Scholarlink

This paper presents a solar energy prediction model for Mecca using Long Short-Term Memory (LSTM) networks. The model integrates feature extraction and dimensionality reduction techniques, including Independent Component Analysis (ICA), Principal Component Analysis (PCA), Common Spatial Pattern (CSP), and Discrete Wavelet Transform (DWT). These methods enhance predictive accuracy by reducing noise and capturing temporal dependencies. Results indicate significant improvement in prediction performance, with lower training and validation losses, demonstrating the model’s potential for renewable energy forecasting in critical regions like Mecca.

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

Solar Energy Prediction in Holy City of Mecca Using LSTM: Enhanced with ICA, PCA, CSP, and DWT

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Hafid Ben Hachour,
  • Said Ziani,
  • Mohamed El Ghmary,
  • Hassan Echoukairi

摘要

This paper presents a solar energy prediction model for Mecca using Long Short-Term Memory (LSTM) networks. The model integrates feature extraction and dimensionality reduction techniques, including Independent Component Analysis (ICA), Principal Component Analysis (PCA), Common Spatial Pattern (CSP), and Discrete Wavelet Transform (DWT). These methods enhance predictive accuracy by reducing noise and capturing temporal dependencies. Results indicate significant improvement in prediction performance, with lower training and validation losses, demonstrating the model’s potential for renewable energy forecasting in critical regions like Mecca.