基于微分代数混杂切换系统理论的混合动力电动汽车优化控制研究

Under the increasingly severe situation of energy depletion and environmental pollution, hybrid electric vehicle (HEV) is the sanitary car with the most feasible low fuel consumption and low emission in the current and future period of time. While HEV is a highly complicated system which includes several working modes, and in each working mode which contains not only differential equation describing the development of variables, but also algebraic equation expressing the relation of variables, and then there are many challenging issues in its optimal control, and it is urgent to establish an appropriate mathematical model for HEV to study. This project aims to investigate the optimal control of HEV by applying the theorem of differential-algebraic hybrid swithched systems. Firstly, explore the kinetic equations and the algebraic constraint of varialbes between various working modes of HEV, and then carry on overall modeling. Seondly, basing on the differential-algebraic hybrid switching system model of HEV, we investagate the stability, switching path design, impulse controllability and so on so as to innovate a new theory and method to solve the optimal control problems of HEV system. Finally, combining with experiment simulation, we gradually investagate the optimal control problems fo HEV system. This project belongs to the interdisciplinary field of power and control science, whose problems orginiate from the great demand of the HEV optimal control and which can promote the development and application of the differential-algebraic hybird switching system's theory.

在能源枯竭和环境污染问题日益严峻的形势下，混合动力电动汽车(HEV)是目前和未来一段时间内最具有可行性的低油耗、低排放的清洁汽车。而HEV是具有高度复杂性的多模态系统，并且在各个模态下既包含描述变量变化的动态方程，又包含表示变量之间约束的代数方程,遂使得其优化控制存在众多挑战性问题,亟需为其建立恰当的数学模型并进行研究。本课题拟在微分代数混杂切换系统理论体系下研究HEV的优化控制问题。首先，通过研究HEV各个工作模态的动力学方程和变量间的代数约束，对其进行整体建模。然后，基于得到的HEV微分代数混杂切换系统模型，研究其多工作模态稳定性、切换路径设计、脉冲能控性等问题，建立解决其优化控制问题的新理论和方法体系。最后，结合实验仿真，不断地探讨HEV系统优化控制问题。本课题属电力、控制领域的交叉前沿课题，问题来自HEV优化控制的实际重大需求，有利于促进微分代数混杂切换系统理论研究和应用的发展。

本项目首先通过对燃油发动机进行系统辨识，得到其近似系统。然后，以并联混合动力电动汽车为例，将其四种工作模态进行简化得到四个子系统，并选择合适的公共变量：扭矩和转速，作为四个子系统进行切换的耦合变量，从而得到了混合动力电动汽车系统模型：微分代数非线性切换系统。然后，基于通常非线性切换系统在任意切换和停留时间满足一定条件的任意切换下的理论结果，通过对微分代数切换系统增加某些限制条件，或者进行反馈预处理，得到了该系统的稳定性结果。基于得到的稳定性结果，结合广义Hamilton系统理论的理论结果，通过设计合适的控制器，研究了微分代数非线性切换系统的控制设计问题，包括鲁棒控制，自适应鲁棒控制，干扰抑制等问题。最后，将所得的理论结果应用到混合动力电动汽车的优化控制问题中，得到了一个鲁棒优化控制策略，数值仿真和近似的实验仿真结果显示在一定的误差范围内上述控制策略是可行的和有效的。