A Study on Interpretable Electric Load Forecasting Model with Spatiotemporal Feature Fusion Based on Attention Mechanism

Driven by the global “double carbon” goal, the volatility of renewable energy poses a challenge to the stability of power systems. Traditional methods have difficulty dealing with high-dimensional nonlinear data, and the single deep learning model has the limitations of spatiotemporal feature decoup...

Full description

Saved in:
Bibliographic Details
Main Authors: Shuaishuai Li, Weizhen Chen
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Technologies
Subjects:
Online Access:https://www.mdpi.com/2227-7080/13/6/219
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850164669873913856
author Shuaishuai Li
Weizhen Chen
author_facet Shuaishuai Li
Weizhen Chen
author_sort Shuaishuai Li
collection DOAJ
description Driven by the global “double carbon” goal, the volatility of renewable energy poses a challenge to the stability of power systems. Traditional methods have difficulty dealing with high-dimensional nonlinear data, and the single deep learning model has the limitations of spatiotemporal feature decoupling and being a “black box”. Aiming at the problem of insufficient accuracy and interpretability of power load forecasting in a renewable energy grid connected scenario, this study proposes an interpretable spatiotemporal feature fusion model based on an attention mechanism. Through CNN layered extraction of multi-dimensional space–time features such as meteorology and electricity price, BiLSTM bi-directional modeling time series rely on capturing the evolution rules of load series before and after, and the improved self-attention mechanism dynamically focuses on key features. Combined with the SHAP quantitative feature contribution and feature deletion experiment, a complete chain of “feature extraction time series modeling weight allocation interpretation and verification” is constructed. The experimental results show that the determination coefficient R<sup>2</sup> of the model on the Australian electricity market data set reaches 0.9935, which is 84.6% and 59.8% higher than that of the LSTM and GRU models, respectively. The prediction error (RMSE = 105.5079) is 9.7% lower than that of TCN-LSTM model and 52.1% compared to the GNN (220.6049). Cross scenario validation shows that the generalization performance is excellent (R<sup>2</sup> ≥ 0.9849). The interpretability analysis reveals that electricity price (average absolute value of SHAP 716.7761) is the core influencing factor, and its lack leads to a 0.76% decline in R<sup>2</sup>. The research breaks through the limitation of time–space decoupling and the unexplainable bottleneck of traditional models, provides a transparent basis for power dispatching, and has an important reference value for the construction of new power systems.
format Article
id doaj-art-eef124f5095c42b6999c01b1dcfcf1d3
institution OA Journals
issn 2227-7080
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Technologies
spelling doaj-art-eef124f5095c42b6999c01b1dcfcf1d32025-08-20T02:21:54ZengMDPI AGTechnologies2227-70802025-05-0113621910.3390/technologies13060219A Study on Interpretable Electric Load Forecasting Model with Spatiotemporal Feature Fusion Based on Attention MechanismShuaishuai Li0Weizhen Chen1School of Electrical and Electronic Engineering, Wuhan Polytechnic University, Wuhan 430048, ChinaSchool of Electrical and Electronic Engineering, Wuhan Polytechnic University, Wuhan 430048, ChinaDriven by the global “double carbon” goal, the volatility of renewable energy poses a challenge to the stability of power systems. Traditional methods have difficulty dealing with high-dimensional nonlinear data, and the single deep learning model has the limitations of spatiotemporal feature decoupling and being a “black box”. Aiming at the problem of insufficient accuracy and interpretability of power load forecasting in a renewable energy grid connected scenario, this study proposes an interpretable spatiotemporal feature fusion model based on an attention mechanism. Through CNN layered extraction of multi-dimensional space–time features such as meteorology and electricity price, BiLSTM bi-directional modeling time series rely on capturing the evolution rules of load series before and after, and the improved self-attention mechanism dynamically focuses on key features. Combined with the SHAP quantitative feature contribution and feature deletion experiment, a complete chain of “feature extraction time series modeling weight allocation interpretation and verification” is constructed. The experimental results show that the determination coefficient R<sup>2</sup> of the model on the Australian electricity market data set reaches 0.9935, which is 84.6% and 59.8% higher than that of the LSTM and GRU models, respectively. The prediction error (RMSE = 105.5079) is 9.7% lower than that of TCN-LSTM model and 52.1% compared to the GNN (220.6049). Cross scenario validation shows that the generalization performance is excellent (R<sup>2</sup> ≥ 0.9849). The interpretability analysis reveals that electricity price (average absolute value of SHAP 716.7761) is the core influencing factor, and its lack leads to a 0.76% decline in R<sup>2</sup>. The research breaks through the limitation of time–space decoupling and the unexplainable bottleneck of traditional models, provides a transparent basis for power dispatching, and has an important reference value for the construction of new power systems.https://www.mdpi.com/2227-7080/13/6/219power load forecastingspatiotemporal feature fusionattention mechanisminterpretability
spellingShingle Shuaishuai Li
Weizhen Chen
A Study on Interpretable Electric Load Forecasting Model with Spatiotemporal Feature Fusion Based on Attention Mechanism
Technologies
power load forecasting
spatiotemporal feature fusion
attention mechanism
interpretability
title A Study on Interpretable Electric Load Forecasting Model with Spatiotemporal Feature Fusion Based on Attention Mechanism
title_full A Study on Interpretable Electric Load Forecasting Model with Spatiotemporal Feature Fusion Based on Attention Mechanism
title_fullStr A Study on Interpretable Electric Load Forecasting Model with Spatiotemporal Feature Fusion Based on Attention Mechanism
title_full_unstemmed A Study on Interpretable Electric Load Forecasting Model with Spatiotemporal Feature Fusion Based on Attention Mechanism
title_short A Study on Interpretable Electric Load Forecasting Model with Spatiotemporal Feature Fusion Based on Attention Mechanism
title_sort study on interpretable electric load forecasting model with spatiotemporal feature fusion based on attention mechanism
topic power load forecasting
spatiotemporal feature fusion
attention mechanism
interpretability
url https://www.mdpi.com/2227-7080/13/6/219
work_keys_str_mv AT shuaishuaili astudyoninterpretableelectricloadforecastingmodelwithspatiotemporalfeaturefusionbasedonattentionmechanism
AT weizhenchen astudyoninterpretableelectricloadforecastingmodelwithspatiotemporalfeaturefusionbasedonattentionmechanism
AT shuaishuaili studyoninterpretableelectricloadforecastingmodelwithspatiotemporalfeaturefusionbasedonattentionmechanism
AT weizhenchen studyoninterpretableelectricloadforecastingmodelwithspatiotemporalfeaturefusionbasedonattentionmechanism