Research On The Follow-up Simulation Loading Control System Based On DSP+FPGA

The servo simulation loading system is one of the main equipments for semi-physical simulation.It can simulate the load changes of the servo system under laboratory conditions and test its performance,which effectively reduces the cost and period.Due to the complex nonlinearity and uncertainty of the loading system and the active motion of the loaded object,its loading performance is seriously affected by surplus torque.For the problems of servo simulation loading system,the main research work of this paper is as follows:This paper introduces the structure and principle of the electric simulation load system,and then establishes the mathematical model of the load simulation system.then the surplus torque and the uncertainty factors of electric load simulation system are analyzed,which establishes the foundation for system identification and control strategy research.The neural network identification strategy of off-line training and on-line adjustment is adopted.The system is identified by the Radical Basis Function(RBF)neural network,then considering the problems of parameters of RBF network,the improved genetic algorithm is used to optimize parameters of the RBF neural network.The optimized parameters obtained from off-line training are taken as the initial values of online identifiers,which accelerates the convergence speed of neural networks.In order to restrain the extra torque of the loading system,the RBF neural network sliding mode controller based on the identification is designed.RBF Neural Network is used to dynamically adjust the switching gain of Sliding Mode Control,so as to weak chattering phenomenon of sliding mode control and obtain strong robustness.In addition,the online identifier provides gradient information to adjust gain for RBF neural network,so that the learning speed of RBF neural network can be carried out correctly.According to the demand of electric simulation loading system,the overall hardware scheme of servo controller is designed based on DSP and FPGA,and the corresponding software scheme of servo controller is designed.Finally,The semi-physical simulation experiment platform is built.On this basis,it is used to verify the effectiveness of the control strategy and servo controller designed in this paper.The experimental results show that the control scheme designed in this paper can meet the performance index of the gun control system and the control accuracy is guaranteed.