• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 추계종합학술대회 논문집
• /
• pp.368-371
• /
• 1999

In this paper, we propose a neural network processor architecture with on-chip learning and with reconfigurability according to the data dependencies of the algorithm applied. For the neural network model applied, the proposed architecture can be configured into either SIMD or SRA(Systolic Ring Array) without my changing of on-chip configuration so as to obtain a high throughput. However, changing of system configuration can be controlled by user program. To process activation function, which needs amount of cycles to get its value, we design it by using PWL(Piece-Wise Linear) function approximation method. This unit has only single latency and the processing ability of non-linear function such as sigmoid gaussian function etc. And we verified the processing mechanism with EBP(Error Back-Propagation) model.

• 대한조선학회논문집
• /
• 제59권2호
• /
• pp.125-133
• /
• 2022

This paper proposes a fault detection method for a Unmanned Surface Vehicle (USV) with overactuated system. Current status information for fault detection is expressed as a scalogram image. The scalogram image is obtained by wavelet-transforming the USV's control input and sensor information. The fault detection scheme is based on Convolutional Neural Network (CNN) algorithm. The previously generated scalogram data was transferred learning to GoogLeNet algorithm. The data are generated as scalogram images in real time, and fault is detected through a learning model. The result of fault detection is very robust and highly accurate.

• International Journal of Control, Automation, and Systems
• /
• 제1권3호
• /
• pp.321-331
• /
• 2003

Experimental software datasets describing software projects in terms of their complexity and development time have been the subject of intensive modeling. A number of various modeling methodologies and modeling designs have been proposed including such approaches as neural networks, fuzzy, and fuzzy neural network models. In this study, we introduce the concept of the Rule-based fuzzy polynomial neural networks (RFPNN) as a hybrid modeling architecture and discuss its comprehensive design methodology. The development of the RFPNN dwells on the technologies of Computational Intelligence (CI), namely fuzzy sets, neural networks, and genetic algorithms. The architecture of the RFPNN results from a synergistic usage of RFNN and PNN. RFNN contribute to the formation of the premise part of the rule-based structure of the RFPNN. The consequence part of the RFPNN is designed using PNN. We discuss two kinds of RFPNN architectures and propose a comprehensive learning algorithm. In particular, it is shown that this network exhibits a dynamic structure. The experimental results include well-known software data such as the NASA dataset concerning software cost estimation and the one describing software modules of the Medical Imaging System (MIS).

• 한국항해항만학회지
• /
• 제32권8호
• /
• pp.589-596
• /
• 2008

선박의 접안운동을 자동화하기 위하여 인공신경망(Artificial Neural Network, 이하 ANN)에 의한 제어를 수행하였다. ANN은 시스템의 비선형성이 표현 가능하므로 접안운동과 같은 비선형성이 강한 조종운동에 적합하다. 입력층과 출력층 사이에 하나 이상의 중간층이 존재하는 다층 인식자(Multi-layer perceptron)를 사용하였고, 교사 데이터(Teaching data)와 역전파(Back-Propagation) 알고리즘을 사용하여 신경망의 출력값과 목표 출력값 사이의 오차가 최소가 되도록 신경망 학습을 수행하였다. 접안 시 저속조종 수학모델을 사용하여 접안 시뮬레이션을 수행하였으며, ANN의 입력층 성분(unit)이 8개인 구조와 6개인 구조의 접안 제어를 비교하였다. 시뮬레이션 결과, 두 ANN에 의하여 접안 경로 선택에 차이가 나타났으나 접안 조건은 모두 만족하였다.

• 한국ITS학회 논문지
• /
• 제17권5호
• /
• pp.173-187
• /
• 2018

본 논문에서는 주차장 관리 시스템의 한 부분으로 차량 계수를 위한 감시 카메라 시스템의 컴퓨터 비전과 심층 학습 기반 기법을 제안하고자 한다. You Only Look Once 버전 2 (YOLOv2) 탐지기를 적용하고 YOLOv2 기반의 심층 컨볼루션 신경망(CNN)을 다른 아키텍처와 두 가지 모델로 구성하였다. 제안 된 아키텍처의 효과를 Udacity의 자체 운전 차량 데이터 세트를 사용하여 설명하였다. 학습 및 테스트 결과, 자동차, 트럭 및 보행자 탐지 시 원래 구조(YOLOv2)의 경우 47.89%의 mAP를 나타내는 것에 비하여, 제안하는 모델의 경우 64.30 %의 mAP를 달성하여 탐지 정확도가 향상되었음을 증명하였다.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2016년도 춘계학술대회
• /
• pp.97-100
• /
• 2016

