Research On Model Reference Adaptive Augmenting Control Of Heavy Lift Launch Vehicle

With the development of China’s aerospace technology and space exploration,the manufacture of heavy lift launch vehicles has become an inevitable requirement for China’s space development strategy.Due to its large mass,thrust and slenderness ratio,low structural quality and multi-engine joint swing,the elastic vibration characteristics is complicated during flight.Furthermore,the sophisticated external environmental disturbances and parameter uncertainty makes traditional gain-scheduled PD control difficult to meet control accuracy.In this paper,a model reference adaptive augmenting control method is studied for heavy lift launch vehicle attitude control,in order to improve the adaptability of traditional gain-scheduled PD control to the flight environment.Firstly,the research status of adaptive augmenting control is investigated,including the development of adaptive gain adjustment algorithm,disturbance compensation control,active load-relief control and control allocation technology.Then the motion characteristics and stress of the heavy lift launch vehicle are analyzed,and the rigid-elastic coupling mathematical model,elastic vibration model and control model of the vehicle are established.According to the frequency change of elastic vibration,the bending filter is designed based on particle swarm optimization(PSO)method and a method of elastic vibration suppression is proposed based on time-of-flight switching.Then an adaptive gain adjustment algorithm is designed to improve the ability of the gain-scheduled PD controller to handle large disturbances,uncertainties,and sudden change of the elastic vibration frequency.Simulation results show that the adaptive gain adjustment algorithm has the advantage on enhancing the nominal controller performance.Aiming at the large attitude error caused by the drop of control gain,a disturbance compensation module is designed based on the extended state observer.For excessively large structural loads when launch vehicle with low structural mass passes through a large dynamic pressure zone,the active load-relief module is designed based on the Lyapunov theorem.For the control allocation of multi-thrust vector engine,the control distribution module is designed based on chained incremental algorithm.Finally,the above three modules are augmented to the model reference gain adaptive control structure to constitute a model reference adaptive augmenting control system.Finally,the simulation is carried out.By comparing the model reference adaptive augmenting control with the conventional gain-scheduled PD control,the following are verified:(1)Adaptive gain adjustment algorithm is able to suppress large disturbances,uncertainties,and the effects of abrupt changes in the elastic vibration frequency by changing the control gain.(2)The disturbance compensation control based on the extended state observer is able to reduce the attitude error caused by the gain reduction;(3)Based on the Lyapunov theorem,active load-relief control is able to reduce the structural load in the large dynamic pressure region.(4)The chain increment algorithm can effectively achieve the torque distribution of the multi-thrust vector engine.