• 산업공학
• /
• 제25권2호
• /
• pp.170-177
• /
• 2012

Process control is essential to operate the semiconductor process efficiently. This paper consider fault classification of semiconductor based cyclic signal for process control. In general, process signal usually take the different pattern depending on some different cause of fault. If faults can be classified by cause of faults, it could improve the process control through a definite and rapid diagnosis. One of the most important thing is a finding definite diagnosis in fault classification, even-though it is classified several times. This paper proposes the method that one-class classifier classify fault causes as each classes. Hotelling T2 chart, kNNDD(k-Nearest Neighbor Data Description), Distance based Novelty Detection are used to perform the one-class classifier. PCA(Principal Component Analysis) is also used to reduce the data dimension because the length of process signal is too long generally. In experiment, it generates the data based real signal patterns from semiconductor process. The objective of this experiment is to compare between the proposed method and SVM(Support Vector Machine). Most of the experiments' results show that proposed method using Distance based Novelty Detection has a good performance in classification and diagnosis problems.

• 한국정보과학회논문지:시스템및이론
• /
• 제31권3_4호
• /
• pp.204-212
• /
• 2004

분산 시스템에서는 단일 시스템보다 높은 결함 발생 확률을 가지기에 기존의 맡은 연구에서는 분산 시스템에서 결함 발생에 대한 맡은 결함 포용 기법들이 연구되어 왔다. 하지만 저장된 결함 포용 정보의 증가에 따른 저장 공간의 부족으로 인해 전체 시스템 성능의 저하를 가져오게 하였다. 시스템 성능의 저하를 막기 위하여 불필요한 결함 포용 정보의 삭제가 필요하게 되었고 이 논문에서는 결함 포용 정보의 쓰레기 처리를 위한 방법을 제안한다. 이에 본 논문에서는 결함 포용 정보의 쓰레기 처리를 담당하는 쓰레기 처리 에이전트, 결함 포용 정보를 유지 관리하는 정보 에이전트, 그리고 전체 에이전트간의 통신 기능을 담당하는 조정 에이전트를 정의 및 설계하고, 쓰레기 처리 에이전트를 이용한 쓰레기 처리 알고리즘을 제안한다. 복귀회복 기법은 독립 검사점(independent checkpoint)기법과 송신자 기반 비관적 매시지 로깅(sender based pessimistic message logging)기법을 사용한다. 제안된 쓰레기 처리 기법에서의 쓰레기 처리, 정보, 조정 에이전트는 프로세스와 동시에 생성되며 정보 에이전트에 프로세스에서 발생하는 검사점과 비결정적인 사건들에 대한 로깅 정보들을 영역 지식으로 구축한다. 그리고 쓰레기 처리 에이전트는 쓰레기 처리 시점을 선정하고 정보 에이전트와 조정에이전트의 협력을 통하여 영역 지식에 구축된 불필요한 결함 포용 정보의 쓰레기 처리를 한다. 제안한 에이전트를 이용한 쓰레기 처리기법의 타당성 증명을 위하여 결함을 발생시켜 복귀 회복 후 쓰레기 처리를 하는 시스템과 하지 않는 시스템의 영역지식을 비교하여 같은 결과를 같는지의 여부를 검사한다

• 인터넷정보학회논문지
• /
• 제9권3호
• /
• pp.103-117
• /
• 2008

일반 고객이 대기업 혹은 중소기업, 증권, 금융, 은행권 등의 서비스를 제공받기 위해서는 보편적으로 인터넷, SMS(Short Message Service), ATM(Automated Teller Machine), DM(Direct Mail), 텔레포니 서비스 등을 사용한다. 특히, 화재 보험사 및 증권, 금융, 은행권의 경우는 QoS(Quality of Service) 보장을 통한 실시간성 제약 조건을 만족해야 한다. 본 논문에서는 고객의 최초 접점으로 이용될 수 있는 CRM(Customer Relationship Management) 환경 하에서 다중 채널 서비스를 위한 결함허용 IVR(Interactive Voice Response)을 설계 구현하였다. 제안한 모델은 대 고객 응대 CRM 모델로 효율적으로 이용될 것으로 사료된다.

