• Journal of Magnetics
• /
• 제16권3호
• /
• pp.234-239
• /
• 2011

Transcranial magnetic stimulation utilizes the method of controlling applied time and changing pulse by output pulse through power density control for diagnosis purposes. Transcranial magnetic stimulation can also be used in cases where diagnosis and treatment are difficult since output pulse shape can be changed. As intensity, pulse range, and pulse shape of the stimulation pulse must be changed according to lesion, the existing sine wave-shaped stimulation treatment pulse poses limitations in achieving various treatments and diagnosis. This study actualized a new method of transcranial magnetic stimulation that applies a 3 Stage 2 Switch( power semiconductor 2EA) for controlling pulse repetition rate by achieving numerous switching control of stimulation coil. Intensity, pulse range, and pulse shape of output can be freely changed to transform various treatment pulses in order to overcome limitations in stimulation treatment presented by the previous sine wave pulse shape. The method of freely changing pulse range by using 3 Stage 2 Switch discharge method is proposed. Pulse shape, composed of various pulse ranges, was created by grafting PFN (Pulsed Forming Network) through AVR AT80S8535 one-chip microprocessor technology, and application in transcranial magnetic stimulation was achieved to study the output characteristics of stimulation treatment pulse according to delaying time of the trigger signal applied in section switch.

• Journal of Magnetics
• /
• 제16권1호
• /
• pp.51-57
• /
• 2011

A power supply for magnetic-stimulation devices was designed via a control algorithm that involved a start current application based on a resonant converter. In this study, a new power supply for magnetic-stimulation devices was designed by controlling the pulse repetition frequency and pulse width. The power density could be controlled using the start-current-compensation and ZCS (zero-current switching) resonant converter. The results revealed a high-repetition-frequency, high-power magnetic-stimulation device. It was found that the stimulation coil current pulse width and that pulse repetition frequency could be controlled within the range of 200-450 ${\mu}S$ and 200-900 pps, respectively. The magnetic-stimulation device in this study consisted of a stimulation coil device and a power supply system. The maximum power of the stimulation coil from one discharge was 130 W, which was increased to 260 W using an additional reciprocating discharge. The output voltage was kept stable in a sinusoidal waveform regardless of the load fluctuations by forming voltage and current control using a deadbeat controller without increasing the current rating at the starting time. This paper describes this magnetic-stimulation device to which the start current was applied.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.350-350
• /
• 2000

Since the heat exchange system, such as the boiler of power plant, gas turbine, and radiator require a high rate heat efficiency and the efficiency of these systems is depended on the control methods. However, it is important f3r operator to understand control system of these systems. In order to properly apply control equipment to these process control systems, such as boiler, any other heat process, or process control system it is necessary to understand the basic aspects and operation principle of the process that relate control, interrelationships of the process characteristics, and the dynamics that are involved. Generally, PID controllers are used in these systems but it is difficult for engineer to understand the complex dynamics and the tuning method because of the coupling action and disturbance in the system loop. In this paper, we design an effective experimental system fur automatic control education and analyze its characteristics through experimental system and industrial plant control software to study how they can team automatic control system by experiments.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.218-223
• /
• 1992

This paper describes a learning method of neural network controllers. The learning method improves the performance of indirect learning mechanism in the neuro-control of nonlinear systems. To precisely identify dynamic characteristics of the plant by utilizing a limited prior information we propose a new energy function which takes advantage of the proportional relationship between outputs of the plant and those of neural networks.

• 전자공학회논문지B
• /
• 제29B권1호
• /
• pp.58-66
• /
• 1992

This paper proposes a neuro-controller for position control of one-link flexible robot arm. Basically the controller consists of a multi-layer neural network and a conventional PD controller. Two controller are parallelly connected. Neural network is traind by the conventional error back propagation learning rules. During learning period, the weights of neural network are adjusted to minimize the position error between the desired hub angle and the actual one. Finally the effectiveness of the proposed approach will be demonstrated by computer simulation.

• 한국지능시스템학회논문지
• /
• 제25권2호
• /
• pp.119-125
• /
• 2015

정수장, 하수처리장, 폐수처리장의 배출수 처리공정에서 고 농도의 슬러지 선별, 이송 및 약품 투입량 조절을 위한 기준으로 슬러지 농도계가 사용되고 있다. 그러나 슬러지에 함유된 이물질이 혼입될 경우 감쇄량이 증가하거나 초음파가 수신부에 전달되지 않아 실제 농도값 보다 높은 값을 출력하거나 헌팅현상이 발생한다. 또한 단일 센서에 슬러지 포착 또는 고장 등의 문제로 배출수 공정 자동화에 어려움이 많았다. 이러한 문제점을 개선하기 위해 초음파 다중빔 농도계를 개발하여 사용하고 있으나 특정 초음파 빔의 농도 측정값에 오류가 발생할 경우 전체 농도시스템의 성능이 떨어지는 단점이 있다. 따라서 본 논문에서는 초음파 다중빔 농도계 간의 신뢰성을 판단하고, 신뢰성이 높은 다중빔 농도계만을 사용하여 슬러지 농도 예측값의 성능 향상방안을 제시하였다. 예측 알고리즘으로는 뉴로-퍼지모델을 적용하였으며 다양한 실험을 통하여 제안된 방법의 타당성을 검증하였다.

