基于重要特征分析的水电站运营成本回归

Regression on Operating Cost of Hydropower Stations Based on the Analysis of Important Features

近年来,我国能源发展环境发生重要变化。在能源转型的大背景下,水电发展重心也需从开发新水电转为对已建水电站的精细化运营管理。为此,以某地区65座水电站运营成本为研究对象,定义了水电站标准支出成本概念,利用皮尔逊相关系数、互信息等方法进行重要特征识别,在此基础上利用列文伯格-马夸尔特方法对成本进行回归。结果表明,上述方法对水电站成本具有较好的拟合效果,为水电站运营效率对标评价、水电企业寻优补差工作提供了新思路。

The environment of China’s energy development has undergone important changes in recent years.In the context of energy transition,the focus of hydropower development also needs to shift from the development of new hydropower to the fine management of existing power stations.In view to this,the costs of 65 hydropower stations in an area are studied.The concept of standard expenditure cost of hydropower stations is defined,and the important features are identified through Pearson linear correlation analysis and mutual information nonlinear correlation analysis.On this basis,the Levenberg-Marquardt method is used to regress the cost.The results show that the proposed method has a good fitting effect on the cost of hydropower stations,which provides a new idea for the benchmarking evaluation of hydropower station’s operation efficiency and the optimization and compensation of hydropower enterprises.