基于深度强化学习的燃料电池混合动力汽车能量管理策略研究

为提高燃料电池混合动力汽车的燃油经济性和燃料电池寿命,该文提出一种基于深度强化学 习(Deep Reinforcement Learning,DRL)的能量管理策略。该策略首先在 DRL 奖励信号中加入寿命因子,通过降低燃料电池功率波动,起到延长燃料电池寿命的效果;其次,通过限制 DRL 的动作空间的方法,使燃料电池系统工作在高效率区间,从而提高整车效率。在 UDDS、WLTC、Japan1015 三个标准工况下进行了离线训练,并在 NEDC 工况下实时应用以验证所提出策略的工况适应性。仿真结果显示,在离线训练中,所提出的策略可以快速收敛,表明其具有较好的稳定性。在燃油经济性方面,与基于动态规划的策略相比,在 3 个训练工况下的差异仅为 5.58%、3.03% 和 4.65%,接近最优燃油经济性;相比基于强化学习的策略,分别提升了 4.46%、7.26% 和 5.35%。与无寿命因子的 DRL 策略相比, 所提出的策略在 3 个训练工况下将燃料电池平均功率波动降低了 10.27%、47.95% 和 10.85%,这有利于提升燃料电池寿命。在未知工况的实时应用中,所提出策略的燃油经济性比基于强化学习的策略提升了 3.39%,这表明其工况适应性。     

燃料电池有轨电车能量管理Pareto多目标优化

节能环保的出行方式得到政府的大力推广, 其中燃料电池混合动力有轨电车由于可无网运行且节能环保而备受关注.为了改善燃料电池/超级电容/动力电池大功率有轨电车的燃料经济性与系统耐久性, 提出一种有轨电车能量管理策略(Energy management strategy, EMS)的多目标优化方法. 首先以氢燃料消耗量和能量源性能衰减率作为评价指标, 建立多目标成本函数. 由于两个指标很难在同一个等式中评价, 设计了基于状态机与非支配排序的能量管理Pareto多目标优化方法, 获得了有轨电车能量管理策略Pareto非劣解集, 并分析了能量管理策略的目标功率参数对性能指标的影响规律, 进而遴选出兼顾燃料经济性与系统耐久性的综合最优解. 结果表明, 与功率跟随策略和基于遗传算法优化策略相比, 该能量管理优化方法的燃料经济性分别提高了29.4 %和2.4 %.     

PEM燃料电池加湿器温度的专家PID控制

质子交换膜(PEM)燃料电池是一种清洁高效的新型能源。PEM燃料电池的湿度对燃料电池的工作性能和使用寿命有着重要的影响。当输出负载变化时,如何采取适当的控制策略控制加湿器出口气体的湿度是燃料电池技术中的一个难题。加湿器出口气体的湿度主要受加湿器温度的影响。然而加湿器的温度系统是一个大时滞非线性系统,采用常规的PID控制无法实现有效的控制。本文基于专家PID控制策略来控制加湿器的温度。仿真和实验结果表明,该算法可以得到良好的控制效果。     

计算机控制汽车电池实验系统

动力电池管理是实现混合动力电动汽车的节能与环保目标的关键技术之一,重点研究动力电池电容量实时计量方法和电池均衡充放电控制方法.提出基于电功率的电池组充放电均衡控制策略,以HEV动力电池组为对象,研制了基于电功率的动力电池均衡控制实验系统.     

动力驱动系统匹配与控制策略研究

能源、有害气体和温室气体排放是影响今后汽车技术发展的三大问题.可外接充电式混合动力电动汽车是指可以使用电力网对动力电池进行充电的混合动力电动汽车,是向最终的清洁能源汽车过渡的最佳方案之一.为实现新一代燃料电池城市客车原理样车的动力性能指标,样车动力驱动系统采用串联式能量混合型结构,基于可外接充电技术,对动力驱动系统的主要部件性能参数进行了合理的匹配理论研究.基于ADVISOR软件建立动力驱动系统仿真模型并提出了不同工作模式和控制策略.分别从纯电动工况、中国典型城市客车车况下对纯电动里程、SOC、动力电池和燃料电池的输出功率进行了仿真研究、动力性能满足性能指标要求,仿真结果验证了匹配理论、控制策略和模型的准确性.     

