• 한국멀티미디어학회논문지
• /
• 제20권2호
• /
• pp.134-143
• /
• 2017

Machine fault diagnosis recovers all the studies that aim to detect automatically faults or damages on machines. Generally, it is very difficult to diagnose a machine fault by conventional methods based on mathematical models because of the complexity of the real world systems and the obvious existence of nonlinear factors. This study develops an automatic machine fault diagnosis system that uses pattern recognition techniques such as principal component analysis (PCA) and artificial neural networks (ANN). The sounds emitted by the operating machine, a drill in this case, are obtained and analyzed for the different operating conditions. The specific machine conditions considered in this research are the undamaged drill and the defected drill with wear. Principal component analysis is first used to reduce the dimensionality of the original sound data. The first principal components are then used as the inputs of a neural network based classifier to separate normal and defected drill sound data. The results show that the proposed PCA-ANN method can be used for the sounds based automated diagnosis system.

• 한국수소및신에너지학회논문집
• /
• 제34권5호
• /
• pp.478-489
• /
• 2023

In this study, fault diagnosis and detection methods developed to ensure the reliability of polymer electrolyte membrane (PEM) hydrogen electrolysis systems have been proposed. The proposed method consists of model development and data generation of the PEM hydrogen electrolysis system, and data-driven fault diagnosis learning model development. The developed fault diagnosis learning model describes how to detect and classify faults in the sensors and components of the system.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2008년도 추계학술대회B
• /
• pp.2070-2075
• /
• 2008

Identification of steady-state is the first step in developing a fault detection and diagnosis (FDD) system. In a complete FDD system, the steady-state detector will be included as a module in a self-learning algorithm which enables the working system's reference model to "tune" itself to its particular installation. In this study, a steady-state detector of a residential air conditioner based on moving windows was designed. Seven representing measurements were selected as key features for steady-state detection. The optimized moving window size and the feature thresholds was suggested through startup transient test and no-fault steady-state test. Performance of the steady-state detector was verified during indoor load change test. From the research, the general methodology to design a moving window steady-state detector was provided for vapor compression applications.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1990년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 26-27 Oct. 1990
• /
• pp.131-134
• /
• 1990

A back-propagation neural network based system for a fault diagnosis of a chemical process is developed. Training data are acquired from FCD(Fault-Consequence Digraph) model. To improve the resolution of a diagnosis, the system is decomposed into 6 subsystems and the training data are composed of 0, 1 and intermediate values. The feasibility of this approach is tested through case studies in a real plant, a naphtha furnace, which has been used to develop a knowledge based expert system, OASYS (Operation Aiding expert SYStem).

• 제어로봇시스템학회논문지
• /
• 제5권1호
• /
• pp.54-61
• /
• 1999

Objective of the research is to put the vast and complex fault tracking knowledge of human experts in semiconductor production line into the knowledge base of computer system. We mined the fault tracking knowledge of domain experts(engineers of production line) for the construction of knowledge base of the expert system. Object oriented fact models which increase the extensibility and reusability have been built. The rules are designed to perform the fault diagnosis of the items in production device. We have exploited the evidence accumulation method to assign check priority in rules. The major contribution is in the overall design and implementation of the nile base and related facts of the expert system in object oriented paradigm for the application of the system in fault diagnosis in semiconductor production line.

• 한국지능시스템학회논문지
• /
• 제12권6호
• /
• pp.503-510
• /
• 2002

본 논문에서는 비선형시스템에서 발생한 고장을 감지하고 분류하기 위해 신경회로망기반 다중고장모델과 통계적기법에 의한 고장진단 방법을 제안한다. 제안한 알고리듬에서는 시스템의 출력과 신경회로망 공칭모델 출력 사이의 오차가 미리 설정한 문턱 값을 넘으면 고장을 감지한다. 고장이 감지되면 고장분류기에서는 각 신경회로망 고장모델 출력과 시스템 출력 사이의 오차를 이용하여 통계적 기법으로 고장을 분류한다. 컴퓨터 시뮬레이션 결과로부터 제안한 고장진단방법이 비선형 시스템에서의 고장감지 및 분류문제에 잘 적용됨을 알 수 있다.

