Research On Finite-time Robust Control For A Class Of Non-affine Nonlinear Large-scale Systems

In practical engineering,there exist a great deal of nonlinear large-scale systems.The stability and control accuracy of the system are often affected by many factors,such as unmodeled dynamics,unknown control directions,unknown disturbances,dead zone and so on.In general,these factors may exist in practical system.If these factors are not further considered,it is difficult to guarantee the ultimate control effect of the systems.Therefore,several types of finite-time robust control problems are studied in the paper,the main content of the research is as follows:1.By means of backstepping control scheme,dynamic signal processing method,neural network adaptive technique and prescribed performance finite time control technique,a decentralized adaptive neural network prescribed performance robust quantization tracking control strategy is present for a class of non-affine nonlinear interconnected systems with unmodeled dynamics and quantized input.The settling time in the proposed control strategy is a design parameter,the convergence rate of tracking errors of each subsystem is further adjusted by tuning the size of settling time.Meanwhile,in the controller design,the problem that prescribed performance is difficult to achieve decentralized control is solved by means of a special mathematical transformation method.Finally,a simulation example is utilized to verify the effectiveness of the proposed method.2.A finite-time robust control problem is studied for a class of non-affine nonlinear interconnected systems with unknown control directions,unmodeled dynamics and unknown dead-zone nonlinearity.And a decentralized adaptive prescribed performance finite-time robust tracking control strategy is proposed.In the control design,the author solves the problem that Nussbaum-gain technique is difficult to achieve finite time control for the first time.The proposed controller consists of two adaptive control parts,one is the adaptive part caused by the change of control directions,the other is the adaptive part caused by unknown functions in the systems.The control problem studied in this chapter is more complex and more suitable for the actual situation.The proposed control strategy guarantees that the tracking error and output signal of each subsystem are semi-globally practically finite-time stable.A numerical simulation and a tripled coupled inverted pendulum system simulation are used to demonstrate the superiority of the proposed method.3.A decentralized adaptive neural network prescribed performance finite time bounded-H? tracking control method is proposed for a class of non-affine nonlinear interconnected systems with saturated input and unknown external disturbances.This method can guarantee that the output signal of each subsystem can track the given reference signal in finite time.Meanwhile,the adverse effects caused by external disturbances can be attenuated.Furthermore,the proposed control strategy avoids the problem of excessive controller output value in nonlinear H? control method.Therefore,this control strategy is more suitable for the actual situation.Finally,a doubled coupled inverted pendulum system simulation is employed to verify the feasibility of the proposed method.