基于随机模型预测控制的插电式混合动力汽车 多目标能量管理策略

为了改善插电式混合动力汽车的燃油消耗和排放, 开展多目标随机模型预测控制策略的研究. 首先, 建立适用于模型预测的多元线性回归的发动机和电池模型, 建立融合燃油消耗和排放的多目标价值函数的模型预测控制, 随后, 基于随机驾驶员模型未来时刻的车速, 结合交通信息并利用动态规划(DP)算法进行参考电荷状态(SOC)优化, 进而建立多目标随机模型预测控制策略. 最后, 通过与DP, MPC等策略进行对比验证, 及给出两组不同权值进行多目标控制效果分析. 结果表明, 该策略的燃油消耗和排放最接近DP的控制效果, 且设置不同权重值可获得相应的控制目标, 说明该策略对提升燃油消耗和排放的多目标性能的有效性.     

含多种分布式电源的微网建模及并网控制仿真

针对微电网中分布式发电的控制策略问题,利用Matlab/Simulink分别搭建了光伏电池、风力发电机、燃料电池及微型燃气轮机等分布式电源的模型,并在此基础上建立了包含上述分布式电源的低压微网系统仿真系统。在并网控制策略方面,光伏电池与风力发电机采用P/Q控制方式,燃料电池与微型燃气轮机采用下垂控制方式。通过仿真研究了该系统模型及控制策略在并网与孤岛模式下的动态特性。结果验证了该系统的可行性和有效性,为建立微网实验平台和示范工程奠定了基础。     

NI尖端技术引领Ford创新潮流

<正> 挑战美国国家仪器公司(NI)为汽车燃料电池系统开发的电子控制设备(ECU),极大地改善了燃料电池系统。使其较之于传统的基于内燃机的传动系统更具竞争力,同时在商业上切实可行,这是NI目前面临的挑战。福特公司(Ford))专注于对燃料电池系统(FCS)的研究,并由此催生了一系列汽车的发展,如全球第一辆全性能燃料电池汽车(P2000)和全球第一辆燃料电池充电式、油电混合动力车(带有HySeries驱动系统的Ford Edge     

并联新能源汽车最优效率实时控制

为了提高并联混合动力汽车驱动系统的实时效率,降低燃油消耗,本文提出一种基于效率最优的协调控制策略.根据不同驱动模式下电池的充放电状态,建立了充放电状态下驱动系统的等效燃油消耗模型,在分析电池效率和发动机效率的基础上,得到驱动系统效率的统一表达式,进而通过建立不同功率需求不同荷电状态下系统最优效率的功率分配系数图谱,设计了系统效率最优的协调控制策略,协调控制策略根据优化的功率分配系数在发动机和电机间进行力矩分配,协调控制策略可以离线计算并实时执行.两种工况循环下的仿真结果表明效率最优控制策略能有效地提高混合动力系统实时效率和燃油经济性.     

双率CARMA模型的广义最小方差自校正控制

针对一类双率采样的CARMA模型,研究了相关的自校正控制问题。基于双率采样以及含有噪声的数据,本文提出一个辅助模型来估计无法采样到的损失输出数据,并进一步采用随机梯度算法来估计模型参数。通过最小化最优预测输出的方差并结合Diophantine方程给出了基于辅助模型的广义最小方差自校正控制(AM-GMVSTC)策略。最后通过一个仿真例子说明提出算法的有效性。     