CNN(Convolutinal Neural Network)을 사용하여 다양한 분야에 대한 심화 학습이 진행되고 있으며 이미지 인식 분야에서 특히 높은 성능을 보이고 있다. 본 논문에서는 5,000,000개 이상의 대규모 한글 문자 데이터베이스를 사용하여 한글을 Convolutional Neural Network에 학습 시킨 후 테스트 정확도를 확인한다. 실험에 사용된 CNN 구조는 AlexNet에 기반하여 새로 만들어진 KCR(Korean Character Recognition)-AlexNet 이며 학습 결과 98% 이상의 테스트 정확도를 보였다. 실험에 사용된 데이터베이스는 대규모 한글 데이터 데이터베이스인 PHD08로 총 2,350개의 한글 문자에 대해 각 문자마다 2,187개의 샘플을 가져 총 5,139,450 개의 데이터가 존재한다. 본 연구를 통해 KCR-AlexNet이 한글 데이터베이스인 PHD08을 학습하는데 우수한 구조임을 보인다.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제49권9호
• /
• pp.512-519
• /
• 2000

A Neural networks has been used for a expert system and fault diagnosis system. It is possible to nonlinear function mapping and parallel processing. Therefore It has been developing for a Diagnosis system of nuclear plower plant. In general Neural Networks is a static mapping but Dynamic Neural Network(DNN) is dynamic mapping.쪼두 a fault occur in system a state of system is changed with transient state. Because of a previous state signal is considered as a information DNN is better suited for diagnosis systems than static neural network. But a DNN has many weights so a real time implementation of diagnosis system is in need of a rapid network architecture. This paper presents a algorithm for RCP monitoring Alarm diagnosis system using Self Dynamic Neural Network(SDNN). SDNN has considerably fewer weights than a general DNN. Since there is no interlink among the hidden layer. The effectiveness of Alarm diagnosis system using the proposed algorithm is demonstrated by applying to RCP monitoring in Nuclear power plant.

• Journal of Electrical Engineering and Technology
• /
• 제12권6호
• /
• pp.2219-2226
• /
• 2017

This paper deals with artificial neural network approach for automatic detection of severity level of stator winding fault in induction motor. The problem is faced through modelling and simulation of induction motor with inter coil shorting in stator winding. The sum of the absolute values of difference in the peak values of phase currents from each half cycle has been chosen as the main input to the classifier. Sample values from workspace of Simulink model, which are verified with experiment setup practically, have been imported to neural network architecture. Consideration of a single input extracted from time domain simplifies and advances the fault detection technique. The output of the feed forward back propagation neural network classifies the short circuit fault level of the stator winding.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1993년도 한국자동제어학술회의논문집(국내학술편); Seoul National University, Seoul; 20-22 Oct. 1993
• /
• pp.14-19
• /
• 1993

In this paper, we proposed an indirect learning and direct adaptive control schemes using neural networks, i.e., composite adaptive neural control, for a class of continuous nonlinear systems. With the indirect learning method, the neural network learns the nonlinear basis of the system inverse dynamics by a modified backpropagation learning rule. The basis spans the local vector space of inverse dynamics with the direct adaptation method when the indirect learning result is within a prescribed error tolerance, as such this method is closely related to the adaptive control methods. Also hash addressing technique, similar to the CMAC functional architecture, is introduced for partitioning network hidden nodes according to the system states, so global neuro control properties can be organized by the local ones. For uniform stability, the sliding mode control is introduced when the neural network has not sufficiently learned the system dynamics. With proper assumptions on the controlled system, global stability and tracking error convergence proof can be given. The performance of the proposed control scheme is demonstrated with the simulation results of a nonlinear system.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제4권2호
• /
• pp.211-216
• /
• 2004

Fuzzy neural networks have been successfully applied to analyze/generate predictive rules for medical or diagnostic data. However, most approaches proposed so far have not considered the weights for the membership functions much. This paper presents a neural network with weighted fuzzy membership functions. In our approach, the membership functions can capture the concentrated and essential information that affects the classification of the input patterns. To verify the performance of the proposed model, well-known Iris data set is performed. According to the results, the weighted membership functions enhance the prediction accuracy. The architecture of the proposed neural network with weighted fuzzy membership functions and the details of experimental results for the data set is discussed in this paper.