• Journal of Welding and Joining
• /
• v.12 no.2
• /
• pp.11-19
• /
• 1994

레이저 가공기술은 재료가공 분야에서 넓은 응용분야를 가지고 있으며, 특히 절단, 용접, 열처리 등의 가공분야에서 고정밀도와 자동화의 용이성으로 인해 생산성이 높은, 고부가가치의 첨단응용 기술로 부각되고 있다. 특히 레이저절단은 타 절단법에 비교되는 절단정도, 열영향, 생산성, 작업 환경등의 각종 우위성으로 박판 및 후판절단분야에서 급속한 보급을 보이기 시작하였다. 현재 대 부분의 레이저 가공기는 CNC화 되어가고 있는 추세이며, 레이저 절단의 경우 생산성증대 및 고 정밀화를 위하여 CAD/CAM인터페이스에 의한 자동화가 필연적인 상황이다. 뿐만아니라 고출력 레이저 발전기를 가공 기본체에 탑재한 탑재형 레이저가공기의 출현으로 대형부재의 절단이 가능 하게 되었으며, 더불어 절단공정의 무인화를 지향하는 각종 시스템이 개발되고 있다. 이와 같은 무인화, 생산성증대, 작업시간단축과 러닝 코스트 및 재료의 절감을 위한 노력의 일환으로 컴 퓨터에 의한 자동 및 반자동 네스팅 시스템의 개발을 들 수 있다. 레이저에 의한 2차원 절단응 용분야에서의 네스팅작업은 설계가 끝난 각 부품의 절단작업의 전단계로서 수행되며, 일반적으로 네스팅공정이 완료되면 절단경로를 결정하고 가공조건과 함께 수치제어공작기계의 제어에 필요한 NC코드를 생성하게 된다. 최근에는 이와 같은 네스팅 시스템이 일부 생산현장에 적용되고 있 으나 이러한 시스템들의 대부분이 외국에서 개발된 것을 수입하여 사용하는 실정이다. 2차원 패턴의 최적자동배치문제는 비단 레이저 절단과 같은 열가공 분야에서 뿐만 아니라 블랭킹 금형, 의류, 유리, 목재등 여러분야에서 응용이 가능하며 패키지의 국산화가 시급한 실정이다. 네스 팅작업은 적용되는 분야에 따라 요구사항과 구속조건이 달라지며 이로 인해 알고리즘과 자료구 조도 달라지게 되나 공통적인 목표는 주어진 영역안에서 겹침없이 배치하면서 버림율을 최소화 하는 것이다. 지난 10여년간 여러 산업의 응용분야에서는 네스팅시스템의 도입이 활발하게 이 루어지고 있는데 수동에 반자동 및 자동에 이르기까지 다양하나 자동네스팅시스템의 경우 배치 효율의 신뢰성이 비교적 부족하기 때문에 아직까지는 생산현장에서 기피하는 실정이다. 배치알 고리즘의 관점에서 볼 때 이러한 문제들은 NP-complete문제로 분류하며 제한된 시간안에 최적의 해를 구하기가 가능한 조합 최적화 문제로 알려져 있다. 따라서 이 글에서는 레이저 절단분야 에서의 네스팅시스템에 관한 개요와 최근의 연구동향 그리고 몇 가지 전형적인 네스팅 알고리 즘들을 소개하고 비교분석을 통해 개선점을 간략하게 논의하고자 한다.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.14 no.5
• /
• pp.1059-1068
• /
• 1990

The minimum weight design for the simply-supported eccentrically stiffened plates subjected to combined loads is studied according to the stiffening configuration. The optimal programming is accomplished by formulating the design requirements in terms of a mathematical programming problem, and by using the gradient projection algorithm. The Huber type equilibrium equation is used as the governing equation for the overall buckling. The overall buckling of stiffened plates and the local buckling of the unstiffened plate between stiffeners and the stiffeners themselves are used as behavior constraints. Results of design examples for the orthogonally stiffening case compared with those of the other study support that the present study is feasible. Design examples for the symmetrically oblique stiffening case are presented and the results indicate that a significant improvement in design efficiency may be achieved through symmetrically oblique stiffening compared to the orthogonal stiffening under the combined loading condition.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.6
• /
• pp.635-642
• /
• 2003

Reinforcement learning got successful results in a lot of applications such as control and scheduling. Various function approximation methods have been studied in order to improve the learning speed and to solve the shortage of storage in the standard reinforcement learning algorithm of Q-Learning. Most function approximation methods remove some special quality of reinforcement learning and need prior knowledge and preprocessing. Fuzzy Q-Learning needs preprocessing to define fuzzy variables and Local Weighted Regression uses training examples. In this paper, we propose a function approximation method, Fuzzy Q-Map that is based on on-line fuzzy clustering. Fuzzy Q-Map classifies a query state and predicts a suitable action according to the membership degree. We applied the Fuzzy Q-Map, CMAC and LWR to the mountain car problem. Fuzzy Q-Map reached the optimal prediction rate faster than CMAC and the lower prediction rate was seen than LWR that uses training example.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.7
• /
• pp.875-880
• /
• 2007

Scientific challenges in the field of MR(magnetorheological) fluids and devices consist in the development of MR devices, the mathematical modeling and simulation of MR devices, and the development of (optimal) control algorithm for MR device systems. To take a maximum advantage of MR fluids in control applications a reliable mathematical model, which predicts their nonlinear characteristics, is needed. A inverse model of the MR device is required to calculate current(or voltage) input of MR damper, which generates required damping force. In this paper, we implemented test a bench for shear mode rotary MR damper and laboratory tests were performed to study the characteristics of the prototype shear-mode rotary MR damper. The direct identification and inverse dynamics modeling for shear mode rotary MR dampers using multi-layer neural networks are studied.

