基于最小二乘支持向量机的短期负荷预测  

Short-Term Load Forecasting Based on Least Squares Support Vector Machines

作　　者：耿艳[1] 韩学山[1] 韩力[1] 

机构地区：[1]山东大学电气工程学院,山东省济南市250061

出　　处：《电网技术》2008年第18期72-76,共5页Power System Technology

基　　金：国家自然科学基金资助项目(50377021;50677036)~~

摘　　要：提出了结合粗糙集(rough sets,RS)理论和遗传算法(genetic algorithm,GA)的最小二乘支持向量机(least squares support vector machines,LS-SVM)短期负荷预测模型和算法。由于影响负荷预测精度的因素众多,该模型采用RS理论进行历史数据的预处理,对各条件属性进行约简分析。属性约简采用GA进行寻优,以确定与负荷密切相关的因素,作为LS-SVM的有效输入变量。在预测过程中,通过GA对LS-SVM的模型参数进行自适应寻优,从而提高负荷预测精度,避免LS-SVM对经验的依赖以及预测过程中对模型参数的盲目选择。采用上述方法对山东电网负荷进行了预测分析,结果证明了该方法的有效性。A short-term load forecasting model and corresponding algorithm that is based on least squares support vector machines （LS-SVM） and integrates with rough sets （RS） theory and genetic algorithm （GA） is proposed. Because there are various factors impacting the accuracy of load forecasting, the .historical data is pre-processed by RS theory and the reduction analysis is applied to condition attributes; the optimization for attribute reduction is implemented by GA to determine factors closely related to load which are taken as effective input variables of LS-SVM. Through adaptively optimizing the model parameters of LS-SVM by GA, the accuracy of load forecasting is improved; the dependence of LS-SVM on experience and the sightless selection of model parameters during the forecasting are avoided. Applying the proposed method to load forecasting of Shandong power grid, the results show that the proposed method is effective.

关 键 词：电力系统 短期负荷预测 支持向量机 粗糙集 遗传算法 

分 类 号：TM715[电气工程—电力系统及自动化]