Data-Driven Approach for Equipment Reliability Prediction Using Neural Network

Abstract:

BP neural network based data-driven method is proposed to predict reliability in this paper. The BP neural network prediction using Gradient Descent Method (GDM), Additional Momentum Gradient Descent Method (AMGDM) and Levenberg-Marquardt Method(L-M) based on numerical optimization theory of training algorithm are compared with different neuron number. The proposed approach is validated via age data collected from computer numerical control (CNC) machine tool in the field. The results from the proposed method show that perfect predicting performance is achieved under considering selecting suitable number of the hidden neurons and training algorithm. Remarks are outlined regarding the fact that BP neural network based on data-driven method is feasible, effective and adequate predicting accuracy can be obtained.

• Periodical: Advanced Materials Research (Volume 411)
• Pages: 563-566
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.411.563
• Online since: November 2011
• Authors: Feng Ding, Xing Ben Han
• Keywords: Error Minimization, Neural Network (NN), Prediction, Reliability

References

[1] Yi-Hui Liang. Combining neural networks and genetic algorithms for predicting the reliability of repairable systems. International Journal of Quality & Reliability Management. vol. 25, no. 2, p.201−210, 2008.

DOI: 10.1108/02656710810846943

[2] Feng Ding, Zhengjia He. Formalization of reliability model for assessment and prognosis using proactive monitoring mechanism. IEEE International Conference on Prognostics and Health Management Conference (PHM '2010). Macao, p.1−6, 2010.

DOI: 10.1109/phm.2010.5413419

[3] Aiwina Heng, Sheng Zhang, Andy C.C. Tan, et al. Rotating machinery prognostics: State of the art, challenges and opportunities. Mechanical Systems and Signal Processing. vol. 23, no. 5, p.724−739, 2009.

DOI: 10.1016/j.ymssp.2008.06.009

[4] S. Lolas, O.A. Olatunbosun. Prediction of vehicle reliability performance using artificial neural networks. Expert Systems with Applications. vol. 34, no. 5, pp.2360-2369, 2008.

DOI: 10.1016/j.eswa.2007.03.014

[5] M. A. Herzog, T. Marwala, P. S. Heyns. Machine and component residual life estimation through the application of neural networks. Reliability Engineering and System Safety. vol. 94, p.479–489, 2009.

DOI: 10.1016/j.ress.2008.05.008

[6] Zhigang Tian, Lorna Wong, and Nima Safaei. A neural network approach for remaining useful life prediction utilizing both failure and suspension histories. Mechanical Systems and Signal Processing. vol. 24, no.5, p.1542−1555, 2010.

DOI: 10.1016/j.ymssp.2009.11.005

[7] Gebraeel N. Z., Lawley M. A. A neural network degradation model for computing and updating residual life distributions. IEEE Transactions on Automation Science and Engineering. vol. 5, p.154−163, 2008.

DOI: 10.1109/tase.2007.910302

[8] Feng Ding, Zhengjia He, Yanyang Zi, et al. Application of support vector machine for equipment reliability forecasting. INDIN 6th IEEE International Conference on Industrial Informatics. Korea, p.526−530, 2008.

DOI: 10.1109/indin.2008.4618157

[9] Zhexue Ge, Zhiqiang Sun, Neural network theory and MATLAB R2007 implementation. Beijing: Publishing House of Electronics Industry, 2007.