• 산업경영시스템학회지
• /
• 제44권4호
• /
• pp.12-22
• /
• 2021

This article suggests the machine learning model, i.e., classifier, for predicting the production quality of free-machining 303-series stainless steel(STS303) small rolling wire rods according to the operating condition of the manufacturing process. For the development of the classifier, manufacturing data for 37 operating variables were collected from the manufacturing execution system(MES) of Company S, and the 12 types of derived variables were generated based on literature review and interviews with field experts. This research was performed with data preprocessing, exploratory data analysis, feature selection, machine learning modeling, and the evaluation of alternative models. In the preprocessing stage, missing values and outliers are removed, and oversampling using SMOTE(Synthetic oversampling technique) to resolve data imbalance. Features are selected by variable importance of LASSO(Least absolute shrinkage and selection operator) regression, extreme gradient boosting(XGBoost), and random forest models. Finally, logistic regression, support vector machine(SVM), random forest, and XGBoost are developed as a classifier to predict the adequate or defective products with new operating conditions. The optimal hyper-parameters for each model are investigated by the grid search and random search methods based on k-fold cross-validation. As a result of the experiment, XGBoost showed relatively high predictive performance compared to other models with an accuracy of 0.9929, specificity of 0.9372, F₁-score of 0.9963, and logarithmic loss of 0.0209. The classifier developed in this study is expected to improve productivity by enabling effective management of the manufacturing process for the STS303 small rolling wire rods.

• 소성∙가공
• /
• 제28권6호
• /
• pp.321-327
• /
• 2019

In this study, a hot and cold forging process was investigated to produce a complex-shaped hub of dual clutch transmission with low material loss and high productivity. The process was designed by the commercial finite element (FE) analysis program, DEFORM-2D (hot forging) and 3D (cold forging). And, the material flow and ductile fracture characteristics were studied to check the surface crack initiation of the specimen. The simulation results indicated that the proposed process could manufacture the complex-shaped hub with no surface crack and high-efficiency compared to the conventional machining process. For verification the numerical results, the hub of the SCM440 was fabricated by the proposed process and the mechanical properties and microstructure evolution were studied. It was demonstrated that the suggested hot and cold forging process might be useful in producing the key components of the automobile industry as a high-efficiency and environmentally friendly process.

• 한국공작기계학회논문집
• /
• 제14권2호
• /
• pp.1-7
• /
• 2005

In order to determine the wear Properties of PECVD ceramic coatings, wear process was evaluated using the coated pin of Falex Tribosystem. Coating materials deposited wear the TiC, TiN and Ti(CN). An experimental process was established to determine the tribological characteristics of friction and wear behavior under the variation of applied load, temperature and sliding distance by the Falex test machine. The experimental results indicate that TiN coating compared with TiC coating on e materials have e excellent friction and wear characteristics. However TiC coating compared i친 TiN coatings have a low friction coefficient with steel and good thermal stability, and Ti(CN) has the excellent anti-wear properly as well as the superiority of extreme pressure property. Compound coating compared wi simple coatings show improved tribological characteristics.

• 한국농업기계학회:학술대회논문집
• /
• 한국농업기계학회 2003년도 하계 학술대회 논문집
• /
• pp.232-237
• /
• 2003

Sound has been rated negatively in many industrial areas as the noise to be removed. In the agricultural process such as rice milling based on grinding the surface of the rice, however, it may be anticipated that the sound can contain the some information on the process. In this research, on the assumption that variation of the shape of the rice kernel during milling may bring the change in frequency characteristic of sound signal, we tried to investigate the relationship between the milling degree in the grinding test mill and the resonance and anti-resonance frequencies (the frequencies at which the sound power has the extreme values). (omitted)

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2001년도 추계학술대회논문집A
• /
• pp.772-777
• /
• 2001

The gas-assisted injection molding process is often perceived to be unpredictable, because of the extreme sensitivity of the gas. Since a slight change in design or process parameters can significantly change the resulting gas penetration, few designers and molders have the level of experience with the new gas-assisted injection molding process required for the development of new parts. This paper is concerned with the unified molding for a thick cosmetic chest by using gas-assisted injection molding. CAE analysis was carried out to design the part and the gas channel without inducing sink marks. And based on the part weight measurement, the processing parameters to control gas penetration percentage were chosen through the method of design of experiments. A thick cosmetic chest was successfully produced using the gas assist technology. The sink mark issue associated with the conventional injection molded parts was resolved. Weight savings and cycle-time reduction were also achieved.

