• 한국정밀공학회지
• /
• 제16권4호통권97호
• /
• pp.229-236
• /
• 1999

The surface, generated by end milling operation, is deteriorated by tool runout, vibration, tool wear and tool deflection, etc. Among them, the effect of tool deflection on surface accuracy is analyzed. Surface generation model for the prediction of the topography of machined srufaces has been developed based on cutting mechanism and cutting tool geometry. This model accounts for not only the ideal geometrical surface, but also the deflection of tool due to cutting force. For the more accurate prediction of cutting force, flexible end mill model is used to simulate cutting process. Computer simulation has shown the feasibility of the surface generation system. Using developed simulation system, the relations between the shape of end mill and cutting conditions are analyzed.

• 한국공작기계학회논문집
• /
• 제12권5호
• /
• pp.53-58
• /
• 2003

Wear debris can be collected from the lubricants of operating machinery and its morphology is directly related to the fiction condition of the interacting materials from which the wear particles originated in lubricated machinery. But in order to predict and estimate working conditions, it is need to analyze the shape characteristics of wear debris and to identify. Therefore, if the shape characteristics of wear debris is identified by computer image analysis and the neural network, The four parameter (50% volumetric diameter, aspect, roundness and reflectivity) of wear debris are used as inputs to the network and learned the friction. It is shown that identification results depend on the ranges of these shape parameters learned. The three kinds of the wear debris had a different pattern characteristic and recognized the friction condition and materials very well by neural network. We resented how the neural network recognize wear debris on driving condition.

• 한국생산제조학회지
• /
• 제22권2호
• /
• pp.268-274
• /
• 2013

Electric discharge machining drill (EDM-drill) is an efficient process for the fabrication of micro-diameter deep metal hole. As there is non-physical contact between tool (electrode) and workpiece, EDM-drill is widely used to machine the hard machining materials such as high strength steel, cemented carbide, titanium alloys. The electro-thermal energy forces the electrode to wear out together with the workpiece to be machined. The electrode wear occurs inside of a machining hole. and It causes hard to monitor the machining state, which leads the productivity and the quality to decrease. Thus, this study presents a methodology to estimated the electrode wear amount while two coefficients (scale factor and shape factor) of the logarithmic regression model are evaluated from the experiment result. To increase the accuracy of estimation model, the linear transformation method is adopted using the differences of initial electrode wear differences. The estimation model is verified through experiment. The experimental result shows that within minute error, the estimation model is able to predict accurately.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1997년도 추계학술대회 논문집
• /
• pp.47-51
• /
• 1997

The high speed machining which can improve the production and quality in machining has been adopted remarkably in dietmold industry. As the speed of machine tool spindle increases, the machinability evaluation and monitoring of high speed machining is necessary. In this study, the machinability of 30, 000rpm class spindle was evaluated by using the developed tool dynamometer and the machining properties of high hardened and toughness materials in high speed were examined. Finally, the in-process monitoring technologies of tool wear were presented through the prediction by the experimental formula and pattern recognition by the neural network.

• 터널과지하공간
• /
• 제34권2호
• /
• pp.143-153
• /
• 2024

TBM 공법은 발파 공법에 비해 굴착 중 소음과 진동 수준이 낮고, 안정성이 높은 터널 굴착 공법이며, 전세계적으로 터널 프로젝트에 TBM 공법을 적용하는 사례가 증가하는 추세이다. 디스크 커터는 TBM의 커터헤드에 장착되는 굴착 도구로 지속적으로 막장면 지반과 상호작용하며, 이때 필연적으로 마모가 발생한다. 본 연구에서는 지질 조건과 TBM 운영파라미터, 머신러닝 알고리즘들을 이용하여 디스크 커터 마모를 정량적으로 예측하였다. 디스크커터 마모 예측의 입력변수 중 UCS 데이터의 수가 다른 기계 데이터 및 마모 데이터에 비해 매우 부족하기 때문에, 먼저 TBM 기계 데이터를 이용하여 전체 구간에 대한 UCS 추정을 진행하고, 완성된 전체 데이터로 마모율 계수 예측을 수행하였다. 마모율 계수 예측 모델의 성능을 비교해 본 결과 XGBoost 모델의 성능이 가장 높게 나타났으며, 복잡한 예측 모델의 해석을 위해 SHapley Additive exPlanation (SHAP) 분석을 진행하였다.

