基于粗糙集和多元线性回归的电力系统中长期负荷预测

基于粗糙集理论和多元线性回归模型,提出一种预测电力系统中长期负荷的新方法。首先采用一种完备的属性约简算法对影响电力负荷的诸多相关因素进行约简,得到与负荷关系最为密切的核心因素。在此基础上,建立多元线性回归模型来预测未来年的负荷。对一个实际电网的负荷进行预测,结果表明所提方法符合中长期负荷预测的特点并具有较高的预测精度。

Mid-long Term Load Forecasting Based on Rough Sets and Multiple Linear Regression

A new mid-long term electric power load forecasting method based on rough set theory and multiple linear regression was proposed. A complete algorithm for the reduction of attributes was used to reduce attributes that influence the electric power load, then the kernel factors were determined. On this basis, the multiple linear regression model for forecasting load was established. The feasibility and effectiveness of the method proposed was verified by load forecasting of one actual electric power network.