Research on electric vehicle charging safety warning model based on back propagation neural network optimized by improved gray wolf algorithm

New energy vehicles have become a global transportation development trend in order to achieve considerable fuel consumption and carbon emission reductions. However, as the number of new energy cars grows, new energy vehicle safety concerns are becoming more evident, posing a major threat to drivers' lives and property and limiting the industry's growth. This paper develops a charging safety early warning model for electric ve-hicles (EV) based on the Improved Grey Wolf Optimization (IGWO) algorithm in order to improve the timeliness and accuracy of charging safety early warning. The greatest voltage of a single battery was chosen as the study goal based on the polarization characteristics of lithium-ion batteries and the equalization features of a vehicle lithium-ion battery pack. The IGWO-BP algorithm is then used to fit the entire EV charging process and anticipate the vehicle's charging condition. At the same time, set the warning threshold and the warning error code. In real time, comparing the EV charging data with the fitted data, computing the residual, and building the EV charging safety warning model based on the residual change. Finally, case analysis is performed using daily charging data from both rapid and slow charging. The findings reveal that the proposed early warning model based on the IGWO-BP algorithm can reliably recognize the abnormal state of EV charging voltage and issue timely warnings.

Automated summary

This paper develops a charging safety early warning model for electric vehicles based on the Improved Grey Wolf Optimization algorithm to improve the timeliness and accuracy of charging safety early warning. The findings reveal that the proposed model can reliably recognize abnormal charging voltage and issue timely warnings.