基于有效度遴选和纵横交叉算法的负荷组合预测.docVIP

第44卷 第504期 电测与仪表 Vol.44 No.504 2007年 第12期 Electrical Measurement Instrumentation Dec.2007 PAGE  PAGE - 1 - 基于有效度遴选和纵横交叉算法的负荷组合预测* 李海亮1，曹彦朝1，孟安波2，卢海明2，郭壮志2 （1.广东电网公司茂名供电局 茂名 525000; 2.广东工业大学自动化学院 广州 510006） 摘要：针对单一模型预测精度不高的问题，提出了基于有效度遴选和纵横交叉算法的负荷组合预测，该方法有效克服了选择单项模型的随机性和权重难以确定的问题。新方法根据预测有效度筛选组合预测单项模型，然后利用组合模型对小波包分解所得各个负荷子序列进行预测，并采用纵横交叉算法求取各单一模型的权值，最后叠加各子序列预测值得到完整预测结果。实例分析中，以广东省某电网实测负荷数据为例，研究结果表明：基于有效度遴选和纵横交叉算法的组合预测方法可操作性强、通用性好，并明显优于各单项预测模型、等权重组合模型和方差倒数组合模型的预测精度。 关键词：纵横交叉算法; 有效度; 小波包变换; 负荷组合预测 中图分类号：TM714 文献标识码：B 文章编号：1001-1390（2015）00-0000-00 A load Combination combination forecasting model for short-term load forecasting based on effective measure selection and crisscross optimization algorithm Li Hailiang, Cao Yanchao, Meng Anbo, Lu Haiming, Guo Zhuangzhi (1. Maoming Power Supply Bureau of Guangdong Power Grid Corporation Maoming, Maoming 525000, Guangdong, China；. 2.College of Automation,Guangdong University of Technology, Guangzhou, 510006, China) Abstract：: There exist are many uncertain factors in electric power system load, a load combination forecasting model based on effective measure selection and crisscross optimization algorithm is proposed in this paper. To addressIn view of the problem of the low prediction of single model , wavelet packet transformation is utilized to divide the original load curve into sub-sequences with different frequencies. On this basis, BP neural network, error feedback weighted time sequence and grey model are combined to form a new prediction modal,whosemodel, whose weight of every individual modal is optimized by crisscross optimization algorithm(algorithm (CSO). Finally, continuous seven daysthe short-term load forecasting with continuous seven days is implemented on certain district power grid in GuangDongGuangdong. The results prove that the proposed model is apparently su