• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.686-689
• /
• 2004

Genetic algorithm is well known for stochastic searching method in imitating natural phenomena. In recent times, studies have been conducted in improving conventional evolutionary computation speed and promoting precision. This paper presents an approach to optimize PID controller gains with the application of modified Genetic Algorithm using Support Vector Machine (SVMGA). That is, we aim to explore optimum parameters of PID controller using SVMGA. Simulation results are given to compare to those of tuning methods, based on Simple Genetic Algorithm and Ziegler-Nicholas tuning method.

• Communications for Statistical Applications and Methods
• /
• 제10권2호
• /
• pp.595-601
• /
• 2003

In this paper we propose an on line training method for classification based on least squares support vector machine. Proposed method enables the computation cost to be reduced and the training to be peformed incrementally, With the incremental formulation of an inverse matrix in optimization problem, current information and new input data can be used for building the new inverse matrix for the estimation of the optimal bias and Lagrange multipliers, so the large scale matrix inversion operation can be avoided. Numerical examples are included which indicate the performance of proposed algorithm.

• Journal of the Korean Data and Information Science Society
• /
• 제25권4호
• /
• pp.951-959
• /
• 2014

Recently Hazelton and Turlach (2009) proposed a weighted kernel density estimator for the deconvolution problem. In the case of Gaussian kernels and measurement error, they argued that the weighted kernel density estimator is a competitive estimator over the classical deconvolution kernel estimator. In this paper we consider weighted kernel density estimators when sample observations are contaminated by double exponentially distributed errors. The performance of the weighted kernel density estimators is compared over the classical deconvolution kernel estimator and the kernel density estimator based on the support vector regression method by means of a simulation study. The weighted density estimator with the Gaussian kernel shows numerical instability in practical implementation of optimization function. However the weighted density estimates with the double exponential kernel has very similar patterns to the classical kernel density estimates in the simulations, but the shape is less satisfactory than the classical kernel density estimator with the Gaussian kernel.

• 산업기술연구
• /
• 제19권
• /
• pp.209-216
• /
• 1999

In this paper, we proposed a modofied genetic algorithm to minimize misclassification rate for determining the codebook. Genetic algorithms are adaptive methods which may be used solve search and optimization problems based on the genetic processes of biological organisms. But they generally require a large amount of computation efforts. GAVQ can choose the optimal individuals by genetic operators. The position of individuals are optimized to improve the recognition rate. The technical properties of this study is that prevents us from the local minimum problem, which is not avoidable by conventional VQ algorithms. We compared the simulation result with Matlab using phoneme data. The simulation results show that the recognition rate from GAVQ is improved by comparing the conventional VQ algorithms.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1994년도 하계학술대회 논문집 A
• /
• pp.307-312
• /
• 1994

In this investigation the PFM(Pulse Frequency Modulation} will be used for optimizing PWM inverter pulse pattern. In traditional the pulse frequency of PWM is kept const. But modulated PWM's frequency in this study, the sinusoidal inverter's performance should be improved. The PWM pulsepatterns are definitely controlled so that the time-integral function of the voltage vectors in the space vector notation may show a circular locus. Further, performance index will be minimized because of minimizing distortion of output current. Finally, we will implement itusingsingle-chip microprocessor.

• 대한전자공학회논문지
• /
• 제27권1호
• /
• pp.105-111
• /
• 1990

In this paper, fast algorithms to generate the codebook of vector quantization in image coding, are proposed. And an efficient algorithm to guess a initial codebook, namely, binary splitting method, is proposed. We generated the initial codebook by binary splitting method and then reduced the searching time using Iterative Optimization algorithm as an alternate to the generalized Lloyd algorithm and several information from binary splitting method. And the searching time and performance can be traded off by varying the searching range. With this proposed algorithm, the computation time can be reduced by a factor of 60 Without any degradation of image quality.

• International Journal of Control, Automation, and Systems
• /
• 제4권5호
• /
• pp.653-659
• /
• 2006

The design of $H_2$-optimal control with regional pole assignment via state feedback in linear time-invariant systems is investigated. The aim is to find a state feedback controller such that the closed-loop system has the desired eigenvalues lying in some desired stable regions and attenuates the disturbance between the output vector and the disturbance vector. Based on a proposed result of parametric eigenstructure assignment via state feedback in linear systems, the considered $H_2$-optimal control problem is changed into a minimization problem with certain constraints, and a simple and effective algorithm is proposed for this considered problem. A numerical example and its simulation results show the simplicity and effectiveness of this proposed algorithm.

• Journal of Power Electronics
• /
• 제2권1호
• /
• pp.67-75
• /
• 2002

This paper presents the control algorithm for maximum efficiency drives of an induction motor system with the high dynamic performance. This system uses a simple model of the induction motor that includes equations of the iron losses. The model, which only requires the parameters of the induction motor, is referred to a field-oriented frame. The minimum point of the input power can be obtained at the steady state condition. The proposed optimal efficiency control algorithm calculates the reference torque and flux currents for the vector control of the induction motors. A 32 bit floating point TMS320C32 DSP chip implements the drive system with the efficiency optimization controller. The results show the effectiveness of the control strategy Proposed for the induction motor drive.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2004년도 추계학술발표논문집(상)
• /
• pp.241-244
• /
• 2004

일반적으로 메쉬(mesh)는 비정규 연결 형태(irregular connectivity)로 되어 있다. 리메싱(remeshing)은 비정규 연결 형태의 메쉬를 정규 연결 형태(regular connectivity)로 바꾸어 주는 작업이다. 메쉬의 기하 정보가 2D 그리드에 저장이 되어 있는 기하이미지(geometry Images)는 비정규 연결 형태의 메쉬를 완전 정규 형태(completely regular connectivity)로 리메싱하는 데 사용된다. 원본 메쉬를 기하 이미지로 생성하는 방법은 변형되는 크기를 최소화 하는 스트레치 메트릭(stretch metric)을 기반으로 이루어 졌다. 이 방법은 리메싱된 메쉬의 언더샘플링(undersampling)을 줄여 주게 된다. 하지만 리메싱 과정에서 생기는 오버샘플링(oversampling)은 줄여 주지 못한다. 본 논문에서는 정점(vertex)의 법선 벡터(normal vector)를 이용하여 기하이미지의 오버샘플링을 줄이는 방법을 제시한다.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2004년도 봄 학술발표논문집 Vol.31 No.1 (B)
• /
• pp.592-594
• /
• 2004

본 논문에서는 SVM의 효과적인 학습 알고리즘인 SMO(Sequential Minimal Optimization)를 하드웨어적으로 설계하고 구현하는 방법을 제시한다. SVM은 Vapnik에 의한 제안된 기계학습 방법으로 음성인식, 문자인식, BT, 보안 등 다양한 응용분야에서 기존의 신경망보다 우수한 성능을 나타내었다. 그러나 SVM은 계산량이 많아 연산속도가 느려지는 단점을 가진다. 이를 개선하기 위해 본 논문에서는 SVM의 학습 알고리즘인 SMO의 핵심인 지수함수와 실수 연산기를 VHDL로 설계하고 Mentor의 ModelSim을 이용하여 시뮬레이션하고 Synopsys의 Design Analyzer를 이용하여 합성하였다. 구현된 칩은 시뮬레이션 결과 약 50MHz의 속도로 동작하며, 이는 소프트웨어적으로 구현된 SMO보다 약 10～20배 빠른 성능을 나타내었다.