PCA-based ANN approach to leak classification in the main pipes of VVER-1000
-
K. Hadad
Abstract
This paper presents a neural network based fault diagnosing approach which allows dynamic crack and leaks fault identification. The method utilizes the Principal Component Analysis (PCA) technique to reduce the problem dimension. Such a dimension reduction approach leads to faster diagnosing and allows a better graphic presentation of the results. To show the effectiveness of the proposed approach, two methodologies are used to train the neural network (NN). At first, a training matrix composed of 14 variables is used to train a Multilayer Perceptron neural network (MLP) with Resilient Backpropagation (RBP) algorithm. Employing the proposed method, a more accurate and simpler network is designed where the input size is reduced from 14 to 6 variables for training the NN. In short, the application of PCA highly reduces the network topology and allows employing more efficient training algorithms. The accuracy, generalization ability, and reliability of the designed networks are verified using 10 simulated events data from a VVER-1000 simulation using DINAMIKA-97 code. Noise is added to the data to evaluate the robustness of the method and the method again shows to be effective and powerful.
Kurzfassung
In dieser Arbeit wird auf der Basis neuronaler Netzwerke ein Fehlerdiagnoseansatz vorgestellt, der die rasche Identifizierung von Rissen und Lecks erlaubt. Die Methode verwendet die Hauptkomponentenanalyse (PCA) um das Ausmaß der Problematik zu reduzieren. Ein solcher Ansatz führt zu einer schnelleren Diagnose und erlaubt eine bessere graphische Darstellung der Ergebnisse. Um die Effektivität des verwendeten Ansatzes zu zeigen werden zwei Methoden zum Training des neuronalen Netzwerks (NN) verwendet. Zuerst wird eine Trainingsmatrix bestehend aus 14 Variablen verwendet um mehrschichtige Perzeptron-Netze (MLP) mit Resilient Backpropagation (RBP) Algorithmen zu trainieren. Durch Anwendung dieser Methode wird ein genaueres und einfacheres Netzwerk gestaltet, bei dem die Eingangsgröße von 14 auf 6 Variable reduziert wird. Die Anwendung der PCA reduziert die Netzwerktopologie erheblich und erlaubt die Verwendung effizienterer Trainingsalgorithmen. Die Genauigkeit, die Fähigkeit zur Verallgemeinerung und die Zuverlässigkeit der gestalteten Netzwerke werden verifiziert mit 10 Ereignisdaten einer WWER-1000 Simulation mit Hilfe des DINAMIKA-97 Codes. Den Daten wird Rauschen zugefügt um die Robustheit des Verfahrens besser bewerten zu können, wobei sich wieder die Effektivität und Leistungsfähigkeit dieser Methode zeigt.
References
1 Lewis, E. E.: Nuclear power reactor safety. Wiley-interscience publication, 1977, p. 105–106Search in Google Scholar
2 Morozov, S. A.;Kovtun, S. N.;Budarin, A. A.;Dvornikov, P. A.;Kudryaev, A. A.;Kondratovich, F. V.;Shutov, P. S.;Shvetsov, D. M.;Konoplev, N. P.: Development of an acoustic leak monitoring system. Atomic Energy103 (2007) 925–93110.1007/s10512-007-0149-1Search in Google Scholar
3 Kiselev, V. A.;Rivkin, E. Yu.: Application of the leak before rupture concept in the analysis of nuclear power plant safety. Atomic Energy25 (1993) 426–430Search in Google Scholar
4 Hashemi-Tilehnoee, M.,Pazirandeh, A.,Tashakor, S.: HAZOP-study on heavy water research reactor primary cooling system. Annals of Nuclear Energy37 (2010) 428–43310.1016/j.anucene.2009.12.006Search in Google Scholar
5 Aleksandrov, P. A.,Kalechits, V. I.;Khozyasheva, E. S.;Chechuev, P. V.: Detecting small leaks in pipes in nuclear power plants by measuring aerosol parameters. Atomic Energy97 (2004) 620–625Search in Google Scholar
6 Zhao, K.;Upadhyaya, B.R.: Model based approach for fault detection and isolation of helical coil steam generator systems using principal component analysis. IEEE Transactions on Nuclear Science53 (2006) 2343–235210.1109/TNS.2006.876049Search in Google Scholar
7 Hadad, K.;Mortazavi, M.;Safavi, A.;Mastali, M.: Enhanced neural network based fault detection of a VVER nuclear power plant with the aid of principal component analysis IEEE Transactions on Nuclear Science55(6) (2008) 3611–3617.Search in Google Scholar
