动平衡状态及非线性时变控制系统设计
文献类型：学位论文
| 作者 | 王庆林 |
| 学位类别 | 工学博士 |
| 答辩日期 | 1998-10-01 |
| 授予单位 | 中国科学院自动化研究所 |
| 授予地点 | 中国科学院自动化研究所 |
| 导师 | 何善堉 ; 裘聿皇 |
| 关键词 | 非线性控制 时变系统控制 平衡状态控制 反馈线性化 模型参考控制 李亚普诺夫方法 nonlinear control time-varying systems control equilibrium control feedback linearization model reference control Lyapunov metho |
| 其他题名 | Dynamic Equilibrium And Nonlinear Time-Varying Control Systems Design |
| 学位专业 | 控制理论与控制工程 |
| 中文摘要 | 非线性时变控制系统的设计不仅是控制领域中最前沿和最具挑战性的课题 之一，也是近年来控制理论平衡状态运动的。对于大范围渐近稳定的系统，系统 状态将渐近于系统的平衡状态。因此对大范围稳定的系统，通过控制系统平衡 状态运动，可以实现对状态和输出的间接控制。动平衡状态概念是平衡状态概 念的扩展，利用它可以将控制系统的调节问题和跟踪问题方便地统一起来。这 为控制系统的设计提供了一条全新的思路。 结合动平衡状态的概念，本章还对系统平衡状态及稳态输出的可控性问题、 稳态误差分析等问题进行了讨论。并给出了相应的结果。 2．基于上述动平衡状态的概念和李亚普诺夫稳定性理论，本文给出了一种 非线性时变系统反馈线性化的直接方法。其基本思想是：首先使控制系统对其 平衡状态大范围渐近稳定，再设计控制律使系统的平衡状态按预定的方式运动。 这一思想在具体实现上采用了模型参考的方法。即首先设计一满足要求的线性 定常系统作为参考模型，然后将模型的状态作为控制系统的动平衡状态，设计 控制律使系统状态对动平衡状态渐近稳定。由于系统状态渐近于其动平衡状态， 从而实现了对系统状态及输出的渐近反馈线性化控制。该方法的一个优点是它 可以较方便地实现控制系统的鲁棒设计。 基于上述的设计思想，论文给出了一类非线性时变控制系统精确线性化和 鲁棒线性化的设计方法。 3．为了解决非线性时变系统反馈线性化设计中的状态估计问题，本文基于 非线性跟踪微分器的概念提出了两种线性跟踪微分器的设计方案。这两种方案 不仅易于实现，而且可按要求的无差度跟踪输入信号。可以方便地获取状态的 微分信号，满足输出反馈的需要。 4．基于动平衡状态的概念，本文给出了一种控制系统稳态设计的方法：稳 态逆系统方法。这一方法可实现系统的稳态无差控制，同时也使复合控制系统 的设计更为方便。这一方法可以看作是逆系统方法的简化，同时又可以看作是 频域传递函数方法的扩展。 5．基于动平衡状态的概念，本文还提出了一种将模型参考控制系统中模型 输出与对象输出同时偏置，以实现无差控制的方法：模型参考逆系统方法。在 满足一定条件的情况下，该方法对非线性时变系统有较好的控制效果。 6．论文对上述各部分的内容均进行了仿真研究，仿真结果证明了上述各种 方法的合理性和有效性。 |
| 英文摘要 | The design of nonlinear time-varying control system is not only one of the most leading and challenging topic, but also one of the emphases and popular points in the field of control theory. Since 1980's, although the great progress has been made in the research of this field, there are still many challenging problems to be solved because of the extreme complexity of nonlinear control system. Dealing with the control problems of nonlinear time-varying system, the contribution of this dissertation mainly includes the following aspects: 1. The affection of input to system is analyzed, and the concept of dynamic equilibrium state is raised. It is considered that what the input of control system controls directly is the system dynamic equilibrium, not the output and states of system. The system equilibrium state changes with the system input. The system state (output) moves relatively to its equilibrium state under the constrain of the system structure. To an asymptotically stable system in the large, the system state will asymptotically converges to the system equilibrium state. So to the asymptotically stable system in the large, the indirect control to state and output can be realized when the movement of the system equilibrium state is controlled. A totally novel idea is provided for the design of the control system. With the concept of dynamic equilibrium state, some problems such as the controllability of system equilibrium state and steady state output, the analysis of steady state error are discussed in this dissertation, and corresponding results are given. 2. On the basis of above concept of dynamic equilibrium state and Lyapunov stability theory, a kind of direct method of feedback linearization for the nonlinear time-varying system is given. The basic idea is, first, the control system is asymptotically stabilized to its equilibrium state in the large. Then, the control law is designed to make the system equilibrium state move in the scheduled way. When this idea is realized concretely, the method of model reference is used. That is, a linear time-invariant system which meets the request is designed to serve as a reference model, then the state of the model is treated as the dynamic equilibrium state of the control system, and the control law is designed to make the system state asymptotically stabilized to dynamic equilibrium state. Because the system state is asymptotically stabilized to its dynamic equilibrium state, the asymptotic feedback linearization control of system state and output is obtained. One advantage of this method is that the robust design of control system can be realized conveniently. Some exact linearization and robust linearization methods for nonlinear time- varying systems are given in this dissertation. 3. In order to solve the problem of state estimation in the design of feedback linearization for nonlinear time-varying system, two kinds of design |
| 语种 | 中文 |
| 其他标识符 | 495 |
| 源URL | [http://ir.ia.ac.cn/handle/173211/5691]  |
| 专题 | 毕业生_博士学位论文 |
| 推荐引用方式 GB/T 7714 | 王庆林. 动平衡状态及非线性时变控制系统设计[D]. 中国科学院自动化研究所. 中国科学院自动化研究所. 1998. |
入库方式： OAI收割
来源：自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。