Failure and Reliability Predictions by Infinite Impulse Response Locally Recurrent Neural Networks

In this paper, Infinite Impulse Response Locally Recurrent Neural Networks (IIR-LRNNs) are employed for forecasting failures and predicting the reliability of engineered components and systems. To theauthors' knowledge, it is the first time that such dynamic modelling technique is used in reliabilityprediction tasks. The method is compared to the radial basis function (RBF), the traditional multilayerperceptron (MLP) model (i.e., the traditional Artificial Neural Network model) and the Box-Jenkinsautoregressive-integrated-moving average (ARIMA). The comparison, made on case studiesconcerning engine systems, shows the superiority of the IIR-LRNN with respect to both the RBF andthe ARIMA models, whereas a similar performance is obtained by the MLP.

[1]  David Zhang,et al.  On the neural network approach in software reliability modeling , 2001, J. Syst. Softw..

[2]  Thong Ngee Goh,et al.  A study of the connectionist models for software reliability prediction , 2003 .

[3]  Enrico Zio,et al.  Simulating the dynamics of the neutron flux in a nuclear reactor by locally recurrent neural networks , 2007 .

[4]  Liang Tian,et al.  Evolutionary neural network modeling for software cumulative failure time prediction , 2005, Reliab. Eng. Syst. Saf..

[5]  Timothy Masters,et al.  Multilayer Feedforward Networks , 1993 .

[6]  Loon Ching Tang,et al.  Application of neural networks in forecasting engine systems reliability , 2003, Appl. Soft Comput..

[7]  Thong Ngee Goh,et al.  A comparative study of neural network and Box-Jenkins ARIMA modeling in time series prediction , 2002 .

[8]  Way Kuo,et al.  An exploratory study of a neural network approach for reliability data analysis , 1995 .

[9]  Robert S. Swarz,et al.  The theory and practice of reliable system design , 1982 .

[10]  Ah Chung Tsoi,et al.  FIR and IIR Synapses, a New Neural Network Architecture for Time Series Modeling , 1991, Neural Computation.

[11]  Chao Ton Su,et al.  Combining time series and neural network approaches for modeling reliability growth , 1997 .

[12]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[13]  Liang Tian,et al.  On-line prediction of software reliability using an evolutionary connectionist model , 2005, J. Syst. Softw..

[14]  Mehdi Ehsan,et al.  Evaluation of power systems reliability by an artificial neural network , 1999 .

[15]  Gregory Levitin,et al.  Robust recurrent neural network modeling for software fault detection and correction prediction , 2007, Reliab. Eng. Syst. Saf..

[16]  Bhaskar D. Rao,et al.  On-line learning algorithms for locally recurrent neural networks , 1999, IEEE Trans. Neural Networks.

[17]  Liang Tian,et al.  Reply to "Comments on 'Evolutionary neural network modeling for software cumulative failure time prediction' by Liang Tian and Afzel Noore, " by Bev Littlewood[Reliability Engineering and System Safety 87 (2005) 45-51] , 2006, Reliab. Eng. Syst. Saf..