• Design & Manufacturing
• /
• 제16권4호
• /
• pp.1-6
• /
• 2022

Fine blanking is a press processing technology that can process most of the product thickness into a smooth surface with a single stroke. In this fine blanking process, shear is an essential step. The punches and dies used in the shear are subjected to impacts of tens to hundreds of gravitational accelerations, depending on the type and thickness of the material. Therefore, among the components of the fine blanking mold (dies), punches and dies are the parts with the shortest lifespan. In the actual production site, various types of tool damage occur such as wear of the tool as well as sudden punch breakage. In this study, machine learning algorithms were used to predict these problems in advance. The dataset used in this paper consisted of the signal of the vibration sensor installed in the tool and the measured burr size (tool wear). Various features were extracted so that artificial intelligence can learn effectively from signals. It was trained with 5 features with excellent distinguishing performance, and the SVM algorithm performance was the best among 33 learning models. As a result of the research, the vibration signal at the time of imminent tool replacement was matched with an accuracy of more than 85%. It is expected that the results of this research will solve problems such as tool damage due to accidental punch breakage at the production site, and increase in maintenance costs due to prediction errors in punch exchange cycles due to wear.

• 한국터널지하공간학회 논문집
• /
• 제20권2호
• /
• pp.453-467
• /
• 2018

쉴드 TBM 공사에서 굴진율 예측과 마모량 예측은 설계 및 시공 단계에서 공사비와 공기를 추정하는데 매우 중요한 요소이다. 암반지반용 TBM의 경우 실험이나 축적된 현장 data를 기반으로 CSM 모델, NTNU 모델 등이 커터 마모량부터 굴진율 예측까지 널리 사용되고 있으나, 토사지반용 TBM은 지반의 복잡성과 정확한 실험방법의 부재로 인해 이를 정확하게 예측할 수 있는 모델이 없는 실정이다. 본 연구에서는 기존에 존재하는 토사지반용 TBM 실험장치들의 단점을 개선하여 TBM 굴착과정을 모사한 실험 장치(Soil Abrasivity Penetration Test, SAPT)를 개발하였다. 회전당 관입 깊이, RPM, 첨가재(foam) 배합비 및 농도 등의 TBM 굴진에 영향을 미치는 주요 변수들에 대한 시험을 실시하여 추력, 토크 등의 변화를 살펴보고 마모량을 측정하였다. 모래(규사) 70%와 점토(일라이트) 30%로 조성된 인공시료에 대한 실험 결과 foam 배합비가 굴진성능과 마모량에 주요한 영향을 끼치는 것으로 나타났다.

• 한국생산제조학회지
• /
• 제5권2호
• /
• pp.38-45
• /
• 1996

This study was focused on the prediction of the approprite tool life by clarifying the correlation between progressive drill wear and AE signal. on drilling SM45C the following results have been obtained; RMSAE, AE CUM-CNTS had a tendency to increase slowly according to wear size, at 1000rpm, 150mm/min However, these increased suddenly in the range of 0.20~0.22mm wear, about 102 holes and had a tendency to go up and down until the drilling was impossible. The sudden increase of AE signals shows that something is wrong and it is closely connected with drill wear and chipping. It also makes the working surface bad From the above results, AE signals could be used to monitor the drill's condition and to determine the right time to change tools.

• 소성∙가공
• /
• 제31권2호
• /
• pp.103-108
• /
• 2022

Systematic quality control based on real time data is required for modern factories. This study introduced a method of predicting punch wear in the trimming process of automobile parts. Based on monitoring data of the mass production process using a bolt-type piezo sensor, it was shown that precursor symptoms of die wear could be predicted from the change in load pattern with respect to production volume. The load pattern that changed according to the wear of the die was verified by numerical analysis.

• Tribology and Lubricants
• /
• 제10권4호
• /
• pp.13-26
• /
• 1994

Effects of resin contents, number of carbonization, graphitization, sliding speed, and oxidation on friction and wear behavior of carbon/carbon composite materials were investigated. Friction and wear tests were carried out under various sliding conditions. An experimental setup was designed and built in the laboratory. Stainless steel disks were used as the counterface material. Friction coefficient, emperature, and wear factor were measured with a data acquisition system. Wear surfaces were observed by the scanning electron microscope. It has been shown that the average friction coefficient was increased with the sliding speed in the range of 1.43~6.10 m/s, but it as decreased in the range of 6.10~17.35 m/s. Specimens prepared by different numbers of carbonization. showed variations in friction coefficient and friction coefficient of the graphitized specimen was the highest. Friction coefficients depended on contribution of the plowing and adhesive components. As the number of carbonization was increased, wear factor was reduced. Wear factor of the graphitized specimens dropped further. In the case of graphitized specimens, sliding speed had a large influence on wear behavior. When the tribological experiments were conducted in nitrogen atmosphere, the wear factor was decreased to two thirds of the wear factor obtained in air. It is obvious that the difference was affected by oxidation. Results of friction and wear tests were applied to a neural network system based on the backpropagation algorithm. A neural network may be a valuable tool for prediction of tribological behavior of the carbon/carbon composite material if ample data are present.