| In the field of flight,the steering gear is an important part of the aircraft.The load simulator is a device for simulating the aerodynamic load received by the aircraft during the actual flight,and used to check whether the control performance of the aircraft meets the standard.Since the load simulator is a passive servo control system,the active motion of the servo will cause a force error in the system,The existence of force error will seriously affect the dynamic loading accuracy of the system.Therefore,it is of great research significance to study the control method suitable for the system and suppress the system's force error in the field of flight.This article takes electric load simulator as the research object.In order to solve the key problem of system force error and satisfy the control accuracy of the loading system,a strategy based on PID control of improved BP neural network is proposed to realize dynamic and high-precision control of the system.The research content of this article is as follows:(1)System design and mathematical modeling.Combined with the domestic and foreign research status of the load simulator,the advantages and disadvantages of the three loading methods of electric,electro-hydraulic and mechanical are compared and analyzed.Because the helicopter steering gear has the characteristics of high loading accuracy and low loading force,determined the scheme of electric servo loading.And analyzed the system composition and working principle,Established mathematical models for the loading mechanism,loading mechanism and sensors of the system.(2)Research on compound method based on displacement feedforward compensation.First,the characteristics of the system are studied.In order to improve the control accuracy of the loading system,a 4-closed-loop control scheme based on traditional PID control is proposed;Then the mechanism and characteristics of system force error are analyzed.Because the feedforward compensation algorithm can compensate for the interference term in advance to achieve the effect of improving the control accuracy of the system.Therefore,for the suppression of system force errors,a feedforward compensation algorithm based on the principle of structural invariance is proposed based on multi-closed-loop control.And the Matlab/Simuink software was used to simulate the above control scheme,and the suppression rate of the force error of the scheme reached 91.4%.Then further analyze the influence of the nonlinear factors of the loading system on the system.(3)Research on BP-PID control method based on nonlinear prediction model.Through simulation analysis,the classic control theory can no longer solve the influence of nonlinear factors on the system.Therefore,according to the BP neural network,it has good self-learning and nonlinear approximation capabilities.Based on the principle of feedforward compensation,a composite control method based on BP neural network is proposed.The BP neural network is used to adjust the PID parameters online,thereby improving the loading accuracy of the system.To further reduce the impact of system parameter perturbation on loading accuracy.It is proposed to use the nonlinear prediction model to improve the weight calculation formula of BP neural network.Simulation results show that the method can effectively suppress the force error of the system and improve the dynamic loading accuracy of the system.Its dynamic loading accuracy is 99.2%,which meets the technical requirements of the steering gear device.(4)System platform construction and experimental verification.According to the technical indicators,the components of the system have been selected,and a three-channel load simulator platform has been built.Using Labview to complete the software design of the system,to achieve the control function of the host computer of the load simulator system.And on this platform,three kinds of loading experiments were carried out.The experimental results show that the load simulator has good control performance. |
PDF Full Text Request