基于退火优化粒子群算法的负荷辨识方法OACSTPCD

In the research of non intrusive load identification based on particle swarm optimization,the randomness of particles is one of the important factors affecting load identification.To solve the problems of low accuracy of i-dentification results caused by particle randomness and easy to fall into local optimal traps,this paper proposes a non-intrusive load identification method based on annealing optimization particle swarm optimization algorithm com-bined with simulated annealing algorithm.Firstly,the characteristics of the household appliance load data in the REDD dataset for power load analysis are extracted;Then,using particle swarm optimization algorithm as the basic framework of simulated annealing algorithm,load identification is carried out for different appliances in each house-hold in the dataset;Finally,the error analysis is carried out with the identified power and the actual power as the standard quantities.In the single objective and multi-objective load identification problems solved by mathemati-cal models,the identification results of different types of algorithms of existing models are compared,and the results show that the proposed optimal particle swarm optimization algorithm has higher identification accuracy and better convergence.