• Nuclear Engineering and Technology
• /
• v.50 no.8
• /
• pp.1306-1313
• /
• 2018

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.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.19 no.11
• /
• pp.94-101
• /
• 2020

In the age of industry 4.0, artificial intelligence is being widely used to realize machinery condition monitoring. Due to their excellent performance and the ability to handle large volumes of data, machine learning techniques have been applied to realize the fault diagnosis of different equipment. In this study, we performed the failure mode effect analysis (FMEA) of an aluminum electrolytic capacitor by using deep learning and big data. Several tests were performed to identify the main failure mode of the aluminum electrolytic capacitor, and it was noted that the capacitance reduced significantly over time due to overheating. To reflect the capacitance degradation behavior over time, we employed the Vanilla long short-term memory (LSTM) neural network architecture. The LSTM neural network has been demonstrated to achieve excellent long-term predictions. The prediction results and metrics of the LSTM and Vanilla LSTM models were examined and compared. The Vanilla LSTM outperformed the conventional LSTM in terms of the computational resources and time required to predict the capacitance degradation.

• Proceedings of the Korean Society for Quality Management Conference
• /
• 2006.11a
• /
• pp.381-388
• /
• 2006

많은 6시그마 혁신을 하는 기업 중에서 IS09000 시리즈나 IS0/TS16949 인증을 받고 있다. 그럼에도 불구하고 6시그마 도입 초기부터 ISO/TS16949와 통합운영을 생각하고 추진하는 곳은 많지 않다. ISO/TS16949에서는 기업전반의 프로세스에 많은 중점을 두고 있다. 그 중에서도 APQP(Advanced Product Quality Planning)은 제품 초기에서 부터 명확한 품질계획 업무 프로세스를 가지고 활동을 하라는 필수 프로세스를 요구하고 있다. 그러나 구체적인 Flow나 단계는 각 기업체의 특성에 맞기고 있다. 6시그마의 Lean DFSS는 이러한 제품개발 프로세스의 구체적인 전개단계를 제공 하고 있다. 여기에 APQP에 서 요건으로 명시한 각 프로세스 단계마다 인풋(Input)요소, 아웃풋(Output) 요소, 타켓(Target), 그 프로세스를 책임질 구성원, 핵심성공요인(KSF), 핵심성공지수(KPI), 단계마다의 FMEA(Failure Mode Effect Analysis) 분석을 대응하면 도입 초기에서 부터 상호 유기적이면서 효과적인 System이 된다.

• IE interfaces
• /
• v.25 no.4
• /
• pp.422-430
• /
• 2012

In this article, we introduced the estimation method by 'Safety Integrity Level'(SIL) for the criterion of safety assurance and performed a case study on a flame scanner. SIL requires probabilistic evaluation of each set of equipment used to reduce risk in a safety related system. FMEDA(Failure Modes, Effects and Diagnostic Analysis) method is widely used to evaluate the safety levels and provides information on the failure rates and failure mode distributions necessary to calculate a diagnostic coverage factor for a part or a component. Basically, two parameters resulting from FMEDA are used for SIL classification of the device : SFF(Safe Failure Fraction) and PFD(Probability of Failure on Demand). In this case study, it is concluded that the flame scanner is designed to fulfill the condition of SIL 3 in the aspect of SFF and PFD.

• Aerospace Engineering and Technology
• /
• v.2 no.1
• /
• pp.44-53
• /
• 2003

The purpose of reliability prediction is to estimate the basic reliability and basic reliability and mission reliability of system and to make a determination of whether these reliability requirements can be achieved with the proposed design. Also, potential design weakness can be identified through the FMECA process. This technical memo summarizes the KOMPSAT-2 reliability and FMECA analysis results.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.23 no.61
• /
• pp.75-87
• /
• 2000

This paper proposes a procedure for designing an accelerated test using SMAT(Stress, (failure) Mechanism and Test) model describing the relation among stress, failure mode/mechanism and test method. In SMAT model the stresses to be applied are derived from the environmental factor analysis, the relative importance of those stresses can be estimated using AHP(Analytic Hierarchy Process) and failure mode/mechanism and test method are derived from the fields failure information and FMEA(Failure Mode and Effect Analysis). By applying the procedure we can make a selection of major factors to cause the failure of assembly and design the accelerated test using DOE(Design of Experiments) The procedure is illustrated with an qualification test case study of washing machine shaft assembly in "A" electric appliance company.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.22 no.3
• /
• pp.68-74
• /
• 2014

As the number of ECUs (Electronic control units) are increasing, reliability and functional stability of a software in an ECU is getting more important. Therefore the application of functional safety standards ISO 26262 is making the software more reliable. Software fault injection test (SFIT) is required as a verification technique for the application of ISO 26262. In case of applying SFIT, an artificial error is injected to inspect the vulnerability of the system which is not easily detected during normal operation. In this paper, the basic concept of SFIT will be examined and the application of SIFT based on ISO26262 will be described.

• Journal of Korean Society for Quality Management
• /
• v.41 no.3
• /
• pp.487-494
• /
• 2013

Purpose: In this study, we propose the precision guided munitions verification methodology using the statistical analysis method has been proposed. and it can be applied to the precision guided munitions quality assurance work. Methods: This modeling is based on Failure Mode and Effects Analysis, Statistical Process Control, Defense Quality Managerment System, Production Readiness Review, Manufacturing Readiness Assesment and so on. Results: The Process Control Modeling that has the following procedures ; searching the critical to quality, statistical analysis by process, verify process. Moreover, the effectiveness of the methodology is verified by applying to the precision guided munitions. Conclusion: To achieve a analysis methods of statistical process control and verify process for precision guided munitions.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.29 no.1
• /
• pp.9-17
• /
• 2006

This paper considers a six sigma project for improving productivity of the brace complement center pillar. The project follows a disciplined process of fife phases: define, measure, analyze, improve, and control. A process map is used to identify process input and output variables. Eleven key process input variables are selected by using X&Y matrix and FMEA, and finally eight vital few input variables are selected from analyze phase. The optimum process conditions of the vital few input variables are jointly obtained by maximizing productivity of the brace complement center pillar using DOE and alternative selection method.

• Proceedings of the Korean Reliability Society Conference
• /
• 2006.05a
• /
• pp.173-180
• /
• 2006

컨테이너 터미널 장비들은 많은 부품들이 매우 복잡하게 구성되어 있으며, 고장 발생 시에 막대한 비용이 발생하게 된다. 본 연구에서는 컨테이너 터미널 장비의 신뢰도, 가용도, 정비도 향상을 위한 관리 시스템을 개발한다. 개발 시스템은 장비 구성 모듈, 장비 운영 관리 모듈, 정비 관리 모듈, 예비품 관리 모듈, 분석 모듈로 구성되어 있다. 장비의 신뢰도, 가용도, 정비도 향상을 위한 기술로 FMEA, 고장현황 분석, 수명 모수 추정법을 사용한다. 또한 산출된 수명 모수와 객체지향 시뮬레이션 모형으로 최적 예방 정비 주기를 결정하고 이를 이용하여 정비인원과 하루 최대 정비 가능 장비대수를 고려한 연간 최적 예방정비 일정을 결정한다.