Shimmy Damping Control Research Of Aircraft's Nose Wheel Based On Magnetorheological Shimmy Damper

The nose-wheel shimmy of the aircraft often occurs at the running stages of take-off and landing.This kind of vibration is harmful and self-excited,and it affects the ride comfort and maneuverability of the aircraft seriously.According to the magnitude of current which is input by the semi-active control system,based on the actual situation of the aircraft's nose-wheel shimmy,Magnetorheological(MR)shimmy damper can output adjustable damping force and control the shimmy in real time at last.During the process,the key problem is how to get the accurate the magnitude of current.In view of this problem,this thesis is mainly carried on the following work.Firstly,the mechanism of nose-wheel shimmy and the factors affecting it are studied.Based on the linear hypothesis,a 3-degree-of-freedom mathematical model of the aircraft's nose-wheel shimmy is established.Determining the structural parameters of the nose landing gear,the critical sliding speed of the aircraft is obtained by using the eigenvalue analysis method.Under the critical sliding speed,the influences what the initial value of each influence factor has on the swing amplitude and frequency are analyzed.Then,the positive and inverse model of the MR shimmy damper are obtained by identification.Based on the experimental data of the damping characteristics of the self-designed MR shimmy damper,the parameters of the modified Bouc-Wen model are identified and the model is verified by simulation.Based the positive model of the MR shimmy damper and the radial basis neural network,the inverse model of MR shimmy damper is constructed,and the generated random sample data are used to train and verify the inverse model.Finally,the effectiveness of forward and inverse model of the MR shimmy damper is verified by simulation.Both of them are added to the front wheel shimmy model of the aircraft,and the nose-wheel shimmy is simulated at different running speeds.The without control,passive control and neural network inverse model control are compared at different running speeds.The results show that the neural networkinverse model can output the magnitude of current accurately and has the best shimmy damping effect.