非线性系统的输入饱和控制研究

发布时间：2018-07-11 11:50

本文选题：非线性系统 + 输入饱和　；参考：《江南大学》2017年硕士论文

【摘要】：在实际的工业生产中,系统不能传输无限大的控制信号,所以饱和现象广泛存在于各种系统中,然而人们在设计控制器时往往会忽略系统的饱和现象,当系统发生饱和时,系统性能显著下降。饱和现象曾引起了许多悲剧,这些悲剧让人们开始意识到饱和的存在及其重要影响。经过国内外学者长时间的探索,饱和约束控制取得了巨大的发展。过去提出的控制方法大多基于线性系统,对含有饱和约束的非线性系统的成果却非常少。此外,鲜有方法能同时处理控制输入的幅值与速率的饱和问题。因此,本文着眼非线性系统的输入饱和现象,结合固态氧化燃料电池(SOFC)、空气弹性系统和变速风力机(VSWT)等对象,展开动态抗饱和策略研究。本文的研究内容和创新点主要分为以下几个部分:1.结合SOFC考虑非线性离散系统,设计动态抗饱和策略。首先简单介绍了SOFC的基本结构和工作原理,并根据实际的控制约束条件来设计约束控制器。利用RBF神经网络来辨识SOFC的系统模型,得到模型输入输出灵敏度函数。基于BP神经网络和PID控制设计自适应控制器,并利用辨识得到的输入输出灵敏度函数设计动态抗饱和补偿器。通过抗饱和补偿器调整控制输入的参考轨迹,达到抑制输入饱和的效果,利用Lyapunov函数进行稳定性分析,最后结合SOFC仿真来验证算法的有效性。2.以气动弹性系统为对象,对多输入多输出非线性离散系统设计抗饱和策略。首先简单介绍了一种典型的气动弹性系统(二自由度机翼)的相关结构,利用紧凑型动态线性化(CFDL)对系统模型进行转换。在线性化模型的基础上,利用多元观测器设计自适应优化控制器,将控制问题转化为线性矩阵不等式(LMI)的约束优化问题。最后通过仿真验证气动弹性系统控制输入在约束范围以内,并且整个过程中都不知道这种气动弹性系统的精确模型。3.以VSWT为对象,研究非线性连续系统的动态抗饱和策略。先简单介绍了一下VSWT的结构和原理,在变换的VSWT模型基础上,设计扩展观测器估计系统未知参数,结合指令滤波器和反步法来设计约束控制器,最后通过VSWT的仿真验证算法的有效性。
[Abstract]:In the actual industrial production, the system can not transmit infinite control signal, so the saturation phenomenon exists widely in various systems. However, people often ignore the saturation phenomenon of the system when designing the controller, when the system saturation occurs, The performance of the system decreased significantly. Saturation has caused many tragedies, which make people realize the existence of saturation and its important influence. After a long time of exploration by scholars at home and abroad, saturation constraint control has made great progress. In the past, most of the control methods were based on linear systems, but few results were obtained for nonlinear systems with saturation constraints. In addition, there are few methods to deal with the saturation of input amplitude and rate simultaneously. Therefore, this paper focuses on the input saturation phenomenon of nonlinear systems, and studies the dynamic anti-saturation strategy combined with solid oxide fuel cell (SOFC), air elastic system and variable speed wind turbine (VSWT). The research content and innovation of this paper are divided into the following parts: 1. Considering the nonlinear discrete system, a dynamic anti-saturation strategy is designed. Firstly, the basic structure and working principle of SOFC are briefly introduced, and the constraint controller is designed according to the actual control constraints. The system model of SOFC is identified by RBF neural network, and the sensitivity function of model input and output is obtained. The adaptive controller is designed based on BP neural network and pid control, and the dynamic anti-saturation compensator is designed by using the input and output sensitivity function. The reference trajectory of the control input is adjusted by the anti-saturation compensator to suppress the input saturation. The stability analysis is carried out by using the Lyapunov function. Finally, the effectiveness of the algorithm is verified by the SOFC simulation. An anti-saturation strategy is designed for multi-input and multi-output nonlinear discrete systems based on Aeroelastic systems. In this paper, a typical Aeroelastic system (2-DOF wing) is introduced briefly, and the model of the system is transformed by using compact dynamic linearization (CFDL). Based on the linearization model, an adaptive optimization controller is designed by using a multivariate observer, and the control problem is transformed into a constrained optimization problem of linear matrix inequality (LMI). Finally, the simulation results show that the control input of the Aeroelastic system is within the constraint range, and the exact Aeroelastic system model .3is not known in the whole process. Taking VSWT as an object, the dynamic anti-saturation strategy for nonlinear continuous systems is studied. Firstly, the structure and principle of VSWT are briefly introduced. Based on the transformed VSWT model, an extended observer is designed to estimate the unknown parameters of the system, and a constraint controller is designed by combining the instruction filter and the backstepping method. Finally, the validity of the algorithm is verified by the simulation of VSWT.
【学位授予单位】：江南大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP273

【参考文献】