Research On The Application Of Parallel Control In A Large - Caliber Rocket Weapon

Large-caliber rocket has great lethality, high accuracy, long range and the price is cheaper than the missile, so it has a wide range of equipment in the national armies. Rocket servo control system constitutes the core of the weapon system, so the design of high precision and rapid response servo control system is the key to ensure the performance of the rocket weapon. Because of the complex battlefield environment, there are many nonlinear and uncertain factors that make the servo system become a nonlinear time-varying system, which increases the difficulty of control.This paper makes research on model identification and control strategies of a rocket AC servo system. It includes the following aspects:A semi-physical simulation platform is built to simulate actual working conditions of a rocket AC servo system. The working principle and main components of the platform are introduced. The PC control interface, hardware circuit of slave computer which is based on TMS320F28335 and servo amplifier circuit are designed.The mathematical model of PMSM and its vector control strategy are introduced. The three loop transfer function of AC servo system is deduced.Using the pseudo random multi-amplitude signal as the input data of the semi-physical simulation system and collecting the output data, then the system is identified by the standard particle swarm optimization (PSO) and the improved particle swarm optimization (IPSO) algorithm of RBF neutral network (RBFNN). At last compare and analyses the two identification methods.According to the nonlinear characteristics of AC servo system of rocket launcher, first a RBF neural network self-tuning PID controller is designed. And then a parallel compound controller is designed, which makes IPSO-RBFNN as the identifier and fuzzy-single neuron PID as the main controller. The performance of the two kinds of controller is verified on the MATLAB/Simulink software platform and the semi-physical simulation platform. The results show that parallel compound control strategy has faster response, better stability, higher accuracy and it is able to meet the requirements of the rocket AC servo system.