Experimental System for Monitoring and Diagnosis of a Static Power Converter

Abstract

The evolution of the computer aided design and control technologies in the field of static power converters leads also to diagnosis systems which include modules based on artificial intelligence. In order to achieve satisfactory reliability, the static power converters must even be assembled by using high reliability devices and/or be conceived as redundant topologies. The paper presents a dedicated monitoring and diagnosis system of the static power converters. It can be used both for the diagnosis of the transistors within an inverter as well as for the analysis and diagnosis of a power bridge rectifier. The diagnosis system dedicated to the power rectifiers uses the fault tree method. Based on the symptoms observed in the behavior of the power rectifiers, the diagnosis and the functional test performed by the developed system is based on consideration of the abnormal comportments and the faults which determine these. The experimental system dedicated to the bridge rectifiers analyses and identifies the causes of the faults which occur during the operation and reduces the displayed results if two sets of values of the sources which determine the same answer, differ by a single input.

Keywords

• fault diagnosis
• static power converters
• redundancy
• reliability theory
• monitoring