Collaborative Control Study Of The Air-Magnetic Suspension Drive System For Fuel Cell Compressors

The fuel cell air compressor is a core power component in hydrogen-oxygen fuel cell systems.Traditional mechanical bearing supports and conventional motor drives are unable to meet the increasing demands of hydrogen energy vehicles.Active magnetic bearings are components that integrate rotation and suspension together.They offer advantages such as fast magnetic levitation response,no mechanical friction,high precision,long lifespan,and a small volume with a high power-to-volume ratio of permanent magnet synchronous motors.Aerodynamic foil bearings generate gas support forces using wedge-shaped structures and have advantages such as no mechanical friction,no external air supply system,and no lubrication system requirement.This study proposes a solution that combines active bearings and aerodynamic foil bearings to address the support and suspension issues in fuel cell compressors.It focuses on the control system issues arising from the new structure of the magnetically and aerodynamically combined supported compressor rotor.Firstly,a survey is conducted on the current support and suspension schemes for fuel cell compressors.A novel structure combining aerodynamic foil bearings and active magnetic bearings is proposed to meet the high-performance design requirements of fuel cell compressors.The gas bearings and magnetic bearings work synergistically,combining the advantages of both to compensate for their respective drawbacks when used independently.Size design and mathematical model derivation are carried out for the active magnetic bearings and aerodynamic foil bearings.Force analysis is performed on the compressor rotor,and the dynamic model of the rotor driven by the magnetically and aerodynamically combined support is derived,laying the foundation for the subsequent control system design.Next,in response to the control issues brought about by the new structure,this study intends to apply coordinated control to the magnetically and aerodynamically combined rotor system.First,the state space equations of the compressor’s magnetically and aerodynamically combined bearings are established.Then,macro-variables for coordinated control are designed,and the coordinated control laws are solved by convergence manifold through coordinated control.The operating conditions of the compressor are classified and discussed,considering various external disturbances.Coordinated control and PID controller simulation models are built in the Simulink/MATLAB software under the same simulation parameters.Simulations and comparative analyses are conducted for the startup and acceleration conditions.The simulation results show that coordinated control has advantages such as fast response,disturbance rejection,strong tracking,and robustness,with control performance superior to PID control.This verifies the possibility of applying coordinated control in the control system of magnetically and aerodynamically combined rotor systems.Furthermore,a parameter tuning method using the tree species algorithm is proposed for coordinated control.First,the tree species algorithm and the commonly used particle swarm algorithm are tested for function optimization.The test results indicate that the tree species algorithm outperforms the particle swarm algorithm in function solving,explaining the reason for selecting the tree species algorithm in this study.Then,tree species algorithm-coordinated algorithm and particle swarm-coordinated algorithm are built in MATLAB.The coordinated control parameters obtained from the two algorithms are assigned to the simulation model of the magnetically and aerodynamically combined bearings.Finally,the simulation results of the coordinated control parameters obtained through empirical methods,tree species algorithm,and particle swarm algorithm are compared and analyzed.The tree species algorithm achieves the best comprehensive control performance among the five controllers,while the empirical method performs the worst.This demonstrates the feasibility of using the tree species algorithm for parameter tuning in coordinated control.Lastly,a rotor control system experimental setup for the magnetically and aerodynamically combined supported rotor is designed and constructed to validate the findings.First,the overall structure of the experimental setup is designed,followed by the development of experimental software and user interface.Next,discrete models for PID control and coordinated control are derived,and the algorithm flows for PID control,coordinated control,and tree species-coordinated control are designed.Finally,under the same experimental conditions,a comparative experiment is conducted to evaluate the suspension,rotation,and disturbance rejection capabilities of the compressor rotor.The results confirm the findings from the simulations,demonstrating the feasibility of applying coordinated control theory in magnetically and aerodynamically combined control systems.Furthermore,the rotor under coordinated control shows superior stability in terms of rotation speed,static performance,and disturbance rejection compared to PID control.The tree species algorithm provides more accurate coordinated control parameters than the empirical method,resulting in better control performance in the experiments.