Research and application of intermittent partial discharge characteristics and easy-warning system for electric equipment

Abstract

Intermittent partial discharge has become an essential threat to the safety and stability of high voltage electrical equipment. In order to reduce the sudden insulation breakdown failure due to the intermittent and latent partial discharge of the operating equipment, this paper analyzes the discharge characteristics of the intermittent partial discharge in the constant voltage mode. The UHF partial discharge spectrum of typical intermittent partial discharges is obtained, and the intermittent discharges are divided into low-risk and high-risk categories by adopting PCA and FCM cluster algorithms. Aiming at the characteristics of intermittent partial discharge and the on-line monitoring method of partial discharge, a partial discharge signal collection method based on dynamic frequency is proposed. An intelligent monitoring system of intermittent partial discharge is designed to acquire partial discharge signals initially locate defects. Then the BP neural network method is applied to automatically evaluate and early warn the internal insulation status. The test concluded that the designed monitoring system for intermittent partial discharge could dynamically track and warn the low-risk and high-risk, and evaluate the insulation status of the equipment.

Keywords

• Long-interval sparse
• Partial discharge
• Dynamic tracking
• Easy-warning
• Intensive care