Scholarlink

The effects of climate change on wind patterns result in a rise in storms, hurricanes, and relax periods. Wind power system efficiency and reliability significantly impacted by these modifications. Researchers are developing more sophisticated wind power forecasting systems by combining additional factors, weather forecasting models, satellite data, Machine Learning (ML), and Deep Learning (DL) approaches are utilized. The present research proposes a Red Deer Optimization algorithm-based Recurrent Neural Network system (RDO-RNN) to forecast wind power data trends. The National Renewable Energy Laboratory (NREL) provided data on wind energy production across certain time periods. The proposed RDO-RNN model exceeds the other models, including the ensemble approach, the Coati Optimization Algorithm with Convolutional Neural Network and Long Short-Term Memory (COA-CNN-LSTM), and the RNN dependent on the Dynamic Fitness Al-Biruni Earth Radius (RNN-DFBER) by achieving lower MAE values of 0.0003, 0.0174, and 0.0002 as well a RMSE of 0.0022, 0.0028 respectively.

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 Red Deer Optimization-Based Deep Learning Approach for Wind Power Forecasting

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Satti Sudha Mohan Reddy,
  • Sriram Anusha,
  • M. R. Tejonidhi,
  • Khalid Nazim Abdul Sattar,
  • K. T. Haneesh Babu

摘要

The effects of climate change on wind patterns result in a rise in storms, hurricanes, and relax periods. Wind power system efficiency and reliability significantly impacted by these modifications. Researchers are developing more sophisticated wind power forecasting systems by combining additional factors, weather forecasting models, satellite data, Machine Learning (ML), and Deep Learning (DL) approaches are utilized. The present research proposes a Red Deer Optimization algorithm-based Recurrent Neural Network system (RDO-RNN) to forecast wind power data trends. The National Renewable Energy Laboratory (NREL) provided data on wind energy production across certain time periods. The proposed RDO-RNN model exceeds the other models, including the ensemble approach, the Coati Optimization Algorithm with Convolutional Neural Network and Long Short-Term Memory (COA-CNN-LSTM), and the RNN dependent on the Dynamic Fitness Al-Biruni Earth Radius (RNN-DFBER) by achieving lower MAE values of 0.0003, 0.0174, and 0.0002 as well a RMSE of 0.0022, 0.0028 respectively.