Purpose
Grey relational analysis (GRA) has already proved itself as an efficient tool for multi-objective optimization of many of the machining processes. In GRA, the distinguishing coefficient (ξ) plays an important role in identifying the optimal parametric combinations of the machining processes and almost all the past researchers have considered its value as 0.5. In this paper, based on past experimental data, the application of GRA is extended to dynamic GRA (DGRA) to optimize two electrochemical machining (ECM) processes.
Design/methodology/approach
Instead of a static distinguishing coefficient, this paper considers dynamic distinguishing coefficient for each of the responses for both the ECM processes under consideration. Based on these coefficients, the application of DGRA leads to determination of the dynamic grey relational grade (DGRG) and grey relational standard deviation (GRSD), helping in initial ranking of the alternative experimental trials. Considering the ranks obtained by DGRG and GRSD, a composite rank in terms of rank product score is obtained, aiding in final rankings of the experimental trials for both the ECM processes.
Findings
In the first example, the maximum material removal rate (MRR) would be obtained at an optimal combination of ECM parameters as electrolyte concentration = 2 mol/l, voltage = 16V and current = 4A, while another parametric intermix as electrolyte concentration = 2 mol/l, voltage = 14V and current = 2A would result in minimum radial overcut and delamination. For the second example, an optimal combination of ECM parameters as electrode temperature = 30°C, voltage = 12V, duty cycle = 90% and electrolyte concentration = 15 g/l would simultaneously maximize MRR and minimize surface roughness and conicity.
Originality/value
In this paper, two ECM operations are optimized using a newly developed but yet to be popular multi-objective optimization tool in the form of the DGRA technique. For both the examples, the derived rankings of the ECM experiments exactly match with those obtained by the past researchers. Thus, DGRA can be effectively adopted to solve parametric optimization problems in any of the machining processes.