The slot effect is a special problem for gap sensor of high-speed maglev train.Aiming at the problem of slot error caused by this

a method is proposed

arrange coils in probe of sensor to detect slot position

use radial basis function（RBF） neural network to set up inverse model of slot effect and compensate the output of senor according to position signal.Simulation result show that the maximum error of compensator output is ±0.18 mm at range of 2

2

0 mm

and can effectively eliminate slot effect

precision of the sensor is increased and meet the requirement of levitation control system of high speed maglev.