集成电路与嵌入式系统2024，Vol.24Issue(9)：81-86,6.DOI:10.20193/j.ices2097-4191.2023.0005
智能电网大数据分析在电力需求预测中的应用
Application of smart grid big data analytics in electricity demand forecasting
摘要
Abstract
An adaptive neuro-fuzzy inference system(ANFIS)-based method for correcting outliers in power demand curves is proposed for outliers in smart grid big data.The method performs big data analysis for smart meters in smart grids to detect and correct zero-value anomalies in timing data.The effectiveness of the proposed method is verified using real power demand data,and the results show that the ANFIS method is able to correct the outliers with higher accuracy compared with the linear interpolation and artificial neural network based methods,with a maximum relative error of only 3.76％.In addition,the standard deviation of the relative error is also smaller at 2.26％.The experiment results show that the ANFIS method fully combines the advantages of fuzzy logic system and neural network,effectively dealing with the outliers in the peak hour power load demand curve well.This provides a valuable reference for further impro-ving the effect of smart grid big data analysis.关键词
大数据分析/智能电表/智能电网/ANN/ANFISKey words
big data analytics/smart meters/smart grid/ANN/ANFIS分类
信息技术与安全科学引用本文复制引用
朱平飞,王宇坤,于喻,白琳..智能电网大数据分析在电力需求预测中的应用[J].集成电路与嵌入式系统,2024,24(9):81-86,6.基金项目
国网北京市电力公司科技项目(B20212220016). （B20212220016）
