An intelligent pulse classification system for electro-chemical discharge machining (ECDM) - a preliminary study

Electro-chemical discharge machining (ECDM) is a hybrid process which combines features of electro-chemical machining (ECM) and electro-discharge machining (EDM). In order to develop a control strategy for this complex process, a preliminary study of a pulse classification system was carried out. By observing the voltage and current waveforms, pulses were classified in to five groups. A feed-forward neural network was trained to classify pulses with various activation functions. Five different activation functions have been employed for comparison. The trained neural networks were simulated. A quantitative analysis was performed to evaluate the performance of pulse classification system.