Materials Today: Proceedings: Divya Marelli, Singh S.K., Sateesh Nagari, Ram Subbiah
Materials Today: Proceedings: Divya Marelli, Singh S.K., Sateesh Nagari, Ram Subbiah
Materials Today: Proceedings: Divya Marelli, Singh S.K., Sateesh Nagari, Ram Subbiah
a r t i c l e i n f o a b s t r a c t
Article history: Wire-cut Electric Discharge Machining (WEDM) is an efficient machining process in numerous applica-
Received 10 December 2019 tions like space craft, defense, transportation vehicles, micro systems, farm machinery. It is used to
Received in revised form 11 January 2020 machine conductive and hard materials like metal matrix composites, ceramic composites and super
Accepted 13 January 2020
alloys. Super alloys exhibit excellent mechanical strength and creep resistance at high temperatures,
Available online xxxx
good surface stability, and corrosion and oxidation resistance. The development of super alloys has pri-
marily been driven by the aerospace and power industries. In this review of papers the optimization of
Keywords:
WEDM machining parameters, responses by deploying different techniques like Taguchi Method,
Wire EDM
ANOVA
ANOVA, GRA, ANN, PCA. These techniques were utilized for examination of impact of wire-cut EDM pro-
Kerf width cess parameters on material removal rate (MRR) and surface roughness (SR) on super alloys. Parameters
TAGUCHI like pulse off time, pulse on time, servo voltage, peak current, kerf width are utilized to optimize the
Super alloys material removal rate (MRR) and surface roughness (SR). Taguchi orthogonal array was selected based
GRA on the process parameters. ANOVA is utilized for advancing parameters with the goal that greatest mate-
PCA rial removal rate and least surface roughness is acquired.
ANN Ó 2020 Elsevier Ltd. All rights reserved.
Selection and of the scientific committee of the 10th International Conference of Materials Processing and
Characterization.
https://doi.org/10.1016/j.matpr.2020.01.306
2214-7853/Ó 2020 Elsevier Ltd. All rights reserved.
Selection and of the scientific committee of the 10th International Conference of Materials Processing and Characterization.
Please cite this article as: D. Marelli, S. S.K., S. Nagari et al., Optimisation of machining parameters of wire-cut EDM on super alloy materials–A review,
Materials Today: Proceedings, https://doi.org/10.1016/j.matpr.2020.01.306
2 D. Marelli et al. / Materials Today: Proceedings xxx (xxxx) xxx
[11]. The performance of electric discharge machining (EDM) Pulse OFF time (Toff)., each parameter at four levels. L16 orthogonal
factors is arbitrary and prompts uneven evacuation of material in array is used for conducting the series of experiments.
each discharge. The designed combination of input parameters and its corre-
Inconel 625 Chemical Composition: Carbon-0.10%, Chromium- sponding surface roughness and kerf width were obtained. Taguchi
20.0–23.0%, Iron 5.00%, Silicon 0.50%, Manganese 0.50%, Sulphur technique was intended to optimize single performance of process
0.015%, Phosphorus 0.015%, Molybdenum 8.00–10.0%, Titanium parameters with high quality and lower cost. The test results are
0.40%, Cobalt 1.00%, Columbium + Tantalum 3.15–4.15%, presently changed into a sign to-commotion (S/N) proportion
Aluminium 0.40%, Nickel Balance [18]. Since surface hardness and kerf width is desired to be
Inconel 750 Chemical Composition: Ni 70% min, Cr 14–17%, Ti least, so Lower the Better characteristic is utilized for S/N ratio
2.25–2.75%, Nb 0.70–1.20%, Mn 1% max, Si 0.50% max, Cu 0.50% estimation. The ideal setting would be the one which could accom-
max, C 0.08% max plish least S/N ratio [19]. Grey theory has been generally utilized in
WEDM is a thermoelectric cutting procedure wherein material engineering analysis, and it uncovers the possibility to solve the
is expelled because of unremitting sparkles created among wire setting of ideal machining parameters related with a procedure
and workpiece isolated by a minute space which reaches from with multiple output parameters.
