Predicting reliability and failures of engine systems by single multiplicative neuron model with iterated nonlinear filters

Failures and reliability prediction in engine systems have attracted much attention over the past decades. Soft computing techniques have been widely studied and successfully applied to engine system reliability prediction with the ability to deal with high non-linearity. However, the development of conventional soft computing techniques involves many difficulties such as the selection of inputs to the network, the selection of network structure and the calculation of model parameters. The single multiplicative neuron (SMN) model is found to be a viable alternative due to its advantages of better approximation capabilities, simpler network structures and faster learning algorithms. Furthermore, nonlinear filters can deal with additive noises and can update model parameters when a new observation data arrives due to their iterative algorithm structure. In this study, novel techniques based on the SMN model with iterated nonlinear filtering online training algorithms for engine systems reliability prediction are proposed. Illustrative examples taken from the previous literature are used to demonstrate the detailed implementation procedures and the corresponding results are compared with the existing developed models. The experimental results reveal that the proposed models result in better prediction results than the existing methods.