P2构型PHEV模型预测能量管理策略研究

Research on the model prediction energy management strategy for P2 configuration PHEV

基于规则的插电式混合动力系统能量管理策略难以实现全局最优,全局优化策略则存在未来功率需求难以获取及无法实时求解等问题。预测能量管理策略通过对未来一段时间内车辆功率需求进行预测,进而在预测时段内采用全局优化算法,从而在保证算法实时性的同时取得接近全局优化的控制效果。车速预测算法是预测能量管理策略的核心和关键,采用适应能力强、计算速度快的径向基神经网络对车辆功率需求进行预测,以提高车速预测的准确性。以P2构型插电式混合动力系统为研究对象,将模型预测控制与动态规划结合,以发动机油耗最小为优化目标对车速预测时域内最优发动机转矩序列进行求解。建立系统仿真模型,对基于规则的能量管理策略和预测能量管理策略进行对比。结果表明:与基于规则的策略相比,在8个NEDC工况下,基于径向基神经网络的预测能量管理策略能耗降低13.8%。

The rule-based energy management strategy for plug in hybrid electric vehicle(PHEV)is difficult to achieve global optimization,while the global optimization strategy has problems such as difficulty in obtaining future power requirements and solving them in real time.The predictive energy management strategy predicts the vehicle power demand in a period of time in the future,and then uses the global optimization algorithm in the prediction period,so as to ensure the real-time performance of the algorithm and achieve the control effect close to the global optimization.Velocity prediction algorithm is the core and key of energy management strategy.In this paper,radial basis function neural network with strong adaptability and fast calculation velocity is used to predict vehicle power demand and improve the accuracy of vehicle velocity prediction.Taking P2 plug-in hybrid electric system as the research object,model predictive control and dynamic programming(DP)are combined to solve the optimal engine torque sequence in the time domain of vehicle prediction with the minimum engine fuel consumption as the optimization objective.The system simulation model is established to compare the rule-based energy management strategy with the predictive energy management strategy.The results show that compared with the rule-based strategy,the energy consumption of the predictive energy management strategy based on radial basis function neural network is reduced by 13.8%under 8 NEDC conditions.