Iterative Learning Control for Active Suspension System of a Railway Vehicle

The paper deals with the automatic control of active force for an active suspension system of railway vehicle. In the first part of the paper, the mathematical model of a simplified active suspension system is derived using the classical Newton laws. The model is represented in the form of a linear first order differential equations system. The control strategy adopted in this paper is Iterative Learning Control (ILC), a control approach intensively used for repetitive processes. In this paper, the repetitive behavior is represented by the body vehicle reaction to track irregularities. From many variants of ILC available in the literature, few simple algorithms were presented: the well-known PD - type algorithms. An heuristic design procedure is implemented for control of a linearized version of the active suspension system. The performances of the proposed control procedure are tested by numerical simulations. The results obtained from the performed simulations show the feasibility of the proposed methods in the case of step-type perturbations.