Research And Design Of Servo Loading System Based On DSP+FPGA

The reliability,accuracy and flexibility of modern information war on the rocket gun control device are put forward higher requirements.In the process of development of gun control device to control the core of the servo system performance test is an important means to improve the overall performance of the gun control device.As the rocket launcher and shooting gun control system load end of the complex and volatile,resulting in excess torque on the performance of gun control devices have a greater impact.In this paper,based on the simulation of the loading environment of the gun control device of a certain type of rocket launcher,and the reason of the superfluous torque of the gun control device.The dynamic servo performance of the servo system is tested and evaluated before the assembly of the weapon,which can effectively shorten the development and production cycle of the gun control device.The main research work of this paper is focused on the following aspects:(1)The structure and working principle of electric servo loading system are introduced,the model of permanent magnet synchronous motor is established by vector control principle,the mathematical model of torque sensor,rotary inertia disk and integral system is deduced.The superfluous torque and the causes of uncertainties and nonlinear factors are analyzed in detail,which lays a theoretical foundation for the selection of system identification strategy,the design of servo controller and the semi-physical simulation.(2)Because of the complex non-linearity of the servo loading system,it is difficult to establish the precise mathematical model of the system.Therefore,the neural network identification strategy of off-line training and on-line adjustment is adopted.In order to guarantee the convergence rate of the neural network and avoid the oscillation of the system,the gray model prediction is used to optimize the RBF neural network.Based on the analysis of the accurate model based feed-forward compensation controller,an adaptive PID controller based on gray model prediction and RBF neural network is designed.In order to verify the validity and practicability of the proposed identification algorithm for restraining the superfluous torque of the electric servo loading system.(3)As the core device for gun control servo controller and the basic realization form in contrast to the processor’s architecture,determine the overall scheme of servo controller based on DSP-TMS320F28335 and FPGA-XC4VLX15 as the main body.According to the respective program processing mechanism of DSP and FPGA,the software scheme of servo controller is established,and several anti-jamming measures are described based on the combat environment of gun control device.(4)This paper build a rocket gun control device of electric servo loading system semi-physical simulation platform,and the proposed control strategy and the design of the servo controller on the platform to verify the experimental results satisfy the type of gun control device for servo system performance requirements,guidance and reference to the actual engineering application.