基于云模型RBF支持向量机的电力系统负荷预测  

Load forecasting of power system based on cloud model and support vector machine with RBF

作　　者：王惠中[1] 刘轲[1] 朱宏毅[2] 

机构地区：[1]兰州理工大学电气工程与信息工程学院,甘肃兰州730050 [2]国网甘肃省电力公司电力科学研究院,甘肃兰州730050

出　　处：《兰州理工大学学报》2016年第4期85-88,共4页Journal of Lanzhou University of Technology

基　　金：国家自然科学基金(50967001)

摘　　要：径向基核函数(RBF)支持向量机被广泛应用于电力系统负荷预测上,然而传统方法在RBF核函数的参数选择上有很多不足之处.为了更精确地选择核函数的参数,提高短期负荷预测的精度,提出一种将云模型和RBF支持向量机相结合的新模型.通过各影响因子的高维云变换确定每个模型的RBF核函数的参数,然后通过模型的加权计算得到最终的预测值.最后,通过与传统模型的仿真对比证明该模型的预测误差比传统模型降低了1.16%,能更好地进行电力系统短期负荷预测.The support vector machine with radial-base kernel function （RBF） is widely applied to power system load forecasting and its traditional method has, however, many shortcomings in parameter selection of RBF. In order to select the parameters of the kernel function more accurately and improve the accu- racy of short-term load forecasting, a new model with combination of cloud model and RBF support vector machine is proposed. Through high-dimensional cloud transform of impact factors, the parameters of RBF of every models are determined. Then the final forecast value is obtained by means of model weighting cal- culation. Finally, it is proved by simulative comparison with traditional model that the prediction error will be reduced by 1.06 % and the short-term load forecasting of power system can well be performed.

关 键 词：云模型 云变换 径向基核函数 负荷预测 

分 类 号：TM715[电气工程—电力系统及自动化]