• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.10
• /
• pp.1231-1238
• /
• 1999

This paper presents an approach for the optimal heating load Identification using Diagonal Recurrent Neural Networks(DRNN). In this paper, the DRNN captures the dynamic nature of a system and since it is not fully connected, training is much faster than a fully connected recurrent neural network. The architecture of DRNN is a modified model of the fully connected recurrent neural network with one hidden layer. The hidden layer is comprised of self-recurrent neurons, each feeding its output only into itself. In this study, A dynamic backpropagation (DBP) with delta-bar-delta learning method is used to train an optimal heating load identifier. Delta-bar-delta learning method is an empirical method to adapt the learning rate gradually during the training period in order to improve accuracy in a short time. The simulation results based on experimental data show that the proposed model is superior to the other methods in most cases, in regard of not only learning speed but also identification accuracy.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.69.2-69
• /
• 2001

This paper presents a new approach to the design of neural control system using digital signal processors in order to improve the precision and robustness. Robotic manipulators have become increasingly important in the field of flexible automation. High speed and high-precision trajectory tracking are indispensable capabilities for their versatile application. The need to meet demanding control requirement in increasingly complex dynamical control systems under significant uncertainties, leads toward design of intelligent manipulation robots. The TMS320C31 is used in implementing real time neural control to provide an enhanced motion control for robotic manipulators. In this control scheme, the networks introduced are neural nets with dynamic neurons, whose dynamics are distributed over all the network nodes.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.40 no.8
• /
• pp.782-791
• /
• 1991

In the fields of pattern recognition such as speech recognition, several new techniques using Artifical Neural network Models have been proposed and implemented. In particular, the Multilayer Perception Model has been shown to be effective in static speech pattern recognition. But speech has dynamic or temporal characteristics and the most important point in implementing speech recognition systems using Artificial Neural Network Models for continuous speech is the learning of dynamic characteristics and the distributed cues and contextual effects that result from temporal characteristics. But Recurrent Multilayer Perceptron Model is known to be able to learn sequence of pattern. In this paper, the results of applying the Recurrent Model which has possibilities of learning tedmporal characteristics of speech to phoneme recognition is presented. The test data consist of 144 Vowel+ Consonant + Vowel speech chains made up of 4 Korean monothongs and 9 Korean plosive consonants. The input parameters of Artificial Neural Network model used are the FFT coefficients, residual error and zero crossing rates. The Baseline model showed a recognition rate of 91% for volwels and 71% for plosive consonants of one male speaker. We obtained better recognition rates from various other experiments compared to the existing multilayer perceptron model, thus showed the recurrent model to be better suited to speech recognition. And the possibility of using Recurrent Models for speech recognition was experimented by changing the configuration of this baseline model.

• Journal of the Korea Concrete Institute
• /
• v.17 no.6 s.90
• /
• pp.1075-1084
• /
• 2005

The compressive strength of concrete is commonly used criterion in producing concrete. However, the tests on the compressive strength are complicated and time-consuming. More importantly, it is too late to make improvement even if the test result does not satisfy the required strength, since the test is usually performed at the 28th day after the placement of concrete at the construction site. Therefore, accurate and realistic strength estimation before the placement of concrete is being highly required. In this study, the estimation of the compressive strength of concrete was performed by probabilistic neural network(PNN) on the basis of concrete mix proportions. The estimation performance of PNN was improved by considering the correlation between input data and targeted output value. Improved probabilistic neural network was proposed to automatically calculate the smoothing parameter in the conventional PNN by using the scheme of dynamic decay adjustment (DDA) algorithm. The conventional PNN and the PNN with DDA algorithm(IPNN) were applied to predict the compressive strength of concrete using actual test data of two concrete companies. IPNN showed better results than the conventional PNN in predicting the compressive strength of concrete.

