance of optimization algorithms for FFF process parameter estimation was
evaluated over 30 independent runs for 100 iterations with population size 20. The results are given
in Tables A4–A6 (see Appendix A) for TS, FS and IS estimation, respectively, which demonstrate that
NNA's mean, median and standard deviation values are much better compared to competitive
algorithms for TS and FS estimation. This proves the enhanced exploration and exploitation
capabilities of NNA for the process parameter optimization of FFF. However, each algorithm is able
to achieve the optimum value for IS estimation, independent of the population size of variables and
Figure 4.
The box plot of simulated algorithm estimations over 30 independent runs for 500 iterations
and population size 40 for (a) TS (b) FS and (c) IS.
The optimum parameter settings to achieve maximum tensile strength of FFF parts are layer
thickness of 0.127 mm, 9.55
orientation angle, 60
raster angle, 0.4064 mm raster width and 0.008 mm
air gap. On the other hand, flexural strength can be maximized at 0.127 mm layer thickness, 0
orientation angle, 60
raster angle, 0.4064 mm raster width and 0.008 mm air gap. In order to obtain
the maximum value of impact strength, the FFF process must be carried out at a layer thickness of
Polymers 2020,12, 2250 13 of 18
0.254 mm, 30
orientation angle, 60
raster angle, 0.5064 mm raster width and 0.0079 mm air gap.
It can be observed that a raster angle of 60
results in the maximization of all three parameters of
mechanical strength, i.e., tensile, flexural and impact strength. Thus, it is recommended to use the
default setting of 60
raster angle for the best mechanical stability of FFF parts. In the case of layer
thickness, a lower value results in better impact and tensile strength, while the opposite phenomenon
is predicted for flexural strength. The air gap is also found to be constant for mechanical stability.
In the case of orientation angle, the horizontal layer deposition (0
) strategy is the most successful in
attaining maximum flexural strength, while 9.55
and 30
result in better tensile and impact strength,
respectively, as suggested by NNA.
4. Confirmatory Experiments
The optimum parameters for each output suggested by NNA must be validated before
recommendation; hence, confirmatory experiments were conducted. The ABS samples were prepared
using commercial P400 material (supplied by Stratasys Inc. Ltd., Eden Prairie, Minnesota, USA) and
an open-source FFF printer (supplied by Prusa Research, Prague, Czech Republic). The technical data
sheet [
23
] of thermoplastic materials was referred to when deciding printing parameters. The three
samples were prepared for tensile strength, flexural strength and impact strength using ASTM D638,
ASTM D790 and ASTM D256, respectively. The test samples were prepared at a constant printing speed,
infill method, extrusion temperature, bed temperature and environmental temperature, as shown in
Table 8.
Table 8. Fixed parameters used for confirmatory experiments.
S. No. Fixed Parameter Value
1 Extrusion temperature 230 C
2 Bed temperature 80 C
3 Ambient temperature 25 C
4 Infill type Rectilinear
5 Printing speed 50 mm/s
Fixed values of temperature were used to eliminate error as the significant impact of temperature
on the bonding strength of ABS layers has been noticed. As the semi-molten thermoplastic beads
are laid down, excessive temperature at the nozzle may increase flowability and cause deformation.
On the other hand, faster cooling may lead to shrinkage, which would exhibit dimensional variation.
Thus, the optimum values of temperature were selected as the thermoplastic polymer layers must
be cooled slowly to avoid deformation and dimensional variability [
19
]. The printing and testing of
samples was performed under controlled environmental conditions and using three replications of
each experiment to avoid random errors (Figure 5).
The variable parameters used for fabrication and experimentation are shown in Table 9. The values
of each parameter are taken from predictions made by NNA for tensile strength, flexural strength
and impact strength. It can be observed that there is minimal variation between the predicted
and experimental value of each output. The maximum variation of 2.92% occurs in sample no. 3,
which undergoes impact testing, and the minimum variation (1.3%) occurs in sample no. 2 during
flexural testing. The average variation of three experiments is 1.94%, which is much less compared to
previous studies. Hence, the NNA predicted results are validated with high accuracy, meaning it is
suitable for solving optimization issues in manufacturing processes.
The NNA results achieved in the present study are compared with the percentage error
(experimental vs. predicted) of previous studies in Figure 6. The response parameter and prediction
tool used by each author is also indicated in the graph. It can be observed that the minimum variation
between experimental and predicted results is achieved in the present study as compared to previous
studies that implemented advanced optimization algorithms. The maximum variation was found for
Polymers 2020,12, 2250 14 of 18
finite element analysis (FEA) [
2
], while the average error was found using artificial neural networks
(ANN) [30] and response surface methodology (RSM) [31].
Polymers 2020, 12, x FOR PEER REVIEW 14 of 19
(a) (b)
Figure 5. (a) FFF printer used for experimentation (b) Universal Testing Machine used for mechanical
testing.
The variable parameters used for fabrication and experimentation are shown in Table 9. The
values of each parameter are taken from predictions made by NNA for tensile strength, flexural
strength and impact strength. It can be observed that there is minimal variation between the
predicted and experimental value of each output. The maximum variation of 2.92% occurs in sample
no. 3, which undergoes impact testing, and the minimum variation (1.3%) occurs in sample no. 2
during flexural testing. The average variation of three experiments is 1.94%, which is much less
compared to previous studies. Hence, the NNA predicted results are validated with high accuracy,
meaning it is suitable for solving optimization issues in manufacturing processes.
Table 9. Variable parameters and results.
Sample
No.
