【作者】 文杰；

【导师】 姜长生；

【作者基本信息】 南京航空航天大学 ， 控制理论与控制工程， 2011， 博士

【摘要】 非仿射非线性系统广泛存在于现实生活中,考虑到实际系统通常具有不确定性,控制非仿射非线性不确定系统是一项创新而又富有挑战性的课题,具有重要的理论意义和应用价值。围绕这一课题,基于模糊逻辑系统的万能逼近特性,系统化地研究了几类非仿射非线性系统的自适应模糊控制方法。针对一类单输入单输出非仿射非线性不确定系统,分别设计自适应模糊状态反馈控制器和输出反馈控制器。在系统状态可测的情况下,用模糊逻辑系统逼近未知函数,模糊参数采用σ自适应律,给出参数的有界性证明,引入鲁棒控制项来消除模糊逻辑系统的逼近误差和系统的外界干扰,使得闭环系统跟踪误差收敛到零。为了抑制控制的抖振性,鲁棒控制项中引入双曲正切函数,设计了与之匹配的自适应参数调整律,并证明了闭环系统所有信号一致最终有界,跟踪误差收敛到零的一个邻域。将算法应用到Duffing-Holmes混沌系统、Sprott电路系统、Genesio混沌系统,验证了算法的有效性。在系统状态不可测的情况下,构造线性误差观测器估计系统输出误差和系统状态,利用估计状态变量和估计输出误差变量设计自适应控制律,并证明了闭环系统的稳定性,控制性能在Duffing-Holmes混沌系统的应用中得到验证。针对一类具有严格反馈形式的非仿射非线性系统设计基于Backstepping方法的自适应模糊控制律,仅要求模糊逻辑系统的逼近误差范数有界,设计自适应律时只考虑对未知参数范数的估计值进行在线调整,这样减少在线调整参数的数目,采用自适应鲁棒控制项消除逼近误差和外界干扰。证明了闭环系统在Lyapunov意义下的稳定性,并将该方法应用到Chua’s电路系统和R?ssler混沌系统中来验证算法的有效性。针对一类严格反馈型多输入多输出非仿射非线性系统设计一种自适应模糊控制器,该系统由多个互联子系统组成,针对每个子系统构造Lyapunov函数以设计相应的控制律,最后设计整个系统的Lyapunov函数,从而保证闭环系统所有信号一致最终有界,跟踪误差收敛到零的一个邻域。将算法应用到Lorenz系统、永磁同步电动机系统、Lü系统、Liu系统以及超混沌R?ssler系统的控制中,并通过仿真分析验证了算法的有效性。针对非仿射非线性时滞系统提出一种自适应模糊控制算法,通过构造适当的Lyapunov- Krasovskii泛函有效消除系统的时滞项,利用双曲正切函数的连续性来解决补偿过程中产生的非线性余留项可能引起奇异性问题,并从理论上证明闭环系统的稳定性。将带时滞的Rssler混沌系统作为仿真对象,进行控制律设计,给出仿真结果,验证算法的有效性。最后,针对控制律设计中的参数选择问题,采用最佳保留遗传算法来选择参数值,将跟踪误差、模糊系统复杂性、控制输入的抖振性作为目标函数,设计相应的遗传算子,得到较优的设计参数,并用数值算例验证算法的有效性。更多还原

【Abstract】 Nonaffine nonlinear systems widely exist in real world. Taking into account that the actual system is often uncertain, how to control the nonaffine nonlinear uncertain systems is an innovative and challenging subject and has important theoretical and practical value. To solve this problem, based on the universal approximation properties of fuzzy logic system (FLS), this thesis systematically studies a series of adaptive fuzzy control strategies for several types of nonaffine nonlinear systems.Adaptive state-feedback and output-feedback fuzzy controllers are developed for a class of single input single output (SISO) nonaffine nonlinear uncertain systems. In the situation that the system states are observable, a fuzzy logic system is employed to approximate the unknown nonlinearity, the adjustable parameters in FLS are updated byσadaptive law and the proof of boundedness of parameters is given. A robust control term is introduced to compensate the approximation error and disturbances to make the tracking error converge to zero. In order to avoid chattering of the control input,a tanh function is used in robust term and the corresponding adaptive law is redesigned. The proposed controller guarantees that all the signals in the closed-loop system are bounded and the tracking error eventually converges to a small neighborhood of zero. The developed design scheme is applied to Duffing-Holmes system, Genesio system and Sprott circuit chaotic system. Simulation results demonstrate the effectiveness of the proposed approach. In additon, it is assumed that the system states are unavailable, then a linear error observer is employed to estimate output errors and states. An adaptive controller is designed by using the observed values of errors and states, so that the whole closed loop system is stable in the sense of Lyapunov. Duffing-Holmes chaotic system is used to emphasize the effectiveness of the approach.Based on backstepping approach, a robust adaptive fuzzy control scheme is presented for a class of strict-feedback nonaffine nonlinear systems. Approximation errors of fuzzy systems are only required norm-bounded, and the online computation quantity is reduced by tuning estimations of the unknown bounds online. An additional adaptive term is employed to compensation approximation errors and disturbances. The stability of the whole closed loop system is proved via Lyapunov method. Chua’s circuit sytem and R?ssler chaotic system is presented to illustrate the feasibility and effectiveness of the proposed control technique.□An adaptive fuzzy control schemes is proposed for a class of multi-input multi-output (MIMO) nonaffine nonlinear systems in block-triangular forms. The MIMO systems consist of interconnected subsystems, Lyapunov function is constructed for each subsystem to design the corresponding control law. Finally, Lyapunov function of the whole system is designed so that the all the signals in the closed-loop system are bounded and the tracking error eventually converges to a small neighborhood of zero. The developed design scheme is applied to Lorenz system, permanent-magnet synchronous motor system, Lüsystem, Liu system and hyperchaotic R?ssler system. Simulation results demonstrate the effectiveness of the proposed approach.An adaptive fuzzy control schemes is proposed for a class of strict-feedback nonaffine nonlinear systems with time delays, the unknown time delays are compensated for using appropriate Lyapunov-Krasovskii functionals, the singularity generated in the process of compensation is overcome by the using of tanh function. The proposed controller guarantees that all closed-loop signals remain bounded. R?ssler chaotic system with time delay is given to illustrate the design procedure and performance of the proposed method.Parameter selection is another important problem for controller design. Thus, in the last part of the thesis, taking the complexity of FLS, tracking error and control oscillation as object functions, an elitist preserved genetic algorithm is employed to search for the optimum parameters by designing corresponding genetic operators. A simulation example is presented to illustrate the feasibility and effectiveness of the proposed control technique.更多还原