基于凸优化的燃料电池汽车节能驾驶研究

Research on Eco-driving of Fuel Cell Vehicles via Convex Optimization

为实现燃料电池汽车在多信号灯场景下的节能驾驶,本文中提出一种基于分层凸优化的快速车速规划和能量管理方法。结合车辆静态氢耗图,运用动态规划获得车辆通过信号灯的最优绿灯窗口,并确定最优行驶路径的搜索区域。建立以车辆需求功率累计最小为优化目标求解车辆加速度的二次规划问题,并运用Matlab/OSQP求解器获取车辆最优行驶路径。根据最优行驶路径,采用基于交替方向乘子法的能量管理策略,实现各动力源输出功率的合理分配。针对9个信号灯场景的仿真结果表明,所提方法的电机工作点平均效率比智能驾驶员模型高10%,氢耗低45%。此外,该方法计算速度快,具备实时优化的潜力。

In order to realize eco-driving of fuel cell vehicles in the multi-signal scenario,a fast speed planning and energy management method based on hierarchical convex optimization is proposed in this paper. Combined with the static hydrogen consumption map of the vehicle,dynamic programming is used to obtain the optimal period of green light for the vehicle while passing through the traffic light,and to determine the search area for the optimal driving route. A quadratic programming problem is established to obtain the acceleration with the minimum accumulated power demand as the optimization objective,and the Matlab/OSQP solver is used to obtain the optimal driving path of the vehicle. According to the optimal driving route,an energy management strategy based on the alternating direction method of multipliers is established to realize the reasonable distribution of the output power of each power source. The simulation results for the scenario with 9 traffic lights show that the average efficiency of the motor operating point of the proposed method is 10% higher,and the hydrogen consumption is 45% lower than that of the intelligent driver model. In addition,the method has fast calculation speed and the potential for real-time optimization.