Regression to fuzziness method for estimation of remaining useful life in power plant components

Abstract Mitigation of severe accidents in power plants requires the reliable operation of all systems and the on-time replacement of mechanical components. Therefore, the continuous surveillance of power systems is a crucial concern for the overall safety, cost control, and on-time maintenance of a power plant. In this paper a methodology called regression to fuzziness is presented that estimates the remaining useful life (RUL) of power plant components. The RUL is defined as the difference between the time that a measurement was taken and the estimated failure time of that component. The methodology aims to compensate for a potential lack of historical data by modeling an expert׳s operational experience and expertise applied to the system. It initially identifies critical degradation parameters and their associated value range. Once completed, the operator׳s experience is modeled through fuzzy sets which span the entire parameter range. This model is then synergistically used with linear regression and a component׳s failure point to estimate the RUL. The proposed methodology is tested on estimating the RUL of a turbine (the basic electrical generating component of a power plant) in three different cases. Results demonstrate the benefits of the methodology for components for which operational data is not readily available and emphasize the significance of the selection of fuzzy sets and the effect of knowledge representation on the predicted output. To verify the effectiveness of the methodology, it was benchmarked against the data-based simple linear regression model used for predictions which was shown to perform equal or worse than the presented methodology. Furthermore, methodology comparison highlighted the improvement in estimation offered by the adoption of appropriate of fuzzy sets for parameter representation.

[1]  Jeffrey W. Griffin,et al.  A Bayesian Prognostic Algorithm for Assessing Remaining Useful Life of Nuclear Power Components , 2010 .

[2]  xing-hui Zhang,et al.  Hidden Markov models in bearing fault diagnosis and prognosis , 2010, 2010 Second International Conference on Computational Intelligence and Natural Computing.

[3]  J. Wesley Hines,et al.  Current Computational Trends in Equipment Prognostics , 2008, Int. J. Comput. Intell. Syst..

[4]  Zhigang Tian,et al.  A neural network approach for remaining useful life prediction utilizing both failure and suspension histories , 2010 .

[5]  L. Bond,et al.  Prognostics and Condition-Based Maintenance: A New Approach to Precursive Metrics , 2004 .

[6]  N. Zerhouni,et al.  Estimation of the Remaining Useful Life by using Wavelet Packet Decomposition and HMMs , 2011, 2011 Aerospace Conference.

[7]  Nancy J. Lybeck,et al.  Prognostics and life beyond 60 years for nuclear power plants , 2011, 2011 IEEE Conference on Prognostics and Health Management.

[8]  Lifeng Xi,et al.  Residual life predictions for ball bearings based on self-organizing map and back propagation neural network methods , 2007 .

[9]  Michael H. Kutner.  Applied Linear Statistical Models , 1974 .

[10]  H. Kaebernick,et al.  Remaining life estimation of used components in consumer products: Life cycle data analysis by Weibull and artificial neural networks , 2007 .

[11]  Lefteri H. Tsoukalas,et al.  Fuzzy and neural approaches in engineering , 1997 .

[12]  Ying Peng,et al.  A prognosis method using age-dependent hidden semi-Markov model for equipment health prediction , 2011 .

[13]  Aitor Arnaiz,et al.  A Bayesian network model integrated in a prognostics and health management system for aircraft line maintenance , 2011 .

[14]  Piero P. Bonissone,et al.  An Instance-Based Method for Remaining Useful Life Estimation for Aircraft Engines , 2008 .

[15]  Lefteri H. Tsoukalas,et al.  Integration of Neural Networks with Fuzzy Reasoning for Measuring Operational Parameters in a Nuclear Reactor , 1993 .

[16]  Antonio Cesar Ferreira Guimarães.  A new methodology for the study of FAC phenomenon based on a fuzzy rule system , 2003 .

[17]  Nagi Gebraeel,et al.  Advanced Prognostics for Aircraft Electrical Power Systems , 2008 .

[18]  A. Jovanovic,et al.  Remaining life management systems: from stand-alone to corporate memory systems and Internet (ALIAS system of MPA Stuttgart) ☆ , 1999 .

[19]  Enrico Zio,et al.  A data-driven fuzzy approach for predicting the remaining useful life in dynamic failure scenarios of a nuclear system , 2010, Reliab. Eng. Syst. Saf..

[20]  Ki-Tae Kim,et al.  Remaining life prediction methods using operating data and knowledge on mechanisms , 1999 .

[21]  Kwang-Il Ahn,et al.  Development of a risk-informed accident diagnosis and prognosis system to support severe accident management , 2009 .

[22]  Steven R. Doctor,et al.  Damage Assessment Technologies for Prognostics and Proactive Management of Materials Degradation , 2009 .