IJSRD - International Journal for Scientific Research & Development| Vol. 3, Issue 03, 2015 | ISSN (online): 2321-0613 All rights reserved by www.ijsrd.com 1457 Process Optimization and Estimation of Oxygen Assisted Wire Electrical Discharge Machining Performance using Artificial Neural Network Mihir S. Rana 1 Dharmin M. Pavagadhi 2 Vijay A. Bhagora 3 Prof. B.C.Khatri 4 Prof. J.B.Valaki 5 1,2,3,4,5 Department of Mechanical Engineering 1,2,3,4 L D College of Engineering, Ahmedabad 5 Government Engineering College, Bhavnagar Abstract— Oxygen assisted WEDM is a thermo- electrical process in which material is removed by a series of sparks between work piece and wire electrode (tool) of material AISI202. The part and wire are immersed in a dielectric (electrically non-conducting) fluid, usually oxygen which also acts as a coolant and flushes the debris away. The material which is to be cut must be electrically conductive. By observing present research based on literature it is identify the gap among them and finely decided that optimization of process parameter of OXYGEN ASSISTED wirecut EDM OF AISI202 using artificial neural network. Wire Electrical Discharge Machining (WEDM) is used where parts are accurately machined with varying hardness or complex shapes, which have sharp edges that are difficulties observed in conventional machining process. This study outlines the development of model and its application to optimize WEDM machining parameters using the Taguchi’s technique which is based on the robust design. Experimentation was performed as per Taguchi’s L’16 orthogonal array. Each experiment has been performed under different cutting conditions of pressure, pulse-on, pulse-off and peak current,. Among different process parameters voltage and flush rate were kept constant. Molybdenum wire having diameter of 0.18 mm was used as an electrode. Three responses namely material removal rate, surface roughness have been considered for each experiment. Based on this analysis, process parameters are optimized. ANOVA is performed to determine the relative magnitude of the each factor on the objectivefunction. Estimation and comparison of responses was done using artificial neural network. Key words: Wedm, Mrr, Surfaceroughness, Ann, Anova I. INTRODUCTION In spite of the advantages, conventional EDM process has certain limitations in Production application, including low material removal rate, long lead time for reshaped tool preparation, large tool wear, environmental concern caused by toxic dielectric disposal, etc. One of the main sources of environmental pollution during the machining processes is the huge amount of supplied cutting fluids. The lubricants are widely considered to be a benefit to cutting operations, but despite the recognition of their advantages, it has also been stated the negative impact and environmental issues associated with their use. To avoid the problems caused by the use of cutting fluids, considerable progress has been made in the last years in the field of near-dry machining. The conversion from conventional processes to minimal quantity lubrication methods demands new tasks classification in the tribology system in order to guarantee the process. Near dry WIRE EDM is one such method to reduce dielectric disposal and to obtain a better finish machining at low pulse discharge energy. In wire electrical discharge machining (WEDM), or wire-cut EDM, a thin single strand metal wire, usually brass, is fed through the work piece, typically occurring submerged in a tank of dielectric fluid. This process is used to cut plates as thick as 300mm and to make punches, tools, and dies from hard metals that are too difficult to machine with other methods. Wire-cutting EDM is commonly used when low residual stresses are desired. Wire EDM may leave residual stress on the work piece that are less significant than those that may be left if the same work piece were obtained by machining. In fact in wire EDM there are not large cutting forces involved in the removal of material. Yet, the work piece may undergo to a significant thermal cycle, whose severity depends on the technological parameters used. Possible effects of such thermal cycles are the formation of a recast layer on the part and the presence of tensile residual stresses on the work piece. If the process is set up so that the energy/power per pulse is relatively little (typically in finishing operations) little change in the mechanical properties of a material is expected in wire-cutting EDM due to these low residual stresses, although material that hasn't been stressed relieved can distort in the machining process. Fig. 1: Schematic diagram of hydro-pneumatic circuit for near-dry WEDM [14] II. LITERATURE REVIEW Qing GAO, Qin-he ZHANG, Shu-peng SU, Jian-hua ZHANG [1] (2007) they optimize EDM process parameters is introduced, which uses Levenberg-Marquardt algorithm and GA together. An ANN model was set up to represent the relationship between MRR and input parameters, which adapted Levenberg-Marquardt algorithm and its network architecture was 3-26-1. It shows that the net has better generalization performance, and convergence speed is faster. GA is used to optimize parameters. MRR is improved by using optimized parameters; it is close to experiment result. With the increase of current, MRR can be improved. MRR can also be improved when we set proper pulse on time and pulse off time with the same current.