Page 1
http://www.iaeme.com/IJMET/index.asp 1569 [email protected]
International Journal of Mechanical Engineering and Technology (IJMET) Volume 9, Issue 10, October 2018, pp. 1569–1584, Article ID: IJMET_09_10_160
Available online at http://www.iaeme.com/ijmet/issues.asp?JType=IJMET&VType=9&IType=10
ISSN Print: 0976-6340 and ISSN Online: 0976-6359
© IAEME Publication Scopus Indexed
OPTIMIZATION OF PROCESS PARAMETERS IN
AUSTENITIC STAINLESS STEEL 316L DURING
PULSED MIG JOINING PROCESS
Dhivyasri G
School of Electrical Engineering, Vellore Institute of Technology, Vellore, India,
Sudha Ramasamy
School of Electrical Engineering, Vellore Institute of Technology, Vellore, India
ABSTRACT
Welding is a technique used to join materials by applying heat. The choosing of
appropriate weld process parameters; weld current, weld voltage and weld speed is
important, to achieve the desired weld bead geometry in a fusion joining process. In the
present study, the bead geometry such as Bead Width (BW), Reinforcement Height (RH)
and Depth of Penetration (DOP) of Pulsed Metal Inert Gas (MIG) welding is carried out
on AISI 316L. Genetic algorithm (GA) based optimization technique has been opted to
obtain the desired combination of process variables and weld bead geometry. Initially,
regression models are developed using the training dataset. The developed GA optimizes
the weld process parameters and weld bead geometry by minimizing the least square error
based objective function. The microstructure examination is performed on the optimal
weld geometry obtained from the optimization technique using optical and scanning
electron microscopy techniques. Energy Dispersive X-ray (EDS) analysis is carried out
to examine the compositional variations in the weld bead. The study attested that the width
of HAZ and depth of the fusion zone increases with increase in welding current (80A to
100A).
Keywords: Pulsed Metal Inert Gas welding; Bead width, Reinforcement Height and Depth of
Penetration and Genetic Algorithm
Cite this Article Dhivyasri Ga and Sudha Ramasamy, Optimization of Process Parameters in
Austenitic Stainless Steel 316l During Pulsed Mig Joining Process, International Journal of
Mechanical Engineering and Technology, 9(10), 2018, pp. 1569–1584.
http://www.iaeme.com/IJMET/issues.asp?JType=IJMET&VType=9&IType=10
1. INTRODUCTION
In the last two decades, the demand for stainless steels in versatile engineering applications are
raised at the rate of 5% [1]. The austenitic stainless steel 316L grade is low carbon, Mo-added
alloy which exhibits excellent combination of toughness and strength. This alloy is highly
resistance to pitting and crevice corrosion and possess excellent welding capability. Such
excellent and rare combination of mechanical properties and high corrosion resistance finds this
Page 2
Dhivyasri Ga and Sudha Ramasamy
http://www.iaeme.com/IJMET/index.asp 1570 [email protected]
alloy in extensive applications including chemical, construction, oil and marine environment.
AISI 316L can be joined by almost all the welding techniques. Although,this alloy exhibits good
weldability, there are few challenges experienced during welding and is mostly influenced by the
welding parameters [2].
The conventional MIG welding does not control the short-circuiting spikes occurring at
reduced arc length which leads to excessive spattering and irregular droplet formation. The
current pulsing in welding is a process in which high peak electric current and low base electric
current are applied at regular time intervals to break away the droplet formation at the electrode
wire tip using electromagnetic pinch force generated by the current pulse. The schematic diagram
of Pulsed MIG Welding is shown in Fig. 1.
