基于DA多重插补法和电力物联网的电能数据缺失修复方法

Power data missing repair method based on DA multiple interpolation method and power Internet of Things

针对电力物联网中电能数据量过多,缺失电能数据修复难度较大的问题,研究基于DA多重插补法和电力物联网的电能数据缺失修复方法。电力物联网利用感知层的电能数据采集终端采集电能数据,所采集电能数据利用通信层传送至应用层,应用层的电能数据缺失修复模块,利用EM插补算法计算电能数据缺失值的初始插补值;将所获取的电能数据插补值作为DA多重插补法的初始值,DA多重插补法利用局部加权回归模型,通过调整电能数据缺失值的预测误差,获取最终电能数据缺失修复结果。实验结果表明,该方法修复电力物联网电能数据的观测误差方差低于0.2,对于短期电能数据与长期电能数据,均具有良好的修复结果。

In view of the problem that there is too much power data in the power Internet of Things and it is difficult to repair the missing power data,the repair method of power data missing based on DA multiple interpolation method and the power Internet of Things is studied.The power Internet of Things uses the power data acquisition terminal of the sensing layer to collect power data,and the collected power data is transmitted to the application layer through the communication layer.The power data missing repair module of the application layer uses the EM interpolation algorithm to calculate the initial interpolation value of the power data missing value;The obtained power data interpolation value is taken as the initial value of the DA multiple interpolation method.The DA multiple interpolation method uses the local weighted regression model to obtain the final power data missing repair result by adjusting the prediction error of the power data missing value.The experimental results show that the observation error variance of this method is less than 0.2 for repairing the power data of the Internet of Things,and it has good repair results for both short-term and long-term power data.