Laye
r
Thickness
(mm)
Orientation
Angle
(°)
Raste
r
Angle
(°)
Raste
r
Width
(mm)
Ai
r
Gap
(mm)
Predicted
Value
Experimental
Value
Error
(%)
1 0.127 9.55 60 0.4064 0.008 17.49 17.23 1.61
2 0.127 0 60 0.4064 0.008 16.26 16.05 1.3
3 0.254 30 60 0.5064 0.0079 2.46 2.39 2.92
The NNA results achieved in the present study are compared with the percentage error
(experimental vs. predicted) of previous studies in Figure 6. The response parameter and prediction
tool used by each author is also indicated in the graph. It can be observed that the minimum variation
between experimental and predicted results is achieved in the present study as compared to previous
studies that implemented advanced optimization algorithms. The maximum variation was found for
finite element analysis (FEA) [2], while the average error was found using artificial neural networks
(ANN) [30] and response surface methodology (RSM) [31].
The comparative analysis indicates the higher efficacy of NNA as compared to other
optimization algorithms and prediction tools. The predicted parameters can be recommended for
commercial production of functional prototypes and end-use components manufactured using FFF.
Moreover, the NNA must be implemented to solve surface roughness and dimensional variability
issues of FFF technology when used for complex designs.
Figure 5.
(
a
) FFF printer used for experimentation (
b
) Universal Testing Machine used for
mechanical testing.
Table 9. Variable parameters and results.
Sample
No.
Layer
Thickness (mm)
Orientation
Angle ()
Raster
Angle ()
Raster
Width (mm)
Air Gap
(mm)
Predicted
Value
Experimental
Value
Error
(%)
1 0.127 9.55 60 0.4064 0.008 17.49 17.23 1.61
2 0.127 0 60 0.4064 0.008 16.26 16.05 1.3
3 0.254 30 60 0.5064 0.0079 2.46 2.39 2.92
Polymers 2020, 12, x FOR PEER REVIEW 15 of 19
Figure 6. Comparison of percentage error of NNA results with previous studies.
5. Conclusions
Despite an efficient additive manufacturing technique, mechanical strength is one of the major
obstructions against the applicability of FFF polymer products for end-use components. Thus, the
selection of optimum parameter settings is required to obtain the best mechanical properties of ABS
parts. The optimization was conducted employing NBA, GWO, DA, SSA, SCA and NNA algorithms
for the maximization of mechanical strength (TS, FS and IS) of parts with varying population size
and number of iterations. From the statistical and convergence results, it was found that the NNA
algorithm was highly competitive with respect to NBA, GWO, DA, SSA and SCA algorithms and was
able to achieve optimum results for FFF products. The endorsed input parameters indicated the
manifestation of enhanced mechanical strength among the test parts. In the future, researchers can
also implement a hybrid metaheuristic algorithm for the enhancement of quality and accuracy along
with a reduction in convergence time.
Author Contributions: Conceptualization, J.S.C., N.M., R.K., S.S. (Sandeep Singh), S.S. (Shubham Sharma), J.S.,
K.V.R., M.M., D.Y.P.; methodology, J.S.C., N.M., R.K., S.S. (Sandeep Singh), S.S. (Shubham Sharma) and M.M.;
investigation, J.S.C., N.M., R.K., S.S. (Sandeep Singh), S.S. (Shubham Sharma); resources, J.S.C., N.M., R.K., S.S.
(Sandeep Singh), S.S. (Shubham Sharma), J.S., K.V.R., M.M., D.Y.P. and S.P.D; writing—original draft
preparation, J.S.C., N.M., R.K., S.S. (Sandeep Singh), S.S. (Shubham Sharma) and M.M.; writing—review and
editing, J.S.C., N.M., R.K., S.S. (Sandeep Singh), S.S. (Shubham Sharma); supervision, J.S.C., N.M., R.K., S.S.
(Sandeep Singh), S.S. (Shubham Sharma) and M.M.; project administration, J.S.C., N.M., R.K., S.S. (Sandeep
Singh), S.S. (Shubham Sharma), J.S., K.V.R., M.M., D.Y.P. and S.P.D. All authors have read and agreed to the
published version of the manuscript.
Acknowledgments: The authors wish to acknowledge the resources of IK Gujral Punjab Technical University,
Main Campus, Kapurthala and Chandigarh University, Mohali for carrying out the research work.
Funding: This research received no external funding.
Conflicts of Interest: The authors declare no conflict of interest.
Appendix A
Table A1. The performance of simulated algorithms for TS estimation over 30 independent runs for
100 iterations and population size 40.
Algorithm Worst Best Average Median Std. Dev.
Figure 6. Comparison of percentage error of NNA results with previous studies.
Polymers 2020,12, 2250 15 of 18
The comparative analysis indicates the higher ecacy of NNA as compared to other optimization
algorithms and prediction tools. The predicted parameters can be recommended for commercial
production of functional prototypes and end-use components manufactured using FFF. Moreover,
the NNA must be implemented to solve surface roughness and dimensional variability issues of FFF
technology when used for complex designs.
5. Conclusions
Despite an ecient additive manufacturing technique, mechanical strength is one of the major
obstructions against the applicability of FFF polymer products for end-use components. Thus,
the selection of optimum parameter settings is required to obtain the best mechanical properties of ABS
parts. The optimization was conducted employing NBA, GWO, DA, SSA, SCA and NNA algorithms
for the maximization of mechanical strength (TS, FS and IS) of parts with varying population size
and number of iterations. From the statistical and convergence results, it was found that the NNA
algorithm was highly competitive with respect to NBA, GWO, DA, SSA and SCA algorithms and
was able to achieve optimum results for FFF products. The endorsed input parameters indicated the
manifestation of enhanced mechanical strength among the test parts. In the future, researchers can
also implement a hybrid metaheuristic algorithm for the enhancement of quality and accuracy along
with a reduction in convergence time.
