Scholarlink

<p>Better forecasting and energy management are needed to improve grid stability and economic performance. Due to the increasing usage of Renewable Energy (RE) in power networks, these methods are needed. A novel hybrid framework using Deep Learning (DL) models, EMD, and uncertainty quantification improves Wind Power Generation (WPG), energy demand, and market price forecasts. The methodology encompasses the acquisition of Supervisory Control and Data Acquisition (SCADA), pre-processing to eliminate discrepancies, and training Neural Networks (NNs) for precise prediction. This study introduces a token-based Peer-to-Peer Energy Trading (P2P-ET) system designed to enhance prosumer energy transactions. The suggested methodology is validated using a real-world dataset from a wind farm in France. The dataset consists of 42,048 samples that were taken at 20-minute intervals over a year. The findings of the experiments show that there is a significant reduction in the errors that occur during forecasting, with improvements in MAPE and RMSE across all prediction tasks combined. The integration of forecasting models with P2P-ET can enhance market efficiency, reduce reliance on centralized energy sources, and cultivate a more resilient and flexible energy ecosystem. The results here demonstrate how the proposed framework can allow the integration of decentralized Renewable Energy Sources (RES) and optimize the functioning of the energy markets.</p>

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

Hybrid Framework for Wind Power Forecasting and Peer-to-Peer Energy Trading To Enhance Grid Stability and Market Efficiency

  • 期刊论文

  • 发表日期：

    2025年12月26日

查看全文
  • Zhipeng Li,
  • Yunxiao Gao,
  • Hao Yang,
  • Shuo Shen,
  • Jun Wang

摘要

Better forecasting and energy management are needed to improve grid stability and economic performance. Due to the increasing usage of Renewable Energy (RE) in power networks, these methods are needed. A novel hybrid framework using Deep Learning (DL) models, EMD, and uncertainty quantification improves Wind Power Generation (WPG), energy demand, and market price forecasts. The methodology encompasses the acquisition of Supervisory Control and Data Acquisition (SCADA), pre-processing to eliminate discrepancies, and training Neural Networks (NNs) for precise prediction. This study introduces a token-based Peer-to-Peer Energy Trading (P2P-ET) system designed to enhance prosumer energy transactions. The suggested methodology is validated using a real-world dataset from a wind farm in France. The dataset consists of 42,048 samples that were taken at 20-minute intervals over a year. The findings of the experiments show that there is a significant reduction in the errors that occur during forecasting, with improvements in MAPE and RMSE across all prediction tasks combined. The integration of forecasting models with P2P-ET can enhance market efficiency, reduce reliance on centralized energy sources, and cultivate a more resilient and flexible energy ecosystem. The results here demonstrate how the proposed framework can allow the integration of decentralized Renewable Energy Sources (RES) and optimize the functioning of the energy markets.