Fast Load Prediction of Distribution Network Equipments with Electric Vehicles Connected

Abstract:
The analysis of the load characteristics of electric vehicle charging stations has strong randomness and volatility, which will have a greater impact on the power grid and microgrid, mainly reflected in the best dispatch and economic, safe, and stable operation. Therefore, it is of great significance to establish an accurate, stable and reasonable load forecasting method for electric vehicle charging stations for power grid and microgrid. This paper focuses on the research of distribution network equipment load forecasting. Aiming at the characteristics of nonlinear, strong randomness and complex historical samples of electric vehicle charging station load, this paper proposes an accurate prediction method based on AP data mining technology and NSGA-II optimized convolutional neural network, which can predict the daily load of distribution network equipment in advance according to historical load data and its influencing factors. Experimental analysis shows that the proposed prediction method has high accuracy and good stability. At the same time, it can be found that the overall peak load of the charging station on weekdays is lower than that on weekends. The average peak load on weekdays is 4.137 MW, and the average peak load on weekends is 6.679 MW, which further verifies that the index factors will have a greater impact on the equipment load of the distribution network.