Scholarlink

The stochastic and changing characteristics of wind speed signals can pose a great challenge to the accuracy of wind speed forecasting. Therefore, this work proposes a wind speed forecasting method founded on singular spectrum analysis (SSA) and ALO-LSSVM. The method firstly uses SSA to decompose the original sequence, then builds the least squares support vector machine (LSSVM) for each part separately, and finally superimposes the forecast outcomes of every component to determine the predicted wind speed value. To improve the prediction performance, the ant-lion algorithm (ALO) is applied in optimize the settings of the LSSVM. The data from a wind farm in southern China are analyzed, and the outcomes reveal that the suggested approach has greater prediction accuracy.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Combined Model Based on the Signal Decomposition Method, Optimization Method, and Machine Learning for Wind Speed Predicting

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Anfeng Zhu,
  • Qiancheng Zhao,
  • Tianlong Yang,
  • Ling Zhou

摘要

The stochastic and changing characteristics of wind speed signals can pose a great challenge to the accuracy of wind speed forecasting. Therefore, this work proposes a wind speed forecasting method founded on singular spectrum analysis (SSA) and ALO-LSSVM. The method firstly uses SSA to decompose the original sequence, then builds the least squares support vector machine (LSSVM) for each part separately, and finally superimposes the forecast outcomes of every component to determine the predicted wind speed value. To improve the prediction performance, the ant-lion algorithm (ALO) is applied in optimize the settings of the LSSVM. The data from a wind farm in southern China are analyzed, and the outcomes reveal that the suggested approach has greater prediction accuracy.