Design an Intelligent Fault Detection System for Spring-Drive Operating Mechanism of SF6 High Voltage Circuit Breaker Using ADAMS

Abstract

High-voltage circuit breakers (HVCBs) are one of the main components of a power system that have a protective function. That is why monitoring and fault diagnosing of HVCBs is essential to prevent damage to other system parts. This paper presents an intelligent fault detection system using machine learning algorithms for a typical EDF, 72.5 kV, SF6 HVCB with a spring drive mechanism. The faults of the drive mechanism appear in the travel curve (TC) of the contacts, which is used in the design of the fault detection model. As collecting experimental data is costly, ADAMS software has been employed to provide various scenarios. The TC in both faulty and healthy modes and the opening and closing process are collected using this model. Subsequently, the database required to train the fault detection model is generated by extracting the appropriate feature from the curves. Afterward, it is possible to compare the performance of machine learning models and provide a suitable model for fault detection. Finally, using the optimum model enables us to detect the state of the HVCBs. In addition to fault detection, the proposed model can identify the source of the fault.