0.025 mm to 0.075 mm in the presence of dielectric which might Presently, the multiple objective optimization issues have been
be deionized water or hydrocarbon oil. The material expulsion changed into a single equivalent objective function optimization
happens because of dissolving and dissipation. WEDM can problem utilizing this methodology. From the grey relational grade
machine any intricate profile with exactness and precision on the values acquired, the means of the grey relational grades at various
material regardless of its hardness. It utilizes a thin wire as an degrees of procedure parameters were determined [2].
instrument with a width in the scope of 0.02–0.3 mm made up of Amitesh Goswami et al demonstrated a cause and effect dia-
copper, metal, tungsten, molybdenum or zinc covered wires with gram for recognizing the potential factors that may influence the
a brass core. It can machine material of thickness going from machining characteristics (MRR and SR) was built. From the cause
1 mm to 500 mm relying on the machine utilized. The most signif- and effect diagram and the literature on WEDM, all out six quanti-
icant exhibition characteristics in WEDM are MRR, surface comple- ties of information parameters were chosen for this examination.
tion, and kerf width [16]. In this work, L27 orthogonal array with six control factors viz.,
Ton, Toff, IP, WF, WT, SV and three associations viz. Ton Toff, Ton IP
and Toff IP have been examined. Modified linear graph is utilized
2. Literature review for the allocation of columns to the input parameters and interac-
tions in the orthogonal array. Signal to noise ratio was acquired
Sachin Ashok Sonawane et al performed a chain of experiments utilizing Minitab 16 programming. Higher is better (HB) for MRR
are executed on 5-axis ePULSE-40 sprint cut WEDM manufactured and lower is better (LB) for SR were taken for acquiring optimum
by Electronica Machine Tool Ltd., India. To ensure parallelism and machining attributes. The S/N proportion can be determined as a
perpendicularity of the workpiece concerned with the machine logarithmic transformation of the loss function
table a dial pointer is utilized. The material applied to the examina- Multi-response optimization using utility concept is used in the
tion is the Nimonic-75 alloy of 30 mm 30 mm 5 mm thick study. Utility concept is different characteristics are combined and
which is clung to the negative extremity. The compound piece of evaluated by term composite index. Such a composite index repre-
Nimonic-75 amalgam is Carbon-0.08–0.15%, Chromium 18–21%, sents the utility of the item. In this paper it is assumed that the
Copper-0.5% max, Iron-5% max, Manganese-1%, Silicon-1%, overall utility is the sum of utilities of individual quality character-
Titanium-0.2–0.6% and equalization Nickel. A square of measure- istics [12].
ment 10 mm 10 mm 5 mm is cut from the workpiece with Naveen Babu et al demonstrated that sheet of 6 mm thickness,
the assistance of a metal wire of 0.25 mm diameter connected to Inconel 750 was analysed in this investigation. To perform the
the active polarity. The debris from the machining region is evacu- experiments L9 orthogonal array of Taguchi’s method was selected.
ated by de-ionized water which is utilized as a dielectric with a The variables which are having significant impact on the perfor-
flushing pressure of 15 kg/cm2. The conductivity of the dielectric mance of electric discharge machining was distinguished. They
is held at a steady value of 20 mS/cm at 22 °C. Testing is done by are (I) Pulse On schedule (ms), Pulse Off time (ms) (iii) Voltage
zero wire offset. The servo feed value is fixed at 2120 units. (V) and (iv) current (Å) and these components were controlled dur-
In this study, six machining factors are considered, they are, ing machining. A series of trial experiments were undertaken to
pulse on time, servo voltage, peak current, pulse off time, wire feed determine the functional range of the factors. As the individual fac-
rate, and cable tension. To decide on the suitable scope of machin- tor range was less, three level plan was utilized. Each bit of the size
ing factors, a series of pilot phase tests were led by moving each is 100 10 mm. For each machining the time taken was noted
factor in turn while keeping different elements stable at some down. Likewise, after each machining weight was gauged. The
value. Considering the thickness of the workpiece and the devia- material removal rate is determined [4].