Author Contributions:
Conceptualization, J.S.C., N.M., R.K., S.S. (Sandeep Singh), S.S. (Shubham Sharma),
J.S., K.V.R., M.M., D.Y.P.; methodology, J.S.C., N.M., R.K., S.S. (Sandeep Singh), S.S. (Shubham Sharma) and
M.M.; investigation, J.S.C., N.M., R.K., S.S. (Sandeep Singh), S.S. (Shubham Sharma); resources, J.S.C., N.M.,
R.K., S.S. (Sandeep Singh), S.S. (Shubham Sharma), J.S., K.V.R., M.M., D.Y.P. and S.P.D.; writing—original draft
preparation, J.S.C., N.M., R.K., S.S. (Sandeep Singh), S.S. (Shubham Sharma) and M.M.; writing—review and
editing, J.S.C., N.M., R.K., S.S. (Sandeep Singh), S.S. (Shubham Sharma); supervision, J.S.C., N.M., R.K., S.S.
(Sandeep Singh), S.S. (Shubham Sharma) and M.M.; project administration, J.S.C., N.M., R.K., S.S. (Sandeep Singh),
S.S. (Shubham Sharma), J.S., K.V.R., M.M., D.Y.P. and S.P.D. All authors have read and agreed to the published
version of the manuscript.
Funding: This research received no external funding.
Acknowledgments:
The authors wish to acknowledge the resources of IK Gujral Punjab Technical University,
Main Campus, Kapurthala and Chandigarh University, Mohali for carrying out the research work.
Conflicts of Interest: The authors declare no conflict of interest.
Appendix A
Table A1.
The performance of simulated algorithms for TS estimation over 30 independent runs for
100 iterations and population size 40.
Algorithm Worst Best Average Median Std. Dev.
NBA 174.6234301 174.9214992 174.9011405 174.921458 7.52 ×102
GWO 174.623429 174.9214992 174.8620095 174.9214386 1.05 ×101
DA 174.6234301 174.9214992 174.8394289 174.9201719 1.18 ×101
SSA 174.6324965 174.9214992 174.8028863 174.8169921 9.99 ×102
SCA 174.4505185 174.9200412 174.6821489 174.6282199 1.24 ×101
NNA 174.8553077 174.9214939 174.9100097 174.9200154 1.98 ×102
Polymers 2020,12, 2250 16 of 18
Table A2.
The performance of simulated algorithms for FS estimation over 30 independent runs for
100 iterations and population size 40.
Algorithm Worst Best Average Median Std. Dev.
NBA 162.1491001 162.6744472 162.5693778 162.6744472 2.14 ×101
GWO 162.674447 162.6744472 162.6744472 162.6744472 3.38 ×108
DA 162.6627456 162.6744472 162.6740571 162.6744472 2.14 ×103
SSA 162.6744472 162.6744472 162.6744472 162.6744472 5.78 ×1014
SCA 162.6744472 162.6744472 162.6744472 162.6744472 5.78 ×1014
NNA 162.6744472 162.6744472 162.6744472 162.6744472 5.78 ×1014
Table A3.
The performance of simulated algorithms for IS estimation over 30 independent runs for
100 iterations and population size 40.
Algorithm Worst Best Average Median Std. Dev.
NBA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
GWO 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
DA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
SSA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
SCA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
NNA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
Table A4.
The performance of simulated algorithms for TS estimation over 30 independent runs for
100 iterations and population size 20.
Algorithm Worst Best Average Median Std. Dev.
NBA 174.6234263 174.9214936 174.8372247 174.9196852 1.27 ×101
GWO 174.6242344 174.921498 174.8359942 174.9200114 1.11 ×101
DA 171.631808 174.9214992 174.1513635 174.6316391 1.04
SSA 174.5267321 174.9214992 174.7720952 174.7517588 1.25 ×101
SCA 174.1631947 174.921241 174.5762605 174.5885636 2.10 ×101
NNA 174.6348368 174.9214885 174.8893958 174.9163291 6.93 ×102
Table A5.
The performance of simulated algorithms for FS estimation over 30 independent runs for
100 iterations and population size 20.
Algorithm Worst Best Average Median Std. Dev.
NBA 162.1491001 162.6744472 162.6044009 162.6744472 1.82 ×101
GWO 162.6734733 162.6744472 162.6743693 162.6744472 2.15 ×104
DA 162.4635226 162.6744472 162.6453716 162.6744472 4.66 ×102
SSA 162.6744472 162.6744472 162.6744472 162.6744472 5.78 ×1014
SCA 162.1491001 162.6744472 162.6569356 162.6744472 9.59 ×102
NNA 162.6744472 162.6744472 162.6744472 162.6744472 5.78 ×1014
Table A6.
The performance of simulated algorithms for IS estimation over 30 independent runs for
100 iterations and population size 20.
Algorithm Worst Best Average Median Std. Dev.
NBA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
GWO 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
DA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
SSA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
SCA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
NNA 1.605823622 1.605823622 1.605823622 1.605823622 2.26 ×1016
Polymers 2020,12, 2250 17 of 18
References
1.