• Nuclear Engineering and Technology
• /
• 제52권10호
• /
• pp.2238-2249
• /
• 2020

In a seismic PSA, dependency among seismic failures of components has not been explicitly modeled in the fault tree or event tree. This dependency is separately identified and assigned with numbers that range from zero to unity that reflect the level of the mutual correlation among seismic failures. Because of complexity and difficulty in calculating combination probabilities of correlated seismic failures in complex seismic event tree and fault tree, there has been a great need of development to explicitly model seismic correlation in terms of seismic common cause failures (CCFs). If seismic correlations are converted into seismic CCFs, it is possible to calculate an accurate value of a top event probability or frequency of a complex seismic fault tree by using the same procedure as for internal, fire, and flooding PSA. This study first proposes a methodology to explicitly model seismic dependency by converting correlated seismic failures into seismic CCFs. As a result, this methodology will allow systems analysts to quantify seismic risk as what they have done with the CCF method in internal, fire, and flooding PSA.

• Korean Chemical Engineering Research
• /
• 제45권1호
• /
• pp.32-38
• /
• 2007

통계적인 공정 제어 기법을 회분식 공정에 적용하여, 일반적인 회분식 공정의 데이터를 통해 보다 빠르고, 손쉽게 공정의 상태를 진단할 수 있는 시스템을 구현해 보았다. 대표적인 회분식 공정의 하나인 반도체 식각공정과 반회분식 스타이렌-부타디엔 고무 생산 공정의 데이터를 이용하여 공정 변수와 공정의 상태간의 연관 관계를 규명할 수 있는 모델을 수립하였으며, 이 모델의 출력(output) 결과를 이용해 통계적 공정 제어 차트를 구성하고, 시간에 따른 공정의 추이를 분석해 이상을 판별해 보았다. 회분식 공정의 다축(multi-way) 데이터를 두개의 축으로 만드는 펼치기(unfolding) 과정을 거쳤으며, 모델링 방법으로는 Support Vector Regression 및 Partial Least Square 등의 다변량 회귀분석 방법을 이용하였다. 또한 에러차트 및 변수 기여도 차트(variable contribution chart)를 이용해 이상의 세기, 형태 및 이상 데이터에 대한 각 변수들의 기여도를 계산해 보았으며, 그 결과 이상의 발생 유무 및 발생시점 뿐만아니라 이상의 세기 및 원인 까지 진단해 볼 수 있는 우수한 성능을 보이는 것을 확인할 수 있었다.

• 한국컴퓨터정보학회논문지
• /
• 제3권2호
• /
• pp.99-104
• /
• 1998

본 논문에서는 다단 논리회로의 고장을 완벽하게 검출할 수 있는 테스트 패턴 생성기를 설계하였다. 이 테스트 기법은 테스트 패턴 생성 논리회로를 사용하여 생성하였다. 생성된 테스트 패턴은 기존의 전체 테스트 방법에 비해 패턴을 크게 감소시켰다. 이 테스트패턴 생성기는 다단 논리회로에서의 모든 고장을 검출할 것으로 본다. 여러 가지 I.C 테스트 방법 중에서 어떤 방법을 선택할 것인지는 고장검출 속도에 영향을 준다. 가장 중요한 것은생산단가이며 설계된 테스트 패턴 생성기는 저가형이다.

• 한국안전학회지
• /
• 제15권2호
• /
• pp.36-45
• /
• 2000

Automated production system is composed of many complicated techniques and it become a very difficult task to control, monitor and diagnose this compound system. Moreover, it is required to develop an effective diagnosing technique and reduce the diagnosing time while operating the system in parallel under many faults occurring concurrently. This study develops a Modular Artificial Neural Network(MANN) which can perform a diagnosing function of multiple faults with the following steps: 1) Modularizing a complicated system into subsystems. 2) Formulating a hierarchical structure by dividing the subsystem into many detailed elements. 3) Planting an artificial neural network into hierarchical module. The system developed is implemented on workstation platform with $X-Windows^{(r)}$ which provides multi-process, multi-tasking and IPC facilities for visualization of transaction, by applying the software written in $ANSI-C^{(r)}$ together with $MOTIF^{(r)}$ on the fault diagnosis of PI feedback controller reactor. It can be used as a simple stepping stone towards a perfect multiple diagnosing system covering with various industrial applications, and further provides an economical approach to prevent a disastrous failure of huge complicated systems.

