• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 1998년도 추계학술대회 논문집
• /
• pp.164-169
• /
• 1998

This paper deals with condition monitoring for tool wear during tuning operation. To develop economic sensing and identification methods for turning processes, sound pressure measurement and digital signal processing technique are proposed. To identify noise sources of tool wear and reject background noise, noise rejection methodology is proposed. features to represent condition of tool wear are obtained through analysis using adaptive filter and FFT in time and frequency domain. By using fuzzy pattern recognition, we extract features, which are sensitive to condition of tool wear, from several features and make a decision on tool wear. The validity of the proposed system is condirmed through the large number of cutting tests in two cutting conditions.

• Nuclear Engineering and Technology
• /
• 제52권4호
• /
• pp.784-790
• /
• 2020

The accelerated corrosion by Flow Accelerated Corrosion (FAC) has caused unexpected rupture of piping, hindering the safety of nuclear power plants (NPPs) and sometimes causing personal injury. For the safety, it may be necessary to select some pipes in terms of condition monitoring and to measure the change in thickness of pipes in real time. Direct current potential drop (DCPD) method has advantages in on-line monitoring of pipe wall thinning. However, it has a disadvantage in that it is difficult to quantify thinning due to various thinning shapes and thus there is a limitation in application. The machine learning approach has advantages in that it can be easily applied because the machine can learn the signals of various thinning shapes and can identify the thinning using these. In this paper, finite element analysis (FEA) was performed by applying direct current to a carbon steel pipe and measuring the potential drop. The fundamental machine learning was carried out and the piping thinning model was developed. In this process, the features of DCPD to thinning were proposed.

• 한국정밀공학회지
• /
• 제16권12호
• /
• pp.119-125
• /
• 1999

This paper gives an overview of the various approaches to designing statistical pattern recognition scheme based on Bayes discrimination rule and the artificial neural networks for rotating machine condition classification. Concerning to Bayes discrimination rule, this paper contains the linear discrimination rule applied to classification into several multivariate normal distributions with common covariance matrices, the quadratic discrimination rule under different covariance matrices. Also we discribes k-nearest neighbor method to directly estimate a posterior probability of each class. Five features are extracted in time domain vibration signals. Employing these five features, statistical pattern classifier and neural networks have been established to detect defects on rotating machine. Four different cases of rotation machine were observed. The effects of k number and neural networks structures on monitoring performance have also been investigated. For the comparison of diagnosis performance of these two method, their recognition success rates are calculated form the test data. The result of experiment which classifies the rotating machine conditions using each method presents that the neural networks shows the highest recognition rate.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제5권9호
• /
• pp.433-440
• /
• 2016

철도 선로전환기는 열차의 진로를 현재의 궤도에서 다른 궤도로 제어하는 장치이다. 선로전환기의 이상 상황은 탈선 등과 같은 심각한 문제를 발생할 수 있기 때문에, 선로전환기의 스트레스를 지속적으로 모니터링 하는 것은 매우 중요하다. 본 논문에서는 선로전환기가 작동할 때 발생하는 소리 정보를 이용하여 선로전환기의 스트레스를 탐지하는 시스템을 제안한다. 제안하는 시스템은 선로전환기의 동작 시 발생하는 소리 데이터로부터 자질 선택방법을 사용하여 스트레스 탐지에 유효한 감소된 차원의 자질 부분집합을 선택한 후, 기계학습의 대표적 모델인 SVM(Support Vector Machine)을 이용하여 선로전환기의 스트레스 상태 여부를 탐지한다. 테스트용 선로전환기를 실제 구동하며 수집한 소리 데이터를 이용하여, 본 논문에서 제안하는 시스템의 성능을 실험적으로 검증한 바 98%를 넘는 정확도를 확인하였다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2005년도 추계학술대회논문집
• /
• pp.513-517
• /
• 2005

The condition monitoring of rotating machinery such as turbines, pumps and compressors, determine what repairs are needed to avoid shutdown and disassembly of the machine in an industrial plant Many diagnosis methods have been developed for use when the machine is running at steady state, the stationary condition. But much information can be gained about a rotor's condition during non-stationary conditions such as run-up and run-down. Order tracking analysis is a powerful tool for analyzing the condition of a rotating machine when its speed changes over time. Powerful OTA using digital signal processing has some advantages(cheap hardware, the powerful methods, the accurate post processing) and also some disadvantages(calculation time, high speed sampling). New OTA tool based on the chirplet transform is similar to the short time Fourier transform. But, it has good resolution at high speed like other OTA methods based STFT and more resolution for constant frequency components than re-sampling OTA.

