Research on Separation Method of Partial Discharge Mixed Signal in Switchgear

When the switchgear is running, there may be multiple insulation defects inside it, resulting in mixed PD signals collected by the partial discharge (PD) sensor, which affects the subsequent PD pattern recognition and positioning. Aiming at this problem, this paper proposes an improved particle swarm optimization (PSO) optimization joint approximate diagonalization (JADE) PD mixed-signal blind source separation algorithm, which optimizes the inertia weight formula of the traditional PSO algorithm to improve the global search of the PSO algorithm. The improved PSO algorithm and the JADE algorithm are combined in the PD mixed signal separation to overcome the shortcoming that the JADE algorithm is easy to fall into the local optimum. Finally, the PD mathematical model of the switchgear is used to construct PD mixed signals with different trigger time differences, and the performance of the algorithm is tested by adding noise of different intensities. The coefficient is significantly higher than other separation algorithms, and the algorithm execution time is shorter, and the accurate signal separation can still be ensured in the case of small trigger time difference and strong noise, indicating that this method can effectively separate mixed PD signals under different working conditions., which provides an important reference for evaluating the operating status of the equipment and arranging the maintenance cycle.