Fault Estimation And Fault-Tolerant Control Of Spacecraft Attitude Control System Based On T-S Model

As an important and expensive precision equipment,spacecraft operates in a complex and harsh space environment,resulting in high system operation and maintenance costs.Once a fault occurs,maintenance is difficult.In particular,when the actuator encounters a fault,the execution effectiveness decreases or even returns to zero,and the remaining actuators need to generate more driving torque to compensate for the failed actuator and maintain control system performance.Decreasing the effectiveness of actuators can exacerbate control performance degradation and may lead to instability of the control system,increasing the risk of space mission failure.In order to enhance system stability and maintainability,and reduce operation and maintenance costs,timely and effective attitude fault estimation and fault-tolerant control technology has received widespread attention from researchers in the aerospace field.In addition,the uncertainty of inertia parameters,spatial interference torque,and other factors make the spacecraft attitude stability problem more complex.This paper focuses on two core issues: fault estimation and fault-tolerant control of spacecraft attitude control systems during large attitude angle maneuvers.The effects of actuator faults,attitude angular velocity sensor faults,parameter perturbations,input saturation constraints,signal quantization,and other factors on the attitude control system of a rigid or flexible spacecraft and their compensation methods are emphatically discussed to ensure the stability and control performance of the attitude system.The main contents are as follows:A new fault estimation method based on Takagi-Sugeno(T-S)fuzzy adaptive observer is proposed for the attitude stability of rigid spacecraft with simultaneous external disturbances,parameter perturbations,and time-varying actuator failures.Firstly,the T-S fuzzy model of the attitude dynamics subsystem under large attitude angle maneuver is constructed.Then,a fuzzy fault estimation observer based on parameter adjustable proportional-integral adaptive law is designed to estimate both the system state and the actuator fault.Based on Lyapunov stability theory and augmentation techniques,the estimation error has been proven to be uniformly ultimately bounded.The elimination method and auxiliary optimization variable design are used to relax the observer design conditions.An intuitive quantitative evaluation performance indicator is introduced to obtain reliable evaluation and decision.Compared to existing observer methods,the proposed method has a better estimation performance index.Finally,numerical simulation analysis of a rigid spacecraft verifies the effectiveness and advantages of the proposed algorithm.To solve the attitude stability problem of flexible spacecraft with actuator faults,external disturbances,input saturation,and installation errors,an anti-disturbance fault estimation and attitude tracking fault-tolerant control scheme is proposed.Unlike traditional methods,both the nonlinear dynamics subsystems and attitude control systems are reconstructed into T-S fuzzy models,where partial nonlinear terms are remained to assist the design of fuzzy controller.The controller has fewer fuzzy rules and lower computational burden.Firstly,a fuzzy fault estimation observer is designed,which can simultaneously obtain estimations of unknown actuator faults and vibration disturbances of flexible attachments,without measuring any dynamic information of the flexible attachment.Then,based on feedback linearization and feedforward compensation,a fuzzy sliding mode fault-tolerant controller is designed to counteract actuator faults and flexible vibration disturbances in real time.Finally,numerical simulation analysis of a flexible spacecraft verifies the effectiveness of the T-S fuzzy model designed and the superiority of the proposed fault estimation and fault-tolerant control strategy.A separate fault estimation method based on T-S fuzzy model is proposed for spacecraft attitude stability in the presence of both angular velocity sensor and actuator faults.Based on the T-S fuzzy model of the attitude kinematics subsystem constructed,a fuzzy iterative learning observer is designed to reconstruct the angular velocity sensor fault.Based on the constructed T-S fuzzy model of the attitude dynamics subsystem and measurement feedforward compensation,a fuzzy adaptive observer is designed to reconstruct the estimation of actuator faults and flexible vibration disturbances.State estimations are used to replace non measurable premise variables.The observer design parameters are adjusted by using regional pole placement and H-infinity optimization techniques,and the estimation error has been proven to asymptotically converge to a small neighborhood of the zero attachment.Finally,numerical simulations of rigid spacecraft and flexible spacecraft are performed to verify the effectiveness and advantages of the proposed algorithm.A quantitative fuzzy fault estimation and sliding mode fault-tolerant control method is proposed to address the attitude stability problem of flexible spacecraft in the presence of both system state and control input signal quantization.Considering a dynamic uniform quantizer and a hysteretic quantizer,a T-S fuzzy model of the attitude dynamics subsystem and control system of a flexible spacecraft suitable for large attitude angle operation mode is constructed.A quantized fuzzy adaptive observer and a quantized fuzzy fault-tolerant controller are designed using the quantized value of the attitude angular velocity as the premise variable.Using H-infinity optimization techniques to reduce quantization errors caused by quantization mechanisms.Based on Lyapunov stability theory and linear matrix inequality techniques,the design process and existence conditions of the observer and controller are given.Finally,the numerical simulation of a flexible spacecraft verifies that the proposed algorithm can achieve simultaneous estimation and separation of actuator and flexible vibration disturbances,and that the observer estimation error,attitude angle and attitude angular velocity tracking error can quickly converge to a small neighborhood near zero.