A digital filter-based approach to the remote condition monitoring of railway turnouts

Railway operations in Europe have changed dramatically since the early 1990s, partly as a result of new European Union Directives. Performance targets have become more and more exacting, due to reductions in state support for railways and the need to increasing traffic. More intensive operations also place greater demands on the hardware of the railway. This is true for both rolling stock and infrastructure subsystems and components, particularly so in the case of the latter where the time available for maintenance is being reduced. The authors of this paper focus on the railway infrastructure, and more specifically on points. These are critical elements whose reliability is key to the operation of the whole system. Using intelligent monitoring systems, it is possible to predict problems and enable quick recovery before component failures disrupt operations. The authors have studied the application of remote condition monitoring to point mechanisms and their operation, and have identified algorithms which may be used to identify incipient failures. In this paper, the authors propose a Kalman filter for the linear discrete data filtering problem encountered when using current sensor data in a point condition monitoring system. The reason for applying Kalman filtering in this study was to increase the reliability of the model presented to the rule-based decision mechanism