Condition Monitoring of DC-Link Capacitors Using Goertzel Algorithm for Failure Precursor Parameter and Temperature Estimation

Abstract

Aluminum electrolytic capacitors are some of the weakest components in a power converter system. Therefore, condition monitoring of capacitors can prevent unexpected system downtime due to capacitor failures. This article proposes a condition monitoring system for electrolytic capacitors used as dc-link capacitors in front-end rectifier-fed three-phase inverters. The proposed methodology is based on equivalent series resistance (ESR) and capacitance estimation using the Goertzel algorithm. The ESR is used as the degradation indicator, whereas the capacitance value is used as both degradation indicator and temperature-sensitive electrical parameter to estimate core or hotspot temperature. Even though the use of capacitance to estimate hotspot temperature has already been proposed, this article tackles the practical difficulties associated with capacitor core-temperature estimation using the capacitance and demonstrates a successful implementation strategy for applying this method in power converters to estimate core temperature online. It also introduces a self-calibration procedure to mitigate ageing-related errors in temperature estimation. An experimental setup is used to verify the proposed conditioning monitoring system. The experimental results prove the effectiveness and accuracy of the proposed technique.