A Particle Filtering-Based Approach for the Prediction of the Remaining Useful Life of an Aluminum Electrolytic Capacitor

Abstract

This work focuses on the estimation of the Remaining Useful Life (RUL) of aluminum electrolytic capacitors used in electrical automotive drives under variable and non-stationary operative conditions. The main cause of the capacitor degradation is the vaporization of the electrolyte due to a chemical reaction. Capacitor degradation can be monitored by observing the capacitor Equivalent Series Resistance (ESR) whose measurement, however, is heavily influenced by the measurement temperature, which, under non-stationary operative conditions, is continuously changing. In this work, we introduce a novel degradation indicator which is independent from the measurement temperature and, thus, can be used for real applications under variable operative conditions. The indicator is defined by the ratio between the ESR measured on the degraded capacitor and the ESR expected value on a new capacitor at the present operational temperature. The definition of this indicator has required the investigation of the relationship between ESR and temperature on a new capacitor by means of experimental laboratory tests. The prediction of the capacitor degradation and its failure time has been performed by resorting to a Particle Filtering-based prognostic technique.