• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06c
• /
• pp.482-486
• /
• 2008

본 논문에서는 하드 디스크 드라이브(Hard Disk Drive, HDD) 생산 공정 과정에서 발생할 수 있는 불량 HDD의 결함 분포에 대해서 패턴을 자동으로 분류해주는 기법을 제시한다. 이를 위해서 표준 패턴 클래스로 분류되어 있는 불량 HDD의 각 클래스의 확률 모델을 GMM(Gaussian Mixture Model)로 가정한다. 실험은 전문가에 의해 분류된 실제 HDD 결함 분포로부터 5가지의 특징 값들을 추출한 후, 결함 분포의 클래스를 표현할 수 있는 GMM의 파라미터(Parameter)를 학습한다. 각 모델의 파라미터를 추정하기 위해 EM(Expectation Maximization) 알고리즘을 사용한다. 학습된 GMM의 분류 테스트는 학습에 사용되지 않은 HDD 결함 분포에서 5가지의 특징 값을 입력 값으로 추정된 모델들의 파라미터 값에 의해 사후 확률을 구한다. 계산된 확률 값 중 가장 큰 값을 갖는 모델의 클래스를 표준 패턴 클래스로 분류한다. 그 결과 제시된 GMM을 이용한 HDD의 패턴 분류의 결과 96.1%의 정답률을 보여준다.

• Journal of the Korean Society for Precision Engineering
• /
• v.22 no.7 s.172
• /
• pp.122-129
• /
• 2005

Shoe with leather upper such as safety and golf shoe requires a roughing process where the upper is roughed fur helping outsole to be cemented well. It is an important and basic process for production of leather shoe but is not automated yet. Thus, there are problems that the defect rate is high and the quality of roughed surface is not uniform. In order to solve such problems, the interest in automation of roughing process is being increased and this paper introduces CAM system for 5-axis automatic roughing machine as one part of automation of roughing process. The CAM system developed interpolates a B-spline curve using points measured from the Roughing Path Measurement System. The B-spline curve is used to generate the tool path and orientation data fer a roughing tool which has not only stiffness but also flexibility to rough the inclined surface efficiently. For productivity, the upper of shoe is machined by side of the roughing tool and tool offset is applied to the roughing tool for machining of inclined surface. The generated NC code was applied to 5-axis polishing machine for the test. The upper of shoe was roughed well along the roughing path data from CAM and the roughed surface was proper fur cementing of the outsole.

• Journal of Korean Society of Water and Wastewater
• /
• v.29 no.1
• /
• pp.123-131
• /
• 2015

A CCTV inspection method has been widely used to assess sewer condition and performance, but Korea lacks a proper decision support system for prioritizing sewer repair and rehabilitation (R&R). The objective of this paper is to introduce the results that we have developed in the Sewer Condition Assessment and Rehabilitation Decision-making (SCARD) Program using MS-EXCEL. The SCARD-Program is based on a standardized defect score for sewer structural and hydraulic assessment. Priorities are ranked based on risk scores, which are calculated by multiplying the sewer severity scores by the environmental impacts. This program is composed of three parts, which are decision-making for sewer condition and performance assessment, decision-making for sewer R&R priority assessment, and decision-making for optimal budget allocation. The SCARD-Program is useful for decision-makers, as it enables them to assess the sewer condition and to prioritize sewer R&R within the limited annual budget. In the future, this program logic will applied to the GIS-based sewer asset management system in local governments.

• Design & Manufacturing
• /
• v.2 no.4
• /
• pp.54-61
• /
• 2008

Warpage, which is one of the molding trouble, acts as possible factor which results in defect in assembly. In this study, a mold was designed to produce specimens with rib parallel to flow direction, specimens with rib perpendicular to flow direction and specimens without rib. This work researched change of warpage according to injection molding condition such as injection pressure, packing pressure, packing time, resin temperature, mold temperature in non-crystalline resins(PC, ABS), crystalline resins(PP, PA66), and 30% glass fiber reinforced-resins(PC, ABS, PP, PA66).Specimens with rib and Crystalline resins show more warpage than specimens without rib and non-crystalline resins, respectively. Glass fiber reinforced-resins and specimens with rib parallel to flow direction show smaller warpage than conventional resins and specimens with rib perpendicular to flow, respectively. Specimens with rib and specimens without rib show reduced warpage as packing time increases. In addition, warpage increase as resin temperature increases. It is found that CAE shows similar tendency with experiment as packing time, resin temperature. when the rib is caused, warpage will reduce and prevent the transformation. product of a irregular form occurs warpage. In the study It'll be basic data that product occurs warpage, preferablity.

