摘要
随着智能电表的大量应用,电力数据量大幅提升,为智能电表大数据分析带来了巨大挑战。为此采用模糊数学理论和人工神经网络方法,建立了自适应神经模糊推理模型。基于由智能电表测得的实际电力大数据,进行了模型的训练和测试,最后利用模型预测了用户在24 h内的电力功耗,并与实际功耗数据进行了对比。研究结果表明所提出的模型总体预测精度为84.02%,其中夜间功耗的预测精度均在90%以上,白天时段由于用电随机性较大预测精度在70%左右。研究结果可为智能电表大数据分析与应用提供参考。
With the extensive application of smart meters,the amount of power data is greatly increased,which brings great challenges to the big data analysis of smart meters.Fuzzy mathematics theory and artificial neural network method are used to establish an adaptive neuro-fuzzy reasoning model.Based on the actual power big data measured by smart meters,the model is trained and tested.Finally,the power consumption of users in 24 hours is predicted by the model and compared with the actual power consumption data.The results show that the overall prediction accuracy of the proposed model is 84.02%,the forecasting accuracy of night consumption is above 90%,and the forecasting accuracy of daytime consumptionis about 70%due to the randomness of electricity consumption.The research results can provide reference for the application of big data analysis on smart meters.
作者
王登峰
李英
杨琦
马广霞
樊博
WANG Dengfeng;LI Ying;YANG Qi;MA Guangxia;FAN Bo(Power Science Research Institute of State Grid Ningxia Electric Power Co.,Ltd.,Yinchuan Ningxia 750011,China)
出处
《电子器件》
CAS
北大核心
2022年第3期727-731,共5页
Chinese Journal of Electron Devices