Rao, R.; Rai, D. Optimization of fused deposition modeling process using teaching-learning-based
optimization algorithm. Eng. Sci. Technol. Int. J. 2016,19, 587–603. [CrossRef]
2.
Hambali, R.; Smith, P.; Rennie, A. Determination of the eect of part orientation to the strength value on
additive manufacturing FFF for end-use parts by physical testing and validation via three-dimensional finite
element analysis. Int. J. Mat. Eng. Innov. 2012,3, 269–281. [CrossRef]
3.
Kim, G.; Oh, Y. A benchmark study on rapid prototyping processes and machines: Quantitative comparisons
of mechanical properties, accuracy, roughness, speed, and material cost. Proc. Inst. Mech. Eng. B J. Eng.
Manuf. 2008,222, 201–215. [CrossRef]
4.
Pandey, P.; Thrimurthulu, K.; Reddy, N. Optimal part deposition orientation in FFF by using a multicriteria
genetic algorithm. Int. J. Prod. Res. 2004,42, 4069–4089. [CrossRef]
5.
Rayegani, F.; Onwubolu, G. Fused deposition modelling (FFF) process parameter prediction and optimization
using group method for data handling (GMDH) and dierential evolution (DE). Int. J. Adv. Manuf. Technol.
2014,73, 509–519. [CrossRef]
6.
Rane, R.; Kulkarni, A.; Prajapati, H.; Taylor, R.; Jain, A.; Chen, V. Post-process eects of isothermal annealing
and initially applied static uniaxial loading on the ultimate tensile strength of fused filament fabrication
parts. Materials 2020,13, 352. [CrossRef]
7.
Ahn, S.; Montero, M.; Odell, D.; Roundy, S.; Wright, P. Anisotropic material properties of fused deposition
modeling ABS. Rapid. Prototyp. J. 2002,8, 248–257. [CrossRef]
8.
Srinivasan, R.; Babu, B.; Annamalai, P.; Sivasankaran, S.; Tharaan, M. An experimental on the influence of
process parameters on the tensile properties of carbon fiber reinforced PLA FFF parts. TEST Eng. Manag.
2019,81, 4606–4611.
9.
Rao, V.; Geethika, V.; Krishnaveni, P. Multi-objective optimization of mechanical properties of aluminium
7075-based hybrid metal matrix composite using genetic algorithm. In Advances in 3D Printing and Additive
Manufacturing Technologies, 1st ed.; Kumar, L., Ed.; Springer: Singapore, 2017.
10.
Natarajan, E.; Kaviarasan, V.; Lim, W.; Tiang, S.; Parasuraman, S.; Elango, S. Non-dominated sorting modified
teaching–learning-based optimization for multi-objective machining of polytetrafluoroethylene (PTFE).
J. Intell. Manuf. 2020,31, 911–935. [CrossRef]
11.
Raju, M.; Gupta, M.; Bhanot, N.; Sharma, V. A hybrid PSO–BFO evolutionary algorithm for optimization of
fused deposition modelling process parameters. J. Intell. Manuf. 2019,30, 2743–2758. [CrossRef]
12.
Garc
í
a-Dom
í
nguez, A.; Claver, J.; Camacho, A.M.; Sebasti
á
n, M.A. Considerations on the applicability of
test methods for mechanical characterization of materials manufactured by FDM. Materials
2020
,13, 28.
[CrossRef] [PubMed]
13.
Yao, T.; Ye, J.; Deng, Z.; Zhang, K.; Ma, Y.; Ouyang, H. Tensile failure strength and separation angle of FDM
3D printing PLA material: Experimental and theoretical analyses. Compos. Part B Eng.
2020
,188, 107894.
[CrossRef]
14.
Wickramasinghe, S.; Do, T.; Tran, P. FDM-based 3D printing of polymer and associated composite: A review
on mechanical properties, defects and treatments. Polymers 2020,12, 1529. [CrossRef] [PubMed]
15.
Zhang, W.; Wu, A.S.; Sun, J.; Quan, Z.; Gu, B.; Sun, B.; Cotton, C.; Heider, D.; Chou, T.-W. Characterization
of residual stress and deformation in additively manufactured ABS polymer and composite specimens.
Comp. Sci. Technol. 2017,150, 102–110. [CrossRef]
16.
Rankouhi, B.; Javadpour, S.; Delfanian, F.; Letcher, T. Failure analysis and mechanical characterization of 3D
printed ABS with respect to layer thickness and orientation. J. Fail. Anal. Prev.
2016
,16, 467–481. [CrossRef]
17.
G
ó
rski, F.; Kuczko, W.; Wichniarek, R. Impact strength of ABS parts manufactured using Fused Deposition
Modeling technology. Arch. Mech. Technol. Autom. 2014,31, 3–12.
18.
Christiyan, K.J.; Chandrasekhar, U.; Venkateswarlu, K. A study on the influence of process parameters on the
Mechanical Properties of 3D printed ABS composite. IOP Conf. Mater. Sci. Eng. 2016,114, 1–8. [CrossRef]
19.
Rodr
í
guez-Panes, A.; Claver, J.; Camacho, A.M. The influence of manufacturing parameters on the mechanical
behaviour of PLA and ABS pieces manufactured by FDM: A comparative analysis. Materials
2018
,11, 1333.
[CrossRef]
20.
Shubham, P.; Sikidar, A.; Chand, T. The influence of layer thickness on mechanical properties of the 3D
printed ABS polymer by fused deposition modeling. Key Eng. Mater. 2016,706, 636–637. [CrossRef]
Polymers 2020,12, 2250 18 of 18
21.