• 드라이브 ㆍ 컨트롤
• /
• 제14권2호
• /
• pp.30-36
• /
• 2017

This paper presents a sliding mode observer-based fault detection algorithm for steering inputs of an all-terrain crane. All-terrain cranes with multi-axles have several steering modes for various working purposes. Since steering angles at the other axles except the first wheel are controlled by using the information of steering angle at the first wheel, a reliable signal of the first axle's steering angle should be secured for the driving safety of cranes. For the fault detection of steering input signal, a simplified crane model-based sliding mode observer has been used. Using a sliding mode observer with an equivalent output injection signal that represents an actual fault signal, a fault signal in steering input was reconstructed. The road steering mode of the crane's steering system was used to conduct performance evaluations of a proposed algorithm, and an arbitrary fault signal was applied to the steering angle at the first wheel. Since the road steering mode has different steering strategies according to different speed intervals, performance evaluations were conducted based on the curved path scenario with various speed conditions. The design of algorithms and performance evaluations were conducted on Matlab/Simulink environment, and evaluation results reveal that the proposed algorithm is capable of detecting and reconstructing a fault signal reasonably well.

• Nuclear Engineering and Technology
• /
• 제55권6호
• /
• pp.2096-2106
• /
• 2023

Rotating machinery is widely applied in important equipment of nuclear power plants (NPPs), such as pumps and valves. The research on intelligent fault diagnosis of rotating machinery is crucial to ensure the safe operation of related equipment in NPPs. However, in practical applications, data-driven fault diagnosis faces the problem of small and imbalanced samples, resulting in low model training efficiency and poor generalization performance. Therefore, a deep convolutional conditional generative adversarial network (DCCGAN) is constructed to mitigate the impact of imbalanced samples on fault diagnosis. First, a conditional generative adversarial model is designed based on convolutional neural networks to effectively augment imbalanced samples. The original sample features can be effectively extracted by the model based on conditional generative adversarial strategy and appropriate number of filters. In addition, high-quality generated samples are ensured through the visualization of model training process and samples features. Then, a deep convolutional neural network (DCNN) is designed to extract features of mixed samples and implement intelligent fault diagnosis. Finally, based on multi-fault experimental data of motor and bearing, the performance of DCCGAN model for data augmentation and intelligent fault diagnosis is verified. The proposed method effectively alleviates the problem of imbalanced samples, and shows its application value in intelligent fault diagnosis of actual NPPs.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제51권9호
• /
• pp.413-422
• /
• 2002

In a power plant, disturbance detection and diagnosis are massive and complex problems. Once a disturbance occurs, it can be either persistent, self cleared, cleared by the automatic controllers or propagated into another disturbance until it subsides in a new equilibrium or a stable state. In addition to the Physical complexity of the power plant structure itself, these dynamic behaviors of the disturbances further complicate the fault monitoring and diagnosis tasks. A data structure called a disturbance interrelation analysis graph(DIAG) is proposed in this paper, trying to capture, organize and better utilize the vast and interrelated knowledge required for power plant disturbance detection and diagnosis. The DIAG is a multi-layer directed AND/OR graph composed of 4 layers. Each layer includes vertices that represent components, disturbances, conditions and sensors respectively With the implementation of the DIAG, disturbances and their relationships can be conveniently represented and traced with modularized operations. All the cascaded disturbances following an initial triggering disturbance can be diagnosed in the context of that initial disturbance instead of diagnosing each of them as an individual disturbance. DIAG is applied to a typical cooling water system of a thermal power plant and its effectiveness is also demonstrated.