A Method for online monitoring the dissolved gas in power transformer oil based on temporal characteristics

Dissolved gas-in-oil analysis (DGA) can reflect the potential faults of a power transformer. In most cases, the fault information produced by operating power transformer has temporal characteristics, and the time-varying trend can reflect the condition and evolution of faults. Integrating this kind of dynamic temporal information analysis with traditional static analysis can make general and exact description for the equipment in condition, fault property, and evolution. We use least square curve fitting method to identify the varying trend of characteristic gas. For online application, sliding window model is used to implement analysis continuously, and recursive least square estimation is used to reduce the calculation complexity. This method can make online condition monitoring and fault diagnosis integrating the static analysis results of three gas ratios and grey correlation degree analysis. Experiments using real DGA data show the effectiveness of the proposed method.