Research On Electro-hydraulic Servo Control Of Projectile Coordination Arm

As an essential component of the automatic loading system,the rotation speed and the position accuracy of the projectile transfer arm would rather affect the reliability of the loading system,the firing speed and firing accuracy of the artillery.However,many factors,such as the non-linear changing of the system parameters and inherent characteristics of the hydraulic system,would have a complicated influence on the position control of the projectile transfer arm during the operation process.Therefore,the research on the precise position control of the projectile transfer arm has practical engineering significance.The thesis mainly includes the following aspects:(1)Not only the basic composition of the projectile transfer arm,but also its' work procedure was introduced at length.And then,the accuracy requirements of the projectile transfer arm,which should be ensured during the work process,were emphasized.Combining the mechanical system equation and the electro-hydraulic system equation of the projectile transfer arm,the dynamic equation of the controlled system was deduced.Moreover,the complexity of the controlled system and the design difficulty of the controller are clarified.(2)The AMESim software were used to build a virtual prototype model of the projectile transfer arm.In order to improve the closeness of the model to the actual system,some pivotal parameters of the model were optimized through the Isight software.Then,a joint simulation platform of MATLAB/Simulink and AMESim was built to trace the planned motion trajectory of the projectile transfer arm,which was controlled by PID arithmetic.(3)Combined with the system dynamics equations,an adaptive sliding mode control method was proposed,when the traditional PID control could not meet the control requirements.Thus,the dynamic error could be effectively reduced and the accuracy of the position precision was ensured.Based on the above research,the RBF neural network algorithm was introduced to approximate the unknown parameters of the system,which effectively improved the robustness of the controller.Furthermore,in order to improve the approximation accuracy of the unknown term by the RBF neural network algorithm,the differential evolution algorithm was used to optimize the network parameters.The simulation results show that the parameter optimization is effective.(4)An experimental platform for the projectile coordination arms was built,and the optimized control algorithm was applied to the actual system.Experimental results show that the optimized RBF neural network adaptive sliding mode controller can guarantee response speed and position accuracy,which meets control requirements with its great robustness.