• Journal of the Korean Data and Information Science Society
• /
• 제15권2호
• /
• pp.449-455
• /
• 2004

Support vector machine (SVM) has been very successful in pattern recognition and function estimation problems for crisp data. This paper proposes a new method to evaluate interval regression models for crisp input-output data. The proposed method is based on quadratic loss SVM, which implements quadratic programming approach giving more diverse spread coefficients than a linear programming one. The proposed algorithm here is model-free method in the sense that we do not have to assume the underlying model function. Experimental result is then presented which indicate the performance of this algorithm.

• East Asian mathematical journal
• /
• 제28권3호
• /
• pp.293-303
• /
• 2012

Sequential quadratic programming (SQP) has been one of the most important methods for solving nonlinear constrained optimization problems. Recently, filter method, proposed by Fletcher and Leyffer, has been extensively applied for its promising numerical results. In this paper, we present and study an active set SQP-filter algorithm for inequality constrained optimization. The active set technique reduces the size of quadratic programming (QP) subproblem. While by the filter method, there is no penalty parameter estimate. Moreover, Maratos effect can be overcome by filter technique. Global convergence property of the proposed algorithm are established under suitable conditions. Some numerical results are reported in this paper.

• Management Science and Financial Engineering
• /
• 제7권1호
• /
• pp.41-56
• /
• 2001

This linear fractional - quadratic bilevel programming problem, in which the leader's objective function is a linear fractional function and the follower's objective function is a quadratic function, is studied in this paper. The leader's and the follower's variables are related by linear constraints. The derivations of the optimality conditions are based on Kuhn-Tucker conditions and the duality theory. It is also shown that the original linear fractional - quadratic bilevel programming problem can be solved by solving a standard linear fractional program and the optimal solution of the original problem can be achieved at one of the extreme point of a convex polyhedral formed by the new feasible region. The algorithm is illustrated with the help of an example.

• Asian-Australasian Journal of Animal Sciences
• /
• 제21권6호
• /
• pp.896-902
• /
• 2008

The objective of this research was to create a new probiotic candy with good flavor and healthy benefits by using the response surface method and a sequential quadratic programming technique. The endpoint was to increase the varieties of dairy products and enhance their market values. In this study, milk was mixed with yogurt cultures (Lactobacillus bulgaricus, Streptococcus thermophilus) and probiotics (L. paracasei, Bifidobacterium longum) and incubated at $37^{\circ}C$ for 20 h. The samples were blended with lyoprotectants (galactose, skim milk powder and sucrose), freeze dried and then mixed with sweeteners (lactose and xylitol) to improve the texture for forming tablets. The processing conditions were optimized in two steps: the first step constructed a surface model using response surface methodology; the second step optimized the model with a sequential quadratic programming procedure. Results indicated that skim milk inoculated with L. delbrueckii subsp. Bulgaricus, S. thermophilus, L. paracasei subsp. paracasei and B. longum and blended with 6.9% of galactose, 7.0% of sucrose and 8.0% of skim milk powder would produce a new probiotic candy with the highest viability of probiotics and good flavor. A relatively higher survival of probiotics can be achieved by placing the probiotic candy product in a glass bottle with deoxidant and desiccant at $4^{\circ}C$. These probiotic counts remained at 106-108 CFU/g after being stored for two months.

• Journal of applied mathematics & informatics
• /
• 제22권1_2호
• /
• pp.133-148
• /
• 2006

This paper proposes a smoothing method for symmetric conic linear programming (SCLP). We first characterize the central path conditions for SCLP problems with the help of Chen-Harker-Kanzow-Smale smoothing function. A smoothing-type algorithm is constructed based on this characterization and the global convergence and locally quadratic convergence for the proposed algorithm are demonstrated.

