Prediction of Arrival Time of Pure Electric Bus Based on FA-BP Algorithm

To establish a suitable pure electric bus arrival time prediction model, this paper takes as the research object. Based on analysis of influencing factors bus, BP neural network model optimized by firefly algorithm (FA-BP model) is established selecting vehicle type, SOC value, battery age, and input conditions. The trained tested using operation data. root mean square error Kalman filter 0.351, 0.059, FA-BP 0.04. results show that in effectively improves accuracy has good reliability feasibility. It can provide some theoretical references for managers basis improving reliability.