短期负荷概率性预测的混沌时间序列方法  

Short-term probabilistic load forecasting using chaotic time series

作　　者：方仍存[1,2] 周建中[1] 张勇传[1] 刘力[1] 

机构地区：[1]华中科技大学水电与数字化工程学院,湖北武汉430074 [2]湖北省电力试验研究院,湖北武汉430077

出　　处：《华中科技大学学报（自然科学版）》2009年第5期125-128,共4页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：科技部水利部公益性行业科研专项资助项目(200701008);国家自然科学基金资助项目(50579022);国家自然科学基金重点资助项目(50539140)

摘　　要：提出了一种基于混沌时间序列的短期负荷概率性预测方法.该方法采用混沌时间序列预测得到确定性预测结果,进而计算局部预测方差,并根据不同置信水平下的历史预测误差样本的双侧分位数估计值,构造概率性预测区间,实现短期负荷的概率性预测.采用北方某电网负荷数据进行了实验,验证了该方法的可行性与有效性.A novel short-term probabilistie load forecasting method was presented. The deterministic forecasting results and local predictive variance were obtained by chaotic time series method. The distribution and the percentiles of history load forecasting errors were estimated. The local predictive variance and the deterministic load forecasting result, the probabilistic load forecasting results could be gained by combining the percentiles. The practicability and validity of the proposed method were tested with the actual data of a northern power grid.

关 键 词：短期负荷预测 概率性预测 混沌时间序列 分位数 置信水平 

分 类 号：TM715[电气工程—电力系统及自动化]