Abstract—In spite of much work done in mapping between the process parameters and performance indicators of electrochemical micro-machining (EMM), very sparse research is available on the optimization of its process parameters. In this article, first, an ANN trained using a hybrid Simulated Annealing (SA) – Levenberg-Marquardt (LM) is developed to map between the process parameters (voltage, feed-rate, and pulse-on time) and performance indicators (inlet and outlet diameters) of EMM. Once the prediction capabilities of the ANN are verified by the use of several testing data sets, the trained ANN is then used as a fitness function to optimize the process parameters of EMM that would lead to the minimization of taper and overcut. The optimization of the process parameters was accomplished using a Genetic Algorithm (GA) based approach. The prediction model was further validates by comparing the tendencies seen in the prediction model to those obtained using partial correlation coefficient.
Index Terms—Electrochemical micro-machining (EMM), genetic algorithm (GA), levenberg-marquardt (LM) algorithm, simulated annealing (SA).
P. Zou is with Donghua University, Songjiang, District, Shanghai 201620, China (e-mail: zoupzp123@gmail.com).
M. Rajora and S.Y. Liang are with Georgia Institute of Technology, Atlanta, Ga, 30332, USA (e-mail: manikrajora@gmail.com, steven.liang@me.gatech.edu).
M. Ma, H. Chen, and W. Wu are with Metal Industries Research & Development Center, Taichung 207 Taiwan, ROC (e-mail: mmy@mail.mirdc.org.tw, hyc@mail.mirdc.org.tw, hyc@mail.mirdc.org.tw).
[PDF]
Cite: Pan Zou, Manik Rajora, Mingyou Ma, Hungyi Chen, Wenchieh Wu, and Steven Y. Liang, "Electrochemical Micro-Machining Process Parameter Optimization Using a Neural Network-Genetic Algorithm Based Approach," International Journal of Materials, Mechanics and Manufacturing vol. 6, no. 2, pp. 82-87, 2018.
![[Click]](http://www.ijmmm.org/img/ijmmm_v7n2_inspec%28IET%29_indexscreenshot.jpg){kind=link}