tion of machining characteristic, MRR about the cutting variables, Subrahmanyam et al performed tests on a wire-cut EDM
the levels of the machining components are fixed. Taguchi’s L27 machine with particular like design fixed column, moving table
orthogonal array is utilized to achieve the analyses, and every trial type with a size of the work piece 250 350 mm, type of interpo-
is performed thrice consequently, overall 81 operations. The lation is linear, power supply 3 stage, AC 415 V, 50 Hz. The work
machining attributes required for the study are surface roughness, material is INCONEL 625 of size 30 15 2 mm plate. Wire diam-
overcut and MRR [1]. eter 0.25 mm made of Brass is utilized for the test. The wire is ten-
Durairaj et al used Taguchi Technique to design the experi- sioned between the lower and upper guide for getting higher
ments [17], orthogonal arrays were generally utilized in planning precision. Dielectric liquid is deionized water with 12 to 16 TDS
experiments. It is utilized to decrease the number of trials should (absolute broke up solids). For measurement of the Surface rough-
have been performed than the full factorial experiment. In view ness, surface roughness analyzer was utilized.
of the machine tool, cutting tool and work piece capability, the pro- During the machining, both of the working surfaces may have
cess parameters and the level for the process parameters were present smooth and irregularities which causes least and the most
selected such as Gap Voltage, Wire Feed, Pulse ON schedule (Ton), extreme gap in the middle of the tool and workpiece [13]. At a
Please cite this article as: D. Marelli, S. S.K., S. Nagari et al., Optimisation of machining parameters of wire-cut EDM on super alloy materials–A review,
Materials Today: Proceedings, https://doi.org/10.1016/j.matpr.2020.01.306
D. Marelli et al. / Materials Today: Proceedings xxx (xxxx) xxx 3
Please cite this article as: D. Marelli, S. S.K., S. Nagari et al., Optimisation of machining parameters of wire-cut EDM on super alloy materials–A review,
Materials Today: Proceedings, https://doi.org/10.1016/j.matpr.2020.01.306
4 D. Marelli et al. / Materials Today: Proceedings xxx (xxxx) xxx
Table 1
Grey Relational Grade value for corresponding levels [2].
gap set voltage (6.68%), peak current (1.45%), wire feed (0.84%) and Naveen Babu et al utilized the ANN method which offers a more
wire tension (0.44%). A low value (1.49%) of experimental error has noteworthy prescience than some other model, such as, linear and
been observed. All the three interactions (Ton Toff, Ton IP and Toff IP) exponential regression. The utilization of ANN is acknowledged by
have seen as critical. So also the outcome for SR has been arranged. researchers as an prediction tool because of its remarkable capabil-
For SR, just four components: pulse on time (66.70%), spark gap set ity of learning algorithm and coordinating of input and output
voltage (11.73%), pulse off time (9.09%) and peak current (3.75%) association, in any event, for other than linear and complicated
have been seen as noteworthy. An interaction has additionally systems [14,15].
been found between pulse on time and peak current. Surface In Particle swarm optimisation (PSO), every particle is viewed
roughness is mostly effected by pulse on time. The brass wire of as an individual among the population in dimensional solution
cutting tool quickens depletion, which prompts in creation of built space. Every particle is initiated with the position and velocity ran-
up layer. This increased built up layer brings about more unpleas- domly and first position of every particle is assessed dependent on
ant surface. Wire feed rate and wire tension strain has irrelevant the target functions [21]. At that point gbest is distinguished
impact on surface harshness. among the outcomes yielded by the all particles in the populations.