Vidakis, N.; Petousis, M.; Vairis, A.; Savvakis, K.; Maniadi, A. A parametric determination of bending and
Charpy’s impact strength of ABS and ABS-plus fused deposition modeling specimens. Prog. Add. Manuf.
2019,4, 323–330. [CrossRef]
22.
Sadollah, A.; Sayyaadi, H.; Yadav, A. A dynamic metaheuristic optimization model inspired by biological
nervous systems: Neural network algorithm. Appl. Soft Comput. 2018,71, 747–782. [CrossRef]
23.
Available online: http://groups.csail.mit.edu/drl/wiki/images/3/3b/MSDS_P400ABSModel_1062730A.pdf
(accessed on 27 January 2020).
24.
Panda, S.; Padhee, S.; Anoop, S.; Mahapatra, S. Optimization of fused deposition modelling (FFF) process
parameters using bacterial foraging technique. Intell. Inf. Manag. 2009,1, 89.
25.
Meng, X.; Gao, X.; Liu, Y.; Zhang, H. A novel bat algorithm with habitat selection and Doppler eect in
echoes for optimization. Exp. Syst. Appl. 2015,42, 6350–6364. [CrossRef]
26. Mirjalili, S.; Mirjalili, S.; Lewis, A. Grey wolf optimizer. Adv. Eng. Soft 2014,69, 46–61. [CrossRef]
27.
Mirjalili, S. Dragonfly algorithm: A new meta-heuristic optimization technique for solving single-objective,
discrete, and multi-objective problems. Neural Comput. Appl. 2015,27, 1053–1073. [CrossRef]
28.
Mirjalili, S.; Gandom, A.; Mirjalili, S.; Saremi, S.; Faris, H.; Mirjalili, S. Salp Swarm Algorithm: A bio-inspired
optimizer for engineering design problems. Adv. Eng. Soft 2017,114, 163–191. [CrossRef]
29.
Mirjalili, S. SCA: A Sine cosine algorithm for solving optimization problems. Know Base Syst.
2016
,96, 120–133.
[CrossRef]
30.
Boschetto, A.; Giordano, V.; Veniali, F. Surface roughness prediction in fused deposition modelling by neural
networks. Int. J. Adv. Manuf. Technol. 2013,67, 2727–2742. [CrossRef]
31.
Samykano, M.; Selvamani, S.K.; Kadirgama, K.; Ngui, W.K.; Kanagaraj, G.; Sudhakar, K. Mechanical property
of FDM printed ABS: Influence of printing parameters. Int. J. Adv. Manuf. Technol.
2019
,102, 2779–2796.
[CrossRef]
©
2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).
... Operation conditions and material determine the optimal thickness of each unfolding powder sheet. Powder Bed Fusion (PBF) is also known as Electron Beam Melting (EBM), Selective Laser Sintering (SLS), Direct Metal Laser Melting (DMLM), SLM, and DMLS [32]. ...
Article
Polymer composite prototypes and final products have been produced using additive manufacturing (AM) or three-dimensional printing (3DP). Due to their exceptional physical, chemical, mechanical, and structural properties, composite or reinforced materials, particularly polymers, are gaining popularity in the fields of engineering and science. High-performance printable polymer composites are urgently required due to the mechanical and functional limitations of printed pure polymer components. Customizable geometry, reduced waste, and accessible material shifting are among the benefits of 3D printing. The market expansion of young industrialists may be restricted by the necessity to recognize the potential of 3DP technology. Furthermore, the literature on this subject must be more specialized or cohesive in order to expand beyond the fundamentals for professionals. This article explores the application of 3D printing matrix composite materials with enhanced properties in the fields of electronics, medical, cars, and aircraft. FDM, PLP, DLP, SLA, SLS, and robocasting are among the AM processes that are addressed in this document. Additional applications for these materials are recommended. Finally, the advantages and disadvantages of 3D printing are highlighted in order to facilitate enhancements.
... The increased adhesion between layers helps prevent cracks from spreading and increases the overall strength of the printed object. The results of this study are comparable to previous studies [3,[53][54][55][56][57][58]. ...
Article
Full-text available
This study investigates and optimizes the tensile, flexural and compressive strengths of polymer parts for local and small industrial components. Extrusion-based additive manufacturing, i.e., fused deposition modeling (FDM) technology is used to manufacture the parts from acrylic butadiene styrene (ABS) material. The samples are produced according to selected process parameters such as road angle, filling percentage, layer size, printing temperature and printing speed, which are varied at three levels. The experiments are designed using a central composite design based on response surface methodology (RSM) in MINITAB software. Mechanical testing is performed using universal testing machines and data are collected for statistical analysis. In addition, hybrid approaches based on artificial intelligence are used for parameter optimization to achieve maximum tensile, bending and compressive strengths. Tensile and bending samples are also subjected to fracture mechanism investigation using scanning electron microscopy (SEM). The results show that the first technique, i.e., response surface methodology and genetic algorithm, resulted in an improvement in tensile, flexural and compressive strength by 2.5%, 7.58% and 8.86%, respectively, compared to the highest values of all experiments. The second approach, i.e., the genetic algorithm for artificial neural networks, provides the tensile, bending and compressive strengths with an improvement of 3.74%, 2.04% and 5.49%, respectively. Similarly, the third technique, adaptive neuro-fuzzy inference system genetic algorithm, yields 4.97%, 6.16%, and 2.62% improvement in tensile, flexural, and compressive strength, respectively. Overall, the results show that the hybrid optimization techniques can provide the desired mechanical strengths of FDM parts for selected applications. The optimized results are confirmed by experiments and scientific methods.
