Research On Intelligent Control Of Weapon Balance And Orientation Electro - Hydraulic Servo System

In modern wars, the requirements for weapon barrel’s control speed, control accuracy and anti-jamming capability continue to increase. At the same time, there are many non-linear disturbances in the process of a weapon barrel’s balancing and positioning. Therefore this paper studies the electro-hydraulic servo system for a weapon’s balancing and positioning to design an intelligent control strategy with fast response, high control precision and good robustness to achieve the nonlinear weapon barrel’s precisely controlling.The main work includes the following aspects:Firstly, it introduces the background, the structure, the composition and the working principle of the electro-hydraulic servo system for a weapons barrel’s balancing and positioning. And study current control algorithm commonly used in weapon barrel’s control system. Analyze the structure and principle of the electro-hydraulic servo system’s hydraulic system, and make the system’s mathematical modeling, and obtain the transfer function. And analyze the system’s nonlinear factors to lay the foundation for future study.Secondly, establish the system identification model based on offline identification. Linear model can not accurately describe the complex nonlinear system. So design two identification models including a BP neural network system identification model and a GA-BP neural network system identification model. After off-line identification training and testing, get two accurate system identification models. Analyze and compare the advantages and disadvantages of these two identification models to select the model with a better control display to be the system’s identification model. The selected identification model will be used for controller’s simulation subsequently, offering neural network the gradient information to adjust the controller’s parameters online.Thirdly, design and study two control strategies for the system. And Select the better one for the experiment. Take the variable structure control as the controller’s designing basis. And use fractional theory and RBF neural network in variable structure control. Then design a neural network variable structure controller and a fractional neural network variable structure controller. Establish systems’simulation models and do the two controllers’simulation research with the system identification model by MATLAB. Analyze the simulation results to determine a controller with a better control performance for the subsequent experimental study.Finally, do experimental research with the selected control strategy and evaluate the controller’s control performance. Introduce the physical model of the system and carry out experiments with the determined controller regarding experimental performance indicators as a guide. According to the experimental results, analyze the completion of the experimental index and the controller’s effectiveness and obtain the experimental results.