• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1997년도 하계학술대회 논문집 B
• /
• pp.561-563
• /
• 1997

In this paper, we identified the Black-box system with serious nonlinerity and fast dynamics using Neural Network. This NN have new structure and learned by RLS. It identify system in real-time without priori data. We use this NN to 7-DOF vehicle identification.

• Bulletin of the Korean Chemical Society
• /
• 제26권3호
• /
• pp.399-408
• /
• 2005

We have utilized neural network (NN) studies to predict impact sensitivities of various types of explosive molecules. Two hundreds and thirty four explosive molecules have been taken from a single database, and thirty nine molecular descriptors were computed for each explosive molecule. Optimization of NN architecture has been carried out by examining seven different sets of molecular descriptors and varying the number of hidden neurons. For the optimized NN architecture, we have utilized 17 molecular descriptors which were composed of compositional and topological descriptors in an input layer, and 2 hidden neurons in a hidden layer.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.581-586
• /
• 2003

This paper presents control designs using neural networks (NN) for a XY positioning table. The proposed neurocontroller is composed of an outer PD tracking loop for stabilization of the fast flexible-mode dynamics and an NN inner loop used to compensate for the system nonlinearities. A tuning algorithm is given for the NN weights, so that the NN compensation scheme becomes adaptive, guaranteeing small tracking errors and bounded weight estimates. Formal nonlinear stability proofs are given to show that the tracking error is small. The proposed neuro-controller is implemented and tested on an IBM PC-based XY positioning table, and is applicable to many precision XY tables. The algorithm, simulation, and experimental results are described. The experimental results are shown to be superior to those of conventional control.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1994년도 하계학술대회 논문집 C
• /
• pp.1540-1542
• /
• 1994

This paper describes an experimental study of pattern recognition of partial discharge for three different discharge sources by using neural network(NN) system. The NN system is three layer feedforward connections and its learning method is a backpropagation algorithm incorporating an external teacher signal. Input information for NN is a statistical parameters of a discharge magnitude and the number of pulse count. After learning three typical input patterns, NN system offers good discrimination between different defects.

• Smart Structures and Systems
• /
• 제30권1호
• /
• pp.1-15
• /
• 2022

This article focuses on stability analysis and fuzzy controller synthesis for large neural network (NN) systems consisting of several interconnected subsystems represented by the NN model. Advanced and fuzzy NN differential inclusion (NNDI) for stability based on the developed algorithm with H infinity can be designed based on the evolved biological design. This representation is constructed using sector linearity for NN models. Sector linearity transforms a non-linear model into a linear model based on proposed operations. New sufficient conditions are realized in the form of LMI (linear matrix inequalities) to ensure the asymptotic stability of the trans-Lyapunov function. This transforms the nonlinear model into a linear model based on multiple rules. At last, a numerical case study with simulations is derived as illustration to prove its feasibility in real nonlinear structures.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.330-333
• /
• 2005

In this study, a method of designing a neurointerface using neural network (NN) is proposed for controlling nonholonomic mobile robots. According to the concept of virtual master-slave robots, in particular, a partially stable inverse dynamic model of the master robot is acquired online through the NN by applying a feedback-error learning method, in which the feedback controller is assumed to be based on a PD compensator for such a nonholonomic robot. A tracking control problem is demonstrated by some simulations for a nonholonomic mobile robot with two-independent driving wheels.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.1330-1333
• /
• 2004

The methodology for modeling building energy consumption is well established for energy saving calculation in the temperate zone both for performance-based energy retrofitting contracts and measurement and verification (M&V) projects. Mostly, statistical regression models based on utility bills and outdoor dry-bulb temperature have been applied to baseline monthly and annual whole building energy use. This paper presents the application of neural networks (NN) to model landlord energy consumption of commercial buildings in Singapore. Firstly, a brief background information on NN and its application on the building energy research is provided. Secondly, five commercial buildings with various characteristics were selected for case studies. Monthly mean outdoor dry-bulb temperature ($T_0$), Relative Humidity (RH) and Global Solar Radiation (GSR) are used as network inputs and the landlord monthly energy consumption of the same period is the output. Up to three years monthly data are taken as training data. A forecast has been made for another year for all the five buildings. The performance of the NN analysis was evaluated using coefficient of variance (CV). The results show that NNs is powerful at predicting annual landlord energy consumption with high accuracy.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2001년도 추계학술대회 학술발표 논문집
• /
• pp.1-4
• /
• 2001

Landsat TM 위성영상을 대상으로 인공신경망 모형과 RBF 신경망 모형의 토지피복분류 정확도를 평가하였다. 토지피복의 특성에 따라 세 개의 연구지역(복합토지이용, 농경지, 도시지역)을 대상으로 RBF 신경망 모형의 입력밴드 조합 및 분류 항목의 변화에 따른 민감도 분석이 수행되었다. 오염부하 원단위의 신뢰구간 및 분포를 추정하기 위하여 붓스트랩기법이 적화하였으며, 특히 토지이용이 다양한 도시지역에서 가장 큰 변화폭을 보였다.

• 대한전기학회논문지
• /
• 제45권2호
• /
• pp.325-328
• /
• 1996

In this paper, we proposed an estimation method of a posterior probability and PDF(Probability density function) using a feed forward neural network and code books of VQ(vector quantization). In this study, We estimates a posterior probability and probability density function, which compose a new parameter with well-known Mel cepstrum and verificate the performance for the five vowels taking from syllables by NN(neural network) and PNN(probabilistic neural network). In case of new parameter, showed the best result by probabilistic neural network and recognition rates are average 83.02%.

• 전력전자학회논문지
• /
• 제7권6호
• /
• pp.509-516
• /
• 2002

본 논문에서는 신경회로망을 이용한 스위치드 릴럭턴스 전동기(Switched Reluctance Motor, SRM)의 간편한 순시토크 추정기법과 스위칭 각 제어에 관해 연구하였다. 간단한 구조와 높은 효율 등의 많은 장점을 가지고 있는 SRM은 부하토크와 회전속도에 따라 운전특성이 달라지므로 최적운전을 위해서는 스위칭 각 제어가 필수적이다. 이러한 스위칭 각은 고정자 및 회전자 극호각, 토크 및 속도 능의 여러 변수들에 따라 달라지기 때문에 적정 스위칭 시점을 결정하는데 있어서 어려움이 있다. 따라서 본 논문에서는 부하토크 및 회전속도에 따라 효율이 가장 높은 적정 스위칭 각을 실험을 통해 선정한 후 신경회로망(Neural Network, NN)을 이용하여 전동기 제어에 적용하는 방안에 관해 고찰하였다. 또한 스위칭 각 제어에 있어서 필수적인 순시토크의 추정에 있어서도 신경회로망을 이용하여 인덕턴스의 비선형적인 특성이 고려되도록 하였다. 구현된 스위칭 각 제어기를 실제 시스템에 적용하였으며, 고효율 측면에서 선정된 스위칭 각 제어기의 동특성을 확인함으로써 스위칭 각 제어기의 적합성을 검증하였다.