电测与仪表2017，Vol.54Issue(19)：67-72,6.
基于信息论与混合聚类分析的短期负荷预测方法研究
Study on short-term load forecasting method based on information theory and mixed cluster analysis
摘要
Abstract
The electricity consumption is affected by many factors in short-term load forecasting .In traditional method of associated factors selection , only the correlation between load and associated factor is considered , and the correla-tion between factors is ignored , which leads to the correlation redundancy .Besides, the Euclidean distance can't measure the similarity of the load curves well in the traditional cluster analysis .So, firstly, cluster analysis based on Euclidean distance mixed with the cosine similarity is made on the load curves .Then, information theory is used to select the optimal associated factors combination from 9 factors, which has considered the the correlation between fac-tors.Finally, load data and its associated factors data to be used are selected according to the classification of the load to be predicted, and the data is adopted to establish the support vector machine ( SVM) model.Through the analysis of the actual sample data in a certain area of Shanghai , the results prove that the average relative error of this method is 1.46%, and 72.72%of the relative errors are below 1%, which has a better practicability .关键词
短期负荷预测/信息论/聚类分析/支持向量机/关联因素选择Key words
short-term load forecasting/information theory/cluster analysis/support vector machine(SVM)/associ-ated factors selection分类
信息技术与安全科学引用本文复制引用
谢真桢,杨秀,张鹏,徐磊..基于信息论与混合聚类分析的短期负荷预测方法研究[J].电测与仪表,2017,54(19):67-72,6.基金项目
国家自然科学基金资助项目(51407114) （51407114）
国家电网公司科技项目(520940150010 （）
52094015001L) （）