Figure 1 Pulsed MIG Welding
The selection of a suitable parameters in pulsed Metal Inert Gas (MIG) process to achieve the
desired weld bead geometry needs lot of experimentation. This necessitates to develop a
methodology for optimization technique of welding parameters since experimental trial runs
involve time and cost. Out of the various optimization techniques, the efficient way of
establishing the weld process parameters is by adopting Genetic algorithm (GA). GA is an
optimization technique based on the principles of biological evolution which is inspired by
Darwin's theory about the survival of fitness in the search space. The fittest individual in any
population has a greater chance to reproduce and survive. GA is widely used to solve non- linear
problems and their application in optimizing the weld parameters are highly emerging [3]. Some
of the research reports have reported the optimization of process parameters using soft computing
techniques in MIG welding and Tungsten Inert Gas (TIG) welding are presented as follows. In
order to relate the weld parameters to the quality of weld bead profile, Ridings et al. presented a
neural network model to envisage the weld bead geometry [4]. Vasudevan et al. developed a GA
based numerical model to obtain the optimal weld parameters in TIG welding, using 304LN and
316LN austenitic stainless steels to attain the target weld bead geometry [5]. It is inferred from
the study that the regression model resulted in good correlation between the calculated and
measured process parameters. Further, these authors presented the soft computing techniques in
modelling and predicting the microstructures of stainless steel welding. A neural network model
is developed to predict the solidification modes in stainless steels and fuzzy logic systems are used
to monitor and control the weld processes. Thus author compared the different soft computing
techniques to analyse the microstructure of the welded stainless steels [6]. Nagesh and Datta
presented a genetic algorithm approach to optimize the process parameter and predict the weld
bead geometry of 1100 aluminium using TIG welding process. This proposed method is effective
in determining the weld bead geometry for TIG welding [7].
Page 3
Optimization of Process Parameters in Austenitic Stainless Steel 316l During Pulsed Mig Joining
Process
http://www.iaeme.com/IJMET/index.asp 1571 [email protected]
Kamal Pal et al. addressed on the modelling and optimization of deposition efficiency in
pulsed MIG welding. GA and differential evolution techniques are deployed in maximising the
deposition efficiency and the latter is found to yield optimal solution [8]. Madhumitha et al.
have applied decision trees to identify weld central line in austenitic stainless-steel joints to find
flaws during ultrasonic testing. Their model was found to be very swift and quantitative [9]. In
a study, the effect of welding parameters on AISI 316 weld joints using ANOVA is examined
by Bharath et al. and the results attest that tensile strength and bend strength of the material are
greatly influenced by the current and weld speed [10]. Shyam Narayan Divakar have employed
GA based Artificial Neural Network (ANN) to measure the weld bead geometries such as front
width, front height, back width and back height and prediction results on aluminium sheet during
TIG welding process [11]. Luis Pe ´rez Pozo et al. optimized the welding parameters using GA
during MIG welding of curved specimens of SAE 1020 steel, for the described objective
functions the weld characteristics obtained less than 6% error [12]. Jerold Jose and Dev Anand
reported on the optimization of weld bead profile during TIG welding of Inconel 718 alloy
using regression analysis and RSM with 95% accuracy level [13].
Zhongmei Gao et al. have proposed a hybrid kriging and genetic model to optimise the weld
process parameter of 316L stainless steel using laser-MIG butt welding. The proposed model is
proven to be more efficient and feasible [14]. Nabendu Ghosh et al. performed parametric
optimization using Grey-Based Taguchi method during MIG welding of SS316L. The quality of
welds are evaluated based on Ultimate Tensile Strength (UTS), Yield Tensile Strength (YTS) and
percentage of elongation [15]. Nabendu Ghosh et al. have performed parametric optimization
using Principal Component analysis (PCA) - Based Taguchi method during MIG welding of
SS316L. The weld quality was evaluated based on UTS, YTS and percentage of elongation.
Optimal parameters were predicted and validated [16].
In this present work, genetic algorithm (GA) based computational model is developed to
determine the optimum process parameters to achieve the target weld bead geometry in 316L
stainless steel material. The Pulsed MIG welding process is done using FD-B6 welding robot
machine. The results indicates closer agreement between experimental and GA predicted
values. The weld bead geometry of the experimentally performed data’s are examined using
microscopic analysis.