• Smart Structures and Systems
• /
• v.20 no.6
• /
• pp.669-682
• /
• 2017

The normalised version of bispectrum, the so-called bicoherence, has often proved a reliable method of damage detection on engineering applications. Indeed, higher-order spectral analysis (HOSA) has the advantage of being able to detect non-linearity in the structural dynamic response while being insensitive to ambient vibrations. Skewness in the response may be easily spotted and related to damage conditions, as the majority of common faults and cracks shows bilinear effects. The present study tries to extend the application of HOSA to damage localisation, resorting to a neural network based classification algorithm. In order to validate the approach, a non-linear finite element model of a 4-meters-long cantilever beam has been built. This model could be seen as a first generic concept of more complex structural systems, such as aircraft wings, wind turbine blades, etc. The main aim of the study is to train a Neural Network (NN) able to classify different damage locations, when fed with bispectra. These are computed using the dynamic response of the FE nonlinear model to random noise excitation.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.49 no.12
• /
• pp.643-650
• /
• 2000

Using neural network approach, the diagnosis of faults in industrial process that requires observing multiple data simultaneously are studied. Two-stage diagnosis is proposed to analyze system faults. By using neural network, the first stage detects the dynamic trend of each normalized date patterns by comparing a proposed pattern. Instead of using neural network, the difference between stored fault pattern and real time data is used for fault diagnosis in the second stage. This method reduces the amount of calculation and saves storing space. Also, we dealt with unknown faults by normalizing the data and calculating the difference between the value of steady state and the data in case of fault. A model of tank reactor is given to verify that the proposed method is useful and effective to noise.

• Journal of Korea Multimedia Society
• /
• v.5 no.5
• /
• pp.523-529
• /
• 2002

The present study describes neural networks for stereoscopic vision, which are applied to identifying human speech. In speech recognition based on stereoscopic vision neural networks (SVNN), the similarities are first obtained by comparing input vocal signals with standard models. They are then given to a dynamic process in which both competitive and cooperative processes are conducted among neighboring similarities. Through the dynamic processes, only one winner neuron is finally detected. In a comparative study, the two-layered SVNN was 7.7% higher in recognition accuracies than the hidden Markov model (HMM). From the evaluation results, it was noticed that SVNN outperformed the existing HMM recognizer.

• Proceedings of the KIEE Conference
• /
• 1995.07b
• /
• pp.705-709
• /
• 1995

The objective of this study is to design a multi-layer neural network which controls the position of excavator's attachment. In this paper, a dynamic controller has been developed based on an error back-propagation(BP) neural network. Since the neural network can model an arbitrary nonlinear mapping, it was used as a commanded feedforward input generator. A PD feedback controller is used in parallel with the feedforward neural network to train the system. The neural network was trained by the current state of the excavator as well as the PD feedback error. By using the BP network as a feedforward controller, no a priori knowledge on system dynamics is need. Computer simulation results demonstrate such powerful characteristics of the proposed controller as adaptation to changing environment, robustness to disturbancen and performance improvement with the on-line learning in the position control of excavator attachment.

• Nuclear Engineering and Technology
• /
• v.51 no.3
• /
• pp.702-708
• /
• 2019

This paper deals with break size estimation of loss of coolant accidents (LOCA) using a nonlinear autoregressive with exogenous inputs (NARX) neural network. Previous studies used static approaches, requiring time-integrated parameters and independent firing algorithms. NARX neural network is able to directly deal with time-dependent signals for dynamic estimation of break sizes in real-time. The case studied is a LOCA in the primary system of Bushehr nuclear power plant (NPP). In this study, number of hidden layers, neurons, feedbacks, inputs, and training duration of transients are selected by performing parametric studies to determine the network architecture with minimum error. The developed NARX neural network is trained by error back propagation algorithm with different break sizes, covering 5% -100% of main coolant pipeline area. This database of LOCA scenarios is developed using RELAP5 thermal-hydraulic code. The results are satisfactory and indicate feasibility of implementing NARX neural network for break size estimation in NPPs. It is able to find a general solution for break size estimation problem in real-time, using a limited number of training data sets. This study has been performed in the framework of a research project, aiming to develop an appropriate accident management support tool for Bushehr NPP.

• Journal of Welding and Joining
• /
• v.18 no.3
• /
• pp.114-121
• /
• 2000

Since resistance spot welding has become one of the most popular sheet metal fabrication processes, a strong emphasis is being put on the quality of the welds. Throughout the years many quality estimation systems have been developed by many researchers to ensure weld quality. In this study, the process variables, which were monitored in the primary circuit of the welding machine, are used to estimate the weld quality with Hopfield neural network. The primary dynamic resistance is vectorized and stored as five patterns in the network. As the welding is done, the dynamic resistance patterns are recognized and the quality is estimated with the proposed method. Due to the primary process variables, it is possible to utilize this algorithms as an in-process real time quality monitoring system.