Key Technologies for a Regional Economic Development Index Based on Electricity Big Data

Electricity big data acts as a vital barometer of regional economic performance. In the transformation process of resource‑based cities, traditional statistical indicators such as GDP and industrial added value suffer from obvious lagging effects and can hardly capture economic fluctuations promptly. In contrast, electricity big data, characterized by high‑frequency and full‑coverage attributes, can serve as an important support for the dynamic monitoring of the regional economy. Taking Lvliang City, a resource‑based city, as the research object, this paper integrates multi‑source time‑series data covering power consumption, macro‑monetary indicators, industrial product prices, and key‑industry statistics. It constructs a candidate model library composed of regularized linear models and ensemble‑learning algorithms, and establishes a closed‑loop forecasting workflow of “data selection‑correlation analysis‑model evaluation‑dynamic correction” to realize adaptive matching of “one‑model‑per‑industry” for different industries. Prediction modeling is performed for Lvliang’s macro‑economy and its key industries, including coal, liquor, aluminum smelting, and iron and steel. Empirical results demonstrate that the model can achieve advanced prediction of economic trends for 3‑6 months. The coefficients of determination (𝑅𝑅2) of the optimal Lasso regression models for the macro‑economy, coal, liquor, and aluminum industries are 0.771, 0.764, 0.859, and 0.560 respectively, and the (𝑅𝑅2) value of the Elastic Net model for the iron‑steel industry is 0.694. The average monthly prediction error for each industry stays within the 3-percentage-point limit, though the coal and aluminum smelting industries approach this error boundary, reflecting heterogeneous prediction performance across sectors. The model can characterize the differentiated-recovery features of various industries. This research can provide support for governments to implement anticipatory regulation and mitigate transformation‑related risks, offer quantitative references for enterprise operation and optimal allocation of regional power resources, and facilitate the coordinated high-quality development of economy‑energy‑environment in Lyuliang.

Authors

Institutions

Publication Details

Journal
EAI Endorsed Transactions on Energy Web
Published
2026-11-19
DOI
https://doi.org/10.4108/ew.14804
Primary Topic
Energy Load and Power Forecasting
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Key Technologies for a Regional Economic Development Index Based on Electricity Big Data

李勇志 Li Yongzhi, Wei Li, Jie Li, Xuefeng Bai et al.
EAI Endorsed Transactions on Energy Web
Energy Load and Power Forecasting
article

Key Technologies for a Regional Economic Development Index Based on Electricity Big Data

李勇志 Li Yongzhi, Wei Li, Jie Li, Xuefeng Bai, Xiaoming Chen, Wen Zhao
article en

Abstract

Electricity big data acts as a vital barometer of regional economic performance. In the transformation process of resource‑based cities, traditional statistical indicators such as GDP and industrial added value suffer from obvious lagging effects and can hardly capture economic fluctuations promptly. In contrast, electricity big data, characterized by high‑frequency and full‑coverage attributes, can serve as an important support for the dynamic monitoring of the regional economy. Taking Lvliang City, a resource‑based city, as the research object, this paper integrates multi‑source time‑series data covering power consumption, macro‑monetary indicators, industrial product prices, and key‑industry statistics. It constructs a candidate model library composed of regularized linear models and ensemble‑learning algorithms, and establishes a closed‑loop forecasting workflow of “data selection‑correlation analysis‑model evaluation‑dynamic correction” to realize adaptive matching of “one‑model‑per‑industry” for different industries. Prediction modeling is performed for Lvliang’s macro‑economy and its key industries, including coal, liquor, aluminum smelting, and iron and steel. Empirical results demonstrate that the model can achieve advanced prediction of economic trends for 3‑6 months. The coefficients of determination (𝑅𝑅2) of the optimal Lasso regression models for the macro‑economy, coal, liquor, and aluminum industries are 0.771, 0.764, 0.859, and 0.560 respectively, and the (𝑅𝑅2) value of the Elastic Net model for the iron‑steel industry is 0.694. The average monthly prediction error for each industry stays within the 3-percentage-point limit, though the coal and aluminum smelting industries approach this error boundary, reflecting heterogeneous prediction performance across sectors. The model can characterize the differentiated-recovery features of various industries. This research can provide support for governments to implement anticipatory regulation and mitigate transformation‑related risks, offer quantitative references for enterprise operation and optimal allocation of regional power resources, and facilitate the coordinated high-quality development of economy‑energy‑environment in Lyuliang.

EAI Endorsed Transactions on Energy WebVol. 13
State Grid Corporation of China (China) (CN), Luliang University (CN)
Energy Load and Power Forecasting
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon