基于模型预测控制的轮毂电驱动汽车制动能量回收

发布时间：2018-06-20 00:54

本文选题：轮毂电驱动汽车 + 制动能量回收　；参考：《吉林大学》2017年硕士论文

【摘要】：伴随着节能减排政策的推广,绿色出行理念深入人心,高效零污染的电动汽车的普及势在必行。虽然电动汽车的充电设施正在逐步完善,但其电池成本、续驶里程的弊端尚不能很好解决。制动能量回收技术属于电动汽车节能环保的关键技术,不但能够回收制动能量提升续驶里程,还能提供一定的制动力矩,减少传统制动系统的磨损和热衰退,提高制动效能及安全性。而合理完善的控制技术是电动汽车安全制动条件下实现能量回收最大化的保障,制动能量回收控制系统的研究具有重要的理论意义和实际的工程应用价值,无论是车企还是相关科研机构都加大了研发力度。然而制动能量回收系统中的控制问题是很复杂的,属于多目标多约束优化问题,工程中实际应用的控制方法并没有达到理想的效果。本文围绕提高能量回收效果、合理分配整车制动力、保证制动安全稳定等核心问题进行研究,针对这些复杂的控制问题,设计了基于模型预测控制的轮毂电驱动汽车制动能量回收控制系统。模型预测控制算法在处理多目标多约束问题方面拥有很大的优势。制动能量回收系统的核心控制问题是分配汽车前后轴制动力,以及协调控制电机和液压制动力,达到安全稳定制动的前提下回收尽可能多能量的目的。为了保证整车制动的制动效能和稳定性,文中引入汽车制动力理想分配曲线和ECE法规限制;车辆制动执行机构存在物理限制,加入了电机和液压制动转矩最大值约束;电机在转速很低时发电能力有限,因此考虑了电机再生制动的转速下限值约束。此外,电池充电SOC上限值约束作为外部阀值约束在模型预测控制算法外部实现。电机和液压制动系统在合理选取的目标函数的作用下协同工作,使得整车能够满足驾驶员制动需求,在制动安全稳定的情况下提高能量回收能力。最后在AMESim环境中搭建的四轮轮毂电驱动汽车高精度整车模型上,验证了所设计控制系统的有效性和优势。本文的主要内容:1.本文首先针对四轮轮毂电驱动汽车的结构特点和动力学方程,在AMESim环境中建立了整车动力学模型。着重对模型中的关键部件进行了参数匹配和动态性能分析,通过仿真标定选定的电机效率map图,应用在模型预测控制系统中,用于实时得到电机当前发电效率进而优化电机的制动转矩;利用真实实验数据证实所选用电池模型的充放电特性符合实际情况;分析了液压制动系统的动态响应效果;最后进行了整车模型功能及动力学合理性验证,模型中考虑了电机制动和液压制动转矩的输出时延的影响,更好的模拟工程实际情况。2.针对制动能量回收系统的特殊性,综合考虑电机发电特性、蓄电池安全充电、制动安全性等因素,引入模型预测算法滚动优化控制的思想,设计基于模型预测控制的制动能量回收控制系统。建立了控制系统的动力学模型,对制动转矩进行集成控制;选定的目标函数包括需求制动转矩的跟踪、能量回收效率及制动转矩波动,分别用于满足驾驶员制动需求、回收能量最大化及良好的制动舒适性;考虑了电机最大制动转矩的时变约束和液压最大制动转矩约束,同时加入了ECE制动法规和电机发电最低转速的限制,并在模型预测控制算法外部加入电池充电最高SOC约束。3.针对控制系统对仿真平台的需求,提出AMESim和Matlab/Simulink联合仿真解决方案。所设计的控制系统在Simulink中实现,结合二者各自的优点建立仿真工况,对所设计模型预测控制系统的制动安全性、稳定性、舒适性、能量回收效果进行仿真测试,验证控制系统有效性,最后通过与制动能量回收模糊控制系统的仿真对比实验,证实模型预测控制的应用能够大幅度提升制动能量回收率。
[Abstract]:With the promotion of energy saving and emission reduction policies, the concept of green travel is deeply rooted in the hearts of the people. The popularization of high efficiency and zero pollution electric vehicles is imperative. Although the charging facilities of electric vehicles are being improved gradually, the cost of battery and the disadvantages of the driving range are not well solved. The key of braking energy recovery is the key to the energy saving and environmental protection of electric vehicles. Technology can not only recover the braking energy and drive mileage, but also provide a certain braking torque, reduce the wear and heat decline of the traditional brake system, improve the braking efficiency and safety. The research of the system has important theoretical significance and practical value of engineering application. Both the car enterprise and the related scientific research institutions have increased the research and development efforts. However, the control problem in the braking energy recovery system is very complex, which belongs to the multi-objective and multi constraint optimization problem. The control method of the actual application in the engineering has not reached the ideal effect. In this paper, the core problems such as improving the energy recovery effect, distributing the vehicle braking force reasonably and ensuring the safety and stability of the brake are studied. In view of these complex control problems, the brake energy recovery control system based on model predictive control is designed. The model predictive control algorithm is used to deal with multi-objective and multi constraint problems. The core control problem of the braking energy recovery system is to allocate the driving force of the front and rear axle of the car and coordinate the control of the motor and hydraulic power to achieve the purpose of recovering as much energy as possible under the premise of safe and stable braking. In order to ensure the braking efficiency and stability of the whole vehicle brake, the automobile brake is introduced in this paper. The force ideal distribution curve and the ECE regulation limit; the vehicle brake actuator has physical restriction, adding the maximum value