| The embedded magnetic levitation balance can not only realize the aerodynamic test of the aircraft model in the wind tunnel test,but also realize the free rock in the wind tunnel test,which better simulates the flight in the real air.As the core part of the embedded magnetic levitation balance,the axial suspension system undertakes the important task of the stable suspension.Its performance monitoring and optimization control have a very important impact on improving the reliability of the system.Based on the embedded magnetic levitation balance,this article carried out the modeling and control,performance monitoring and evaluation,online optimization framework and implementation of the axial suspension system.Finally,the effectiveness of the scheme and algorithm in the single suspension system is verified in the experimental platform.The main contents of the article are as follows:(1)The structure of the embedded magnetic levitation balance and its radial and axial suspension system is proposed after analyzing the functional requirements of the embedded magnetic levitation balance.After establishing a six-degree-of-freedom magnetic levitation model and analyzing the coupling of radial and axial control,the dynamic model of the axial suspension system is deduced,and the nominal controller of the axial levitation system is designed.(2)The residual generator is designed and the residual signal is analyzed to meet the needs of axial suspension system performance monitoring.Based on the physical model and the Luenberger observer,the applicability of the residual generator in the performance monitoring of the axial suspension system is studied.In order to realize the performance evaluation of the system,the implementation of the residual generator based on the data is analyzed,and the evaluation standard based on the residual is established to evaluate the operation of the system in real time.The applicability of the residual generator based on lowest missed detection rate in the suspension system is studied,and the effectiveness of this residual generator is verified by the operation data of the suspension system in maglev train.(3)Aiming at the performance optimization of the axial suspension system,the feasibility analysis of the residual-driven performance optimization framework is studied.Combined with the formula of parameter update based on the gradient descent method,the algorithm of parameter update is summarized.The suspension model is built in Simulink,and the framework and parameter update algorithm are verified through simulation.(4)The hardware design and software design of the single suspension experimental platform is realized.The effectiveness of the data-based residual generator,the residualbased performance monitoring scheme and the residual-driven control performance optimization framework in the single suspension system are verified. |