8 Hadad, K.;Pourahmadi, M.;Majidi-Maraghi, H.: Fault diagnosis and classification based on wavelet transform and neural network. Progress in Nuclear Energy53 (2011) 41–4710.1016/j.pnucene.2010.09.006Search in Google Scholar
9 Bykov, M. A.;Zaitsev, S. I.;Alekhin, G. V.;Kurbaev, S. A.;Egorov, A. P.;Gusev, V. I.: The development of the TRAP-97 software system. A sensitivity analysis. Thermal Engineering53 (2006) 53–6110.1134/S0040601506010095Search in Google Scholar
10 Santosh, T. V.;Gopika, V.;Saraf, R. K.;Ghosh, A.K.;Kushwaha, H. S.: Application of artificial neural networks to nuclear power plant transient diagnosis. Reliability Engineering and System Safety92 (2007) 1468–147210.1016/j.ress.2006.10.009Search in Google Scholar
11 Zare, A.;Nematollahi, M. R.;Hadad, K.;Mozaffari, M. A.: Typical steam generator tube rupture (SGTR) effect on thermo-hydraulic parameters of VVER-1000 primary loop. ICENES 2007, Istanbul, TurkeySearch in Google Scholar
12 Samani, N.;Moghadam, M. G.;Safavi, A. A.: A simple neural network model for the determination of aquifer parameters. Journal of Hydrology340 (2007) 1–1110.1016/j.jhydrol.2007.03.017Search in Google Scholar
13 Afifi, A. A.;Clark, V.: Computer aided multivariate analysis. Chapman and Hall, London, 199610.1007/978-1-4899-3342-3Search in Google Scholar
14 Gaitani, N.;Lehmann, C.;Santamouris, M.;Mihalakakou, G.;Patargias, P.: Using principal component and cluster analysis in the heating evaluation of the school building sector. Applied Energy87 (2010) 2079–208610.1016/j.apenergy.2009.12.007Search in Google Scholar
© 2012, Carl Hanser Verlag, München
Articles in the same Issue
- Contents/Inhalt
- Contents
- Summaries/Kurzfassungen
- Summaries
- Technical Contributions/Fachbeiträge
- Simulation of HALDEN IFA-650 loss-of-coolant accidents tests with TRACE
- On the evaluation of a fuel assembly design by means of uncertainty and sensitivity measures
- Best estimate plus uncertainty analysis of LBLOCA for Indian PHWR
- Remarks on boiling water reactor stability analysis – part 1: theory and application of bifurcation analysis
- Upgrading (V)HTR fuel elements for generationIV goals by SiC encapsulation
- Investigation of the Ru-43LV fuel behaviour under LOCA conditions in a CANDU reactor
- PCA-based ANN approach to leak classification in the main pipes of VVER-1000
- Analytical investigation of the properties of the neutron noise induced by vibrating absorber and fuel rods
- Diffusion approximation for certain scattering parameters of the Anli-Güngör phase function
- Calculation of beta induced Bremsstrahlung exposure from therapeutic radionuclide 198Au in tissues, DNA and RNA
- Dosimetric aspects of 103Pd radioactive stent source
- Radiological significance of coal, slag and fly ash samples from the Eastern Black Sea region
Articles in the same Issue
- Contents/Inhalt
- Contents
- Summaries/Kurzfassungen
- Summaries
- Technical Contributions/Fachbeiträge
- Simulation of HALDEN IFA-650 loss-of-coolant accidents tests with TRACE
- On the evaluation of a fuel assembly design by means of uncertainty and sensitivity measures
- Best estimate plus uncertainty analysis of LBLOCA for Indian PHWR
- Remarks on boiling water reactor stability analysis – part 1: theory and application of bifurcation analysis
- Upgrading (V)HTR fuel elements for generationIV goals by SiC encapsulation
- Investigation of the Ru-43LV fuel behaviour under LOCA conditions in a CANDU reactor
- PCA-based ANN approach to leak classification in the main pipes of VVER-1000
- Analytical investigation of the properties of the neutron noise induced by vibrating absorber and fuel rods
- Diffusion approximation for certain scattering parameters of the Anli-Güngör phase function
- Calculation of beta induced Bremsstrahlung exposure from therapeutic radionuclide 198Au in tissues, DNA and RNA
- Dosimetric aspects of 103Pd radioactive stent source
- Radiological significance of coal, slag and fly ash samples from the Eastern Black Sea region
