摘要
负荷预测对合理安排发电计划,维持电网安全稳定运行具有重要意义。文中提出自下而上的区域负荷预测方法,逐线逐站展开预测工作。将深度搜索引入布谷鸟搜索算法,提高其优化精度,然后求解最小二乘支持向量机的参数,对某市某区域日有功负荷进行预测,各线、各站的预测结果与基于布谷鸟-最小二乘支持向量机、粒子群-最小二乘支持向量机、BP神经网络预测进行比较,表明该方法更加稳定、准确。最后,将文中提出方法与整体负荷预测相比较,实验结果表明,其预测结果具有可靠性,可为合理安排发电计划、电网扩建等提供可靠依据。
Load forecasting is of great significance to the reasonable arrangement of power generation plan and the maintenance of safe and stable operation of power grid.In this paper,a bottom-up regional load forecasting method is proposed,and the forecasting work is carried out line by line and station by station.In this paper,depth search is introduced into cuckoo search algorithm to improve its optimization accuracy,and then,it is used to solve the parameters of least squares support vector machine(LSSVM)to predict the daily active power load of a certain area in a city.The prediction results of each line and station are compared with those based on cuckoo LSSVM,PSO-LSSVM and BP neural network,showing the method can be more stable and accurate.Finally,the proposed method is compared with the overall load forecasting,and the experiment results show that the prediction results are reliable,which provides a reliable basis for the reasonable arrangement of power generation plan and power grid expansion.
作者
刘兴栋
邹一男
熊小东
程超
杨宏宇
LIU Xing-dong;ZOU Yi-nan;XIONG Xiao-dong;CHENG Chao;YANG Hong-yu(State Grid Chongqing Shiqu Power Supply Company,Chongqing 400000,China;Shanghai Proinvent Information Techndogy Co.,Ltd.,Shanghai 200000,China)
出处
《信息技术》
2021年第8期41-47,共7页
Information Technology
基金
国家自然科学基金(5177070051)。