... Through advanced computational multi-objective optimization, we achieved superior mechanical properties, an ultra-smooth surface, and a consistent model design [85,86]. To refine laser settings, one could leverage both the minimum and peak values ascertained from this computational method. ...
... It gave the desirability of 0.68. Although we could get minimum Ra, T and E at other points, since we had to make only one sample for further post-processing, thus we chose this optimization where we could get the balanced optimized sample with enhanced mechanical properties and better surface quality [78,79]. This result supports the discussion on the input parameters' effect on the responses [80,81]. ...
... 3D printing enables the near net shape manufacture of customized products [1][2][3] and potential to improve sustainability metrics in manufacturing [4,5]; extrusion of thermoplastics represents one of the least expensive modes of additive manufacture [6,7], especially with respect to initial capital investment. However, the surface finish [8] and mechanical properties from material extrusion additive manufacturing (MEAM) of thermoplastics tend to be inferior to traditional plastic manufacture [9,10]. Significant efforts have gone into the optimization of print conditions [11,12] and formulating plastics [13][14][15] to improve mechanical performance. ...
Article
Full-text available
Non-destructive characterization of 3D-printed parts is critical for quality control and adoption of additive manufacturing (AM). The low-cost driver for AM of thermoplastics, typically through material extrusion AM (MEAM), challenges the integration of real-time, operando characterization and control schemes that have been developed for metals. Here, we demonstrate that the surface topology determined from optical profilometry provides information about the mechanical response of the printed part using commercial ABS filaments through calibration-based correlations. The influence of layer thickness on the tensile properties of MEAM ABS was examined. Surface topology was converted into amplitude spectra using fast Fourier transforms. The scatter in the tensile strength of the replicate samples was well represented by the differences in the amplitude of the two fundamental waves that describe the periodicity of the printed roads. These results suggest that information about previously printed layers is transferred to subsequent layers that can be resolved from optical profilometry and offers the potential of a rapid, non-destructive post-print characterization for improved quality control.
Article
Purpose Polymethyl methacrylate (PMMA) is a remarkable biocompatible material for bone cement and regeneration. It is also considered 3D printable but requires in-depth process–structure–properties studies. This study aims to elucidate the mechanistic effects of processing parameters and sterilization on PMMA-based implants. Design/methodology/approach The approach comprised manufacturing samples with different raster angle orientations to capitalize on the influence of the filament alignment with the loading direction. One sample set was sterilized using an autoclave, while another was kept as a reference. The samples underwent a comprehensive characterization regimen of mechanical tension, compression and flexural testing. Thermal and microscale mechanical properties were also analyzed to explore the extent of the appreciated modifications as a function of processing conditions. Findings Thermal and microscale mechanical properties remained almost unaltered, whereas the mesoscale mechanical behavior varied from the as-printed to the after-autoclaving specimens. Although the mechanical behavior reported a pronounced dependence on the printing orientation, sterilization had minimal effects on the properties of 3D printed PMMA structures. Nonetheless, notable changes in appearance were attributed, and heat reversed as a response to thermally driven conformational rearrangements of the molecules. Originality/value This research further deepens the viability of 3D printed PMMA for biomedical applications, contributing to the overall comprehension of the polymer and the thermal processes associated with its implementation in biomedical applications, including personalized implants.
Article
Additive manufacturing (AM) is a preferred industrial manufacturing method for modeling and rapid prototyping of physical systems. The final product in AM must have appropriate mechanical properties, such as flexural strength and be of good quality. The selection of printing parameters is essential for this reason. In this study, three critical printing parameters, such as layer thickness (100-200-300 µm), raster angle (0-30-60°), and infill density (40-60-80%) were examined. The analysis of variance method was used to look at the relationship between these parameters and the flexure strength of samples fabricated using the fused deposition modeling technique with polyethylene terephthalate glycol material. The experimental design process was performed using Taguchi L9 orthogonal design. Fuzzy logic-based modeling was applied to estimate the flexural strength. The results demonstrated that the infill density is the most important parameter affecting flexural strength compared to the other parameters. The highest strength of 57.76 MPa was achieved when the layer thickness, raster angle, and infill density were set to 100 µm, 60°, and 80%, respectively. The fuzzy logic provided a high-accuracy estimation of the flexural strength with a maximum percentage error of 2.65%. Consequently, it was determined that the model and experimental results were in agreement.
Article
Full-text available
Fused deposition modelling (FDM) is one of the fastest-growing additive manufacturing methods used in printing fibre-reinforced composites (FRC). The performances of the resulting printed parts are limited compared to those by other manufacturing methods due to their inherent defects. Hence, the effort to develop treatment methods to overcome these drawbacks has accelerated during the past few years. The main focus of this study is to review the impact of those defects on the mechanical performance of FRC and therefore to discuss the available treatment methods to eliminate or minimize them in order to enhance the functional properties of the printed parts. As FRC is a combination of polymer matrix material and continuous or short reinforcing fibres, this review will thoroughly discuss both thermoplastic polymers and FRCs printed via FDM technology, including the effect of printing parameters such as layer thickness, infill pattern, raster angle and fibre orientation. The most common defects on printed parts, in particular, the void formation, surface roughness and poor bonding between fibre and matrix, are explored. An inclusive discussion on the effectiveness of chemical, laser, heat and ultrasound treatments to minimize these drawbacks is provided by this review.
