考虑模型误差估计的AMT起步控制研究

发布时间：2018-03-04 10:17

本文选题：电控机械式自动变速器(AMT)　切入点：起步控制　出处：《吉林大学》2016年硕士论文　论文类型：学位论文

【摘要】：电控机械式自动变速器(Automated Manual Transmission,简称AMT)是在手动机械变速器的基础上安装自动执行机构和电控单元,实现离合器和换挡操作自动化的一类自动变速器。由于AMT保留了手动变速器基本的变速器结构,与其他类型的自动变速器相比,其具有结构简单、易加工、成本低、传动效率高等优势,符合我国汽车行业发展需求,被广泛应用到商用车和微型轿车。起步控制是AMT控制问题的重点与难点。由于干式离合器的存在,AMT的起步控制很难同时满足平稳起步减小冲击度和快速起步减小离合器滑摩功的要求,这也制约着AMT的应用推广。另外,车辆本身的参数和行驶工况,例如离合器摩擦片的摩擦系数、车重、路面状况、道路坡度等都是随时变化的,这也给AMT起步控制带来很大困难。为了解决这些AMT起步控制难题,本文提出了带有模型误差估计的优化控制策略,该控制策略利用基于线性二次型优化控制(Linear Quadratic Regulator,简称LQR)的算法对起步过程中的发动机转速、离合器输入输出转速差、传动系的冲击进行优化,保证车辆快速平稳起步,同时,该算法设计了模型误差估计器对车辆参数和工况变化带来的控制模型误差进行估计,该估计值反馈给控制模型使车辆在参数和工况改变的情况下依然可以顺利完成起步。针对该起步控制策略的研究,本文做了以下方面的工作:1、对AMT的基本知识、发展情况进行了学习,着重研究了应用在AMT起步的控制算法。分析了干式离合器的结构、工作原理和AMT起步控制的评价指标,在此基础上对常用的AMT起步控制策略进行了分析。2、通过对AMT工作原理、起步过程和控制算法的分析,确定了本文研究的主要内容:考虑模型误差估计的AMT起步优化控制策略,并进行了控制算法的理论设计。3、利用专业软件AMESim和AMTLAB/Simulink分别搭建了整车仿真模型和AMT起步控制算法模型,通过AMESim-Simulink联合仿真的方式对算法进行了验证,同时也对不带有模型误差估计的起步优化控制算法进行仿真,与本算法的仿真结果进行比较。4、利用d SPACE工具搭建了实车试验环境,对算法进行了实车测试,测试结果表明该算法不仅可以满足不同起步意图的起步过程,而且可以在车辆参数和工况发生变化时顺利完成起步,并且起步过程满足快速性平顺性等评价要求,证明了算法的实用价值。
[Abstract]:Automatic mechanical transmission (Automated Manual Transmission, referred to as AMT) is the installation of automatic actuator and electronic control unit based on manual mechanical transmission, realize a kind of automatic transmission clutch and shift operation automation. Because the AMT retains the basic manual transmission gearbox structure, compared with other types of automatic transmission and it has the advantages of simple structure, easy processing, low cost, high transmission efficiency, in line with the development needs of the automotive industry in China, has been widely applied to commercial vehicles and mini cars. Starting control is AMT control key and difficult problem. Because the dry clutch, the starting control of AMT is difficult to meet the smooth start to reduce the friction work of clutch requirements and reduce the impact of fast, it also restricts the popularization and application of AMT. In addition, the parameters of the vehicle itself and driving conditions, such as friction clutch The coefficient of friction, friction plate weight, road conditions, road slope are subject to change, it also gives AMT starting control very difficult. In order to solve these problems AMT starting control, optimal control strategy is proposed in this paper with model error estimation, linear quadratic optimal control of two based on the control strategy and (Linear Quadratic Regulator, referred to as LQR) algorithm to the process of starting the engine speed, the clutch input and output speed difference, the impact of transmission system optimization, ensure the vehicle fast smooth start, at the same time, the algorithm design of the control model of model error estimator of vehicle parameters and conditions are estimated, the estimated value of feedback to control the vehicle model in changing the parameters and conditions of the case can still be successfully started. According to the research of the control strategy of the start, this paper made the following work 1, basic knowledge of AMT, the development of the study focuses on the control algorithm used in AMT started. Analysis of the structure of dry clutch, AMT working principle and starting control of the evaluation index, based on the commonly used AMT starting control strategy of.2, the working principle of the AMT the starting process, analysis and control algorithm, and determine the main contents of this paper: the AMT starting control strategy optimization model error estimation is considered, and the theoretical design of.3 control algorithm, using professional software AMESim and AMTLAB/ Simulink respectively to build a simulation model of vehicle starting control algorithm and AMT model by AMESim-Simulink joint simulation mode the correctness of the algorithm, but also to start with optimal control algorithm simulation model error estimation, and simulation results of this algorithm were compared by using D SPA.4. CE tools to build a real vehicle test environment, the algorithm for vehicle test, the test results show that this algorithm can not only meet the starting process of different starting intention, but also in the vehicle parameters and the working condition is changed smoothly starting and starting process to meet the fast ride comfort evaluation requirements, proves the practicality of the proposed algorithm.

【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U463.212

【参考文献】