多能源电动汽车中的Fuzzy控制策略及仿真

多能源电动汽车的能量存储系统由锌空电池、镍氢电池和超大电容三种能量存储元件组成。锌空电池为负载提供基本能量。镍氢电池工作在中级能量区，并回收下坡和刹车过程中的能量。超大电容工作在尖峰负载区，为大加速度过程提供能量，在短时间内可以实现能量回收。该文在多能源电动汽车的模型基础上，针对能量管理系统(EMS)提出了一种模糊控制策略。EMS模糊控制策略的输入包括所需功率、镍氢电池的SOC和超大电容的SOC，模糊控制策略的输出包括三个能量存储元件的分配功率因子，每个输入和输出有不同的模糊量。仿真结果表明：模糊控制策略比简单查表控制策略在续驶里程、燃料经济性和效率等方面均有所改善。     

Optimization management of hybrid energy source of fuel cell truck based on model predictive control using traffic light information

Energy optimization management can make fuel cell truck (FCT) power system more efficient, so as to improve vehicle fuel economy. When the structure of power source system and the torque distribution strategy are determined, the essence is to find the reasonable distribution of electric power between the fuel cell and other energy sources. The paper simulates the assistance of the intelligent transport system (ITS) and carries out the eco-velocity planning using the traffic signal light. On this basis, in order to further improve the energy efficiency of FCT, a model predictive control (MPC)-based energy source optimization management strategy is innovatively developed, which uses Dijkstra algorithm to achieve the minimization of equivalent hydrogen consumption. Under the scenarios of signalized intersections, based on the planned eco-velocity, the off-line simulation results show that the proposed MPC-based energy source management strategy (ESMS) can reduce hydrogen consumption of fuel cell up to 7\% compared with the existing rule-based ESMS. Finally, the Hardware-in-the-Loop (HiL) simulation test is carried out to verify the effectiveness and real-time performance of the proposed MPC-based energy source optimization management strategy for the FCT based on eco-velocity planning with the assistance of traffic light information.     

Fuel efficient power management strategy for fuel cell hybrid powertrains

A real time control strategy for fuel cell hybrid vehicles is proposed. The objective is to reduce the hydrogen consumption by using an efficient power sharing strategy between the fuel cell system (FCS) and the energy buffer (EB). The energy buffer (battery or supercapacitor) is charge-sustained (no plug-in capabilities). The real time control strategy is derived from a non-causal optimization algorithm based on optimal control theory. The strategy is validated experimentally with a hardware-in-the-loop (HiL) test bench based on a 600 W fuel cell system.     

Combined passivity based control and optimal control for energy management of fuel cell/battery hybrid system

The energy management of hybrid electric vehicles is becoming an interesting topic for many researchers. Furthermore, the wise choice of the energy management strategy allows not only the best distribution of the power between the used sources, but also it allows reduction of consumption, increase in the lifetime of the sources, and improves the autonomy of the hybrid electric vehicle. The autonomy is guaranteed by the optimization of the embedded sources. In this study, the hybrid system consists of combining the fuel cell as the main source with the battery as the auxiliary source. The novelty of the proposed energy management strategy for the studied hybrid system is the combination between interconnection and damping assignment‐passivity based control and the Hamiltonian Jacobi Bellman method. The stability proof is given and the efficiency of the proposed strategy is proved by the experimental work, where the obtained results show the good and adequate results to the proposed scenario.     

Hardware-in-the-loop simulation for the design and verification of the control system of a series–parallel hybrid electric city-bus

Hybrid electric buses have been a promising technology to dramatically lower fuel consumption and carbon dioxide (CO₂) emission, while energy management strategy (EMS) is a critical technology to the improvements in fuel economy for hybrid electric vehicles (HEVs). In this paper, a suboptimal EMS is developed for the real-time control of a series–parallel hybrid electric bus. It is then investigated and verified in a hardware-in-the-loop (HIL) simulation system constructed on PT-LABCAR, a commercial real-time simulator. First, an optimal EMS is obtained via iterative dynamic programming (IDP) by defining a cost function over a specific drive cycle to minimize fuel consumption, as well as to achieve zero battery state-of-charge (SOC) change and to avoid frequent clutch operation. The IDP method can lower the computational burden and improve the accuracy. Second, the suboptimal EMS for real-time control is developed by constructing an Elman neural network (NN) based on the aforementioned optimal EMS, so the real-time suboptimal EMS can be used in the vehicle control unit (VCU) of the hybrid bus. The real VCU is investigated and verified utilizing a HIL simulator in a virtual forward-facing HEV environment consisting of vehicle, driver and driving environment. The simulation results demonstrate that the proposed real-time suboptimal EMS by the neural network can coordinate the overall hybrid powertrain of the hybrid bus to optimize fuel economy over different drive cycles, and the given drive cycles can be tracked while sustaining the battery SOC level.     

