Chandramouli S Journal of Engineering Research and Application www.ijera.com ISSN : 2248-9622, Vol. 8, Issue 6 (Part -IV) June 2018, pp 54-58 www.ijera.com DOI: 10.9790/9622-0806045458 54 | Page Prediction and Optimization of EDM Process Parameter Using Artificial Neural Network and Genetic Algorithm Chandramouli S *, Eswaraiah ** *(Department of Mechanical Engineering,Kakatiya Institute of Technology & Science Warangal, India ** (Department of Mechanical Engineering,Kakatiya Institute of Technology & Science Warangal, India Corresponding Auther : Chandramouli S ABSTRACT The objective of the paper is to develop empirical models and prediction of machining quality for Electrical Discharge Machining of Precipitation Hardened Stainless Steel (PH Steel) with copper tungsten electrode. The important process input parameters such as peak current, pulse on time, pulse off time and tool lift time are selected to predict the machining qualities of Material Removal Rate and Surface Roughness . Taguchi experimental design L27 orthogonal array was used to formulate the experimental design. The empirical models have been developed to predict the Material Removal Rate and Surface Roughness using Regression Analysis and Artificial Neural Network (ANN). Back propagation algorithm with experimental data used to train ANN. Prediction capability of ANN model and regression models are verified with experimental data. According to results, the ANN model is better performed as compared to the regression model to predict the MRR and SR for a given range of process parameters of EDM. Finally non-dominated sorting genetic algorithm (NSGA-II) applied to obtain non-dominated solution set to achieve the maximum material removal rate and minimum surface roughness. Keywords - PH Steel, copper tungsten electrode, MRR, SR, ANN, NSGA-II -------------------------------------------------------------------------------------------------------------------------------------- Date of Submission: 14-06-2018 Date of acceptance: 29-06-2018 --------------------------------------------------------------------------------------------------------------------------------------- I. INTRODUCTION Electrical Discharge Machining (EDM) is one of the most extensively used nonconventional material removal processes. The EDM efficiency is measured in terms of machining characteristics viz. material removal rate, tool wear rate and surface roughness. EDM is a complex manufacturing process and improving a material removal rate and surface roughness are still challenging problems[4]. When new and advanced materials appear in the field it has not been possible to use existing models and hence experimental investigations are always required. Undertaking frequent tests is also not economically justified.Bharti et al. [1] Experiments have been carried out on die-sinking EDM by taking Inconel 718 as workpiece machined with copper. Artificial neural network with back propagation algorithm has been used to model EDM process. Controlled elitist non-dominated sorting genetic algorithm has been employed in the trained network and a set of pareto-optimal solutions is obtained. Sameh [2] developed model for EDM process. concluded that the total average prediction error of experimental results with that values predicted from the developed neural network model prediction was calculated as 4.4616 %. Well-trained neural network models provide fast, accurate and consistent results, making them superior to all other techniques. Bhavesh et al [3] for modeling Neural Network Toolbox with Mat lab 7.1 has been used. The neural network based process model has been generated to establish relationship between input process conditions and process responses. Panchal et al [4] presented a research work on Effect of process parameters has been examined for Copper electrode in Die Sinking EDM process of SS 440C using ANN. MRR decreases and Surface Quality increases and when Flushing speed increases, MRR increases. Shiba Narayan Sahu et al [5]. studied performance of EDM the machining parameters discharge current , pulse duration , duty cycle and voltage were used as model input variables during the development of the models. Krishna Mohana Rao and Hanumantha Rao [6] Work is aimed at optimizing the hardness of surface produced in die sinking electric discharge machining by considering the simultaneous affect of various input parameters. Multi perceptron neural network models were developed using Neuro solutions package. Genetic algorithm concept was used to optimize the weighting factors of the network. Das et al [7] In this research the prediction of surface roughness in Electrical Discharge Machining of SKD 11 Tool steel reported results indicate that the proposed model can satisfactorily predict the surface roughness in EDM. Ashikur Rahman Khan et al [8] proposed multi-layer RESEARCH ARTICLE OPEN ACCESS