Research On Improving The Performance Of Electric Load Simulator

Load simulator is an important equipment of half hardware-in-loop simulation system. It is often used to simulate the aerodynamic torque load on the rudder of aircraft in the flight. The object of research is electric load simulator. In this dissertation, a new method for improving the system's performance of the load simulator was proposed.In this dissertation, the main technical problem, the basic operational principle and basic structure of the load simulator were analyzed. The system's mathematical model was established using the physical modeling method and it's accuracy was tested by experiments. Based on the system mathematical model,the electric load simulator's static performance and the principle of surplus torque was analyzed in detail,and the factors that cause the surplus torque were analyzed. Linear PID controller was introduced to expand the frequency band.Structure invariance principle was introduced to eliminate surplus torque. The simulation results shew that the controller can enable the rapidity and tracking accuracy. Because of the electric load simulator has the characters of non-linearity such as friction torque,the real practical effect of that control method was reduced. So the optimal control theory with neural network was applied in the control system in order to eliminate the disadvantages of traditional control method. The simulation results shew that the hybrid controller can improve the performance of the system more remarkably than the Linear PID controller. Surplus torque was eliminated effectively and the dynamic loading performance was improved remarkably.In this dissertation, the experimental rig of the electric load simulator was designed and the experiments of the dynamic loading performances based on the controller were done. In the experiment, the experiments results were got with the step signal as its input. The experiments'results show this controller was valid.