• 대한산업공학회지
• /
• 제32권2호
• /
• pp.74-81
• /
• 2006

In this paper, we deal with the separation of data by concurrently determined, piecewise nonlinear discriminant functions. Toward the end, we develop a new $l_1$-distance norm error metric and cast the problem as a mixed 0-1 integer and linear programming (MILP) model. Given a finite number of discriminant functions as an input, the proposed model considers the synergy as well as the individual role of the functions involved and implements a simplest nonlinear decision surface that best separates the data on hand. Hence, exploiting powerful MILP solvers, the model efficiently analyzes any given data set for its piecewise nonlinear separability. The classification of four sets of artificial data demonstrates the aforementioned strength of the proposed model. Classification results on five machine learning benchmark databases prove that the data separation via the proposed MILP model is an effective supervised learning methodology that compares quite favorably to well-established learning methodologies.

• Journal of applied mathematics & informatics
• /
• 제7권1호
• /
• pp.183-193
• /
• 2000

In this paper we use measure theory to solve a wide range of second-order boundary value ordinary differential equations. First, we transform the problem to a first order system of ordinary differential equations(ODE's)and then define an optimization problem related to it. The new problem in modified into one consisting of the minimization of a linear functional over a set of Radon measures; the optimal measure is then approximated by a finite combination of atomic measures and the problem converted approximatly to a finite-dimensional linear programming problem. The solution to this problem is used to construct the approximate solution of the original problem. Finally we get the error functional E(we define in this paper) for the approximate solution of the ODE's problem.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2005년도 춘계학술발표대회
• /
• pp.333-336
• /
• 2005

웹구현을 위한 통합솔루션 환경과 재사용성 확장성의 확실한 장점을 지닌 COM+ 기술은 많은 연구를 통해 계속된 진화가 이루어지고 있다. 한편, COM+의 유용성을 이해하면서도 제어하고 관리할 수 있는 프로그램 개발과 유지보수 단계에서의 어려움은 상존한다. 본 논문에서는 CR다이어그램을 통해 운용을 모니터링 화하여 디버깅시 구조파악을통한 오류검출이 가능하도록 하였으며, 세부적으로 COM+를 그 실체인 Entity와 속성(Attribute)과 관계(Relationship)로 표현하여 구체화 하였다.

• 한국정밀공학회지
• /
• 제13권1호
• /
• pp.131-142
• /
• 1996

Most industrial robots cannot be off-line programmed to carry out a task accurately, unless their kinematic model is suitably corrected through a calibration procedure. However, normal calibration is an expensive and time-consumming precedure due to the highly accurate measurement equipment required and due to the significant amount of data that must be collected. This paper presents a simple and economic procedure to improve the efficiency of robot calibration especially for arc welding application. To simplify the measurement process, an automotic data measurement algorithm as well as a simple measurement device are developed. Also, a calibration algorithm which can automatically identify the independent model parameters to be estimated is presented. To demonstrated the simplicity and the effectiveness of the procedure, experimental studies and computer simulations are performed and their results are discussed.

• 대한전기학회논문지
• /
• 제38권10호
• /
• pp.805-810
• /
• 1989

This paper presents a method for planning reactive power compensation such as shunt capacitors and reacters so as to maintain bus voltage in acceptable range during steady state operation in power system. The algorithm in this paper decomposes the problem into reactive power planning module for the compensation of bus voltage and load flow module for adjusting the error resulted from the linear approximation. A planning technique is based on linear programming to minimize the amount of added reactive power compensation in each case. Transformer tap settings and generator voltages are adjusted to minimize the compensation. The constraints are the operation limits of the control variables and bus voltages. The result of one sample system is presented to confirm the practical use of the proposed algorithm.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1995년도 추계학술대회 학술발표 논문집
• /
• pp.211-217
• /
• 1995

Conventional linear control fails to provide precise positioning of a control object under the influence of friction, deadzone, saturation, etc. This paper proposes a high-precision control scheme for a precise point-to-point positioning system, called an X-Y table, even under the same influences above. The proposed scheme is composed of a fuzzy precompensator and a PD controller. The fuzzy precompensator is employed to improve the performance of the PD controller. Its fuzzy rules are obtained from experimental evolutionary programming (EP), not from an expert. The effectiveness of the scheme is demonstrated by experiments on the X-Y table. with a positioning error of within 1$\mu\textrm{m}$.

• Industrial Engineering and Management Systems
• /
• 제3권1호
• /
• pp.63-70
• /
• 2004

In this paper, we consider the adjustment of convergence on Color Display Tube (CDT). Convergence is a measure of how well the red, green and blue beams are physically aligned with each other to strike the same area on the screen. When misconvergence (convergence error) occurs, one way of compensating it is to attach several ferrite sheets on the inner part of Deflection Yoke (DY). We suggest an optimization model of misconvergence compensation process and report test results for 81 DY samples. As a result, more than 90% of the samples could be made to satisfy the required convergence criteria.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2003년도 춘계학술대회 논문집 센서 박막재료 반도체 세라믹
• /
• pp.194-199
• /
• 2003

Depth error correction effect for maladjusted stereo cameras with calibrated pixel distance parameter is presented. The proposed neural network technique is the real time computation method based theory of inter-node diffusion for searching the safety distances from the sudden appearance-objects during the work driving. The main steps of the distance computation using the theory of stereo vision like the eyes of man is following steps. One is the processing for finding the corresponding points of stereo images and the other is the interpolation processing of full image data from nonlinear image data of objects. All of them request much memory space and time. Therefore the most reliable neural-network algorithm is derived for real-time matching of objects, which is composed of a dynamic programming algorithm based on sequence matching techniques.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년도 제37회 하계학술대회 논문집 C
• /
• pp.1589-1590
• /
• 2006

A method to determine functional relationships between the variable physical dimensions of an antenna and the antenna performance characteristics is presented. By applying multi-expression programming (MEP) to this data set, optimization with regard to a given criteria can be subsequently performed on the functions instead of performing repealed electromagnetic simulations. The functionals are trained on an initial population of simulation samples and refined using a point-wise error estimate to identify design parameters for subsequent samples. Additionally, the depth of the MEP tree is adjusted for increased accuracy as the data set is deemed sufficient.

• Computers and Concrete
• /
• 제24권4호
• /
• pp.295-302
• /
• 2019

Evolutionary algorithms based on conventional statistical methods such as regression and classification have been widely used in data mining applications. This work involves application of gene expression programming (GEP) for predicting compressive strength of fly ash geopolymer concrete, which is gaining increasing interest as an environmentally friendly alternative of Portland cement concrete. Based on 56 test results from the existing literature, a model was obtained relating the compressive strength of fly ash geopolymer concrete with the significantly influencing mix design parameters. The predictions of the model in training and validation were evaluated. The coefficient of determination ($R^2$), mean (${\mu}$) and standard deviation (${\sigma}$) were 0.89, 1.0 and 0.12 respectively, for the training set, and 0.89, 0.99 and 0.13 respectively, for the validation set. The error of prediction by the model was also evaluated and found to be very low. This indicates that the predictions of GEP model are in close agreement with the experimental results suggesting this as a promising method for compressive strength prediction of fly ash geopolymer concrete.