Research on feature extraction and fault diagnosis of mechanical vibration and sound signal of Disconnector

As an important switching device in power system, disconnector plays a role of protection and control in the system. However, the research and application of disconnector on-line monitoring technology is still less. In order to improve its operation reliability, it is necessary to study its condition monitoring and fault diagnosis methods. The action of the disconnector will be accompanied by a series of mechanical, electrical, acoustic and other information. Through data collection and theoretical analysis, the vibration signal, acoustic signal and opening coil current of the operating mechanism box of the disconnector are selected to monitor and analyze the disconnector in this paper. A physical test platform is built to study its fault diagnosis method. The random error empirical mode decomposition method is used to analyze the distribution of the vibration signal of the disconnector, and the vibration signal in the characteristic frequency band is extracted for the mechanical fault diagnosis of the disconnector. Through the simulation, the diagnostic schemes of mechanism jamming and loosening are obtained.