A LightGBM-Based Power Grid Frequency Prediction Method with Dynamic Significance–Correlation Feature Weighting
Accurate grid frequency prediction is essential for maintaining the stability and reliability of power systems. However, the complex dynamic characteristics of grid frequency and the nonlinear correlations among massive time series data make it challenging for traditional time series prediction meth...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/18/13/3308 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849319894461448192 |
|---|---|
| author | Jie Zhou Xiangqian Tong Shixian Bai Jing Zhou |
| author_facet | Jie Zhou Xiangqian Tong Shixian Bai Jing Zhou |
| author_sort | Jie Zhou |
| collection | DOAJ |
| description | Accurate grid frequency prediction is essential for maintaining the stability and reliability of power systems. However, the complex dynamic characteristics of grid frequency and the nonlinear correlations among massive time series data make it challenging for traditional time series prediction methods to balance efficiency and accuracy. In this paper, we propose a Dynamic Significance–Correlation Weighting (D-SCW) method, which generates dynamic weight coefficients that evolve over time. This is achieved by constructing a joint screening mechanism of feature time series correlation analysis and statistical significance test, combined with the LightGBM gradient-boosting decision tree (GBDT) framework; accordingly, high-precision prediction of grid frequency time series data is realized. To verify the effectiveness of the D-SCW method, this study conducts comparative experiments on two actual grid operation datasets (including typical scenarios with wind/photovoltaic (PV) installations, accounting for 5–35% of the grid); additionally, the Spearman’s rank correlation coefficient method, mutual information (MI), Lasso regression, and the feature screening method of recursive feature elimination (RFE) are selected as the baseline control; root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) are adopted as assessment indicators. The results show that the D-SCW-LightGBM framework reduces the root mean squared error (RMSE) by 5.2% to 10.4% and shortens the dynamic response delay by 52% compared with the benchmark method in high renewable penetration scenarios, confirming its effectiveness in both prediction accuracy and computational efficiency. |
| format | Article |
| id | doaj-art-cdccc34b8f8648bfb7a3d48d54f3d0cf |
| institution | Kabale University |
| issn | 1996-1073 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj-art-cdccc34b8f8648bfb7a3d48d54f3d0cf2025-08-20T03:50:17ZengMDPI AGEnergies1996-10732025-06-011813330810.3390/en18133308A LightGBM-Based Power Grid Frequency Prediction Method with Dynamic Significance–Correlation Feature WeightingJie Zhou0Xiangqian Tong1Shixian Bai2Jing Zhou3School of Electrical Engineering, Xi’an University of Technology, Xi’an 710048, ChinaSchool of Electrical Engineering, Xi’an University of Technology, Xi’an 710048, ChinaSchool of Electrical Engineering, Xi’an University of Technology, Xi’an 710048, ChinaInternational of Faculty of Social Sciences and Liberal Arts, University College Sedaya, Kuala Lumpur 56000, MalaysiaAccurate grid frequency prediction is essential for maintaining the stability and reliability of power systems. However, the complex dynamic characteristics of grid frequency and the nonlinear correlations among massive time series data make it challenging for traditional time series prediction methods to balance efficiency and accuracy. In this paper, we propose a Dynamic Significance–Correlation Weighting (D-SCW) method, which generates dynamic weight coefficients that evolve over time. This is achieved by constructing a joint screening mechanism of feature time series correlation analysis and statistical significance test, combined with the LightGBM gradient-boosting decision tree (GBDT) framework; accordingly, high-precision prediction of grid frequency time series data is realized. To verify the effectiveness of the D-SCW method, this study conducts comparative experiments on two actual grid operation datasets (including typical scenarios with wind/photovoltaic (PV) installations, accounting for 5–35% of the grid); additionally, the Spearman’s rank correlation coefficient method, mutual information (MI), Lasso regression, and the feature screening method of recursive feature elimination (RFE) are selected as the baseline control; root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) are adopted as assessment indicators. The results show that the D-SCW-LightGBM framework reduces the root mean squared error (RMSE) by 5.2% to 10.4% and shortens the dynamic response delay by 52% compared with the benchmark method in high renewable penetration scenarios, confirming its effectiveness in both prediction accuracy and computational efficiency.https://www.mdpi.com/1996-1073/18/13/3308grid frequency predictionDynamic Significance–Correlation WeightingLightGBMtime series prediction |
| spellingShingle | Jie Zhou Xiangqian Tong Shixian Bai Jing Zhou A LightGBM-Based Power Grid Frequency Prediction Method with Dynamic Significance–Correlation Feature Weighting Energies grid frequency prediction Dynamic Significance–Correlation Weighting LightGBM time series prediction |
| title | A LightGBM-Based Power Grid Frequency Prediction Method with Dynamic Significance–Correlation Feature Weighting |
| title_full | A LightGBM-Based Power Grid Frequency Prediction Method with Dynamic Significance–Correlation Feature Weighting |
| title_fullStr | A LightGBM-Based Power Grid Frequency Prediction Method with Dynamic Significance–Correlation Feature Weighting |
| title_full_unstemmed | A LightGBM-Based Power Grid Frequency Prediction Method with Dynamic Significance–Correlation Feature Weighting |
| title_short | A LightGBM-Based Power Grid Frequency Prediction Method with Dynamic Significance–Correlation Feature Weighting |
| title_sort | lightgbm based power grid frequency prediction method with dynamic significance correlation feature weighting |
| topic | grid frequency prediction Dynamic Significance–Correlation Weighting LightGBM time series prediction |
| url | https://www.mdpi.com/1996-1073/18/13/3308 |
| work_keys_str_mv | AT jiezhou alightgbmbasedpowergridfrequencypredictionmethodwithdynamicsignificancecorrelationfeatureweighting AT xiangqiantong alightgbmbasedpowergridfrequencypredictionmethodwithdynamicsignificancecorrelationfeatureweighting AT shixianbai alightgbmbasedpowergridfrequencypredictionmethodwithdynamicsignificancecorrelationfeatureweighting AT jingzhou alightgbmbasedpowergridfrequencypredictionmethodwithdynamicsignificancecorrelationfeatureweighting AT jiezhou lightgbmbasedpowergridfrequencypredictionmethodwithdynamicsignificancecorrelationfeatureweighting AT xiangqiantong lightgbmbasedpowergridfrequencypredictionmethodwithdynamicsignificancecorrelationfeatureweighting AT shixianbai lightgbmbasedpowergridfrequencypredictionmethodwithdynamicsignificancecorrelationfeatureweighting AT jingzhou lightgbmbasedpowergridfrequencypredictionmethodwithdynamicsignificancecorrelationfeatureweighting |