Henceforth, in Wire EDM of Nimonic 80A, higher machining Based gbest, the pbest particle is fixed. Next velocity is added to
rates couldn’t be acquired without giving up surface quality. the p best particle and made new population which characterizes
The examination of the microstructure of machined work surface the new (second) position of all particles. On the other hand gbest
was performed for appraisal of the surface quality utilizing WEDM is distinguished among the outcomes yielded by the new made
process. The Specimens were seen with scanning electron micro- population which depends on ongoing pbest particle. This
scope (Hitachi S-3400 N) with an accelerating voltage of 10.0 kV. procedure is proceeded until the stop criteria take an action on
Three examples were chosen for microstructure perception, one the algorithm either number of iterations or no changes on the g
example was machined at trial condition comparing to high info best value
vitality where beat on time is set at its most significant level, beat At long last the model is accomplished with RMSE and assur-
off time being at least level, while top current has middle of the road ance coefficient for MRR is 0.0053881 and 0.99995 and for SR is
level. Another two examples were machined at exploratory condi- 0.0038324 and 0.99992 respectively. To execute the optimisation
tion relating to low and moderate input energy rate (Figs. 6–8). in PSO the accompanying parameters were utilized [22]. The
Please cite this article as: D. Marelli, S. S.K., S. Nagari et al., Optimisation of machining parameters of wire-cut EDM on super alloy materials–A review,
Materials Today: Proceedings, https://doi.org/10.1016/j.matpr.2020.01.306
D. Marelli et al. / Materials Today: Proceedings xxx (xxxx) xxx 5
Fig. 6. a) Microstructure of the sample machined at low input energy rate, b) Microstructure of the sample machined at high input energy rate c) Microstructure of the
machined sample [3].
Fig. 7. Multi Response Data (Utility) Main Effects Plots (MRR and SR) [3].
Fig. 9. Experimental no. vs experimental value of MRR and SR [4].
Please cite this article as: D. Marelli, S. S.K., S. Nagari et al., Optimisation of machining parameters of wire-cut EDM on super alloy materials–A review,
Materials Today: Proceedings, https://doi.org/10.1016/j.matpr.2020.01.306
6 D. Marelli et al. / Materials Today: Proceedings xxx (xxxx) xxx
Fig. 12. (a) Main effects plot for Means; (b) Main effects plot for S/N ratios [5].
Fig. 13. (a) Main effects plot for means; (b) Main effects plot for S/N ratios [5].
material removal rate (MRR), surface roughness (SR) of Inconel 625 Experimental examination on wire electrical discharge machin-
machined work piece with Brass Wire tool utilizing Taguchi ing of Stainless Steel (SS304) was performed utilizing brass wire
strategy of 0.25 mm. Taguchi technique, the optimized input parameter
F-ratio builds up whether the process parameter is critical or combinations to get the minimum surface roughness are 40 V
not at a specific confidence level. The higher value of F-ratio shows gap voltage, 2 mm/min wire feed, 6 ls pulse on time, 10 ls
that any little variation of the process parameter can make a huge pulse off time and correspondingly optimized conditions to
effect on the performance characteristics. Increment in pulse off get the minimum kerf width are 50 V gap voltage, 2 mm/min
time builds MRR because increasing pulse off time number of elec- Wire Feed, 4 ls pulse on time, 6 ls pulse off time.
trons striking the work surface in a discharge thus increases the Based on the Grey relational analysis, the optimised input
erosion of material from the work surface per discharge. As per parameter combinations to get both the minimum surface
this, Pulse-off time is seen as the most critical factor influencing roughness and the nominal kerf width are 50 V gap voltage,
MRR with the contribution of 61.90% (Figs. 12,13). 2 mm/min wire feed, 4 ls pulse on time and 4 ls pulse off time.
