Haohan Hu, Hongbo Guo, Li Zhang, Wanlong Liu, Ning Li, Yan Li
{"title":"基于电力大数据的用户电力分布及负荷预测研究","authors":"Haohan Hu, Hongbo Guo, Li Zhang, Wanlong Liu, Ning Li, Yan Li","doi":"10.1109/ICCEAI52939.2021.00032","DOIUrl":null,"url":null,"abstract":"According to the new power system reform, the power sales market has become an emerging industry in the power industry. For a single high-power user, more and more detailed energy consumption analysis is required. At present, the in-depth analysis of consumer energy by various market entities has produced certain results, but rigorous academic research is scarce. According to the actual situation of the electricity sales market, this article applies the relevant principles of machine learning to electricity users. Combine the collected user power big data to extract various user energy characteristics in multiple dimensions. Use a variety of load forecasting algorithms to simulate user portraits and apply them to feature engineering. The use of non-dimensional, binarization, dimensionality reduction and other methods has improved the main influencing factors of user energy consumption. According to the energy distribution diagram, a class of load forecasting methods suitable for current electricity market entities expanded. Finally, an example used to verify the effectiveness of the research results. The load forecasting of users through the forecasting algorithm shows that the average error result is 2.65%, and the error of the overall forecast result is generally 2% to 7%. Ensure the reliability of the forecasting method.","PeriodicalId":331409,"journal":{"name":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Research of User Power Profile and Load Forecast Based on Power Big Data\",\"authors\":\"Haohan Hu, Hongbo Guo, Li Zhang, Wanlong Liu, Ning Li, Yan Li\",\"doi\":\"10.1109/ICCEAI52939.2021.00032\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"According to the new power system reform, the power sales market has become an emerging industry in the power industry. For a single high-power user, more and more detailed energy consumption analysis is required. At present, the in-depth analysis of consumer energy by various market entities has produced certain results, but rigorous academic research is scarce. According to the actual situation of the electricity sales market, this article applies the relevant principles of machine learning to electricity users. Combine the collected user power big data to extract various user energy characteristics in multiple dimensions. Use a variety of load forecasting algorithms to simulate user portraits and apply them to feature engineering. The use of non-dimensional, binarization, dimensionality reduction and other methods has improved the main influencing factors of user energy consumption. According to the energy distribution diagram, a class of load forecasting methods suitable for current electricity market entities expanded. Finally, an example used to verify the effectiveness of the research results. The load forecasting of users through the forecasting algorithm shows that the average error result is 2.65%, and the error of the overall forecast result is generally 2% to 7%. Ensure the reliability of the forecasting method.\",\"PeriodicalId\":331409,\"journal\":{\"name\":\"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-08-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICCEAI52939.2021.00032\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCEAI52939.2021.00032","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Research of User Power Profile and Load Forecast Based on Power Big Data
According to the new power system reform, the power sales market has become an emerging industry in the power industry. For a single high-power user, more and more detailed energy consumption analysis is required. At present, the in-depth analysis of consumer energy by various market entities has produced certain results, but rigorous academic research is scarce. According to the actual situation of the electricity sales market, this article applies the relevant principles of machine learning to electricity users. Combine the collected user power big data to extract various user energy characteristics in multiple dimensions. Use a variety of load forecasting algorithms to simulate user portraits and apply them to feature engineering. The use of non-dimensional, binarization, dimensionality reduction and other methods has improved the main influencing factors of user energy consumption. According to the energy distribution diagram, a class of load forecasting methods suitable for current electricity market entities expanded. Finally, an example used to verify the effectiveness of the research results. The load forecasting of users through the forecasting algorithm shows that the average error result is 2.65%, and the error of the overall forecast result is generally 2% to 7%. Ensure the reliability of the forecasting method.