2. EXPERIMENTAL WORK
2.1. Base Metal and Welding
AISI 316L stainless steel sheet of 2 mm thick with the dimensions 150mm x 150mm is used as
the candidate metal for the study, whose chemical composition is presented in Table 1. MIG
welding process is adopted using DAIHEN’s FD-B6 MIG welding robot and PLC based FD11
manipulator controller. The filler wire is SS308 of 1.2mm diameter. The shielding gas used is the
mixture of 98% Argon and 2% O2 whose flow rate is 18 litres/minute. The experimental setup is
shown in the Fig. 2.
Page 4
Dhivyasri Ga and Sudha Ramasamy
http://www.iaeme.com/IJMET/index.asp 1572 [email protected]
Figure 2 Experimental setup with FD-B6 MIG welding robot
The FD11 manipulator controller consists of a PLC programmer in which the user can
program the robotic movements to perform welding. The panel screen is shown in Fig. 3 in which
the user can set the input parameters.
Figure 3 setting the input parameters in PLC panel
2.2. Design of Experiments
In this experimentation, the process parameters namely the welding current (I), voltage (V) and
weld speed (WS) are varied as they significantly influence on the weld bead geometry such Bead
Width (BW), Reinforcement Height (RH) and Depth of Penetration (DOP). The geometry of the
weld bead is shown in Fig. 4. Then 15 trained data samples and 5 test data samples are adopted
to perform GA optimization for FD-B6 MIG welded samples. The experimental data are listed in
Table 1. The experimental data is optimized using GA algorithm therefore the target weld bead
geometry which can be achieved for specific process parameters is obtained.
Page 5
Optimization of Process Parameters in Austenitic Stainless Steel 316l During Pulsed Mig Joining
Process
http://www.iaeme.com/IJMET/index.asp 1573 [email protected]
Figure 4 Weld bead geometry
3. OPTIMIZATION METHODOLOGY
The optimizing methodology for pulsed MIG welding process parameters with GA is done by
developing the regression models correlating the process parameters with the weld bead
geometry. Then a GA is developed by means of MATLAB toolbox, where the regression model
is used to evaluate the objective function.
Table 1 Experimental Data of 316L welded in FD-B6 MIG welding robot
Trial No.
Experimental Data Experimental Results
Weld Speed
(cm/min)
Welding
Current (A)
Welding
Voltage (V) DOP (mm) BW (mm) RH (mm)
Training Dataset
1 55 70 15 1.0108 3.2233 1.547
2 55 73 14.9 1.0106 3.2456 1.658
3 55 75 14.1 1.0106 3.2600 1.7248
4 55 77 14.3 0.9761 3.2664 1.7786
5 55 80 14.4 1.0090 3.2674 1.7568
6 55 83 14.5 1.1321 3.267 1.7456
7 55 85 14.6 1.1677 3.672 1.6149
8 55 87 14.6 1.1752 3.7145 1.71
9 55 90 14.7 1.1872 3.7365 1.7228
10 55 93 14.8 1.1645 3.7298 1.9856
11 55 95 14.9 1.1511 3.6776 2.0664
12 55 97 14.9 1.1401 3.7546 2.0114
13 55 100 14.9 1.3367 3.8895 1.8565
14 55 105 15.1 1.2369 3.9871 1.8457
15 55 110 15.2 1.2245 3.9962 1.8341
Test Dataset
16 55 80 14.2 1.0103 3.2606 1.7058
17 55 85 14.5 1.0587 3.2358 1.7589
18 55 90 14.7 1.1872 3.7445 1.7181
19 55 95 14.9 1.1511 3.6798 2.0574
20 55 100 14.9 1.3258 3.7695 1.8665
Page 6
Dhivyasri Ga and Sudha Ramasamy
http://www.iaeme.com/IJMET/index.asp 1574 [email protected]
3.1. Regression Model for Pulsed MIG Welded SS316L
Regression models are developed using multiple regression method. The standard regression
model notation is given by,
3.2. Development of Genetic Algorithm
Genetic Algorithm (GA) is developed using MATLAB tool to optimize the pulsed MIG welded
process parameters during the welding of SS316L. The flow chart in Fig. 5 depicts the procedure
involved during the execution of GA. The search space defines the input parameter ranges for
weld current (I), weld voltage (V) and weld speed (WS) as shown in Table 2 within which optimal
solution is identified by the GA.
