Research On Model Predictive Control Of Flexible Spacecraf

In order to maintain the energy supply during space flight and to accomplish the corresponding space missions,modern spacecraft need to carry flexible appendages such as solar sails,robotic arms,antennas,etc.With the gradual implementation of major spaceflight projects,spacecraft are developing toward multi-function and mega-scale,and the mass and rotational inertia of the carried flexible structures occupy a large proportion of the whole star,so flexible vibration has an increasing influence on attitude.In addition,various uncertainties in the space environment can also interfere with the attitude,while modern space missions require high accuracy for spacecraft attitude control.Therefore,it is a challenging research topic to design a control strategy that can effectively suppress the vibration of large flexible attachments and meet the attitude control requirements.In this thesis,considering the control moment limitation and the influence of flexible attachment vibration on the attitude of spacecraft,we design the fuzzy model predictive attitude control strategy,the self-triggered model predictive attitude control strategy,the self-triggered model predictive attitude control strategy based on the Tube invariant set,and the full-order sliding mode active vibration control strategy to address the attitude control problem of large flexible spacecraft.The effectiveness of the designed control strategies is verified by numerical simulations.Before designing attitude controllers and active vibration controllers,accurate and effective mathematical models are inevitably required.Therefore,this thesis first states the reference coordinate system required to establish the mathematical model of a large flexible spacecraft,and then introduces the global modal method and details the dynamical and kinematic models of a large flexible spacecraft based on Hamilton’s principle.Considering the attitude control problem of the large flexible spacecraft,adaptive fuzzy control and generalized predictive control are combined to design a fuzzy generalized predictive control strategy for the large flexible spacecraft.Based on the dynamics model of the large flexible spacecraft,a nonlinear model predictive control law is designed using Taylor expansion,which avoids tedious calculations in the online optimization process of predictive control and thus effectively reduces the computational complexity.After that,the adaptive fuzzy control is used to approximate the uncertainty perturbation caused by the large flexible attachment vibration on the attitude control,and the adaptive law of fuzzy parameters is derived.The simulation results show that the designed control strategy has a good suppression effect on the large flexible attachment vibration and can control the attitude angle to achieve fast tracking of the desired value.Considering the attitude control problem of large flexible spacecraft with constrained control moments,a model predictive attitude control strategy based on a selftriggering mechanism is designed to solve the optimization problem only at the triggering moment to obtain the optimal control sequence,which is applied to the actual system using the first few steps of the optimal control sequence before reaching the next triggering moment.Thus,the frequency of solving the optimization problem is reduced and the online calculation time of the control strategy is shortened.Simulation results show that the selftriggering model predictive control strategy can control the large flexible spacecraft to reach the desired attitude rapidly.In order to deal with the influence of large flexible attachment vibration on the attitude control system more effectively,a self-triggered model predictive controller based on the Tube invariant set is designed.Firstly,the self-triggered model predictive control is designed for the nominal system of the large flexible spacecraft,and the optimization problem for the nominal system is solved only at the trigger moment to obtain the optimal control sequence for the nominal system.After that,the model predictive control algorithm based on the Tube invariant set is introduced to design the control law of the actual system.Finally,a full-order sliding mode active vibration controller is designed in order to suppress the flexible attachment vibration.The simulation results show that the designed control algorithm can make the vibration modes attenuate to zero quickly and have good tracking performance for attitude angle commands.In summary,the fuzzy model prediction controller,the self-triggered model prediction controller,the self-triggered model prediction controller based on the Tube invariant set,and the full-order sliding mode active vibration controller are designed for the attitude control of large flexible spacecraft,and then numerical simulations are performed.The simulation results show that the designed attitude controller and active vibration controller can make the flexible attachment vibration of spacecraft attenuate quickly and control the attitude angle to reach the desired value quickly and accurately,and finally the advantages and disadvantages of the designed control strategy are analyzed.