The application of Matlab to railway signalling system fault modelling

The paper presents a review of Matlab/Simulink-based techniques used in the development of fault models of signalling railway systems. The data used to develop the models was derived, both in normal and fault modes, from operational railway monitoring equipment using four data signals to characterise the electrical and mechanical behaviour of a points mechanism. The main objective of the work was to regenerate three of the data signals from a single monitored signal, thus reducing the number of sensors required to identify and predict a system failure. Matlab/Simulink based static curve fitting and dynamic system identification techniques were applied to regenerate the data, and in each case the robustness and accuracy of the fault model are discussed. Finally, the signal regeneration is demonstrated using a Simulink model combining dynamic and linear models