• 제어로봇시스템학회논문지
• /
• 제20권3호
• /
• pp.261-269
• /
• 2014

This paper provides an overview of the basics and recent studies of Lyapunov-based nonlinear adaptive control, the aim of which is to improve or maintain the performance and stability of the closed-loop system by cancelling out the presumable uncertainties in the nonlinear system dynamics. The design principles are essentially based on Lyapunov's direct method. In this survey, we provide a comprehensive overview of Lyapunov-based nonlinear adaptive control techniques with simplified effective design examples, which are to be elaborated as related recent results are gradually shown. The scope of the survey contains research on singularity problems in adaptive control, the techniques to deal with linearly and nonlinearly parameterized uncertainties, robust neuro-adaptive control, and adaptive control methodologies combined with various nonlinear control techniques such as sliding-mode control, back-stepping, dynamic surface control, and optimal/$H_{\infty}$ control.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 춘계학술대회
• /
• pp.249-251
• /
• 2022

본 연구에서는 뇌파를 이용하여 드론을 제어하기 위한 기계학습을 논의한다. 드론의 이륙과 전진, 후진, 좌측 이동 그리고 우측 이동을 제어대상으로 정의하였고 이를 제어하기 위한 뇌파의 신호를 전두엽을 대상으로 하는 Fp1·Fp2 2채널 건식 전극(NeuroNicle FX2) 뇌파 측정장비를 통하여 5.19초동안 각 제어대상과 연관된 행동의 운동 심상을 눈을 뜬 상태에서 측정(Sampling Rate 250Hz, Cutoff Frequency 6~20Hz) 하였다. 측정된 뇌파신호에 대해 매틀랩의 분류학습기를 이용해서 삼중 계층 신경망, 로지스틱 회귀커널, 비선형 3차 SVM 학습을 실시하였으며 학습결과 로지스틱 회귀 커널 학습에서 드론의 이륙과 전진, 후진, 좌측 이동 그리고 우측 이동을 위한 가장 높은 정확도를 가지고 있음을 클래스 참양성률로 확인할 수 있었다.

• Wind and Structures
• /
• 제22권6호
• /
• pp.635-662
• /
• 2016

This paper deals with the comparison of different control strategies for the Induction generators in wind energy conversion system. Mainly, two types of induction machines, Self excited induction generator (SEIG) and doubly Fed Induction generators (DFIG) are studied. The different control strategies for SEIG and DFIG are compared. For SEIG, Electronic load Controller mechanism, Static Compensator based voltage regulator are studied. For DFIG the main control strategy namely vector control, direct torque control and direct power control are implemented. Apart from these control strategies for both SEIG and DFIG to improve the performance, the ANFIS based controller is introduced in both STATCOM and DTC methods. These control methods are simulated using MATLAB/SIMULINK and performances are analyzed and compared.

• 생물환경조절학회지
• /
• 제3권1호
• /
• pp.42-51
• /
• 1994

대다수 농산물과 마찬가지로 건조표고의 등급판정은 외관특징에 주로 의존한다. 표고 갓의 전후면에 걸친 복잡하고 다양한 외관특징들로 인하여 표고의 등급판정은 임의로 추출한 표고샘플에 대하여 전문가가 수작업으로 판정하고 있으며, 선별작업 역시 전적으로 수작업에 의존하고 있다. 단순한 반복작업으로 보이는 농산물의 등급판정은 사실 시각과 촉각을 위시한 고도의 감각신경계를 통하여 상호 복잡하게 얽혀 들어오는 정보를 지능적으로 처리하는 고기능의 작업이다. 농산물의 경우, 외관특성을 비롯한 물성은 종류별로 그 경계치를 일괄적으로 명확하게 규정할 수 없기 때문에 대개는 오차를 포함한 통계적 접근에 의하여 규정하고 있다. 따라서 농산작업에 있어서는 농산물 물성이 갖는 모호성을 효율적으로 처리할 수 있는 가변적인 작업구조 및 정보처리가 필수적으로 요구된다. 본 연구에서는 인간 뇌의 정보처리 기능을 부분적으로 구현할 수 있는 인공신경회로망을 컴퓨터 시각 시스템에 적용하여 단순 기하도형의 분류 및 표고의 등급판정을 성공적으로 수행하였다. 회로망 입력으로는 컴퓨터시각 시스템을 이용하여 건조표고의 정성적 외관특징을 자동으로 추출한 후 정량화한 특징점 값들을 이용하였다. 신경회로망의 학습은 표본 추출한 등급표고와 이들의 정량적 특징점 값들을 입출력 쌍으로 하여 수행하였다. 학습한 회로망의 등급판정 성능시험은 표본추출한 미지의 표고에 대한 컴퓨터 영상 특징점 값들을 입력하여 수행하였다.