Research on Real-Time Algorithm of Sensor Data of Mobile Energy Storage Equipment in Distribution Network

A new algorithm for real-time data processing is presented, which combines adaptive filtering with deep learning. Firstly, the algorithm can eliminate the noise of sensor data by adaptive filtering to guarantee the reliability of the data. Then, CNN and RNN are applied to analyze and forecast the time sequence data, and the charging and discharging policy is adjusted in real-time. In this paper, the algorithm is compared with the traditional Kalman filter and the general deep learning method. The experiment results indicate that the algorithm is superior in many aspects, especially in the processing time, the precision of forecast and the efficiency of data processing. In particular, the response time is reduced by 30%, and the forecast precision is increased 12%. Data processing efficiency has been increased 20%. All these experiments prove that this algorithm has great potential in improving the ability of real time data processing and optimizing energy management.