ON-LINE FAULT DETECTION AND LOCALIZATION IN ELECTRICAL DC-DRIVES BASED ON PROCESS PARAMETER ESTIMATION AND STATISTICAL DECISION METHODS

Abstract The monitoring of technical processes has become an increased interest due to the higher demands on reliability and safety. This paper presents a method for fault detection and -localization in elecrical d.c.-drives based on a continuoustime parameter estimation procedure, implemented on a digital micro-process-computer. Using a recursive estimation procedure, the model parameters of the differential equations describing the dynamic behaviour of the d.c.-drive are estimated. Then the physical process coefficients P of ^the drive (armature resistance and -inductance, magnetic flux linkage, friction coefficient, etc.) are calculated. A subsquent modified statistical Bayestest is used to decide between (M+1) hypothesis {H.,i=0,…,M}: H describes the normal operating d.c.-drive, and H. involves a fault-state of type i. In order to demonstrate the applicability of the method, the paper will be completed by simulation studies based on a real d.c.-motor- pump system with a power consumption of 4 kW at 3000 rev/min.