• Journal of Korea Foundry Society
• /
• v.30 no.6
• /
• pp.241-246
• /
• 2010

The viscosity of foam metal is necessary to get the pores, but it is difficult to manufacture net-shape foam, because the fluidity decreases by increasing viscosity. In this study, the A356 alloy which has good fluidity and less defect was selected and fabricated to foam metal. To understand about effect of thickening agent on foaming and mechanical properties, quantity of thickening agent was changed. The pore size, porosity and distribution of foam metal were measured by i-solution program. And compression test were performed by UTM. In case of 3.0wt% Ca in thickening agent, it is found that most of foam consist of homogeneous shape, and the growth height had the highest value of 204 mm in the all fabricated foams. The porosity was 93% and compressive strength was 3.1 MPa. In the microstructure, the $Al_2Si_2Ca$ intermetallic compound and Ti were observed. The vickers hardness value rose with increasing viscosity value.

• Nuclear Engineering and Technology
• /
• v.42 no.5
• /
• pp.546-551
• /
• 2010

The objective of this research is to estimate the crack location and size of a carbon steel pipe by using a laser ultrasound guided wave for the wall thinning evaluation of an elbow. The wall thinning of the carbon steel pipe is one of the most serious problems in nuclear power plants, especially the wall thinning of the carbon steel elbow caused by Flow-Accelerated Corrosion (FAC). Therefore, a non-destructive inspection method of elbow is essential for the nuclear power plants to operate safely. The specimens used in this study were carbon steel elbows, which represented the main elements of real nuclear power plants. The shape of the wall thinning was an oval with a width of 120mm, a length of 80mm, and a depth of 5mm. The L(0,1) and L(0,2) modes variation of the ultrasound guided wave signal is obtained from the response of the laser generation/air-coupled detection ultrasonic hybrid system represent the characteristics of the defect. The trends of these characteristics and signal processing were used to estimate the size and location of wall thinning.

• Journal of Welding and Joining
• /
• v.35 no.3
• /
• pp.44-51
• /
• 2017

In the present study, AA 5083-O plates are joined by friction stir welding technique. A universal milling machine was used to perform the welding process of the work-pieces which were fixed on the proper position by a vice. The joints were friction stir welded by two tools with different pin profiles; cylindrical threaded pin and tapered smooth one at different rotational speed values; 400 rpm and 630 rpm, and different welding speed values; 100 mm/min and 160 mm/min. During FSW of each joint, the temperature was measured by infra-red thermal image camera. The welded joints were inspected by visually as well as by the macro- and microstructure evolutions. Furthermore, the joints were tested for measuring the hardness and the tensile strength to study the effect of changing the FSW parameters on the mechanical properties. The results show that increasing the rotational speed results in increasing the peak temperature, while increasing the welding speed results in decreasing the peak temperature for the same tool pin profile. Defect free welds were obtained at lower rotational speed by the threaded tool profile. Moreover, the threaded tool pin profile gives superior mechanical properties at lower rotational speed.

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.450-453
• /
• 2000

In this paper, an automated segmentation algorithm is proposed for MR brain images using T1-weighted, T2-weighted, and PD images complementarily. The proposed segmentation algorithm is composed of 3 steps. In the first step, cerebrum images are extracted by putting a cerebrum mask upon the three input images. In the second step, outstanding clusters that represent inner tissues of the cerebrum are chosen among 3-dimensional (3D) clusters. 3D clusters are determined by intersecting densely distributed parts of 2D histogram in the 3D space formed with three optimal scale images. Optimal scale image best describes the shape of densely distributed parts of pixels in 2D histogram. In the final step, cerebrum images are segmented using FCM algorithm with it’s initial centroid value as the outstanding cluster’s centroid value. The proposed segmentation algorithm complements the defect of FCM algorithm, being influenced upon initial centroid, by calculating cluster’s centroid accurately And also can get better segmentation results from the proposed segmentation algorithm with multi spectral analysis than the results of single spectral analysis.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2008.06a
• /
• pp.121-122
• /
• 2008

Lithium Niobate (LN:LiNbO3) is a compound of niobium, lithium and oxygen. The characteristics of LN are piezoelectricity, ferroelectricity and photoelectricity, and which is widely used in surface acoustic wave (SAW). To manufacture LN device, the LN surface should be a smooth surface and defect-free because of optical property, but the LN material is processed difficult by traditional processes such as grinding and mechanical polishing (MP) because of its brittleness. To decrease defects, chemical mechanical polishing (CMP) was applied to the LN wafer. In this study, the suitable parameters scuh as pressure and relative velocity, were investigated for the LN CMP process. To improve roughness, the LN CMP was performed using the parameters that were the highest removal rate among process parameters. And, evaluation of optical property was performed by the optical reflectance and non-linear characteristic.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.31 no.7
• /
• pp.450-454
• /
• 2018

Because of the rapidly changing environment and high uncertainties, the semiconductor industry is in need of appropriate forecasting technology. In particular, both the cost and time in the test process are increasing because the process becomes complicated and there are more factors to consider. In this paper, we propose a prediction model that predicts a final "good" or "bad" on the basis of preconditioning test data generated in the semiconductor test process. The proposed prediction model solves the classification and regression problems that are often dealt with in the semiconductor process and constructs a reliable prediction model. We also implemented a prediction model through various machine learning algorithms. We compared the performance of the prediction models constructed through each algorithm. Actual data of the semiconductor test process was used for accurate prediction model construction and effective test verification.