The optimized value of surface roughness obtained through sin-
gle response optimization has been as low as 0.16 mm with per-
4. Conclusion cent contribution of pulse-on time (66.70%) and spark gap set
voltage (11.73%) have been found to dominate other factors
Principal Component Analysis (PCA) based technique take into such as pulse-off time (9.09%), peak current (3.75%), wire feed
consideration association between various quality outputs and (0.22%) and wire tension (0.18%). Only one interaction (pulse-
converts this into uncorrelated components known as principal on time peak current) has been found to be significant for
components. These principal components reduce the number of roughness as response.
aspects and trims down the convolution of the multi- The microstructure investigation of the samples machined at
characteristic problems. experimental condition corresponding to high energy input rate
The optimal settings of the machining parameters based on the has revealed a strong co-relation between the surface quality
principal component analysis for the WEDM of Nimonic-75 and energy input rate. The samples machined at high energy
alloy obtained are pulse-on time 110 ms, pulse-off time 51 ms, input condition exhibited rougher surface with lot of built-
servo voltage 40 V, peak current 230 Amp., wire feed rate edge layers, whereas the better surface quality was obtained
5 m/min and cable tension 8 g (Figs. 9–11). under low energy input conditions.
Please cite this article as: D. Marelli, S. S.K., S. Nagari et al., Optimisation of machining parameters of wire-cut EDM on super alloy materials–A review,
Materials Today: Proceedings, https://doi.org/10.1016/j.matpr.2020.01.306
D. Marelli et al. / Materials Today: Proceedings xxx (xxxx) xxx 7
The model was constructed between the process factors and [5] M. Subrahmanyam, T. Nancharaiah, Optimization of process parameters in
wire-cut EDM of Inconel 625 using Taguchi’s approach, Mater. Today. Proc.
output values by artificial neural network and the model was
(2019), https://doi.org/10.1016/j.matpr.2019.05.449.
found that the root mean square error is as low as for MRR [6] Aerospace materials: past, present and future (2012). Introduction to
(0.0053881) and SR (0.0038324). Whereas the determination aerospace materials, 15–38.
coefficient for MRR and SR is 0.99995 and 0.99992 respectively. [7] J.H. Weber, in encyclopedia of materials:science and technology, 2001.
[8] E.O. Ezugwu, Key improvements in the machining of difficult-to-cut aerospace
The model trained by ANN was suitably incorporated with the alloys, Int. J. Mach. Tools Manuf. 45 (2005) 1353–1367.
evolutionary computational techniques of particle swarm opti- [9] E.O. Ezugwu, J. Bonney, Y. Yamane, An overview of the machinability of
mization for optimizing the MRR and SR aeroengine alloys, J. Mater. Process. Technol. 134 (2003) 233–253.
[10] Manoj Kumar Sorabh, Neeraj Nirmal, A literature review on optimization of
It is evident that when working on wire electric discharge machining parameter in wire EDM, Int. J. Latest Res. Sci. Technol. 2 (1) (2013)
machine with brass wire and INCONEL 625 as work piece, the 492.
above stated parameters gave the maximum material removal [11] D. Priyaranjan Sharma, S. Narendranath Chakradhar, Effect of wire diameter on
surface integrity of wire electrical discharge machined Inconel 706 for gas
rate and minimum surface roughness. turbine application, J. Manuf. Process. 24 (2016) 170–178.
[12] W.B. Derek, Analysis for optimum decisions, Wiley, New York, 1982.
[13] A.V.S. Ramprasad, Koona Ramji, G.l. Datta, An experimental study of wire EDM
CRediT authorship contribution statement on Ti-6A1-4VAlloy, Sci. Direct Procedia Mater. Sci. 5 (2014) 2567–2576.
[14] C.R. Alavala, Fuzzy Logic and Neural Networks: Basic Concepts and
Applications, New Age International, Daryaganj, Delhi, 2008.
Marelli Divya: Conceptualization, Methodology, Writing - [15] J.L. Pilkington, C. Prestonb, R.L. Gomesa, Comparison of response surface
original draft, Writing - review & editing. S.K. Singh: Supervision. methodology (RSM) and artificial neural networks (ANN) towards efficient
N. Sateesh: Data curation, Visualization, Investigation, Supervision. extraction of artemisinin from Artemisia annual, Ind. Crop. Prod. 58 (2014)
15–24.
