关键词: 道路坡度识别, ＢＰ神经网络, 控制策略, 模式切换

Abstract: Mode switching is one of the core technologies of hybrid vehicles, which determines vehicle ride comfort and passenger comfort. Aiming at the problem that the current mode switching of hybrid electric bus (HEB) does not take into account the actual road gradient,this paper uses longitudinal dynamics algorithm to identify the road slope based on the actual road driving condition of the hybrid electric bus on a tourist route in Luoyang Cityand then uses the BP neural network algorithm to predict the speed in the future. According to the known road gradient and the future speed of the vehicle, the demand torque is obtained using the vehicle dynamics formula. The mode status is determined according to the required torque and SOC value, so that the mode switching can be completed in advance when the engine is involved in the mode, and the problem of insufficient power can be avoided when the vehicle is going uphill. The experimental simulation results show that the control strategy can make the vehicle complete the mode witching ahead of time when it is about to go uphill, improve the driving performance, and avoid the problem of insufficient uphill power.