Neural sliding mode control on suspension gap for single electromagnetic guiding actuator of linear elevator

The maglev guiding technology is applied to linear elevator system. Consider the system parameter variation and extra disturbance for maglev guiding actuator system influence. A neural network based adaptive sliding mode control method is proposed to control suspension altitude for maglev guiding system of linear elevator. Adopting RBF neural network and utilizing its learning function to compensate uncertain parameters of the single electromagnetic suspension device of linear lift adaptively could replace the switching part of conventional sliding mode control and eliminate the chattering phenomenon with high-frequency. The proportional and differential controller is designed as one parallel control part, which improves the convergence of neural network, and enhances system stability. The stability of the system was proved by lyapunov theory. Matlab Simulation results show that the proposed control scheme shows good tracking performance and strong robustness.