Study On Model Predictive Control Of Maglev Multi-electromagnet Suspension System

Maglev can satisfy people's requirements for high-quality travel with its high speed,comfort,environment protection and energy saving.Suspension control is the core of the maglev train.At present,domestic and overseas scholars use some advanced control algorithms for maglev suspension control,but the study using model predictive control(MPC)is still relatively lacking.MPC is a very mature control method which has been successfully applied in many industrial fields.Especially,it has great advantages at handling objective optimal control with constraints.At present,most of the studies on suspension control take the single electromagnet as a study object,but the states coupling between the two ends of the suspension module cannot be prevented by carrying out the single-electromagnet suspension control scheme,to a large extent,which will also affects the overall performance of the suspension system.Therefore,this thesis mainly studies maglev multi-electromagnet suspension decoupling control scheme through taking the electromagnetic-suspension maglev suspension module as a study object.On this basis,the decoupled model is studied by MPC to improve the suspension control performance.The concrete study contents are as follows:Firstly,the nonlinear model of multi-electromagnet suspension based on voltage control is described based on modeling and analysis of the maglev suspension module.The stability of the model is analyzed by using the nonlinear control theory and MATLAB simulation tool.Meanwhile,the coupling between the two ends of the suspension module is analyzed with the established nonlinear model.Secondly,the feasibility of decoupling control of multi-electromagnet suspension nonlinear model is analyzed.On this basis,the maglev multi-electromagnet suspension decoupling model is set up by using the feedback linearization decoupling control method based on differential geometry,and this decoupling ability is analyzed through MATLAB simulation.The traditional PID control method is used to control the decoupled system,and the control performance is simulated and analyzed.Thirdly,based on the theory of predictive control,the maglev predictive model for multi-electromagnet suspension decoupling system is established through adding disturbance variables,and the unconstrained rolling optimization is carried out.In order to further improve the control performance,the constrained MPC controller for the suspension system is designed,considering the constraints of suspension gap,velocity and acceleration.According to the requirements of real-time control,the constrained optimization algorithm based on the path-following method is studied.At last,the effectiveness of the control algorithm proposed in this thesis is verified by simulating the two cases with disturbances and without disturbances through MATLAB simulation.The simulation results indicate that the constrained suspension decoupling predictive controller based on the path-following method is feasible and reasonable.Compared with the PID control algorithm,it is verified that the constrained predictive control algorithm can make the suspension system have faster convergence and higher stability.