电力系统及其自动化学报2011，Vol.23Issue(1)：34-38,5.
短期负荷预测的聚类组合和支持向量机方法
Application of Clustering Combination and Support Vector Machine in Short-term Load Forecasting
摘要
Abstract
In order to improve the accuracy of power system short-term load forecasting,a method of short-term load forecasting based on clustering combination and support vector machine is proposed. First, the standardized data are trained through SOM network and the initial clustering center are acquired,then the initial clustering centre is used as the input values of the C-means algorithm,and the best number of the clustering is obtained through DB index,the samples of similar days are acquired through training. Finally, support vector machine using appropriate parmaeters and kernel function are constructed and the load was forecasted point by point. The results showed that the method has a distinct advantage than simple support vector machine algorithms.关键词
短期负荷预测/聚类组合/SOM网络/C-均值/相似日Key words
short-term load forecasting/ clustering combination/ SOM network/ C-means/ similar day分类
信息技术与安全科学引用本文复制引用
梁建武,陈祖权,谭海龙..短期负荷预测的聚类组合和支持向量机方法[J].电力系统及其自动化学报,2011,23(1):34-38,5.基金项目
国家自然科学基金资助项目(60173041) （60173041）
