摘要
针对电力系统年用电量增长的特点,提出一种基于加权最小二乘支持向量机(LS-SVM)的电力负荷预测模型。与标准LS-SVM的电力预测方法比较,该模型能通过设置训练样本权重比例,实现样本优化选择,达到历史数据"重近轻远"的学习效果,从而能有效提高预测精度。通过具体实例验证,WLS-SVM模型预测精度明显优于标准LS-SVM模型,说明本文模型实现容易,鲁棒性好,预测精度高。
According to the speciality of electric power consumption development, an adaptive prediction model based on the weighted least squares support vector machine (WLS-SVM) was applied to load forecast of power system. Compared with standard LS-SVM prediction method, the model could implement the optimized selection, determine the weight proportions by history data and enhance forecasting precision. In the same condition, the results show that the prediction accuracy of the weighted LS-SVM is higher than the st...
出处
《微计算机信息》
北大核心
2008年第4期312-314,共3页
Control & Automation
基金
江西省教育厅科技项目(2007328号)资助
关键词
加权最小二乘支持向量机
回归
电力负荷
预测
Weighted least squares support vector machine（WLS-SVM）
regression
power load
forecasting