• 재활복지공학회논문지
• /
• 제5권1호
• /
• pp.59-64
• /
• 2011

인공 중이 (IMEHD, implantable middle ear hearing device)용 플로팅 매스 트랜스듀서는 제작 과정상에서 발생하는 구조적 오차 및 제작 결합시의 문제점들로 인해 설계 시 의도와 다른 진동 특성을 가질 수 있게 되고, 이렇게 제작된 진동 트랜스듀서는 향후 실제 환자에게 이식되었을 때의 진동체 진동 성능을 예측하기 어려운 문제점이 있다. 본 연구에서는 제작된 플로팅 매스형 트랜스듀서에 대해 설계값과의 비교 평가 및 향후 컴퓨터 모델화된 청각 중이 집중 소자 모델에 적용이 가능하도록 기 제작된 진동체의 집중 소자 모델 파라미터들을 추정할 수 있는 방법을 제안하였으며, 제안된 방법을 LabVIEW 기반의 그래픽 유저 인터페이스 소프트웨어로 구현하였다. 제안된 방법은 실제 제작된 플로팅 매스형 진동체의 진동 변위 및 위상 데이터를 이용하여 해당 진동체의 집중 모델 파라미터들을 순차적 이차 프로그래밍 (SQP, sequential quadratic programming) 방법으로 추정하는 방식이다. 구현된 방법을 이용하여 측정 데이터에 가해지는 잡음의 양에 따른 플로팅 매스형 진동 트랜스듀서의 집중 모델 파라미터 추정 시의 발생 오차 변화를 컴퓨터 시뮬레이션을 통해 조사하여 본 방법의 파라미터 추정 정확도를 확인하였다. 또한, 실제로 제작한 인공중이용 진동 트랜스듀서에 대해 측정된 진동 변위 및 위상 데이터를 이용하여 진동체의 집중 모델 파라미터들을 추정하였으며, 기존 설계값들과 비교를 통해 제안된 방법의 유용성을 평가하였다.

• 대한전자공학회논문지SD
• /
• 제43권6호
• /
• pp.23-29
• /
• 2006

본 논문에서는 Quadratic Programming(QP)을 이용한 효과적인 광역배치 기법을 제안한다. QP를 이용한 배치 기법의 단점인 셀 밀집 문제를 해결하기 위해 추가적인 힘(additional force)과 Grid Pre-warping 기법을 사용하였다 추가적인 힘은 밀집도에 기반한 값으로 적절한 힘을 구하기 위해 새로운 밀집도 함수를 고안하였다. Grid Pre-warping은 셀 좌표 사이의 상대적인 순서에 따라 셀들을 전체 영역에 재배치하는 기법이다. 새로운 밀집도 함수를 통해 구해진 추가적인 힘과 Pre-warping 기법을 반복적으로 적용하여 셀들이 효과적으로 분산된 광역배치를 얻었다. 이렇게 얻어진 배치를 "middle-down" 방식의 배치기인 Mongre의 초기배치로 적용하여 최종 상세배치 결과를 얻었다 제안하는 기법을 적용한 실험결과, FM 기법을 이용한 광역배치에 비해 향상된 결과를 보였으며, 우수한 배치기로 알려진 FengShui, Dragon과도 비교할만한 결과를 얻었다.

• International Journal of Control, Automation, and Systems
• /
• 제4권3호
• /
• pp.318-324
• /
• 2006

The mobile manipulator is expected to work in partially defined or unstructured environments. In our global/local approach to path planning, joint trajectories are generated for a desired Cartesian space path, designed by the global path planner. For a local path planner, inverse kinematics for a redundant system is used. Joint displacement limit for the manipulator links is considered in the motion planner. In an event of failure to obtain feasible trajectories, the task cannot be accomplished. At the point of failure, a deviation in the Cartesian space path is obtained and a replanner gives a new path that would achieve the goal position. To calculate the deviation, a nonlinear optimization problem is formulated and solved by standard Sequential Quadratic Programming (SQP) method.

• Journal of Electrical Engineering and Technology
• /
• 제9권5호
• /
• pp.1520-1526
• /
• 2014

This paper presents a probabilistic method to evaluate the total transfer capability (TTC) by considering the sequential quadratic programming and the uncertainty of weather conditions. After the initial TTC is calculated by sequential quadratic programming (SQP), the transient stability is checked by time simulation. Also because power systems are exposed to a variety of weather conditions the outage probability is increased due to the weather condition. The probabilistic approach is necessary to evaluate the TTC, and the Monte Carlo Simulation (MCS) is used to accomplish the probabilistic calculation of TTC by considering the various weather conditions.

• 대한원격탐사학회지
• /
• 제24권3호
• /
• pp.257-266
• /
• 2008

This study presents an approach to synthesize multispectral images at a higher resolution by exploiting a high-resolution image acquired in panchromatic modality. The synthesized images should be similar to the multispectral images that would have been observed by the corresponding sensor at the same high resolution. The proposed scheme is designed to reconstruct the multispectral images at the higher resolution with as less color distortion as possible. It uses a regression model of the second order to fit panchromatic data to multispectral observations. Based on the regression model, the multispectral images at the higher spatial resolution of the panchromatic image are optimized by a quadratic programming. In this study, the new method was applied to the IKONOS 1m panchromatic and 4m multispectral data, and the results were compared with them of several current approaches. Experimental results demonstrate that the proposed scheme can achieve significant improvement over other methods.