欢迎访问《太阳能学报》官方网站,今天是 分享到:
ISSN 0254-0096 CN 11-2082/K

太阳能学报 ›› 2022, Vol. 43 ›› Issue (8): 273-280.DOI: 10.19912/j.0254-0096.tynxb.2020-1312

• • 上一篇    下一篇

基于多目标优化和误差修正的短期风速预测

李嘉文, 盛德仁, 李蔚, 陈坚红   

  1. 浙江大学能源工程学院,杭州 310027
  • 收稿日期:2020-12-04 出版日期:2022-08-28 发布日期:2023-02-28
  • 通讯作者: 盛德仁(1960—),男、硕士、教授,主要从事能源系统运行优化方面的研究。ShengDR@zju.edu.cn

SHORT-TERM WIND SPEED PREDICTION BASED ON MULTI-OBJECTIVE OPTIMIZATION AND ERROR CORRECTION

Li Jiawen, Sheng Deren, Li Wei, Chen Jianhong   

  1. College of Energy Engineering, Zhejiang University, Hangzhou 310027, China
  • Received:2020-12-04 Online:2022-08-28 Published:2023-02-28

摘要: 提出一种多目标优化、误差修正的短期风速混合预测模型。首先对原始风速数据进行分解,降低序列的非线性,利用一种有效的多目标优化算法优化ELM神经网络,保证预测精度和稳定性。最后采用深度学习网络LSTM对初始预测结果进行误差校正,为克服超参数选取困难,利用乌鸦算法对层神经元数量进行优化。以中国华中某风电场实际数据为例进行分析,结果表明该方法具有较高的预测精度。

关键词: 风速, 预测, 多目标优化, 误差修正, 深度学习

Abstract: This paper presents a multi-objective optimization and error correction model for short-term wind speed prediction. Firstly, decompose the original wind speed data to reduce the nonlinearity of the sequence. An effective multi-objective optimization algorithm is used to optimize the ELM neural network, which ensures the accuracy and stability of the prediction. Finally, the deep learning network LSTM is used to correct the error of the initial prediction results. In order to overcome the difficulty of super parameter selection, crow algorithm is used to optimize the number of neurons in layer. An example is given to analyze the data of a wind farm in central China, and the results show that the method has high prediction accuracy.

Key words: wind speed, prediction, multi-objective optimization, error correction, deep learning

中图分类号: