Development of an intelligent process model for EDM

A hybrid process model for EDM based on finite-element method and Gaussian process regression

“…The detailed procedure to build the single spark EDM model is reported by Joshi [16]. Figure 5 shows the temperature distribution obtained for one typical machining condition in this study.…”

“…Joshi and Pande introduced an intelligent model [16] by use of FEM and artificial neural network (ANN) to optimize the EDM process. The results from their model were compared with the experimental data reported by DiBitonto et al [2].…”

“…To analyze the effect of key process parameters on MRR and Ra and to optimize these parameters for machining process, Taguchi method was used in the experimental researches by many researchers [31][32][33][34][35]. Joshi et al [16] and Wang et al [36] performed a hybrid approach for modeling and optimization of EDM process using ANN to obtain optimal process parameters. Mandal et al [37] developed back propagation neural network (BPNN) based on the experimental data and optimized the process parameters by the method of nondominated sorting genetic algorithm II (NSGA II).…”

“…Both of these models are simplistic to regard as the heat flux applied on the discharge channel. Thus, Gaussian distribution for heat flux is needed to approximate the heat from the plasma [15,16,19].…”

“…4). Joshi et al [16] predicts the shape of crater cavity as shallow-bowl-shaped, and calculates the volume of crater cavity by dividing it into number of cylindrical discs. The image processing techniques are used to calculate the volume of crater cavity by Marafona and Chousal [7].…”