Attitude Maneuver Control Method For Flexible Spacecraft Based On Universal Kriging-MPSO

With the continuous development of space technology,high pointing accuracy and stability require higher attitude control stability of spacecraft for earth observation,laser communication and space exploration.In order to obtain enough energy under weak solar pressure and serve as the core energy supply component of spacecraft,solar panels are gradually developing towards large-scale and lightweight.The coupling effect between the central rigid body and the elastic modes of the solar panel of this type of spacecraft has a strong nonlinear dynamic characteristic.When the spacecraft performs large-angle attitude maneuver,elastic vibration will be generated in the flexible attachment,which will seriously affect the attitude stability of the spacecraft.This paper proposes a rigid-flexible coupled dynamic model and predicts the running state using a discrete variable data set to address this problem.An optimal control method based on Universal Kriging–MPSO is proposed.The effectiveness of the proposed control method is verified through numerical simulation.The specific research contents are summarized as follows:?1?A rigid-flexible coupled dynamic model of a flexible spacecraft was presented using the Euler–Lagrange equation.A central rigid body dynamic model was established using the Euler method,a solar array dynamic model was constructed using the Lagrange method,and their coupling relationship was explained.?2?A universal kriging spacecraft operation state prediction method was proposed to address the inaccurate operation status prediction of spacecraft under the vibration of flexible solar panels and the external environment.Two standard test functions were selected,and sample points were selected through Latin hypercube sampling and were divided into two groups,namely,training and test sample sets.The sample points were fitted using universal kriging,second-order kriging,kriging,neural network,and polynomial response surface methods.The comparison of results indicates that the prediction accuracy of universal kriging can reach 98.67%.The prediction performance of neural network and universal kriging was compared in accordance with the actual data of the spacecraft operating status.Compared with neural network,the training time of universal kriging is reduced by 99%,and its prediction accuracy reaches more than 99%.?3?Aiming at the problem of acquiring the optimal output torque of the actuator,this paper proposes an optimal control method based on improved particle swarm optimization.On the basis of traditional particle swarm optimization,global–local information sharing terms,probabilistic sudden jump characteristics,and elimination mechanism are introduced and applied to spacecraft attitude control.The performance of the improved particle swarm optimization is verified through the simulation results of 26 standard test functions.?4?A universal kriging prediction model was proposed on the basis of the improved particle swarm optimization control method to meet the control stability requirements of flexible spacecraft.The spacecraft operation state was predicted in advance,with the minimum prediction deviation and the actuator torque as targets.The particle swarm optimal control method obtains the actuator output torque.The numerical simulation analysis shows that the proposed optimal predictive control method can complete a large angle attitude maneuver of 30°yaw within 42 s,and 15°yaw,-20°roll,and 20°pitch within 38 s,and the stability is improved by an order of magnitude.?5?The virtual prototype model of flexible spacecraft is established,and the joint simulation test of spacecraft coupling dynamic response and attitude maneuver control is carried out.The validity of Universal Kriging—MPSO control method is verified.The joint simulation results of virtual prototype are in good agreement with the theoretical analysis results.This paper proposes an optimal predictive control method based on universal kriging and improved particle swarm optimization.Under the free vibration of a flexible solar panel,the universal kriging algorithm is used to predict the spacecraft operating state in advance.The relevant information is substituted into the improved particle swarm optimal controller to obtain the actuator output torque while performing the spacecraft attitude maneuver and compensating the influence of solar panel vibration on the platform attitude,thereby achieving large angle maneuvering stability control with good control performance and attitude stability of±4.706×10^-3rad/s.Compared with other control methods,the proposed optimal control system has high attitude stability,and the corresponding results can contribute to the development of the large flexible spacecraft technology.