Hierarchical optimize scheduling strategy for orderly charging of electric vehicles based on PTM and CNN-Transformer-LightGBM model

Abstract

With the significant increase in the penetration rate of electric vehicles, uncoordinated charging of EV clusters can exacerbate peak-to-valley differences, resulting in a "peak on peak" phenomenon. This paper proposes a hierarchical optimize scheduling strategy for orderly charging based on the Probability Transition Matrix algorithm (PTM) using CNN-Transformer-LightGBM. Firstly, we establish electric vehicle load prediction models for CNN-Transformer-LightGBM separately, and use the inverse variance method to weight and combine the two models into a CNN-Transformer-LightGBM composite model; To optimize the continuous parameters within the model, TPM is used for hyperparameter optimization to achieve optimal charging control. Simulation results indicate that the proposed strategy effectively reduces the grid load's peak-to-valley difference by 43 % and decreases comprehensive grid load variance by 32 %.