��������� = �0 + �1 ∗ � + �2 ∗ + �3 ∗ � + �4 ∗ �2 + �5 ∗ 2 + �6 ∗ �2
+ �7 ∗ ∗ � + �8 ∗ � ∗ � (1)
Where �0, �1, �2, �3, �4, �5, �6, �7, �8 are the estimated co-efficient. I is welding current, V is
Welding voltage and WS is Weld speed.
The relationship between three weld-bead geometry parameters and the process variables are
estimated as follows.
��� = 0.311 + 0.0512 ∗ � + 0.1312 ∗ + 0.2017 ∗ � + 0.4932 ∗ �2
− 0.5110 ∗ 2 + 0.0021 ∗ �2 + 0.0017 ∗ ∗ � + 0.0112 ∗ �
∗ � + 0.0098 ∗ ∗ (2)
� = 0.3617 + 0.0745 ∗ � + 0.2698 ∗ + 0.1693 ∗ � − 0.0002 ∗ �2
− 0.0063 ∗ 2 + 0.0013 ∗ �2 − 0.0012 ∗ ∗ � + 0.0123 ∗ �
∗ � + 0.0076 ∗ ∗ � (3)
�� = −0.3063 + 0.0042 ∗ � + 0.1012 ∗ + 0.0102 ∗ � + 0.0003 ∗ �2
− 0.013 ∗ 2 + 0.0012 ∗ �2 − 0.0102 ∗ ∗ � + 0.0006 ∗ �
∗ � + 0.0213 ∗ ∗ � (4)
Page 7
Optimization of Process Parameters in Austenitic Stainless Steel 316l During Pulsed Mig Joining
Process
http://www.iaeme.com/IJMET/index.asp 1575 [email protected]
Figure 5 Procedure of GA optimization
There are more than one process variable that influences the target parameters; DOP, BW and
RH. The objective function is one that directs the solution to convergence. According to
Vasudevan et al. the least-square error minimization is commonly preferred objective function
[5]. In this present work, the sum of the least-square errors of the weld-bead geometries is
preferred as the objective function which is known by the equation,
Where,
���� - Objective function
���( ),�( ), ��( ) – are the values of DOP, BW and RH of the ith individual respectively
The GA attempts to maximize the solution though the objective function gets minimized.
Thus, a fitness index is given to the solutions were objective functions higher value corresponds to
lower fitness index of the solution. Since the convergence rate of GA is influenced by population
size, crossover type, number of generation, mutation rate, crossover rate and trial-and-error
Page 8
Dhivyasri Ga and Sudha Ramasamy
http://www.iaeme.com/IJMET/index.asp 1576 [email protected]
approach is used to arrive at the best fit of GA parameters [5]. With this view, the variation
performed in GA parameters in illustrated in the Table 2 and Table 3.
Table 2 Range of GA parameters
GA Parameters Variation range
Size of population 50-500
No. of generation 100-1000
Crossover rate 0.55-0.90
Mutation rate 0.001-0.009
Table 3 GA parameters for optimizing the Pulsed MIG joining process
GA Parameters Value
Size of population 100
No. of generation 200
Crossover rate 0.75
Mutation rate 0.006
Crossover type Single point
The minimum and maximum values of the variables I, V, WS are specified independently.