Ram Subbiah: Supervision, Validation.
[16] K.H. Ho, S.T. Newman, S. Rahimifard, R.D. Allen, State of the art in Wire
electrical discharge machining (WEDM), Int. J. Mach. Tools Manuf. 44 (12–13)
(2004) 1247–1259.
Declaration of Competing Interest [17] V.S. Gadakh, Parameteric Optimization of Wire Electrical Discharge Machining
using Topsis Method, Adv. Prod. Eng. Manage. 7 (3) (2012) 157.
The authors declare that they have no known competing finan- [18] S.V. Sapakal, M.T. Telsang, Parametric Optimization of MIG Welding using
Taguchi Design Method, Int. J. Adv. Eng. Res. Stud. 1 (IV) (2012) 28.
cial interests or personal relationships that could have appeared [19] Saurav Datta, Siba Sankar Mahapatra, Modeling, simulation and parametric
to influence the work reported in this paper. optimization of wire EDM process using response surface methodology
coupled with grey-Taguchi technique, Int. J. Eng. Sci. Technol. 2 (5) (2010) 162.
[20] J. Singh, M.R. Singh, Review on effects of process parameters in wire cut EDM
References and wire electrode development, Int. J. Innovat. Res. Insci. Technol. 2 (2)
(2016).
[21] S. Bharathi Raja, N. Baskar, Particle swarm optimization technique for
[1] Sachin Ashok Sonawane, M.L. Kulkarni, Optimization of machining parameters
determining optimal machining parameters of different work piece
of WEDM for Nimonic-75 alloy using principal component analysis integrated
materials in turning operation, Int. J. Adv. Manuf. Technol. 54 (2011) 445–463.
with Taguchi method, J. King Saud Univ. Eng. Sci. 30 (3) (2018) 250–258,
[22] Xuesong Yan, Can Zhang, Wenjing Luo, Wei Li, Wei Chen, Hanmin Liu, Solve
https://doi.org/10.1016/j.jksues.2018.04.001.
traveling salesman problem using particle swarm optimization algorithm, Int.
[2] M. Durairaj, D. Sudharsun, N. Swamynathan, Analysis of process parameters in
J. Comput. Sci. 9 (6) (2012).
wire EDM with stainless steel using single objective taguchi method and multi
[23] R.T. Yang, C.J. Tzeng, Y.K. Yang, M.H. Hsieh, Optimization of wire electrical
objective grey relational grade, Procedia Eng. 64 (2013) 868–877, https://doi.
discharge machining process parameters for cutting tungsten, J. Adv. Manuf.
org/10.1016/j.proeng.2013.09.163.
Tech. (2011).
[3] Amitesh Goswami, Jatinder Kumar, Optimization in wire-cut EDM of Nimonic-
[24] S. Sarkar, S. Mitra, B. Bhattacharyya, Parametric optimisation of wire electrical
80A using Taguchi’s approach and utility concept, Eng. Sci. Technol. Int. J. 17
discharge machining of g titanium aluminide alloy through an artificial neural
(2014) (2014) 236–246.
network model, Int. J. Adv. Manuf. Tech. 27 (2006) 501–508.
[4] K. Naveen Babu, R. Karthikeyan, A. Punitha, An integrated ANN – PSO approach
[25] M. Gostimirovic, P. Kovac, M. Sekulic, B. Skoic, Influence of discharge energy on
to optimize the material removal rate and surface roughness of wire cut EDM
machining characteristics in EDM, J. Mech. Sci. Technol. 26 (1) (2012) 173–
on INCONEL 750, Mater. Today. Proc. 19 (2019) 501–505, https://doi.org/
179.
10.1016/j.matpr.2019.07.643.
Please cite this article as: D. Marelli, S. S.K., S. Nagari et al., Optimisation of machining parameters of wire-cut EDM on super alloy materials–A review,
Materials Today: Proceedings, https://doi.org/10.1016/j.matpr.2020.01.306