| As the motion control subsystem in a rocket system,the performance of the thrust vector servo mechanism(TVSM)will directly affect the control accuracy and reliability of the whole rocket system in the launching process.To develop a TVSM with excellent performances,it usually gets tested by using a class of electro-hydraulic load simulators in the ground tests.Therefore,the simulation accuracy on the real environmental load conditions and also the real-time performance of a load simulator will directly determine the accuracy and validity of TVSM performance test results,and then further affect the reliability of the rocket launch.In this work,the modeling for force loading dynamics and the inherent coupling characteristics of the load simulator system are first discussed,and then the corresponding nonlinear loading control strategies are designed to achieve the high accuracy simulated force loading.First,a nonlinear multi-disturbance coupled force loading model is proposed based on the actual mechanical structure of the electro-hydraulic load simulator.In this model,various disturbance factors,such as the position disturbance from the tested TVSM,the friction existed in the force loading cylinder and the mechanical backlash of the transmission mechanism,are considered.Furthermore,the mechanism of extra force generation is also explained theoretically.By comparing simulation results with actual engineering phenomenons,the validity of the above model is verified,which lays a theoretical foundation for the following analysis of influences from nonlinear coupling disturbances and the design of nonlinear force control strategies.Aiming at the problem of effective compensation for the influences of friction in the force loading cylinder and the mechanical backlash,how to obtain accurate mathematical descriptions of friction and backlash in the actual load simulator system is studied.First,considering the dynamic characteristic of friction and the discontinuous characteristic of backlash,an improved generalized Maxwell slip friction model and a quasi linear backlash model are proposed respectively,which are especially constructed for the parameter identification method design.Next,a friction identification method based on particle swarm optimization algorithm and a backlash identification method combining the second-order sliding mode observer with recursive least squares algorithm are designed accordingly,which make the identification for the parameters in nonlinear models more accurate.By using above identification methods and experimental data,the parameters of friction and backlash in the actual simulator system are both obtained,and the influences of parameter uncertainties on the loading performance are also analyzed,which further improves the multi-disturbance coupled force loading model and provides the feasibility for the following nonlinear disturbance compensator design.To solve the problem that how to improve the force loading accuracy and the dynamic performance under influences of nonlinear disturbances,an improved adaptive terminal sliding mode control strategy and an almost disturbance decoupling saturation control strategy are designed respectively based on the above-mentioned multi-disturbance coupled force loading model.First,from the point of view of the robustness enhancement for the loading process,an improved adaptive terminal sliding mode control strategy based on the speed observer is proposed.This strategy can not only suppress the external disturbances including TVSM’s position disturbance and mechanical backlash,but also ensure the system dynamic performance and stability by its finite time convergence characteristic.Meanwhile,its adaptive laws can realize the accurate compensation for the influence of friction parameter uncertainties.On the other hand,from the point of view of the external disturbance decoupling,an anti-windup almost disturbance decoupling saturation control strategy is proposed next for enhancing the engineering practicability.This method is basically designed based on the analysis result of the coupling characteristic between the position disturbance and the loading force,as well as the differential geometry theory.By choosing appropriate control parameters,the force tracking error under this control strategy can be reduced to a designed range without considering the form and boundary of external disturbance,which implies its simplicity.In addition,the saturation compensation auxiliary subsystem can effectively decrease the control signal oscillation,which helps eliminate the response lag caused by the signal saturation from hardware limitations and further improve the loading dynamic performance.At last,the effectiveness of above two control strategies in improving the loading accuracy and dynamic performance has both been verified by simulations,respectively.In order to prove the feasibility of above two force loading strategies in engineering applications,they have both been verified in the actual load simulator system.Then,on the basis of their effectivenesss verification,performance differences between proposed nonlinear control strategies and the widely used PID strategy,as well as performance differences between these two nonlinear control strategies are both compared and summarized,which provides a guidance on how to select the appropriate control algorithm for different practical engineering conditions in the load simulator field. |