Initially 100 population is selected to perform the first iteration, with every individual
representing a set of process variables. In this present work, the maximum value
among all the independent variables is considered as 110 (which is the highest value of
current), then the length of the chromosome in every gene is considered as 8 (as 28=256). The
number of variables is 3, therefore the length of the chromosome is 24 bits (3 × 8). Thus 200
number of generations is specified. Then, using the objective function all the chromosomes were
evaluated for their fitness. Roulette Wheel Selection (RWS) method is used to choose the
chromosome with highest fitness among the available chromosomes until total population of next
generation is generated [7]. Single Point Crossover (XSOP) is performed on the particular
chromosomes to produce offspring. Mutation is then performed after crossover, with the mutation
rate of 0.006 to avoid perturbations. Then using objective function the offspring’s are checked for
the fitness. Next 100 chromosomes are selected by mixing the chromosomes of offspring and
parents and selecting the best 100 chromosomes based on fitness index. The recently chosen 100
chromosomes are used for subsequent iterations. The iterations are sustained until no further
change in the optimized variables is observed.
3.3. Experimental Validation
In the experimentation, limited target weld bead geometries are selected from the experimental
data, and GA is applied to optimize Pulsed MIG welding process parameters. Whenever the GA
is executed, it resulted in various sets of process variables that all produced the same set of
targeted weld bead geometry. Nabendu Ghosh et al. identified that weld bead geometry target
can be attained by different combinations of weld process variables; i.e., different combination
of weld current, weld voltage and weld speed, with every combination capable of resulting the
same targeted weld bead geometry [15].
Page 9
Optimization of Process Parameters in Austenitic Stainless Steel 316l During Pulsed Mig Joining
Process
http://www.iaeme.com/IJMET/index.asp 1577 [email protected]
Figure 6 GA predicted and experimental result for DOP
Figure. 7 GA predicted and experimental result for BW
Page 10
Dhivyasri Ga and Sudha Ramasamy
http://www.iaeme.com/IJMET/index.asp 1578 [email protected]
Figure 8 GA predicted and experimental result for RH
From the Fig. 6, Fig. 7 and Fig. 8 it is clearly noted that a closed agreement is observed
between the predicted value and experimental values of DOP, BW and RH obtained from the
regression equations. Thus, all the regression models is observed to have a good correlation
between weld bead geometry and the process parameters (I, V and WS) with the error percentage
less than 0.04.
Table 4 Evaluation of actual and predicted pulsed MIG process parameters for 316L stainless steel
welds.
Parameters
Case 1 Case 2 Case 3
Actual
Value
Predicted
Value
Error
%
Actual
Value
Predicted
Value
Error
%
Actual
Value
Predicted
Value Error %
Welding
current (A) 80 80 0 85 85 0 90 90 0
Welding
Voltage (V) 14.2 14.2 0 14.5 14.5 0 14.7 14.7 0
Weld Speed
(cm/min) 55 55 0 55 55 0 55 55 0
DOP (mm) 1.0103 0.9837 0.0266 1.0587 1.0587 0 1.1872 1.1607 0.0265
BW (mm) 3.2606 3.2605 0.0001 3.2358 3.1716 0.0642 3.7445 3.7443 0.0002
RH (mm) 1.7058 1.7049 0.0009 1.7589 1.7587 0.0002 1.7181 1.7016 0.0165
Parameters
Case 4 Case 5
Actual Value Predicted
Value Error % Actual Value Predicted Value Error %
Welding current (A) 95 95 0 100 100 0
Welding Voltage (V) 14.9 14.9 0 14.9 14.9 0
Weld Speed (cm/min) 55 55 0 55 55 0
DOP (mm) 1.1511 1.1510 0.0001 1.3258 1.3209 0.0049
BW (mm) 3.6798 3.6789 0.0009 3.7695 3.7585 0.0385
RH (mm) 2.0574 2.0572 0.0002 1.8665 1.8288 0.0377
Page 11
Optimization of Process Parameters in Austenitic Stainless Steel 316l During Pulsed Mig Joining
Process
http://www.iaeme.com/IJMET/index.asp 1579 [email protected]
For experimental validation, bead on trials were performed on 316L sheets with process
variables mentioned in case 1, case 2, case 3, case 4 and case 5. The samples were then cross-
sectioned and etched to carry out microscopic examination to measure the experimental values
of DOP, BW and RH. GA is applied to predict their corresponding combinations of welding
current, voltage and weld speed and their resultant DOP, BW and RH. The obtained results are
presented in Table 4. It may be noted that a good agreement is observed between predicted and
experimental values with very less error of 0.04%. This ability of GA makes it more superior to
Artificial Neural Networks (ANN) and regression models resulting in high accuracy.
