Design of Partial Discharge Detection and Anti-Interference System for Distribution Network Cable Joints Based on Edge Calculation

As one of the key infrastructures in modern society, the reliability and stability of power distribution network are very important for maintaining normal life and industrial operation. As one of the fragile links in the distribution network, the real-time monitoring of cable joints' health status and partial discharge detection (PDD) are very important for preventing potential faults in advance. In this paper, a design of PDD and anti-interference system for distribution network cable joints based on edge calculation(EC) is proposed, aiming at improving the reliability and anti-interference of cable joints. In this paper, the detection unit based on EC is developed, and the sensor constantly monitors the cable joint and transmits the data to EC unit. The EC unit executes the PDD algorithm. If an abnormal discharge signal is detected, it will generate an alarm and transmit the data to the central monitoring system. Particle swarm optimization (PSO) algorithm is used to optimize the generalized regression neural network (GRNN) model. Finally, the statistical characteristic parameters are used as the input of the neural network to identify the discharge type. In order to deal with all kinds of interference sources in power system, we propose an anti-interference system design. Through signal filtering, data cleaning and machine learning model training, the system effectively reduces the influence of interference on system performance and improves the accuracy of PDD. The experimental results show that our system has excellent performance under different environmental conditions, which can accurately detect the partial discharge of cable joints and effectively deal with various interference sources.