基于灰色新陈代谢和神经网络的中期风力发电容量预测  

Medium Term Wind Power Generation Capacity Prediction Based on Grey Metabolic-Neural Network

作　　者：钟宏宇 ZHONG Hongyu(Tonghua Power Supply Company,State Grid Jilin Electric Power Co.,Ltd.,Tonghua 134001,China)

机构地区：[1]国网吉林省电力有限公司通化供电公司,吉林通化134001

出　　处：《电器与能效管理技术》2022年第9期80-84,共5页Electrical & Energy Management Technology

基　　金：辽宁省高等学校优秀人才支持计划资助(LJQ2014136);教育部科学技术研究重大项目资助(212033)。

摘　　要：提出了一种基于灰色理论和神经网络的中期风力发电容量预测方法,目的是利用灰色理论模型的指数增长规律和混沌神经网络模型的非线性学习能力实现风力发电容量的中期预测,为电网调度提供发电依据。首先,设计了灰色新陈代谢预测模型,仿真了灰色新陈代谢模型预测的月发电量,并与实际发电量曲线对比;其次,采用线性加权法对灰色新陈代谢模型、混沌神经网络模型得到的预测结果进行优化组合,仿真了组合模型,并与实际值的曲线对比;最后,得到组合模型预测值的预测误差最小的结论,有效提高了风电场发电容量中期预测的精度。A medium term wind power generation capacity prediction method based on Grey theory and neural network is proposed.The purpose is to use the exponential growth law of the grey theory model and the nonlinear learning ability of the chaotic neural network model to realize the wind power generation capacity prediction of the wind farm,and provide the power generation basis for the power grid dispatching.First,the grey metabolism prediction model is designed,and the monthly power generation curve predicted by the grey metabolism model is simulated and compared with the actual power generation curve.Secondly,the prediction results of grey metabolism model and chaotic neural network model are optimized by the linear weighting method,and the combined model is simulated and compared with the actual curve.Finally,the conclusion that the prediction error of the combined model prediction value is the smallest is obtained,which can effectively improve the accuracy of the mid-term prediction of wind farm power generation capacity.

关 键 词：灰色新陈代谢 混沌神经网络 中期预测 组合预测 

分 类 号：TM743[电气工程—电力系统及自动化]