Vol. 1 No. 2 (2024): Future scientist
Articles

A Study on Regional Power Load Forecasting Using an ARIMA-CNN-LSTM Residual Correction Model

PDF
  • Chaozheng Wang
Categories

Published 2024-09-02

Keywords

  • Power Load Forecasting,
  • ARIMA-CNN-LSTM,
  • Hybrid Model,
  • Time Series Analysis,
  • Energy Managementt

How to Cite

Wang, C. (2024). A Study on Regional Power Load Forecasting Using an ARIMA-CNN-LSTM Residual Correction Model. International Journal of Youth Science and Technology Innovation, 1(2). Retrieved from https://ijysti.org/index.php/main/article/view/9

Copyright (c) 2024 Chaozheng Wang

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Abstract

This paper presents a hybrid ARIMA-CNN-LSTM model for accurate regional power load forecasting. The model leverages the ARIMA model’s ability to capture linear trends and the CNN-LSTM network’s strength in modeling nonlinear dependencies and temporal patterns. Experimental results demonstrate that the hybrid model outperforms the standalone ARIMA model, achieving significant improvements in forecasting accuracy. With an RMSE of 3504.08, an MAE of 1466.66, and an R-squared value of 0.9902, the hybrid model proves to be an effective tool for energy management and grid planning. Its balance between accuracy and computational efficiency makes it suitable for real-time forecasting applications.

PDF