Impact Factor:

4.2

DOI number:

10.1016/j.epsr.2025.112593

Affiliation of Author(s):

College of Shipbuilding Engineering, Harbin Engineering University

Journal:

Electric Power Systems Research

Place of Publication:

Switzerland

Funded by:

Key Research Project in Universities of Henan Province (No.24B480012, No.25A450004), Key Specialized

Key Words:

Short-term electric load; Variational modal decomposition (VMD); Whale optimization algorithm (WOA); Sparrow search algorithm (SSA); Convolutional neural network (CNN); Bi-directional long short-term memory neural network (BiLSTM); Uncertainty analysis

Abstract:

This study proposes an innovative short-term power load hybrid prediction model. The model deeply integrates the WOA-VMD data decomposition technology, multi-dimensional uncertainty analysis method, and Sparrow Search Algorithm (SSA) optimization strategy. First, the power load data and meteorological data are selected and decomposed by the WOA-VMD method. Then, the statistical techniques are used to conduct multi-dimensional analysis for each modal, i.e., time-features, domain features, and digital features are extracted. This not only enhances the two-way integration of feature information, but also classifies the modal components. Specifically, the non-periodic components are input into the SSA-CNN-BiLSTM model, and the periodic components are input into the CNN-BiLSTM model. Meanwhile, the SSA is used to complete the optimal search of model parameters. Furthermore, the influence of meteorological factors are further integrated by jointly put into the prediction model. Finally, the output results of each model are reconstructed. Experimental results show that compared with traditional models, the hybrid model can more accurately capture the load variation characteristics, effectively reduce the prediction error, and significantly improve the goodness of fit. It provides reliable technical support for power departments to scientifically plan the construction and transformation of distribution networks.

Note:

Key Research Project in Universities of Henan Province (No.24B480012, No.25A450004), Key Specialized Research and Development Breakthrough Program in Henan Province (No.2421022 40051)

Co-author:

Xin-Yu Yang,Xin-Xin Hua,Guo-Feng Fan,Yong-Jing Wang,Anantkumar J. Umbarkar

First Author:

Li-Ling Peng

Indexed by:

Journal paper

Correspondence Author:

Wei-Chiang Hong

Document Code:

112593

Discipline:

Engineering

Document Type:

J

Volume:

254

ISSN No.:

0378-7796

Translation or Not:

no

Date of Publication:

2026

Included Journals:

SCI

Links to published journals:

https://www.sciencedirect.com/science/article/pii/S0378779625011800?via%3Dihub