The Application Of Improved PSO Algorithm In Aggregate Load Modeling

Load models for power system Digital simulation play significant pole on the power system planning,design,operation and control.However, the load modeling problem has not been solved due to its characteristics of randomness, time variance and nonlinear. Inaccurate load model is one of the important factors that restricted the power system simulation accuracy, so the overly optimistic or pessimistic analysis system results may bring huge losses to power system planning and operation. Therefore, to establish the pracfical load model has great significance.The significance and the cument sitation of load modeling research are summarized in the paper. The basic theoretical knowledge of the load modeling is desctibed considering the model structure and model parameter identification.On this basis, the basic PSO algorithm theory and implementation are deeply studied. In order to solve the problem of setting parameters and premature situation, the two improved algorithm of chaos PSO and adaptive PSO are researched in the paper. The adaptive PSO algorithm is applied for the static power function load model parameter identification. Comparing with the basic PSO algorithm, this algorithm can enhance the identification accuracy of load model and convergence in the degree. The chaos PSO algorithm is applied for the difference equation load model parameter identification, then interpolation and extrapolation capability are verified. Simulation results test good generalization capability and feasibility of adaptive PSO algorithm of the difference equation load model.Furthermore, a chaos adaptive PSO algorithm is presented and is applied for the third-order induction paralleled with constant impedance model parameter identification. Generalization capability of aggregate load modeis verified. The results show that the validity of chaotic adaptive PSO algorithm and practical feature of the aggregate load model.