Discharge Quality Estimation and Significant Feature Identification in Micro-Electrical Discharge Machining

We present a method for estimating the quality of a measured discharge train in Micro-EDM processes. In contrast to generic EDM, Micro-EDM is characterized by a higher frequency of discharges and is used for machining smaller features, thus making regulation more difficult. In this paper, a system similar to a fuzzy rule based classification is used to identify good and bad disharges based on discharge quality. In addition to traditional machining quality issues, we include parameters that illustrate the degree of control of the voltage pulse input over the spark gap. To aid the formulation of quick closed loop control schemes for Micro-EDM, a method for identifying the parameter/sensor that separates best the good and bad discharges, is presented. The use of this method in closed loop control schemes for Micro-EDM is discussed.