A connectionist model for lock coil rope fault prediction

In this paper, a neural network-based method of wire rope fault prediction in a system is proposed. This method is developed based on past observations on various wire rope parameters for lock coil rope, for example number of rope used in the system, period of test, number of faults, etc. To capture the data from various systems, with a view to improving the prediction accuracy, we have designed a multi-layer perceptron network (MLP) to realise better performance.