constraint of the motor and hydraulic braking torque; the generator has limited power generation ability when the speed is very low, so the lower limit limit of the motor regenerative braking is considered. In addition, the limit limit of the battery charge SOC is used as the external threshold constraint. The external realization of the model predictive control algorithm. The motor and hydraulic brake system work together under the function of the reasonable target function, making the whole vehicle meet the driver's braking demand and improve the energy recovery ability under the condition of safe and stable braking. Finally, the high precision whole four wheel hub electric drive car built in the AMESim environment is high precision. On the vehicle model, the effectiveness and advantages of the designed control system are verified. The main contents of this paper are as follows: 1. firstly, the structure characteristics and dynamic equations of the four wheel hub electric drive vehicle are first set up in the AMESim environment, and the key parts in the model are analyzed and the parameters matching and dynamic performance analysis are carried out, through which the key parts of the model are analyzed. The map diagram of the selected motor efficiency is simulated and calibrated. It is applied to the model predictive control system to obtain the current power efficiency of the motor and optimize the braking torque of the motor. The real experimental data is used to verify the charge discharge characteristics of the selected battery model, and the dynamic response of the hydraulic brake system is analyzed. Finally, the dynamic response of the hydraulic brake system is analyzed. The function and dynamics of the vehicle model are verified, and the effect of the output delay of the motor braking and the hydraulic braking torque is considered in the model, and a better simulation of the engineering actual situation.2. is given to the particularity of the braking energy recovery system, and the factors such as the electric generator characteristics, the battery safety charging, the braking safety and so on are introduced, and the model is introduced into the model. The idea of rolling optimization control is predicted and the braking energy recovery control system based on model predictive control is designed. The dynamic model of the control system is established, and the braking torque is integrated. The selected target functions include the tracking of the braking torque, the efficiency of energy recovery and the fluctuation of braking torque, which are used to satisfy driving respectively. At the same time, the maximum braking torque of the motor and the maximum braking torque are taken into consideration. At the same time, the ECE braking regulation and the minimum motor power generation speed limit are added, and the maximum SOC constraint.3. for battery charging is added to the model predictive control algorithm for control. In order to meet the requirements of the simulation platform, the AMESim and Matlab/Simulink joint simulation solutions are proposed. The designed control system is implemented in Simulink, and the simulation conditions are established by combining the advantages of the two parties. The simulation test is made for the braking safety, stability, comfort and energy recovery effect of the designed model predictive control system, and the verification control is carried out. The effectiveness of the system is made. Finally, through the simulation comparison experiment with the fuzzy control system of braking energy recovery, it is proved that the application of model predictive control can greatly improve the braking energy recovery rate.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U469.72;TP273

【相似文献】