5. RESULTS AND DISCUSSION
The weld bead geometry is estimated using Pulsed MIG welding process for various process
parameters. Optimal process parameter is determined using GA achieves the desired weld
geometry. For the optimal results listed on Table 4 experiment is done and microscopic analysis
are performed. The manually obtained hardware results have a good correlation with the GA
performed results. Thus GA is potential tool in experimental Pulsed MIG welding optimization.
Table 1 Welded samples of different current ratings
Welding
Current (A) Weld Image
80
85
90
95
100
Table 5 illustrates the welded samples of AISI 316L under various process parameters such
as, welding current from 80A increasing five intervals periodically till 100A. Thus welding
voltage are automatically tuned to set around 14.2V to 14.9V in DM 350 - Invertor DC power
source. The welding speed is set constant as 55 cm/min to reduce spatter and obtain regular bead
geometry.
The microscopic images of the experimentally validated samples in case 1, case 2, case 3, case
4 and case 5 are presented in Table 2.
Page 12
Dhivyasri Ga and Sudha Ramasamy
http://www.iaeme.com/IJMET/index.asp 1580 [email protected]
Table 2 Experimental results
Sl.
No Welding Current (A)
Microscopic Images
DOP
(mm)
BW
(mm)
RH
(mm)
1
80
1.010
3.260
1.705
2
85
1.059
3.236
1.759
3
90
1.187
3.745
1.718
4
95
1.151
3.679
2.057
Page 13
Optimization of Process Parameters in Austenitic Stainless Steel 316l During Pulsed Mig Joining
Process
http://www.iaeme.com/IJMET/index.asp 1581 [email protected]
5
100
1.336
3.769
1.866
BP = Bead Penetration; BH = Bead Height/ RH = Reinforcement Height; BW = Bead Width
The macrophotographs of austenitic stainless steel 316L is utilized for the experimental
analysis is presented in Table 6. The weld samples are rubbed with stainless steel brush applying
gentle pressure to elude the undesirable heat generation on materials and cleaned with acetone to
removal the oil residues, dirt, oxides and moistures. Narsimhachary et al. reports that change in
DOP depends on the melting of the material [17]. It is observed that with the increase in welding
current, the DOP and bead width also increases. For the current of 100A, the value of DOP and
bead width are 1.336 mm and 3.769mm respectively. While reinforcement height is 2.057 mm
for a welding current of 95A. Thus the targeted weld bead geometry of DOP and BW is obtained
at welding current of 100A and reinforcement height at 95A of welding current is obtained to
have the best targeted fitness values with respect to that of the predicted values.
5.1. SEM analysis
In Fig. 9 shows different focus of weldment like; base material, heat affected zone and fusion
zone for sample 1 (80A), sample 2 (85A), sample 3 (90A), sample 4 (95A) and sample 5 (100A).
The influence of welding current (80A, 85A, 90A, 95A and 100A) variations are clearly
presented, increase in the current value resulted in fast melting of the filler metals. This variations
are clearly observed in fusion zone, is presented in Fig. 9. From the Fig. 9 the coarse serrated
morphology appeared in the fusion boundary. The equiaxed dendrites and columnar dendrites
can be observed in the welding zone. In addition, columnar dendrites adjacent to the fusion line
developed in normal-to-isotherm line direction. Based on the proportion temperature, grade to
cooling rate of the microstructures solidification mode is contingent. The high heat input and low
cooling of the base material causes increased grain size.
