ADRC-based Control Technology, A Tank Gun Control System Applications

Currently, the performance of our tank gun control system still lags behind that of those developed countries'. Therefore, it is important to research on the performance of tank gun control system. Based on the development of a new type of tank gun, this paper focuses on the research of the identification and control strategy of its gun control system.Firstly, the overall design of tank gun is introduced:permanent magnet synchronous motor are used at horizontal and vertical direction, and the all-electric position servo system is taken. Then, the design of semi-physical simulation test bench is introduced. On the basis of this, the input-output data of the system is proposed.Then, methods to construct model of the gun control system are detailed. And back-propagation neural network (BP) and genetic algorithm have been studied. In order to overcome the limitation of BP algorithm, such as easy converging to a local minimum point and slowly training velocity, one method for optimizing BP algorithm is proposed. Considering the structure and parameters optimization of BP algorithm, so another method which based on hierarchical genetic algorithm (HGA) is also proposed. This algorithm has much better performance. The experiment results prove the validity of this method.Finally, the auto-disturbances rejection controller (ADRC) combining the genetic algorithm is applied in the gun control system. ADRC is able to suppress the vibration by the robustness of itself and optimize the controller's parameters by the ability of global optimization of genetic algorithm. In this paper, the principle of ADRC is proposed, and a set of simulation of PID and ADRC are carried out. The results show that ADRC has excellent effects in controlling typical input, and can fulfill the dynamic and static capability of the system, and achieves precise control of gun control system.