Speed planning and energymanagement strategy of hybrid electric vehicles in a car-following scenario

The development of intelligent connected technology has brought opportunities and challenges to the design of energy management strategies for hybrid electric vehicles. First, to achieve car-following in a connected environment while reducing vehicle fuel consumption, a power split hybrid electric vehicle was used as the research object, and a mathematical model including engine, motor, generator, battery and vehicle longitudinal dynamics is established. Second, with the goal of vehicle energy saving, a layered optimization framework for hybrid electric vehicles in a networked environment is proposed. The speed planning problem is established in the upper-level controller, and the optimized speed of the vehicle is obtained and input to the lower-level controller. Furthermore, after the lower-level controller reaches the optimized speed, it distributes the torque among the energy sources of the hybrid electric vehicle based on the equivalent consumption minimum strategy. The simulation results show that the proposed layered control framework can achieve good car-following performance and obtain good fuel economy.     

多能源电动汽车中的功率比较控制策略及仿真

在多能源电动汽车的模型基础上，针对能量管理系统(EMS)该文提出了一种功率比较控制策略。EMS的功率比较控制策略是通过比较实际所需功率和能量存储元件的最大提供和吸收功率来达到分配功率的目的。同时提出了最大提供功率和吸收功率的确定方法。仿真结果表明：功率比较控制策略比简单查表控制策略在续驶里程、燃料经济性和效率等方面均有所改善。     

Real‐time optimal energy management for a fuel cell/battery hybrid system

An energy management strategy is proposed for a class of fuel cell/battery hybrid systems. In such hybrid systems, a fuel cell system is the main power source, and a lithium‐ion battery is the auxiliary power source. In order to manage the system power at the next moment in a reasonable way, a load current filter with bounded estimation errors is designed to estimate the load current. Then, a real‐time optimal energy management algorithm is proposed to optimize economy consumption of the hybrid system. By taking current change rate of the fuel cell and the state of charge into consideration and taking reasonable model simplifications, the optimization problem can be described as a quadratic programming problem. Then a general purpose solver is proposed to solve the quadratic programming problem based on the alternating direction method of multipliers. The efficiency of the proposed solver is much faster than computing interior point method or active set method. Simulation results in MATLAB/SIMULINK are carried out to validate the significant effectiveness and efficiency of the proposed management strategy.     

混合动力电动汽车中利用决策树CART算法的能源管理方案

针对混合动力电动汽车(HEV)氮氧化物( )排放的问题,提出了一种基于决策树CART算法的柴油混合动力能源管理策略。首先,提出了一种结合决策树与回归树的分类算法(Classification and Regression Tress,CART),针对类别和变量特征,从一个或多个预测变量中预测出个例的趋势变化关系;然后,通过控制发动机和电动机之间的扭矩分配,引入了额外的自由度以调整从纯燃料经济性情况到纯 限制情况的优化权衡;最后,采用基于软件在环路和硬件在环仿真的方法,从而根据动力系统配置了解系统性能,并调整所提出的能源管理策略。实验结果表明,提出的柴油混合动力能源管理策略中, 的减少对燃料消耗的影响,且可以通过选择最佳工作点和限制发动机动力来限制 排放的潜力。相比其他几种较新的同类方案,提出的方案在同等燃料消耗的情况下 排放量更小,在燃料消耗略有下降的情况下,可以显着降低 。