Levitation control scheme for the hybrid Maglev system based on neuron-PID control

The Maglev train can reach the speed of 500 km per hour, and it is one of the most promising conveyance tools, which has the advantages of high speed, low noise, high security and so on. The levitation control scheme is the key problem of the Maglev system, and the propulsion system has to be designed on the base of stable levitation. To lower the suspension power, permanent magnets are added into the levitation system to assistant the magnetic coil. Its air gap is enlarged and the energy consumed by the levitation subsystem is significantly reduced due to the permanent magnets. Novel scheme combined neuron control and the conventional PID control is proposed, which has good online learning ability. The adaptive neuron is used to regulate the parameters of the PID controller by self-learning and associative searching method according to the dynamic characteristics of the control plant. The suggested strategy is verified by experiments, which is model-free, and can realize fast, precise and flexible suspension. Its performance is better than that of state controller.