CONDITION MONITORING USING EMPIRICAL MODELS: TECHNICAL REVIEW AND PROSPECTS FOR NUCLEAR APPLICATIONS

The purpose of this paper is to extensively review the condition monitoring (CM) techniques using empirical models in an effort to reduce or eliminate unexpected downtimes in general industry, and to illustrate the feasibility of applying them to the nuclear industry. CM provides on-time warnings of system states to enable the optimal scheduling of maintenance and, ultimately, plant uptime is maximized. Currently, most maintenance processes tend to be either reactive, or part of scheduled, or preventive maintenance. Such maintenance is being increasingly reported as a poor practice for two reasons: first, the component does not necessarily require maintenance, thus the maintenance cost is wasted, and secondly, failure catalysts are introduced into properly working components, which is worse. This paper first summarizes the technical aspects of CM including state estimation and state monitoring. The mathematical background of CM is mature enough even for commercial use in the nuclear industry. Considering the current computational capabilities of CM, its application is not limited by technical difficulties, but by a lack of desire on the part of industry to implement it. For practical applications in the nuclear industry, it may be more important to clarify and quantify the negative impact of unexpected outcomes or failures in CM than it is to investigate its advantages. In other words, while issues regarding accuracy have been targeted to date, the concerns regarding robustness should now be concentrated on. Standardizing the anticipated failures and the possibly harsh operating conditions, and then evaluating the impact of the proposed CM under those conditions may be necessary. In order to make the CM techniques practical for the nuclear industry in the future, it is recommended that a prototype CM system be applied to a secondary system in which most of the components are non-safety grade. Recently, many activities to enhance the safety and efficiency of the secondary system have been encouraged. With the application of CM to nuclear power plants, it is expected to increase profit while addressing safety and economic issues.

[1]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[2]  K. Gross,et al.  Sequential probability ratio test for nuclear plant component surveillance , 1991 .

[3]  Soon Heung Chang,et al.  Thermal power estimation by fouling phenomena compensation using wavelet and principal component analysis , 2000 .

[4]  Man Gyun Na,et al.  ESTIMATION OF BREAK LOCATION AND SIZE FOR LOSS OF COOLANT ACCIDENTS USING NEURAL NETWORKS , 2004 .

[5]  W. Härdle.  Nonparametric and Semiparametric Models , 2004 .

[6]  Jie Lin,et al.  Nuclear power plant transient diagnostics using artificial neural networks that allow 'don't-know' classifications , 1995 .

[7]  K. Gross,et al.  Sensor fault detection in nuclear power plants using multivariate state estimation technique and support vector machines. , 2000 .

[8]  Thomas Parisini,et al.  Keynote paper: Fault diagnosis and neural networks: A power plant application , 1995 .

[9]  E. Davis,et al.  Lessons learned from the U.S. nuclear power Plant on-line monitoring programs , 2005 .

[10]  Jiro Wakabayashi,et al.  Application of adaptive Kalman filtering technique for the diagnostic system of nuclear power plants , 1980, 1980 19th IEEE Conference on Decision and Control including the Symposium on Adaptive Processes.

[11]  Imre Pázsit,et al.  Sequential tests for failure detection , 1990 .

[12]  Theodora Kourti,et al.  Multivariate SPC Methods for Process and Product Monitoring , 1996 .

[13]  Seong Soo Choi,et al.  Thermal Performance Test through On-line Turbine Cycle Performance Monitoring in Nuclear Power Plants , 1999 .

[14]  Robert E. Uhrig,et al.  COMPUTATIONAL INTELLIGENCE IN NUCLEAR ENGINEERING , 2005 .

[15]  Sang Ryul In,et al.  TOKAMAK REACTOR SYSTEM ANALYSIS CODE FOR THE CONCEPTUAL DEVELOPMENT OF DEMO REACTOR , 2008 .

[16]  Manfred Morari,et al.  Level control in the steam generator of a nuclear power plant , 2000, IEEE Trans. Control. Syst. Technol..

[17]  R. W. King,et al.  Pattern-recognition software for plant surveillance , 1987 .

[18]  Daniel Hodouin,et al.  Use of a novel autoassociative neural network for nonlinear steady-state data reconciliation , 1997 .

[19]  Hamparsum Bozdogan,et al.  Statistical Data Mining and Knowledge Discovery , 2004 .

[20]  M. Kramer.  Nonlinear principal component analysis using autoassociative neural networks , 1991 .

[21]  Jacek M. Zurada,et al.  Introduction to artificial neural systems , 1992 .

[22]  P. van der Laan,et al.  Statistical quality management , 1992 .

[23]  Soon Heung Chang,et al.  Application of neural networks to a connectionist expert system for transient identification in nuclear power plants , 1993 .

[24]  Soon Heung Chang,et al.  Development of Nuclear Power Plant Online Monitoring System Using Statistical Quality Control , 2006 .

[25]  Man Gyun Na.  A Neuro-Fuzzy Inference System for Sensor Failure Detection Using Wavelet Denoising, PCA and SPRT , 2001 .

[26]  J. Wesley Hines,et al.  MSET PERFORMANCE OPTIMIZATION THROUGH REGULARIZATION , 2005 .

[27]  Carlo Vitanza.  Overview of the OECD–Halden reactor project☆ , 2001 .

[28]  Lindsey Jordan Bignell,et al.  Analysis of adhesive tape activation during reactor flux measurements , 2008 .

[29]  Belle R. Upadhyaya,et al.  Application of Neural Networks for Sensor Validation and Plant Monitoring , 1990 .

[30]  Attila Rácz,et al.  Detection of small leakages by a combination of dedicated Kalman filters and an extended version of the binary sequential probability ratio test , 1993 .

[31]  Paolo F. Fantoni,et al.  Experiences and applications of PEANO for online monitoring in power plants , 2005 .

[32]  J. Wesley Hines,et al.  Process and equipment monitoring methodologies applied to sensor calibration monitoring , 2007, Qual. Reliab. Eng. Int..

[33]  D. Simon.  Optimal State Estimation: Kalman, H Infinity, and Nonlinear Approaches , 2006 .

[34]  P. F. Fantoni,et al.  A pattern recognition-artificial neural networks based model for signal validation in nuclear power plants , 1996 .

[35]  Ali Cinar,et al.  Statistical process monitoring and disturbance diagnosis in multivariable continuous processes , 1996 .

[36]  Paolo F. Fantoni,et al.  Multiple-Failure Signal Validation in Nuclear Power Plants Using Artificial Neural Networks , 1996 .

[37]  David J. Kukulka.  An Evaluation of Heat Transfer Surface Materials Used in Fouling Applications , 2005 .