基于K-means++与ELM的短期风电功率预测模型研究  

Research on short-term wind power forecasting model based on K-means++and ELM

作　　者：陈天阳 钱政[1] 荆博 韩妙荃 CHEN Tianyang;QIAN Zheng;JING Bo;HAN Miaoquan(School of Instrumentation and Optoelectronic Engineering,Beihang University,Beijing 100191,China)

机构地区：[1]北京航空航天大学仪器科学与光电工程学院,北京100191

出　　处：《电测与仪表》2024年第6期45-50,共6页Electrical Measurement & Instrumentation

基　　金：国家自然科学基金资助项目(61573046);教育部长江学者和创新团队发展计划项目(IRT1203)。

摘　　要：风能的波动性对风电产业的迅速发展带来了巨大挑战,准确可靠的短期风电功率预测对满足电网调度以及降低度电成本具有重要意义。文中提出了一种基于K-means++聚类分析和极限学习机ELM的短期风电功率预测方法,同时使用数值天气预报(NWP)数据与SCADA系统的历史监测数据,实现了对未来72 h的短期风电功率预测。文中通过K-means++聚类算法将NWP数据划分为数量不等的簇,使用ELM对每个簇的数据分别建立NWP数据与SCADA功率数据间的映射模型。完成模型训练后,根据数据与各聚类中心点之间的距离选择最佳预测模型。实验结果表明,与常用的经典模型相比,其预测结果精度更高,具有更高的预测性能。The volatility of wind energy has brought great challenges to the rapid development of the wind power industry.Accurate and reliable short-term wind power forecasting is essential to meet the requirement of power grid dispatching and reduce the cost per kilowatt hour of the electricity.This paper introduces a short-term wind power forecasting method based on K-means++cluster analysis and ELM,meanwhile,the 72-hour wind power forecast is realized by using SCADA data and NWP data.Firstly,K-means++clustering algorithm is applied to divide the NWP data into clusters of varying numbers.Then,ELM model is used to establish a mapping model between NWP data and SCADA power data for each cluster data.The best forecasting model is selected based on the distance between the data and the center point of each cluster after completing model training.The experimental results show that,compared with typical wind power forecast model,the proposed model has better performance in prediction accuracy.

关 键 词：K-means++聚类 ELM 短期 功率预测 NWP 

分 类 号：TM614[电气工程—电力系统及自动化]