AI - BASED CONDITION MONITORING AND FAULT DIAGNOSIS OF POWER TRANSFORMERS USING MULTI-SOURCE DIAGNOSTIC DATA

Abstract

Power transformers play a crucial role in regional and national electrical power transmission and distribution systems.  Transformer failures can be catastrophic and pose a serious threat to public safety and the economy. Extensive research has been conducted to develop reliable transformer failure mode diagnostics. Traditional failure mode diagnostics usually rely on individual transformer diagnostic methods, for example, Dissolved Gas Analysis (DGA), measurement of partial discharge, assessment of transformer oil, etc. These methods, however, may not capture all transformer failures. In this study, the use of Artificial Intelligence (AI) and other advanced technologies for condition monitoring and diagnostics of power transformers is explored. The use of other diagnostic methods, for example measurement of transformer partial discharge, assessment of transformer oil, thermal imaging, and other electrical diagnostic tests are combined with the proposed AI-based methods to develop a comprehensive diagnostic framework. The focus of this research is to evaluate the potential of AI and other advance technologies in providing reliable diagnostic methods for transformer faults. Special emphasis is placed on assessment of diagnostics in a real time environment.