Toward Stable Power Systems in Complex Environments: A Hybrid Forecasting Model Validated Across Two Typical Regions in China
Jinfeng Zhao, Changqi Tao, Jiahao Fu, Yong Wang
Keywords:
Power prediction, Deep learning, Stability
Abstract:
This study explores the application of artificial-intelligence technology in power prediction, demonstrating that deep learning-based high-precision forecasting can provide critical technical support for power system stability in complex environments. By integrating deep-learning methods, this study proposes a hybrid approach combining complete ensemble empirical mode decomposition with adaptive noise and variational mode decomposition to uncover intrinsic patterns in photovoltaic power sequences. A convolutional neural network – bidirectional long short-term memory – Transformer prediction model is constructed, and the slime mold algorithm is introduced to optimize its hyperparameters. The model is validated using heterogeneous datasets from two geographically distinct regions—the East China Plain and Central China Hills—facilitating cross-climate and cross-terrain experimentation. The impact of improved prediction accuracy on power-system stability is illustrated through theoretical analyses and practical cases from the two regions. Results show that the integrated signal decomposition and deep-learning model maintains high prediction accuracy under complex climatic and topographic conditions, contributing to power system stability. A key innovation of this research lies in the construction of an intelligent forecasting framework that provides a reliable decision-support tool for power enterprises in the two regions, mitigating the operational risks caused by dispatch deviations and enhancing the system stability and management efficacy during the energy-structure transition.
Fulltext download:
 Toward Stable Power Systems in Complex Environments: A Hybrid Forecasting Model Validated Across Two Typical Regions in China [PDF file] [Filesize: 2.19 MB]
 10.7441/joc.2026.03.11
Zhao, J., Tao, C., Fu, J., & Wang, Y. (2026). Toward stable power systems in complex environments: A hybrid forecasting model validated across two typical regions in China. Journal of Competitiveness, 18(3), 323–358. https://doi.org/10.7441/joc.2026.03.11
|