• 융합신호처리학회논문지
• /
• 제23권2호
• /
• pp.84-90
• /
• 2022

공작기계 상태 진단은 기계의 상태를 자동으로 감지하는 프로세스이다. 실제로 가공의 효율과 제조공정에서 제품의 품질은 공구 상태에 영향을 받으며 마모 및 파손된 공구는 공정 성능에 보다 심각한 문제를 일으키고 제품의 품질 저하를 일으킬 수 있다. 따라서 적절한 시기에 공구가 교체될 수 있도록 공구 마모 진행 및 공정 중 파손 방지 시스템 개발이 필요하다. 본 논문에서는 공구의 적절한 교체 시기 등을 진단하기 위해 딥러닝 기반의 계층적 컨볼루션 신경망을 이용하여 5가지 공구 상태를 진단하는 방법을 제안한다. 기계가 공작물을 절삭할 때 발생하는 1차원 음향 신호를 주파수 기반의 전력스펙트럼밀도 2차원 영상으로 변환하여 컨볼루션 신경망의 입력으로 사용한다. 학습 모델은 계층적 3단계를 거쳐 5가지 공구 상태를 진단한다. 제안한 방법은 기존의 방법과 비교하여 높은 정확도를 보였고, 실시간 연동을 통해 다양한 공작기계를 모니터링할 수 있는 스마트팩토리 고장 진단 시스템에 활용할 수 있을 것이다.

• 대한기계학회논문집A
• /
• 제28권6호
• /
• pp.867-875
• /
• 2004

In order to inspect burr geometry and hole quality in micro-drilling processes, a cost-effective method using an image processing and shape from focus (SFF) methods on the machine tool is proposed. A CCD camera with a zoom lens and a novel illumination unit is used in this paper. Since the on-machine vision unit is incorporated with the CNC function of the machine tool, direct measurement and condition monitoring of micro-drilling processes are conducted between drilling processes on the machine tool. Stainless steel and hardened tool steel are used as specimens, as well as twist drills made of carbide are used in experiments. Validity of the developed system is confirmed through experiments.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1996년도 춘계학술대회 논문집
• /
• pp.232-236
• /
• 1996

This paper deals with an on-the-machine measurement method for roughness of ground surface by using flux ratio of scattered lights. A sensor and control unit is developed so as to e applied to surface grinding processes. The performance of the sensor is compared with that of stylus. The experimental investigation shows that not onlythe sensor has good performance as a surface roughness sensor but alsothe sensor is very useful for monitoring grinding condition in order to detect ill-conditioned grinding or dressing time.

• 한국생산제조학회지
• /
• 제4권2호
• /
• pp.38-45
• /
• 1995

The diagnosis and monitoring system of abnormal cutting condition is necessary to realize precision machining proces and factory automation, which are final goal of metal cutting in order to develop this system, theimage processing technique has been investigated in machining process. In theis paper, the measurement system of tool wear using computer vision is designed to detect the wear pattern by non-contact and direct method and get the realiable wear information about cutting tool. We measured the area of the side and front part of the diamond core dril which is used in 40kHz ultrasonic vibration machine.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1994년도 추계학술대회 논문집
• /
• pp.26-31
• /
• 1994

The analysis of acoustic emission signals generated during machining has been proposed as a technique for studying both the fundamentals of the cutting process and process and as a methodology for detecting tool fracture on line. In this study, AE signals detected during End Milling were applied as the experimental test to sensing tool fracture on the CNC vertical milling machine. Because automatic monitoring of the cutting condition is one of the most important technologies in machining, the in-process detection of cutting tool life including fracture has been investigated by performing experimental test.