Article
Full-text available
It is discovered in this investigation that there exist two different failure modes and a special separation angle which is the demarcation point of the two different failure modes when FDM (Fused Deposition Modelling) 3D printing materials fail under a tensile load. In order to further understand the mechanical properties of FDM 3D printing materials and promote the use of FDM 3D printing materials, their tensile failure strengths at different printing angles and separation angles are measured and analysed theoretically. A new separate-modes of transversely isotropic theoretical failure model is established to predict the tensile failure strength and separation angle of FDM 3D printing PLA (polylactic acid) material based on the hypothesis of transverse isotropy and the classical separate-modes failure criterion. During this research, the tensile specimens designed according to the current test standard ISO (527-2-2012) for plastic-multi-purpose specimens are fabricated in 7 different printing angles (0∘, 15∘, 30∘, 45∘, 60∘, 75∘, 90∘) and three levels of printing layer thickness (0.1 mm, 0.2 mm, 0.3 mm). Experimental results show that the tensile failure strength increases with the increase of the printing angle or the decrease of the layer thickness. Meanwhile, inter-layer failure tends to occur when the printing angle is small and in-layer failure tends to occur when the printing angle is big. In comparison with the results predicted by the established theoretical model, all values of the Generalized-Relative-Root-Mean-Square Error are close to zero and the experimental separation angles are also between 45∘ and 60∘. So the predictive capacity of the theoretical model is affirmed by experimental results.
Article
Full-text available
Fused filament fabrication (FFF) is one of the most popular additive manufacturing (AM) techniques used to fabricate polymeric structures. However, these polymeric structures suffer from an inherent deficiency of weak inter-laminar tensile strength. Because of this weak inter-laminar strength, these parts fail prematurely and exhibit only a fraction of the mechanical properties of those fabricated using conventional means. In this paper, we study the effect of thermal annealing in the presence of an initially applied static uniaxial load on the ultimate tensile strength of parts fabricated using FFF. Tensile specimens or dogbones were fabricated from an acrylonitrile butadiene styrene (ABS) filament with a glass transition temperature (Tg) of 105 °C; these specimens were then isothermally annealed, post manufacture, in a fixture across a given range of temperatures and static loads. Tensile testing was then performed on these specimens to gauge the effect of the thermal annealing and static loading on inter-laminar tensile strength by measuring the ultimate tensile strength of the specimens. A design of experiments (DOE) approach was followed to calculate the main and interaction effects of the two factors (temperature and static loading) on the ultimate tensile strength, and an analysis of variance was conducted. Cross-sectional images of the specimens were studied to observe the changes in the mesostructure of the specimens that led to the increase in inter-laminar strength of the parts. The results show that temperature plays a dominant role in increasing the ultimate tensile strength and an 89% increase in the average ultimate tensile strength was seen corresponding to an annealing temperature of 160 °C. A change in the mesostructure of the parts is seen, which is characterized by an increase in bond length and void coalescence. These results can be helpful in studying the structural strength of 3D printed parts, and thus could eventually guide the fabrication of components with strength comparable to those of conventional manufacturing techniques.
Article
Full-text available
The lack of specific standards for characterization of materials manufactured by Fused Deposition Modelling (FDM) makes the assessment of the applicability of the test methods available and the analysis of their limitations necessary; depending on the definition of the most appropriate specimens on the kind of part we want to produce or the purpose of the data we want to obtain from the tests. In this work, the Spanish standard UNE 116005:2012 and international standard ASTM D638-14:2014 have been used to characterize mechanically FDM samples with solid infill considering two build orientations. Tests performed according to the specific standard for additive manufacturing UNE 116005:2012 present a much better repeatability than the ones according to the general test standard ASTM D638-14, which makes the standard UNE more appropriate for comparison of different materials. Orientation on-edge provides higher strength to the parts obtained by FDM, which is coherent with the arrangement of the filaments in each layer for each orientation. Comparison with non-solid specimens shows that the increase of strength due to the infill is not in the same proportion to the percentage of infill. The values of strain to break for the samples with solid infill presents a much higher deformation before fracture.
Article
Full-text available
A non-dominated sorting modified teaching–learning-based optimization (NSMTLBO) is proposed to obtain the optimum solution for a multi-objective problem related to machining Polytetrafluoroethylene. Firstly, an experimental design is done and the L27 orthogonal array with three-level of cutting speed (Vc) , feed rate (f), depth of cut (ap) and nose radius (Nr) is formulated. A CNC turning machine is used to perform experiments with cemented carbide tool at an insert angle of 80° and the response variables known as surface finish and material removal rate are measured. A response surface model is rendered from the experimental results to derive the minimization function of surface roughness (Ra) and maximization function of material removal rate (MRR). Both optimization functions are solved simultaneously using NSMTLBO. A fuzzy decision maker is also integrated with NSMTLBO to determine the preferred optimum machining parameters from Pareto-front based on the relative importance level of each objective function. The best responses Ra = 2.2347 µm and MRR = 96.835 cm3/min are predicted at the optimum machining parameters of Vc = 160 mm/min, f = 0.5 mm/rev, ap = 0.98 mm and Nr = 0.8 mm. The proposed NSMTLBO is reported to outperform other six peer algorithms due to its excellent capability in generating the Pareto-fronts which are more uniformly distributed and resulted higher percentage of non-dominated solutions. Furthermore, the prediction results of NSMTLBO are validated experimentally and it is reported that the performance deviations between the predicted and actual results are lower than 3.7%, implying the applicability of proposed work in real-world machining applications.
