Global Model Updating And Sensitivity Analysis On The Mechatronic System Of The Maglev Elastic Supporting Structure

The maglev system is a kind of complex mechatronic system. Modeling method and modeling accuracy are concerned by science researchers and engineers all the time. In this thesis, with the purpose of study on the dynamic characteristics of the global mechatronic system which are coupled by the structure system, control system and electromagnetic system, co-simulation method, global model updating and global sensitivity analysis about the Mechatronic System of the Maglev Elastic Supporting Structure(MS-MESS) are carried out. The importance of the study is to promote the accuracy of modeling and co-simulation, and to push the development of modeling and co-simulation technology of mechatronic systems eventually.The main research contents of this thesis are as follows:First, study the co-modeling and co-simulation method of the maglev mechatronic system and summarize the process of this method, based on software of ADAMS and MATLAB/SIMULINK. The virtual modal test is carried out based on virtual prototype; Feasibility of this method is testified.Second, establishing a co-simulation model of the MS-MESS based on the hierarchy method of the model updating. During the modeling proses, mechanical structure system model is established by using model updating based on modal frequency from tier to tier. Due to the electromagnetic actuator machined error, installation error and etc. which makes the theory of electromagnetic force is different to the real value, the parameters of electromagnetic force are updated based on frequency response function(FRF) method. After the model calibration, the model’ simulated results are in good agreement with the experimental results.Three, the uncertainties of the model parameters in mechanical structure subsystem are studied based on the model validation and are estimated by using Kriging Response Surface Model(RSM). The reliability of the estimation is verified by experiment.Four, the global RSM of the MS-MESS is established, the input characteristics include eight parameters which are from structural system or control system. The output characteristics are the modal frequencies. Sobol’s sequence is found to have more advantages for the Global Sensitivity(GS) analysis by comparing with the Latin hypercube sampling method. Calculating the GS index and evaluating the first-order sensitivity indexes and the total sensitivity indexes. Then the uncertainty parameters are selected according to their sensitivity indexes and the response characteristic probabilities of the global mechatronic system are predicted.