• Nuclear Engineering and Technology
• /
• 제35권5호
• /
• pp.454-470
• /
• 2003

In this paper, a preliminary study for development of a fault diagnosis is presented for monitoring and diagnosing faults in the agitator driving system of a high temperature reduction reactor. In order to identify a fault occurrence and classify the fault cause, vibration signals measured by accelerometers on the outer shroud of the agitator driving system are firstly decomposed by wavelet transform (WT) and the features corresponding to each fault type are extracted. For the diagnosis, the fuzzy ARTMAP is employed and thereby, based on the features extracted from the WT, the robust fault classifier can be implemented with a very short training time - a single training epoch and a single learning iteration is sufficient for training the fault classifier. The test results demonstrate satisfactory classification for the faults pre-categorized from considerations of possible occurrence during experiments on a small-scale reduction reactor.

• 대한전기학회논문지
• /
• 제39권1호
• /
• pp.16-21
• /
• 1990

A Knowledge based expert system is a computer program that emulates the reasoning process of a human expert in a specific problem domain. Expert system has the potential to solve a wide range of problems which require knowledge about the problem rather than a purely analytical approach. This papaer presents the application of knowledge based expert system to power system fault diagnosis. The contents of expert system develpped in this paper is judgement of fault section from a given alarm sets and production of all possible hypothesis for the single fault. Both relay failures and circuit breaker failures are considered simultaneously. Although many types of relay are used in actual system, experts recognize ones as several typical signals corresponding to the fault types. Therefore relays are classified into several types. The expert system is written in an artificial intelligence language "PROLOG" . Best-first search method is used for problem solving. Both forward chaining and backward chaining schemes are used in reasoning process. The application to a part of actual power system proves the availability of the developed expert system.

• 전기전자학회논문지
• /
• 제27권4호
• /
• pp.601-609
• /
• 2023

본 논문에서는 DC 계통의 지락고장시 고속 고장진단을 위해 FPGA를 이용한 인공지능기반 고장진단 방법을 제안한다. 인공지능 알고리즘을 고장진단에 적용시 많은 연산량과 대용량의 실시간 데이터 처리가 요구된다. 또한 DC 계통에서의 고장 및 사고는 고장 전류의 빠른 상승률로 인하여 DC 차단기가 고속 차단능력이 필요하다. 인공지능기반 고속 고장진단이 가능한 FPGA를 사용하여 DC 차단기가 더 빠르게 동작함으로써, DC 차단기의 차단용량을 줄일 수 있다. 따라서 본 논문에서는 Matlab Simulink를 이용하여 DC계통의 고장 모의를 통해 고장데이터를 수집하여 지능형 고속 진단 알고리즘 구현하였으며, FPGA에 지능형 고속고장 진단 알고리즘을 적용 및 성능검증을 하였다.

• 한국정밀공학회지
• /
• 제22권10호
• /
• pp.80-89
• /
• 2005

In this paper, time-frequency analysis using wavelet packet transform and advanced-MDSA (Multiple Dimensional Spectral Analysis) which based on wavelet packet transform is applied fur fault source identification and diagnosis of early detection of fault non-stationary sound/vibration signals. This method is analyzing the signal in the plane of instantaneous time and instantaneous frequency. The results of ordinary coherence function, which obtained by wavelet packet analysis, showed the possibility of early fault detection by analysis at the instantaneous time. So, by checking the coherence function trend, it is possible to detect which signal contains the major fault signal and to know how much the system is damaged. Finally, It is impossible to monitor the system is damaged or undamaged by using conventional method, because crest factor is almost constant under the range of magnitude of fault signal as its approach to normal signal. However instantaneous coherence function showed that a little change of fault signal is possible to monitor the system condition. And it is possible to predict the maintenance time by condition based maintenance for any stationary or non-stationary signals.