Data processing technology of distribution station area combined with convolutional neural network

As the end of the distribution network closest to the user, the station area is very important to ensure the stability and safety of the station area. But now the station area has more problems in data collection and storage than the Central Asian side. This makes it more difficult to optimize the regulation of the Taiwan area. Aiming at the problems of abnormal sampling period, abnormal sampling reference time, data noise, and missing data in the data of the station area, this paper proposes a data processing method combining convolutional neural network and mathematical methods to solve these problems in the station area, and constructs a station The accurate time series of regional data was obtained, and the feasibility of the method was verified by simulation.