【作者】 姜伟；

【导师】 李寿涛；

【作者基本信息】 吉林大学 ， 控制工程（专业学位）， 2016， 硕士

【摘要】 汽车产业是我国发展的重要支柱,根据公安部交管局发布的数据,截止2014年底,我国机动车保有量达2.64亿辆,其中汽车1.54亿辆,而且还在增长。汽车业发展给我国环境带来了很大压力,废气、噪声对我们的生活及环境造成了严重危害。同时,汽车消耗了全球石油的百分之五十以上,石油资源也日趋紧张。在这样严峻的背景下,电动汽车越来越受到世界各国的高度重视[1]。和传统汽车相比,电动汽车最主要的优点是可以通过电机的再生制动技术在车辆制动或者减速时回收能量,实现节能低污染的性能。电机作为再生制动技术的关键因素,一直是人们研究的热点。永磁同步电机以其结构简单、功率密度高、体积小、维修保养方便等特点引起了人们的青睐,并被广泛的应用在汽车领域,并且我国是稀土资源的大国,应用永磁同步电机更具优势。因此,本文对车用永磁同步电机的再生制动技术进行深入地探讨分析。本文针对基于永磁同步电机的再生制动控制策略进行分析研究,在确保安全的情况下,使其尽可能多的回收制动能量。主要内容如下:本文首先综述了电动汽车以及车用永磁同步电机的发展现状,以及再生制动能量回收控制的主要研究策略。其次在对永磁同步电机基本结构及其工作原理进行研究的基础上,建立了表贴式永磁同步电机的数学模型,运用定向磁场控制的方法,对电机的转矩进行跟踪控制。通过仿真分析证明其有较好的跟踪特性,为研究永磁同步电机的再生制动过程打下基础。根据表贴式永磁同步电机在磁场定向控制下的稳态模型,对永磁同步电机的再生制动过程进行分析,推导出电机输入功率模型,并联合电池的等效电路模型,分析再生制动过程中电机与电池之间的功率损失,以此建立再生制动的功率损耗模型。采用再生制动能量瞬时优化策略,以再生制动过程中的功率损耗最小为指标,对电机转矩进行优化控制,在保证电机和电池安全的前提下,间接的达到能量回收最大化。对ADVISOR软件中的制动模块进行修改,用本文的瞬时优化策略建立S函数并嵌入制动模块中,在NEDC工况下对其进行仿真实验分析。仿真结果表明,采用瞬时优化策略的结果要优于ADVISOR中的策略。采用基于动态规划算法的全局的优化策略对再生制动过程中的功率损失进行分析,并在ECE工况下,分别采用全局和瞬时优化策略进行对比仿真实验分析。结果表明,当制动时间较短时,瞬时优化策略略低于基于动态规划的全局优化策略;当制动时间较长时,基于动态规划的全局优化策略要明显优于瞬时优化策略。针对动态规划算法计算量大,无法用于实时控制的问题,本文提出了基于模型预测框架的动态规划算法。首先利用当前的制动强度采用线性预测的方法预测在车辆制动或减速时未来的车速,然后以制动过程中功率损失最小为目标,用动态规划的方法对预测区间里的电机转矩进行滚动优化控制,仿真结果表明采用基于模型预测框架的动态规划算法的效果基本上能和采用动态规划算法的效果保持一致,但计算时间却大大减少。对ADVISOR软件中的制动模块进行修改,用基于模型预测框架的动态规划算法建立S函数并嵌入制动模块中,在NEDC工况下对其进行仿真实验分析。仿真结果表明,采用动态规划和模型预测控制结合的策略要优于ADVISOR中的策略,证明本文提出的方法能充分利用电机的制动转矩,并在很大程度上减少了动态规划的计算时间。更多还原

【Abstract】 The automotive industry is playing an important role for Chinese development. According to the Ministry of Public Security Traffic Management Bureau released data, a total of 260 million vehicles in China by 2014, including 154 million cars, but also on the rise. Automotive industry brought a lot of pressure on the environment. Emissions, noise caused serious harm to our life and environment. Meanwhile, the car consumes more than fifty percent of the world’s oil, and oil resources are becoming increasingly strained. In this grim backdrop, the electric car attracted more and more attention around the world. And compared to conventional car, the main advantage of electric vehicles is that energy can be recovered by regenerative braking technology when braking or decelerating. The advantage achieved energy-saving and low pollution performance. The regenerative braking technology as a key element had been a hot research topic in the study of people. Permanent magnet synchronous motor with its simple structure, high power density, small size, easy maintenance etc attracted people’s attention, and have been widely used in automobile industry. Our country is a big country of rare earth resources. The application of permanent magnet synchronous motor has the advantage. In this paper, the regenerative braking technology of permanent magnet synchronous motor had been explored in depth.In this paper, the strategy of regenerative braking of permanent magnet synchronous motor is studied. In the safety of circumstances, make its recycling braking energy as much as possible. The main contents are as follows:This paper explained the development of electric vehicles and permanent magnet synchronous motor and the main research strategy of regenerative braking energy recovery control. Secondly, on the basis of permanent magnet synchronous motor structure and working principle established the mathematical model of surface mounted permanent magnet synchronous motor. And using the method of directional magnetic field control tracking controlled the torque and rotational speed of motor. And simulation analysis shows that it has better tracking characteristics. It lays a theoretical basis for the research of process of permanent magnet synchronous motor regenerative braking.According to steady-state model under the directional control of surface mounted permanent magnet synchronous motor, the process of the regenerative braking of permanent magnet synchronous motor is analyzed and the input power model of the motor is deduced, combined with the battery equivalent circuit model to analyze the regenerative braking during power loss between the motor and the battery to establish a power consumption model of regenerative braking. With the transient optimization strategy of regenerative braking energy control motor torque during power loss is minimized as the index. In the safety of the battery and motor circumstances, achieve maximize energy recovery indirectly. Modified brake module in ADVISOR software, and establish S function based on instantaneous optimization strategy. The simulation results show that instantaneous optimization strategy is superior to the strategy in ADVISOR.The global optimization strategy based on dynamic programming analyzes the power loss in the process of regenerative braking. In the ECE conditions, Simulation analysis was carried out on the global and instantaneous optimization strategy respectively. The result shows that, when the braking time is shorter, transient optimization strategy is slightly lower than global optimization strategy based on dynamic programming; when braking time is long, the global optimization strategy based on dynamic programming is superior to transient optimization strategy.In view of the large amount of calculation, the dynamic programming algorithm cannot be used for real-time control problem. We propose a model to predict the dynamic programming algorithm based on the framework. Firstly, use braking strength to predict the speed when braking or decelerating for a period of time. Secondly, in braking process of minimum power loss as the goal, use dynamic programming to take rolling optimization control of the motor torque. The simulation results show that the effect of the dynamic programming algorithm based on model prediction framework can be basically consistent with the effect of dynamic programming algorithm, but the calculation time is greatly reduced. Modified brake module in ADVISOR software, and establish S function based on model prediction framework of the dynamic programming algorithm. The simulation results show that dynamic programming is combined with model predictive control strategy is superior to the strategy in ADVISOR. The policy proves the proposed method can make full use of the braking torque of the motor, and largely reduces the computation time of dynamic programming.更多还原