• 마이크로전자및패키징학회지
• /
• 제23권4호
• /
• pp.25-29
• /
• 2016

The semiconductor industry has been developed mainly by micronization process due to many advantages of miniaturization of devices. Mass production of semiconductors of 10 nm class has been started recently, and it is expected that the technology generation of 5 nm & 7 nm technology will come. However, excessive linewidth reduction affects physical limits and device reliability. To solve these problems, new process technology development and new concept devices are being studied. In this review, we introduce the next generation technology and introduce the advanced research for the new concept device.

• 소성∙가공
• /
• 제10권5호
• /
• pp.402-410
• /
• 2001

The gas-assisted injection molding process is often perceived to be unpredictable, because of the extreme sensitivity of the gas. Since a slight change in design or process parameters can significantly change the resulting gas penetration, few designers and molders have the level of experience with the new gas-assisted injection molding process required for the development of new parts. This paper is concerned with the unified molding for a thick cosmetic chest by using gas-assisted injection molding. CAE analysis was carried out to design the part and the gas channel without inducing sink marks. And based on the part weight measurement, the processing parameters to control gas penetration percentage were chosen through the method of design of experiments. A thick cosmetic chest was successfully produced using the gas assist technology. The sink mark issue associated with the conventional injection molded parts was resolved. Weight savings and cycle-time reduction were also achieved.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제14권1호
• /
• pp.61-69
• /
• 2014

With the extreme ultraviolet (EUV) lithography, the performance limit of chemically amplified resists has recently been extended to 16- and 11-nm nodes. However, the line edge roughness (LER) and the line width roughness (LWR) are not reduced automatically with this performance extension. In this paper, to investigate the impacts of the EUVL mask and the EUVL exposure process on LER, EUVL is modeled using multilayer-thin-film theory for the mask structure and the Monte Carlo (MC) method for the exposure process. Simulation results demonstrate how LERs of the mask transfer to the resist and the exposure process develops the resist LERs.

• 한국분말야금학회:학술대회논문집
• /
• 한국분말야금학회 2006년도 Extended Abstracts of 2006 POWDER METALLURGY World Congress Part 1
• /
• pp.205-206
• /
• 2006

A ball-shape alumina arc-tube for low-wattage lamp was developed by the PIM process. An ultra high purity translucentgrade alumina powder was used. In injection molding process, a hot-runner type mold was developed. The translucent-grade alumina powder was extremely sensitive to contamination so that the injection molding condition and atmosphere control in the furnace should be taken care of with extreme caution. Contamination sources were pinpointed with EPMA. The arc-tube was molded in half and two halves were bonded in the middle by a new bonding technique at room temperature developed in this study.

• 한국지능시스템학회논문지
• /
• 제19권4호
• /
• pp.574-579
• /
• 2009

복잡한 비선형 회귀문제들에서 최적의 신경망을 구축하기 위해서는 구조의 선정 및 노이즈에 의한 과잉학습(overtraining)등에 따른 많은 문제들이 있다. 본 논문에서는 flexible incremental 알고리즘을 이용하여 단계적으로 최적의 신경망을 구축하는 방법을 제안한다. Flexible incremental 알고리즘은 예측 잔류오차를 최소화하기 위해 단계적으로 추가되어지는 은닉노드 개수를 검증데이터를 이용하여 신축성 있게 조절하고, 빠른 학습을 위하여 ELM (Extreme Learning Machine)을 이용한다. 제안된 방법은 신경망의 구축과정에서 사용자의 어떠한 관여 없이도 빠른 학습과 적은 수의 은닉노드들에 의한 범용 근사화 (universal approximation)가 가능한 신경망의 구축이 가능한 장점을 가지고 있다. 다양한 종류의 벤치마크 데이터들을 이용한 실험 결과를 통하여 제안된 방법이 실제 회귀문제들에서 우수한 성능을 가짐을 확인하였다.