Research On Control Method Of Launch Dynamics For Multiple Launch Rocket System

Multiple Launch Rocket System(MLRS)has been paid increasingly extensive attention throughout the word due to its advantages of fast firing speed,strong fire power,low cost and high mobility.Poor firing precision due to the orientation of the launcher for MLRS departing from that intended in adjusting angle and salvo firing processes have always restricted the MLRS development for a long time.It has become an important development direction in the research field of MLRS to reduce the firing deviation in the processes of adjusting angle and salvo firing,and improve the firing precision by launch dynamics control technology.For a complex mechanical system such as MLRS,the major difficulty in designing an appropriate control strategy that can achieve the high-precision control is how to guarantee the robustness and stability of the control system under the occurrence of uncertainties,nonlinearities and motor delay.The national key basic research project,i.e.,the national 973 project is combined and a new type 122mm MLRS developed by the project group is taken as the research object in this dissertation.To deal with the problem of the strong nonlinearity,the significant uncertainty and the input delay of the actuator,a launch dynamics control scheme for the MLRS using a permanent magnet synchronous motor as the actuator and a fiber optic gyro as a sensor is proposed through in-depth exploration of the theory and technology of the dynamics control.This paper preliminarily verifies the feasibility and effectiveness of the designed control algorithm.The main contents and academic contributions of the thesis are as follows:(1)Based on the launch dynamics theory of multibody system,a rocket-launcher coupled launch dynamics model of MLRS is established.A PID controller is designed for the elevation and azimuth mechanisms of MLRS,and the developed model is used to check the control performance.The simulation results show that with the launch dynamics control technology,the firing deviation and vibration level of MLRS are greatly reduced,and the firing precision is significantly improved.(2)In order to facilitate the in-depth study of the launch dynamics control method,the rocket-launcher coupled dynamics model of the MLRS is simplified,and the dynamics equation of the simplified model is established based on the second-type Lagrange principle.Based on the simplified model,the computed torque method for the nonlinear control of MLRS is studied.The PID-computed torque controller is designed using the inner-loop and outer-loop strategies.The Routh criterion is used to analyze the stability of the proposed control system.(3)Directing at the issue that the computed torque control algorithm is sensitive to the uncertainty,i.e.,the robustness of such kind of controller is poor;a computed torque controller integrated with a radial basis function neural network is presented for the control of MLRS.First,the dynamic model of motor-mechanism coupling system for MLRS is established employing the preceding vector control theory and reduced model of MLRS.Then,in order to compensate for the lumped uncertainty due to parametric variations and un-modeled dynamics,a radial basis function neural network estimator is developed to adapt the uncertainty based on Lyapunov stability theory.Furthermore,the closed-loop system stability and the convergence of the estimated lumped uncertainty are mathematically proven.Finally,the simulated results demonstrate the effectiveness of the proposed control system and show that the proposed controller is robust with regard to the uncertainty.(4)To deal with the problem of the PMSM with time delay,a dynamic recurrent neural network predictor-based control is developed.The uncertain nonlinear MLRS dynamics is linearized by introducing this network and the feedback linearization technique.Then a modified Smith predictor is utilized to compensate for the time delay based on such linearized system.The simulated results demonstrate the effectiveness of the proposed controller and show that it has a good robustness comparing with other controllers regarding variations of the time delay.(5)Considering the effect of the random error and the environmental noise on the accuracy of fiber optic gyro in the actual firing process of MLRS,a novel adaptive Kalman filtering algorithm is presented.A current statistical model of the launcher for the MLRS is established,and the adaptive tuning algorithms for the process and measurement noise covariance matrices are proposed on the basis of the standard Kalman filtering algorithm.The results from simulations and the vibration response test under the pulse excitation validate the effectiveness of the proposed filtering algorithm,which lays a foundation for the engineering practice of the launch dynamics control of MLRS.(6)Compared with the simulation results and the results of the previous experimental data,the effectiveness of the control methods for MLRS based on the simplified dynamics model is preliminarily verified.A research on the control method for launch dynamics of MLRS has been carried out and the problem of poor robustness of the control system has been solved in the dissertation.The obtained conclusions and results possess a certain reference value for the controller design of the MLRS in the processes of adjusting angle and salvo firing,and provide a potential theoretical method and tool to improve the firing accuracy of MLRS at a low cost.