Page 14
Dhivyasri Ga and Sudha Ramasamy
http://www.iaeme.com/IJMET/index.asp 1582 [email protected]
Figure 9 SEM Results
5.2. Energy Dispersive X-Ray Analysis
For the case 1, case 2, case 3, case 4 and case 5 samples EDX analysis is carried out which are
presented in Fig. 10 (a, b, c, d and e) shows peaks of carbon, iron, chromium and nickel. The
Table 7 presents the elemental composition of carbon, chromium, nickel and iron.
Table 7 Elemental composition of samples
Sample No. Elemental wt. %
C Cr Fe Ni
1 13.92 23.84 56.01 6.03
2 13.86 23.99 55.26 6.97
3 13.73 24.02 54.84 7.14
4 13.62 25.25 53.30 7.83
5 13.5 31.02 47.05 8.04
Elemental chemical analysis by EDX is executed on five different current values, presenting
the innumerable distribution of elements in Fig. 10. It is observed finely that, when welding
current ratings are increased, the elemental composition nickel and chromium increases with
decreasing the iron and carbon weight composition. In general, Nickel increases the ductility and
Page 15
Optimization of Process Parameters in Austenitic Stainless Steel 316l During Pulsed Mig Joining
Process
http://www.iaeme.com/IJMET/index.asp 1583 [email protected]
toughness of the steel materials. The content of carbon needs to be preferably low, such that
weldability of steel is improved [18]. Where else in Table 7, carbon content is observed to be
reducing which in turn produces no major change of chemical compositions is resulted in EDX
spectrum. Thus based on the welding current chemical compositions are varied within limits.
(a) (b)
(c) (d)
(e)
Figure 10 EDX pattern of 316L samples (a) 80A (b) 85A (c) 90A (d) 95A (e) 100A
6. CONCLUSIONS
It is essential to identify the suitable weld process parameters; weld current, weld voltage and
weld speed in demand to attain the desired weld bead geometry in a Pulsed MIG joining process.
The following conclusions are obtained as:
• From the experimental data, regression models are constructed and resultant plots
indicates a good agreement between predicted value and experimental values of DOP,
BW and RH obtained from the Genetic Algorithm optimization.
• Then the developed Genetic Algorithm based optimisation gave best solution for size
of population – 100; No. of generation – 200; crossover rate – 0.75; mutation rate
–
• 0.006 and mutation type – single point cross over.
• Experimental validation through microscopic analysis indicates close agreement with
the predicted values obtained from the results of GA. It was observed that increase in
welding current resulted in increased DOP and BW.
• From the EDX results it may be noted that with increase in welding current the
elemental composition of carbon and iron reduces while the elemental composition of
chromium and nickel increases. At the same time, there is no major amount of chemical
changes observed in the material, the original state of the material has been retained
austenitic property after welding.
• For advanced welding processes like robotic Pulsed-MIG welding, instead of wasting
resources by trial and error approach optimization techniques like GA can be opted to
estimate good combinations of process variables to achieve good target weldments.
ACKNOWLEDGEMENT
The authors express special thanks to Dr. N. Siva Shanmugam, Department of Mechanical
Engineering, NIT Trichy for providing the research lab facilities to perform the experimentation
work.
Page 16
Dhivyasri Ga and Sudha Ramasamy
http://www.iaeme.com/IJMET/index.asp 1584 [email protected]
REFERENCES
[1] N. R. Baddoo, “Stainless steel in construction: A review of research, applications, challenges
and opportunities,” J. Constr. Steel Res., vol. 64, no. 11, pp. 1199–1206, 2008.
[2] P. S. Korinko and S. H. Malene, “Considerations for the weldability of types 304L and 316L
stainless steel,” Journal of Failure Analysis and Prevention, vol. 1, no. 4. pp. 61– 68, 2001.
[3] M. Saraswat, “Genetic Algorithm for optimization using MATLAB,” vol. 4, no. 3, pp. 155–
159, 2013.
[4] G. E. Ridings, R. C. Thomson, and G. Thewlis, “Prediction of multiwire submerged arc weld
bead shape using neural network modelling,” Sci. Technol. Weld. Join., vol. 7, no. 5, pp. 265–
279, 2002.