Article
Full-text available
This paper presents experimental results of standard three-point bending tests, for fused deposition modeling (FDM) specimens made of acrylonitrile–butadiene–styrene (ABS), as well as ABS plus. Both types of specimens were built with different building strategies. Furthermore, specimens of the aforementioned materials were subjected to standard Charpy’s impact test. The Charpy’s tests have been conducted using notched and unnotched specimens. Hereby, the flexural and impact strength of the specimens were quantitatively determined. Moreover, the findings of the implemented parametric analysis were discussed and correlated with the ones of the corresponding literature for bulk and FDM built materials. In addition, the fracture areas of the specimens were examined on a scanning electron microscope (SEM) to determine the failure patterns of the filaments’ strands. The research findings proved that ABS and ABS-plus FDM-printed specimens exhibit inferior flexural strength and bending capacity when compared to the bulk material. Finally, the derived impact strength results proved that ABS and ABS-plus FDM-printed specimens display a decreased impact strength when compared with the ones of the corresponding bulk materials.
Article
Full-text available
Fused deposition modeling (FDM) technology works with specialized 3D printers and production-grade thermoplastics to build robust, durable, and dimensionally stable parts with the best accuracy and repeatability of any other available 3D printing technology. FDM is one of the highly used additive manufacturing technology due to its ability to manufacture very complex geometries. However, the critical problems with this technology have been to balance the ability to produce esthetically appealing products with functionality and properties at the lowest cost possible. In this study, three major process parameters such as layer height, raster angle, and infill density have been considered to study their effects on mechanical properties of acrylonitrile butadiene styrene (ABS) as this material is widely used industrial thermoplastic in FDM technology. The test results show a clear demonstration of the considered factors over the mechanical variables measured. Response surface methodology is used for the validation of the experimental data and the future prediction of the test results. It was found that the optimum parameters for 3D printing using ABS are 80% infill percentage, 0.5 mm layer thickness, and 65° raster angle. The achieved experimental ultimate tensile strength, elastic modulus, yield strength, fracture strain, and toughness (energy absorption) are 31.57 MPa, 774.50 MPa, 19.95 MPa, 0.094 mm/mm, and 2.28 Jm⁻³, respectively. Mathematical equation has been developed using surface response methodology which can be used to predict the ABS tensile properties numerically and also to predict the optimum parameter for ultimate properties.
Article
Full-text available
This paper presents a comparative study of the tensile mechanical behaviour of pieces produced using the Fused Deposition Modelling (FDM) additive manufacturing technique with respect to the two types of thermoplastic material most widely used in this technique: polylactide (PLA) and acrylonitrile butadiene styrene (ABS). The aim of this study is to compare the effect of layer height, infill density, and layer orientation on the mechanical performance of PLA and ABS test specimens. The variables under study here are tensile yield stress, tensile strength, nominal strain at break, and modulus of elasticity. The results obtained with ABS show a lower variability than those obtained with PLA. In general, the infill percentage is the manufacturing parameter of greatest influence on the results, although the effect is more noticeable in PLA than in ABS. The test specimens manufactured using PLA perform more rigidly and they are found to have greater tensile strength than ABS. The bond between layers in PLA turns out to be extremely strong and is, therefore, highly suitable for use in additive technologies. The methodology proposed is a reference of interest in studies involving the determination of mechanical properties of polymer materials manufactured using these technologies.
Article
Full-text available
In this research, a new metaheuristic optimization algorithm, inspired by biological nervous systems and artificial neural networks (ANNs) is proposed for solving complex optimization problems. The proposed method, named as neural network algorithm (NNA), is developed based on the unique structure of ANNs. The NNA benefits from complicated structure of the ANNs and its operators in order to generate new candidate solutions. In terms of convergence proof, the relationship between improvised exploitation and each parameter under asymmetric interval is derived and an iterative convergence of NNA is proved theoretically. In this paper, the NNA with its interconnected computing unit is examined for 21 well-known unconstrained benchmarks with dimensions 50–200 for evaluating its performance compared with the state-of-the-art algorithms and recent optimization methods. Besides, several constrained engineering design problems have been investigated to validate the efficiency of NNA for searching in feasible region in constrained optimization problems. Being an algorithm without any effort for fine tuning initial parameters and statistically superior can distinguish the NNA over other reported optimizers. It can be concluded that, the ANNs and its particular structure can be successfully utilized and modeled as metaheuristic optimization method for handling optimization problems.
Article
Full-text available
Fused Deposition Modeling (FDM), a well known 3D printing technology is widely used in various sorts of industrial applications because of its ability to manufacture complex objects in the stipulated time. However, the proper selection of input process parameters in FDM is a tedious task that directly affects the part performance. Here, in this work, the research efforts have been made to optimize the FDM process parameters in order to find out the best parameter setting as per the mechanical and surface quality perspectives by using hybrid particle swarm and bacterial foraging optimization (PSO-BFO) evolutionary algorithm. Taguchi L18 orthogonal array was used for the development of Acro-Nitrile Butadiene Styrene (ABS) based 3D components by considering layer thickness, support material, model interior and orientation as a process parameters. Further, the relationships among selected FDM process parameters and output responses such as hardness (H), flexural modulus (FM), tensile strength (TS) and surface roughness (Ra) were established by using linear multiple regression. Then, the effects of individual process parameters on selected response parameters were examined by signal to noise (S/N) ratio plots. Finally, a multi-objective optimization of process parameters has been performed with hybrid PSO-BFO, general PSO and BFO algorithm, respectively. The overall results reveal that the layer thickness of 0.007 mm, support material type sparse, part orientation of 60 degrees and model interior of high density helps in achieving desired performance level.