Artificial Neural Network Training Algorithms in Modeling of Radial Overcut in EDM

This chapter describes with the comparison of the most used back propagations training algorithms neural networks, mainly Levenberg-Marquardt, conjugate gradient and Resilient back propagation are discussed. In the present study, using radial overcut prediction as illustrations, comparisons are made based on the effectiveness and efficiency of three training algorithms on the networks. Electrical Discharge Machining (EDM), the most traditional non-traditional manufacturing procedures, is growing attraction, due to its not requiring cutting tools and permits machining of hard, brittle, thin and complex geometry. Hence it is very popular in the field of modern manufacturing industries such as aerospace, surgical components, nuclear industries. But, these industries surface finish has the almost importance. Based on the study and test results, although the Levenberg-Marquardt has been found to be faster and having improved performance than other algorithms in training, the Resilient back propagation algorithm has the best accuracy in testing period.

EDM of D2 Steel: Performance Comparison of EDM Die Sinking Electrode Designs

Electric discharge machining (EDM) of tool steel (D2 grade) has been performed using different tool designs to produce through-holes. Machining performance has been gauged with reference to machining time, hole taper angle, overcut, and surface roughness. Inaccuracies and slow machining rate are considered as the most common limitations of the electric discharge machining (die-sinking). Traditionally, a cylindrical tool is used to form circular holes through EDM. In this study, the hole formation is carried out by changing the tool design which is the novelty of the research. Two-stage experimentation was performed. The newly designed tools substantially outperformed a traditional cylindrical tool, especially in terms of machining time. The main reason for the better machining results of modified tools is the sparking area that differs from the traditional sparking. Comparing against the performance of a traditional cylindrical tool, the newly designed tools offer a considerable reduction in the machining time, radial overcut, and roughness of the inside surfaces of machined holes, amounting to be approximately 50%, 30.6%, and 38.7%, respectively. The drop in the machining time along with a condensed level of radial overcut and surface roughness can shrink the EDM limitations and make the process relatively faster with low machining inaccuracies.

Fabrication of array of micro circular impressions using different electrolytes by maskless electrochemical micromachining

Maskless electrochemical micromachining (EMM) is a prominent technique for producing the array of micro circular impressions. A method for producing the array of micro circular impressions on stainless steel workpiece applying maskless electrochemical micromachining process is presented. The experimental setup consists of maskless EMM cell, electrode holding devices, electrical connections of electrodes and constricted vertical cross flow electrolyte system to carry out the experimental investigation. One non-conductive masked patterned tool can produce more than twenty six textured samples with high quality. A mathematical model is developed to estimate theoretically the radial overcut and machining depth of the generated array of micro circular impressions by this process and corroborate the experimental results. This study provides an elementary perceptive about maskless EMM process based on the effects of EMM process variables i.e. pulse frequency and duty ratio on surface characteristics including overcut and machining depth for NaCl, NaNO3 and NaNO3 + NaCl electrolytes. From the experimental investigation, it is observed that the combined effect of lower duty ratio and higher frequency generates the best array of micro circular impressions using the mixed electrolyte of NaNO3 + NaCl with mean radial overcut of 23.31 µm and mean machining depth of 14.1 µm.

Influence of Different Powder-Suspended Dielectric on The EDM Characteristics of Inconel 825

The present work aims to study the influence of various powder-suspended dielectrics viz. aluminium (Al), graphite (C), and silicon (Si) on several EDM performance characteristics namely material removal rate (MRR), surface roughness (Ra), and radial overcut (ROC) of Inconel 825. Results indicate that the powder properties like thermal conductivity, electrical conductivity, density, and hardness have a major impact on the machining performance and the quality of the machined surface. It has been observed that the aluminium powder particles dispersed in EDM oil yield highest material removal rate as compared to the other powders whereas the silicon powder particles provide a better surface finish and least radial overcut.