摘要
为解决增程式电动汽车实时能量分配最优的问题,提出了一种基于车速预测的能量管理策略。通过建立增程式电动汽车模型,利用实车试验数据开发城市道路环境下的典型行驶工况,建立了基于小波神经网络的车速预测模型,确定了输入历史车速信息长度、未来时域等关键参数。以模型预测控制理论为基础,结合动态规划算法,建立基于车速预测的增程式电动汽车能量管理策略。仿真结果表明该能量管理策略的燃油消耗相比动态规划(DP)策略增加了6.08%,电能消耗增加了0.26%,但是相比定点能量管理策略燃油消耗降低了40.19%,经济成本减少了28.86%,能够提高车辆的燃油经济性。
To optimize the real time energy distribution of extended range electric vehicles,the paper proposes an energy management strategy based on vehicle speed prediction.By establishing a model of an extended range electric vehicle and synthesizing typical driving conditions of electric vehicles on urban roads with the data collected from real world tests,a vehicle speed prediction model based on wavelet neural networks is constructed and the key parameters such as the length of input historical vehicle speed information and future time domain are determined.On the basis of the model predictive control theory and by the dynamic programming algorithm,an energy management strategy for extended range electric vehicles by predicting the vehicle speed is built.The simulation results show that compared with Dynamic programming(DP)strategy,this energy management strategy increases fuel consumption by 6.08%and power consumption by 0.26%,and that compared with the fixed point energy management strategy,fuel consumption is reduced by 40.19%and economic cos by 28.86%,indicating that this strategy can improve fuel economy of extended range electric vehicles.
作者
李瑞明
余强
吴煜锴
LI Ruiming;YU Qiang;WU Yukai(School of New Energy Vehicle,Xi’an Vocational University of Automobile,Xi’an 710038,China;School of Automobile Engineering,Chang’an University,Xi’an 710064,China)
出处
《西安工业大学学报》
CAS
2023年第5期447-459,共13页
Journal of Xi’an Technological University
基金
交通部重点实验室开放基金项目(300102229507)。
关键词
能量管理策略
增程式电动汽车
车速预测
小波神经网络
energy management
extended range electric vehicles
speed prediction
wavelet neural network