基于PSO优化和LIBSVM的中长期负荷预测

Mid-long Term Load Forecasting Based on Particle Swarm Optimization Algorithm and Library for Support Vector Machines

提出一种基于粒子群算法(Particle Swarm Optimization,PSO)和LIB SVM组合模型的中长期负荷预测,其中PSO算法用于对LIB SVM参数的寻优。对山西某地区的中期负荷预测实际算例仿真分析表明,所提出的方法预测精度明显优于传统的基于结构风险最小化理论的支持向量机(Support Vector Machine,SVM)方法,且收敛速度较快,因此,该算法用于中长期电力负荷预测是可行的。

This paper proposes a new model for MTLF based on Particle Swarm Optimization algorithm （PSO） and Library for Support Vector Machines （LIB SVM）, and PSO algorithm is employed to optimize the parameters of LIB SVM. A case study of PSO-LIB SVM on mid-long term load prediction of an actual area of Shanxi was done, and the simulation results showed that the proposed model which is superior in convergence rate can offer more accurate forecasting resuhs than traditional SVM method does. Therefore, it is feasible for the new model to be applied in mid-long term load forecasting of electric power system.