Fault-tolerant Control of Electromagnetic Levitation System Based on Diversified Basis Functions

In the context of continuous operation under strong magnetic fields, it is common for the electromagnet levitation system of maglev trains to lose part of their effectiveness. Additionally, the parameters of the maglev train may change over time, leading to modeling uncertainties. These challenges pose difficulties for parameter adaptation and fault tolerance in electromagnetic levitation control. To address these issues, a neuroadaptive fault-tolerant control algorithm based on a diversified basis function neural network is proposed. This method is designed to be robust to modeling uncertainties and capable of fault tolerance for electromagnet faults. Based on Lyapunov stability theory, it is proved that the proposed control method is uniformly ultimately bounded (UUB) and continuous. Both theoretical analysis and numerical simulation demonstrate the effectiveness of the proposed strategy.