Self-organizing Neural adaptive tracking control of high speed trains subject to unexpected traction-breaking failures

Abstract

This paper develops an adaptive control scheme for position and velocity tracking control of high speed trains under uncertain system nonlinearities and actuator failures. Neural networks with self-organizing capabilities are integrated into control design, where the number of the neurons can be adjusted online automatically, so as not only to avoid the problem inherent in the NN with fixed structure but also to deal with system uncertainties containing of nonlinear in-train forces, traction-braking nonlinearities, as well as the unknown actuation faults. As such, the resultant control algorithms are able to achieve high precision train speed and position tracking under varying operation railway conditions, as validated by theoretical analysis and numerical simulations.