• 대한음성학회:학술대회논문집
• /
• 대한음성학회 2003년도 5월 학술대회지
• /
• pp.79-82
• /
• 2003

In this paper, we proposed Backgorund Model Set algorithm for the speaker verification to improve the shortcoming of calculating process in conventional confidence measure(CM). CM is to display relative likelihood between recognized models and unrecognized models. Unrecognized models is known as antiphone models. Calculate probability and standard deviation using all phonemes at process that compose antiphone model. At this process, antiphone CM brought bad result. Also, recognition time increases. In order problem, we studied about method to reconstitute average and standard deviation taking BMS algorithm using antiphoneme that near phoneme of CM calculation.

• Current Optics and Photonics
• /
• 제4권6호
• /
• pp.516-523
• /
• 2020

The normal low-cost manufacturing process for freeform vehicle side mirrors causes deviations from the design curvature. Here, an improved manufacturing process is proposed, combining photorealistic ray-traced visualization of each deviation and subsequent analysis of its deviated reflective scene compared to that of the original design. The proposed mechanism successfully highlights the overlap and mismatch regions of deviated reflected scenes with reference to the desired reflective scenes. We benchmarked the robustness of freeform mirror manufacturing by evaluating the 10, 20, and 30% root-mean-square (RMS) deviated curvature, and concluded that the acceptable deviation needs to be below RMS20% to avoid mismatched regions that could mislead the driver.

• 전기전자학회논문지
• /
• 제27권3호
• /
• pp.327-332
• /
• 2023

전통적인 실리콘 기반 반도체 소자 보다 높은 성능과 다양한 활용성으로 차세대 반도체 후보로 높은 관심 받고 있는 CNTFET은 CNT 배치와 같은 CNTFET만의 고유한 공정 편차가 아직 성숙되지 않아 상용화에 어려움을 겪고 있다. 이러한 어려움을 극복하고자 반복적인 회로 구성으로 공정 편차의 영향을 적게 받는 회로를 MOSFET-CNTFET 기반 하이브리드 회로로 구현하여 CNTFET 의 장점을 취하고 단점을 보완하고자 하는 수많은 연구들이 지속적으로 수행되어 왔다. 본 논문에서는 하이브리드 SRAM의 성능이 기존의 MOSFET SRAM 또는 CNTFET SRAM에 존재하는 반도체 공정 변화에 의해 얼마나 변화될 수 있는지를 비교하였다. 시뮬레이션 결과, CNT 밀도를 32nm 당 7개에서 9개 사이로 유지할 수 있다면, hybrid SRAM은 기존 MOSFET SRAM보다 읽기 동작에서 그리고 쓰기 동작에서 공정 편차에 대한 강건성이 각각 약 2.6배 그리고 약 1.1배 있음을 보여준다.

• Journal of Mechanical Science and Technology
• /
• 제18권7호
• /
• pp.1074-1085
• /
• 2004

There are many intermediate web guides in cold rolling mills process such as CRM (cold rolling mill), CGL (continuous galvanizing line), EGL (electrical galvanizing line) and so on. The main functions of the web guides are to adjust the center line of the web (strip) to the center line of the steel process. So they are called CPC (center position control). Rapid process speed cause large deviation between the center position of the strip and the process line. Too much deviation is not desirable. So the difference between the center position of the strip and the process line should be compensated. In general, the center position control of the web is obtained by the hydraulic driver and electrical controller. In this paper, we propose modelling and several controller designs for web-guide systems. We model the web and guide by using geometrical relations of the guide ignored the mass and stiffness of the web. To control the systems, we propose PID controllers with their gains tuned by the Ziegler-Nichols method, the H$\_$$\infty$/ controller model-matching method, and the coefficient diagram method (CDM). CDM is modified for high order systems. The results are verified by computer simulations.

• Journal of the Korean Data and Information Science Society
• /
• 제18권3호
• /
• pp.773-782
• /
• 2007

The statistical process control charts are very extensively used for monitoring of process mean, deviation, defect rate or reject rate. In this paper we consider a control chart to monitor the process reject rate in the high yield process, which is based on the observed cumulative probability of the number of items inspected until r defective items are observed. We first propose selection of the optimal value of r in the CPC-r charts, and also consider the usefulness of the chart in high yield process such as semiconductor or TFT-LCD manufacturing process.

