Speed tracking control for electric multiple unit based on unmodeled dynamics compensation

In view of environment uncertain, complicated running mode and strong nonlinear characteristics, this paper describe EMU as integrating model composed of linear adaptive model and unmodeled dynamics compensation model. According to the traction/ brake characteristic curve and the actual running data of EMU, recursive least-squares method is used to develop linear adaptive model, adaptive neurofuzzy inference system (ANFIS) is adopted to build unmodeled dynamics compensation model. Adaptive control algorithm for running speed is proposed based on integrating model for EMU in order to ensure the high-precision speed tracking of EMU. Through the simulation study on CRH380A running data, the effectiveness of the proposed method is justified.