• Journal of the Korean Society for Railway
• /
• v.18 no.4
• /
• pp.279-288
• /
• 2015

Using the Vehicle Modeling Function, which can model various 3D nose shapes, nose shape optimization is performed to reduce the aerodynamic drag of the KTX Sancheon. 2D characteristic shapes of the KTX Sancheon nose were extracted and a base model of the KTX Sancheon was constructed for design optimization using the Vehicle Modeling Function. The design space was constructed with the base model and does not violate the shape constraints of commercial trains. Through nose shape optimization with the Broyden-Fletcher-Goldfarb-Shanno algorithm, the aerodynamic drag of the optimized shape was reduced by 6% compared to that of the base model. The longer nose and sharper edge of the optimized shape weaken the vortices behind the last car and can reduce the aerodynamic drag.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.2
• /
• pp.55-62
• /
• 2016

Recently, users choosing a restaurant basedon information provided by blogs are increasing significantly. However, those of most blogs are unreliable since domestic restaurant blogs are occupied by advertising postings written by 'power bloggers'. Thus, in order to ensure the reliability of blogs, it is necessary to filter the advertising blogs which are sometimes false or exaggerated. In this paper, we propose the method of distinguishing the advertising blogs utilizing an automatic classification technique. In the proposed technique, we first manually collected advertising restaurant blogs, and then analyzed features which are commonly found in those blogs. Using the extracted features, we determined whether a given blog is advertising one applying automatic classification algorithms. Additionally, we select the features and the algorithm which guarantee optimal classification performance through comparative experiments.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.2
• /
• pp.236-241
• /
• 2009

Information integration for distributed and heterogeneous data sources is still an open challenging, and schema matching is critical in this process. This paper presents and approach to automatic elements matching between XML application schemas using similarity measure and relaxation labeling. The semantic modeling of XML application schema has also been presented. The similarity measure method considers element categories and their properties. In an effort to achieve an optimal matching, contextual constraints are used in the relaxation labeling method. Based on the semantic modeling of XML application schemas, the compatible constraint coefficients are devised in terms of the structures and semantic relationships as defined in the semantic model. To examine the effectiveness of the proposed methods, an algorithm for XML schema matching has been developed, and corresponding computational experiments show changes of calculated values.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.6
• /
• pp.665-670
• /
• 2014

A SVM is a kind of binary classifier in order to find optimal hyperplane which separates training data into two groups. Due to its remarkable performance, the SVM is applied in various fields such as inductive inference, binary classification or making predictions. Also it is a representative black box model; there are plenty of actively discussed researches about analyzing trained SVM classifier. This paper conducts a study on a method that is automatically detecting the line-shaped echoes, sun strobe echo and radial interference echo, using the SVM algorithm because the line-shaped echoes appear relatively often and disturb weather forecasting process. Using a spatial clustering method and corrected reflectivity data in the weather radar, the training data is made up with mean reflectivity, size, appearance, centroid altitude and so forth. With actual occurrence cases of the line-shaped echoes, the trained SVM classifier is verified, and analyzed its characteristics using the decision tree method.

• Journal of Drive and Control
• /
• v.12 no.1
• /
• pp.1-8
• /
• 2015

The stability of friction characteristics and thermal management for a dry type dual clutch transmission (DCT) are inferior to those of a wet clutch. Too high temperature resulting from frequent engagement of DCT speeds up degradation or serious wear of the pressure plate or burning of the clutch disk lining. Even though it is significantly important to estimate the temperature of a dry double clutch (DDC) in real-time, few meaningful study of the thermal model of DDC has been known yet. This study presented a thermal analytical model of lumped parameters for a DDC by analyzing its each component firstly. Then a series of experimental test was carried out on the test bench with a patented temperature telemetry system to validate the proposed thermal model. The thermal model, whose optimal parameter values were found by optimization algorithm, was also simulated on the experimental test conditions. The simulation results of DDC temperature show consistency with the experiment, which validates the proposed thermal model of DDC.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.6
• /
• pp.1294-1301
• /
• 2014

In this paper, separate probabilistic distribution models for voiced and unvoiced speech are estimated and utilized to improve speaker recognition performance. Also, in addition to the conventional mel-frequency cepstral coefficient, skewness, kurtosis, and harmonic-to-noise ratio are extracted and used for voiced speech intervals. Two kinds of scores for voiced and unvoiced speech are linearly fused with the optimal weight found by exhaustive search. The performance of the proposed speaker recognizer is compared with that of the conventional recognizer which uses mel-frequency cepstral coefficient and a unified probabilistic distribution function based on the Gassian mixture model. Experimental results show that the lower the number of Gaussian mixture, the greater the performance improvement by the proposed algorithm.