[5] M. Vasudevan, A. K. Bhaduri, B. Raj, and K. P. Rao, “Genetic-algorithm-based
computational models for optimizing the process parameters of A-TIG welding to
[6] achieve target bead geometry in type 304 L(N) and 316 L(N) stainless steels,” Mater. Manuf.
Process., vol. 22, no. 5, pp. 641–649, 2007.
[7] M. Vasudevan, “Soft computing techniques in stainless steel welding,” Mater.
Manuf.Process, vol. 24, no. 2, pp. 209–218, 2009.
[8] D. S. Nagesh and G. L. Datta, “Genetic algorithm for optimization of welding variables for
height to width ratio and application of ANN for prediction of bead geometry for TIG welding
process,” Appl. Soft Comput. J., vol. 10, no. 3, pp. 897–907, 2010.
[9] K. Pal, S. Bhattacharya, and S. K. Pal, “Optimisation of weld deposition efficiency in pulsed
MIG welding using hybrid neuro-based techniques,” Int. J. Comput. Integr. Manuf., vol. 24,
no. 3, pp. 198–210, 2011.
[10] P. Madhumitha, S. Ramkishore, K. Srikanth, and P. Palanichamy, “Application of Decision
trees for the identification of weld central line in austenitic stainless steel weld joints,” 2014
Int. Conf. Comput. Power, Energy, Inf. Commun., pp. 400–406, 2014.
[11] P. Bharath, V. G. Sridhar, and M. S. Kumar, “Optimization of 316 Stainless Steel Weld Joint
Characteristics using Taguchi Technique,” Procedia Eng., vol. 97, pp. 881–891, Jan. 2014.
[12] I. Design and S. N. Divakar, “Prediction of Weld Bead Geometry and its Optimization During
TIG Welding Process Prediction of Weld Bead Geometry and its Optimization During TIG
Welding Process,” National Institute of Technology, Rourkela, 2015.
[13] L. Pérez Pozo, Z. Fernando Olivares, and A. Orlando Durán, “Optimization of welding
parameters using a genetic algorithm: A robotic arm-assisted implementation for recovery of
Pelton turbine blades,” Adv. Mech. Eng., vol. 7, no. 11, pp. 1–17, 2015.
[14] P. J. Jose and M. D. Anand, “Optimization of Weld Bead Profile Parameters in TIG Welding
Process for Inconel 718 Alloy Using RSM and Regression Analysis,” vol. 9, no. 4, pp. 1953–
1962, 2016.
[15] Z. Gao et al., “Parameters optimization of hybrid fiber laser-arc butt welding on 316L stainless
steel using Kriging model and GA,” Opt. Laser Technol., vol. 83, pp. 153– 162, Sep. 2016.
[16] N. Ghosh, P. K. Pal, and G. Nandi, “Parametric Optimization of MIG Welding on 316L
Austenitic Stainless Steel by Grey-based Taguchi Method,” Procedia Technol., vol. 25,pp.
1038–1048, Jan. 2016.
[17] Ghosh, P. Kumar Pal, G. Nandi, and R. Rudrapati, “Parametric Optimization of Gas metal arc
welding process by PCA based Taguchi method on Austenitic Stainless Steel AISI 316L,”
Mater. Today Proc., vol. 5, no. 1, pp. 1620–1625, Jan. 2018.
[18] Narsimhachary, D., K. Dutta, S. M. Shariff, G. Padmanabham, and A. Basu. "Mechanical and
microstructural characterization of laser weld-brazed AA6082- galvanized steel joint."
Journal of Materials Processing Technology (2018).
[19] Samanta, S.K., S.K. Mitrs, and T. K. Pal. "Effect of rare earth elements on microstructure and
oxidation behaviour in TIG weldments of AISI 316L stainless steel." Materials Science and
Engineering: A 430, no. 1-2 (2006): 242-247.