Support vector ensemble for incipient fault diagnosis in nuclear plant components

The randomness and incipient nature of certain faults in reactor systems warrant a robust and dynamic detection mechanism. Existing models and methods for fault diagnosis using different mathematical/statistical inferences lack incipient and novel faults detection capability. To this end, we propose...

Full description

Bibliographic Details
Main Authors: Abiodun Ayodeji, Yong-kuo Liu
Format: Article
Language:English
Published: Elsevier 2018-12-01
Series:Nuclear Engineering and Technology
Online Access:http://www.sciencedirect.com/science/article/pii/S1738573317306988
id doaj-8008c25d1f5946249c7f557726246dad
record_format Article
spelling doaj-8008c25d1f5946249c7f557726246dad2020-11-25T00:45:00ZengElsevierNuclear Engineering and Technology1738-57332018-12-0150813061313Support vector ensemble for incipient fault diagnosis in nuclear plant componentsAbiodun Ayodeji0Yong-kuo Liu1Fundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, Harbin, Heilongjiang 150001, China; Nuclear Power Plant Development Directorate, Nigeria Atomic Energy Commission, Abuja, NigeriaFundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, Harbin, Heilongjiang 150001, China; Corresponding author.The randomness and incipient nature of certain faults in reactor systems warrant a robust and dynamic detection mechanism. Existing models and methods for fault diagnosis using different mathematical/statistical inferences lack incipient and novel faults detection capability. To this end, we propose a fault diagnosis method that utilizes the flexibility of data-driven Support Vector Machine (SVM) for component-level fault diagnosis. The technique integrates separately-built, separately-trained, specialized SVM modules capable of component-level fault diagnosis into a coherent intelligent system, with each SVM module monitoring sub-units of the reactor coolant system. To evaluate the model, marginal faults selected from the failure mode and effect analysis (FMEA) are simulated in the steam generator and pressure boundary of the Chinese CNP300 PWR (Qinshan I NPP) reactor coolant system, using a best-estimate thermal-hydraulic code, RELAP5/SCDAP Mod4.0. Multiclass SVM model is trained with component level parameters that represent the steady state and selected faults in the components. For optimization purposes, we considered and compared the performances of different multiclass models in MATLAB, using different coding matrices, as well as different kernel functions on the representative data derived from the simulation of Qinshan I NPP. An optimum predictive model - the Error Correcting Output Code (ECOC) with TenaryComplete coding matrix - was obtained from experiments, and utilized to diagnose the incipient faults. Some of the important diagnostic results and heuristic model evaluation methods are presented in this paper. Keywords: Fault diagnosis, Support vector machine, Error correcting output code, Reactor coolant systemhttp://www.sciencedirect.com/science/article/pii/S1738573317306988
collection DOAJ
language English
format Article
sources DOAJ
author Abiodun Ayodeji
Yong-kuo Liu
spellingShingle Abiodun Ayodeji
Yong-kuo Liu
Support vector ensemble for incipient fault diagnosis in nuclear plant components
Nuclear Engineering and Technology
author_facet Abiodun Ayodeji
Yong-kuo Liu
author_sort Abiodun Ayodeji
title Support vector ensemble for incipient fault diagnosis in nuclear plant components
title_short Support vector ensemble for incipient fault diagnosis in nuclear plant components
title_full Support vector ensemble for incipient fault diagnosis in nuclear plant components
title_fullStr Support vector ensemble for incipient fault diagnosis in nuclear plant components
title_full_unstemmed Support vector ensemble for incipient fault diagnosis in nuclear plant components
title_sort support vector ensemble for incipient fault diagnosis in nuclear plant components
publisher Elsevier
series Nuclear Engineering and Technology
issn 1738-5733
publishDate 2018-12-01
description The randomness and incipient nature of certain faults in reactor systems warrant a robust and dynamic detection mechanism. Existing models and methods for fault diagnosis using different mathematical/statistical inferences lack incipient and novel faults detection capability. To this end, we propose a fault diagnosis method that utilizes the flexibility of data-driven Support Vector Machine (SVM) for component-level fault diagnosis. The technique integrates separately-built, separately-trained, specialized SVM modules capable of component-level fault diagnosis into a coherent intelligent system, with each SVM module monitoring sub-units of the reactor coolant system. To evaluate the model, marginal faults selected from the failure mode and effect analysis (FMEA) are simulated in the steam generator and pressure boundary of the Chinese CNP300 PWR (Qinshan I NPP) reactor coolant system, using a best-estimate thermal-hydraulic code, RELAP5/SCDAP Mod4.0. Multiclass SVM model is trained with component level parameters that represent the steady state and selected faults in the components. For optimization purposes, we considered and compared the performances of different multiclass models in MATLAB, using different coding matrices, as well as different kernel functions on the representative data derived from the simulation of Qinshan I NPP. An optimum predictive model - the Error Correcting Output Code (ECOC) with TenaryComplete coding matrix - was obtained from experiments, and utilized to diagnose the incipient faults. Some of the important diagnostic results and heuristic model evaluation methods are presented in this paper. Keywords: Fault diagnosis, Support vector machine, Error correcting output code, Reactor coolant system
url http://www.sciencedirect.com/science/article/pii/S1738573317306988
work_keys_str_mv AT abiodunayodeji supportvectorensembleforincipientfaultdiagnosisinnuclearplantcomponents
AT yongkuoliu supportvectorensembleforincipientfaultdiagnosisinnuclearplantcomponents
_version_ 1725271990546202624