Research On Robust Fault Tolerant Control Technology For Complex Vehicles

At present, with the rapid development and progress of the society, the scale of the complexcontrol system becomes larger and larger; the demands for the safety, reliability and maintainability ofthese complex control systems are increasing. In order to reduce the number of catastrophic accidents,reduce the environment pollution, reduce the losses of social economic assets and protect the lives ofthe people, the technology about fault diagnosis and fault tolerant control of complex control systemsprovides an alternative road for solving the problems described above. In the past few decades, thetechnology about fault diagnosis and fault-tolerant control of the complex control system has madeconsiderable progress. Especially in the past ten years, along with the quick development of thecomputer network, pattern recognition, machine learning and all kinds of advanced control algorithms,many new methods and technologies are introduced into the fields of fault diagnosis and fault tolerantcontrol, which greatly enrich the research contents of fault diagnosis and fault-tolerant controltechnology.Currently, United States, Russia, France, United Kingdom, Japan and other aerospace powercountries have put a lot of manpower and material resources for developing the flight technology ofncomplex vehicles, they have achieved lots of research results, even some key technologies haveentered the engineering verification phase. China, as a world power country, should vigorouslydevelop the relational key technologies for technical reserves. Based on the existing research results,this study is further investigating the technology about active robust fault tolerant control forfixed-wing aircraft and multi-rotor vehicles. This research is divided into the following five sections:First: For near space vehicl attitude control system, based on neural network and commandfiltered backstepping techniques, a robust fault-tolerant control method is proposed, first, near spacevehicle attitude control system mathematical model is given, on this basis, the state equation of NSVattitude control system under the modeling errors caused by uncertainty, external disturbances andcontrol surfaces faults is considered. The fault tolerant control design involves two main units, one isauxiliary system design, the other is controller design using the auxiliary system. The neural networkis used to ensure robustness for auxiliary system. And the stability of the closed-loop system isrigorously proved by Lyapunov theorem. Finally, in the Matlab simulation of NSV attitude controlsystem, the simulation results show that the proposed method can make the tracking performance forflight control system with uncertain and external disturbances under control surface damage.Second: For the near space vehicle attitude control system, and to consider the flight control surface deflection position and velocity saturation constraints, based on the dynamic control allocationmethod, a control reallocation method is proposed under control surfaces stuck and damage faults.The pseudo-controller is designed using dynamic inversion method. The overall fault-tolerant controlarchitecture is still mentioned in the framework of the second chapter. First, we give the attitudemodel, which is caused by modeling errors; external disturbances; the control surfaces stuck anddamage faults. And an adaptive sliding mode observer is designed, which implicits faults anddisturbances information. In order to achieve fault tolerant control, the pseudo-dynamic controller andcontrol allocation can be accommodated using above information. Finally, Matlab simulation resultsshow the effectiveness of the proposed method for NSV attitude control system.Third: For near space vehicl attitude control system, considering the actuator control loopdynamics, based on decentralized fault-tolerant control framework, actuators damage and stuck faultdetection and identification unit is designed for the flight control system, and observer-based auxiliarysystem unit is also designed. The auxiliary system implicits control surface damage faults anddisturbances information. A reconfigurable fault-tolerant controller is used to achieve fault-tolerant.First, we give the NSV attitude control system under actuator stuck, lose of effectiveness, and controlsurface damages faults. Second, a multi-observer is designed for actuator fault detection andidentification using a decision-making mechanism to determine current actuator failure modes. Then,an adaptive sliding mode observer is designed for implicit control surface damages and interferenceinformation. The reconfigurable controller can achieve fault tolerant using the information of actuatorFDI unit and adaptive sliding mode observer. And the stability of the closed-loop system is rigorouslyproved by Lyapunov theorem. Finally, Matlab simulation results show the effectiveness of theproposed method for NSV attitude control system.Fourth: For non-affine nonlinear flight control systems, a nonaffine nonlinear robust FTCmethod is proposed. Here, the fault tolerance control framework of Chapter II is used, and on thisbasis, observer-based auxiliary system is designed to implicit fault parameteres and disturbancesinformation. Then reconfigurable controller is designed using the dynamic of observer-based auxiliarysystem, which can chieve nonaffine nonlinear systems robust adaptive fault-tolerant. Theoreticalanalysis and simulation results show the effectiveness of the proposed method.Fifth: Based on proposed fault tolerant control framework of second chapter and proposedadaptive sliding mode observer of the third chapter, a fault-tolerant command filtered backstepping isdesigned. And further application of the algorithm in the3-DOF Hover platform, verify theframework design algorithm practical application value.