{"title":"基于集合学习和特征选择的配电网线损预测研究","authors":"Ke Zhang, Yongwang Zhang, Jian Li, Zetao Jiang, Yuxin Lu, Binghui Zhao","doi":"10.3389/fenrg.2024.1453039","DOIUrl":null,"url":null,"abstract":"IntroductionAccurate prediction of line losses in distribution networks is crucial for optimizing power system planning and network restructuring, as these losses significantly impact grid operation quality. This paper proposes a novel approach that combines advanced feature selection techniques with Stacking ensemble learning to enhance the effectiveness of distribution network loss analysis and assessment.MethodsUtilizing data from 44 substations over an 18-month period, we integrated a Stacking ensemble learning model with multiple feature selection methods, including correlation coefficient, maximum information coefficient, and tree-based techniques. These methods were employed to identify the key predictors of power loss in the distribution network.ResultsThe proposed model achieved a Mean Absolute Percentage Error (MAPE) of 3.78% and a Root Mean Square Error (RMSE) of 1.53, demonstrating a substantial improvement over traditional linear regression-based prediction methods. The analysis revealed that historical line loss and line active power were the most influential predictive variables, while the inclusion of time-related features further refined the model's performance.DiscussionThis study highlights the efficacy of combining multiple feature selection methods with Stacking ensemble learning for predicting power loss in 10 kV distribution networks. The enhanced accuracy and reliability of the proposed model offer valuable insights for electrical engineering applications, potentially contributing to more efficient and sustainable energy distribution systems. Future research could explore the applicability of this approach to other distribution network voltage levels and investigate the incorporation of additional environmental and network-specific factors to further improve power loss prediction.","PeriodicalId":12428,"journal":{"name":"Frontiers in Energy Research","volume":null,"pages":null},"PeriodicalIF":2.6000,"publicationDate":"2024-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Research on line loss prediction of distribution network based on ensemble learning and feature selection\",\"authors\":\"Ke Zhang, Yongwang Zhang, Jian Li, Zetao Jiang, Yuxin Lu, Binghui Zhao\",\"doi\":\"10.3389/fenrg.2024.1453039\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"IntroductionAccurate prediction of line losses in distribution networks is crucial for optimizing power system planning and network restructuring, as these losses significantly impact grid operation quality. This paper proposes a novel approach that combines advanced feature selection techniques with Stacking ensemble learning to enhance the effectiveness of distribution network loss analysis and assessment.MethodsUtilizing data from 44 substations over an 18-month period, we integrated a Stacking ensemble learning model with multiple feature selection methods, including correlation coefficient, maximum information coefficient, and tree-based techniques. These methods were employed to identify the key predictors of power loss in the distribution network.ResultsThe proposed model achieved a Mean Absolute Percentage Error (MAPE) of 3.78% and a Root Mean Square Error (RMSE) of 1.53, demonstrating a substantial improvement over traditional linear regression-based prediction methods. The analysis revealed that historical line loss and line active power were the most influential predictive variables, while the inclusion of time-related features further refined the model's performance.DiscussionThis study highlights the efficacy of combining multiple feature selection methods with Stacking ensemble learning for predicting power loss in 10 kV distribution networks. The enhanced accuracy and reliability of the proposed model offer valuable insights for electrical engineering applications, potentially contributing to more efficient and sustainable energy distribution systems. Future research could explore the applicability of this approach to other distribution network voltage levels and investigate the incorporation of additional environmental and network-specific factors to further improve power loss prediction.\",\"PeriodicalId\":12428,\"journal\":{\"name\":\"Frontiers in Energy Research\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":2.6000,\"publicationDate\":\"2024-08-23\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Frontiers in Energy Research\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://doi.org/10.3389/fenrg.2024.1453039\",\"RegionNum\":4,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ENERGY & FUELS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers in Energy Research","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.3389/fenrg.2024.1453039","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENERGY & FUELS","Score":null,"Total":0}
Research on line loss prediction of distribution network based on ensemble learning and feature selection
IntroductionAccurate prediction of line losses in distribution networks is crucial for optimizing power system planning and network restructuring, as these losses significantly impact grid operation quality. This paper proposes a novel approach that combines advanced feature selection techniques with Stacking ensemble learning to enhance the effectiveness of distribution network loss analysis and assessment.MethodsUtilizing data from 44 substations over an 18-month period, we integrated a Stacking ensemble learning model with multiple feature selection methods, including correlation coefficient, maximum information coefficient, and tree-based techniques. These methods were employed to identify the key predictors of power loss in the distribution network.ResultsThe proposed model achieved a Mean Absolute Percentage Error (MAPE) of 3.78% and a Root Mean Square Error (RMSE) of 1.53, demonstrating a substantial improvement over traditional linear regression-based prediction methods. The analysis revealed that historical line loss and line active power were the most influential predictive variables, while the inclusion of time-related features further refined the model's performance.DiscussionThis study highlights the efficacy of combining multiple feature selection methods with Stacking ensemble learning for predicting power loss in 10 kV distribution networks. The enhanced accuracy and reliability of the proposed model offer valuable insights for electrical engineering applications, potentially contributing to more efficient and sustainable energy distribution systems. Future research could explore the applicability of this approach to other distribution network voltage levels and investigate the incorporation of additional environmental and network-specific factors to further improve power loss prediction.
期刊介绍:
Frontiers in Energy Research makes use of the unique Frontiers platform for open-access publishing and research networking for scientists, which provides an equal opportunity to seek, share and create knowledge. The mission of Frontiers is to place publishing back in the hands of working scientists and to promote an interactive, fair, and efficient review process. Articles are peer-reviewed according to the Frontiers review guidelines, which evaluate manuscripts on objective editorial criteria