引用本文: | 杨茂,董骏城,罗芫,赵伟男.基于近似熵的电力系统负荷预测误差分析[J].电力系统保护与控制,2016,44(23):24-29.[点击复制] |
YANG Mao,DONG Juncheng,LUO Yuan,ZHAO Weinan.Study of power system load forecasting errors based on approximate entropy[J].Power System Protection and Control,2016,44(23):24-29[点击复制] |
|
摘要: |
为深入探究负荷时间序列预测误差的影响因素,提高负荷预测精度,提出近似熵算法,用于定量刻画负荷时间序列的规律性,全面认识负荷预测误差的成因。采用近似熵算法对负荷时间序列进行分析,确定其规律性的强弱。在此基础上,针对负荷时间序列的规律性与预测误差之间的关系进行研究。算例分析结果表明,近似熵算法可以有效刻画负荷时间序列的规律性,且负荷时间序列的规律性与其预测误差之间有着较强的相关性,证明了方法的正确性和有效性。 |
关键词: 负荷时间序列 负荷预测 规律性 近似熵 预测误差 |
DOI:10.7667/PSPC152071 |
投稿时间:2015-11-27修订日期:2016-02-25 |
基金项目:国家重点基础研究发展计划项目(973计划)(2013CB228201);国家自然科学基金(51307017);吉林省科技发展计划(20140520129JH);吉林省产业技术研究与开发专项项目(2014Y124) |
|
Study of power system load forecasting errors based on approximate entropy |
YANG Mao,DONG Juncheng,LUO Yuan,ZHAO Weinan |
(School of Electrical Engineering, Northeast Dianli University, Jilin 132012, China;Fushun Power Supply Bureau, State Grid Liaoning Electric Power Ltd., Fushun 113000, China;Daqing Power Supply Bureau, State Grid Heilongjiang Electric Power Ltd., Daqing 163000, China) |
Abstract: |
To further explore the influence factors of load time sequence forecasting errors and improve the accuracy of load forecasting, an approximate entropy algorithm is proposed for characterizing the regularity of load time series quantitatively and recognizing the causes for load forecasting errors fully. Approximate entropy algorithm is applied to analyze load time series and determine the regularity of it. Based on the above research, this paper carries out the related research according to the relationship between the regularity and forecasting errors of load time series. The results show that, approximate entropy algorithm can describe the regularity of load time series effectively, and there is a strong correlation between regularity of load time series and load forecasting errors, proving the correctness and validity of the method. This work is supported by National Key Basic Research Program (973 Program) (No. 2013CB228201) and National Natural Science Foundation of China (No. 51307017). |
Key words: load time series load forecasting regularity approximate entropy forecasting errors |