Multi-objective Particle Swarm Algorithm Based Prediction of Maintenance Work Hours in Distribution Substation

As the hub part of the power system, the overhaul work of substations is crucial to maintain the normal operation of substations. Equipment damage can be caused by insufficient or inaccurate data and difficulties in variable selection. Aiming at the above problems, this study explores a multi-objective particle swarm algorithm based maintenance labor hour prediction method for distribution network substations. The method takes the maintenance working hours as the objective function, and considers the time period in the maintenance process, the mutually exclusive relationship between maintenance work and the resources of the maintenance team as the constraints, and establishes a flexible hybrid AC and DC distribution network substation maintenance working hours prediction model. Meanwhile, the coefficient optimization process of multi-objective particle swarm algorithm is designed for accomplishing the accurate prediction of substation maintenance working hours. The experimental results show that the multi-objective particle swarm algorithm is able to consider multiple optimization objectives at the same time, such as minimizing the maintenance cost, shortening the maintenance time, and improving the quality of maintenance, so as to obtain a more comprehensive and reasonable optimization scheme. The particle swarm algorithm has a good global search capability, which can quickly locate the approximate area of the optimal solution; at the same time, it also has a local search capability, which can carry out a fine search near the optimal solution and improve the accuracy of the solution.