• 한국환경과학회지
• /
• 제30권12호
• /
• pp.983-991
• /
• 2021

The role of the distribution basin role is to apportion incoming raw water to the primary sedimentation basin as part of the water treatment process. The purpose of this study was to calculate the amount of water in the distribution basin using computational fluid dynamics (CFD) analysis and to find a way to improve any non-uniformity. We used the Taguchi method and the minitab tool as optimization methods. The results of the CFD calculation showed that the distribution flow had a deviation of 5% at the minimum inflow, 10% at the average inflow, and 22% at the maximum inflow. At maximum flow, the appropriate heights of the 7 weirs(C, D, A, B, E, F, G) were 40 mm, 20 mm, 20 mm, 0, 0, 0, and 20 mm, respectively, according to the Taguchi optimization tool. Here, the maximum deviation of the distribution amount was 9% and the standard deviation was 23.7. The appropriate heights of the 7 weirs, according to the Minitab tool, were 40 mm, 20 mm, 20 mm, 0, 0, 0, and 20 mm, respectively, for weirs C, D, A, B, E, F, and G. Therefore, the maximum deviation of the distribution amount was 8% and the standard deviation was 17.1, which was slightly improved compared to the Taguchi method.

• 품질경영학회지
• /
• 제31권2호
• /
• pp.183-193
• /
• 2003

A special cause producing temporary deviation in the underlying process can influence on process adjustment in responsive feedback control system. In this paper, the impact of special causes on the EWMA(Exponentially Weighted Moving Average) forecasts and the process adjustment that is based on the EWMA forecasts are derived. For some special causes with patterned type of contamination, the influence of the causes on the output process are explicitly investigated. A data set, contaminated by a special cause of level shift, is analyzed to evaluate the impact numerically.

• 한국CDE학회논문집
• /
• 제18권5호
• /
• pp.359-366
• /
• 2013

In the lower arm design process, a tolerance optimization of the variance of design variables should be preceded before manufacturing process, since it is very cost-effective compared to a strict management of tolerance of products. In this study, a design of experiment (DOE) based on response surface model (RSM) was carried out to find optimized design variables of the lower arm, which can meet a given requirement of probability constraint for the process capability index (Cpk) of the weight and maximum stress. Then, the design space was explored by using the central composite design method, in which the 2nd order Taylor expansion was applied to predict a standard deviation of the responses. The optimal solutions satisfying the probability constraint of the Cpk were found by considering both of the mean value and the standard deviation of the design variables.

• 제어로봇시스템학회논문지
• /
• 제15권6호
• /
• pp.634-640
• /
• 2009

The plate manufacturing processes are composed of the reheating furnace, finishing mill, cooling process and hot leveling. The finishing rolling mill (FM) as a reversing mill has produced the plate steel through multiple pass rolling. The automatic gauge control (AGC) is employed to maintain the thickness tolerance. The high grade products are forming greater parts of the manufacturing and customers are requiring strict thickness margin. For this reason, the advanced AGC method is required instead of the conventional AGC based on the PI control. To overcome the slow response performance of the conventional AGC and the thickness measurement delay, a fuzzy AGC based on the thickness deviation and its trend is proposed in this paper. An embedded controller with the fuzzy AGC has been developed and implemented at the plate mill in POSCO. The fuzzy AGC has dynamically controlled the roll gap in real time with the programmable logic controller (PLC). On line tests have been performed for the general and TMCP products. As the results, the thickness deviation range (maximum - minimum of the inner plate) is averagely from 0.3 to 0.1 mm over the full length. The fuzzy AGC has improved thickness deviation and completely satisfied customer needs.

• KSBB Journal
• /
• 제27권1호
• /
• pp.40-44
• /
• 2012

The laboratory animal group separation is dividing animal population into subgroups, which have similar average and standard deviation values among the subgroups, based on the biological characteristics such as body weight, glucose level in blood, etc. Although group separation is very important and initial step in experimental design, it needs a labor intensive process for researchers because of making similar average and standard deviation values among the subgroups using the raw biological characteristics. To reduce the labor cost and increase the efficiency of animal grouping, we developed a web program named as laboratory animal group separation (LAGS) program. This LAGS uses biological characteristics of population, number of group, and the number of elements per each subgroup as input data. The LAGS automatically separates the population into each subgroup that has similar statistical data such as average and standard deviation values among subgroups. It also provides researchers with the extraordinary data generated in the process of grouping and the final grouping results by graphical display. Through our LAGS, researchers can validate and confirm results of laboratory animal group separation by just a few mouse clicks.