• Journal of Industrial Technology
• /
• v.16
• /
• pp.197-205
• /
• 1996

In this paper, we propose a new neural architecture. We synthesize the architecture from a combination of structures known as MRCCN (Multi-resolution Radial-basis Competitive and Cooperative Network) and BPN (Backpropagation Network). The proposed neural network is able to improve the learning speed of MRCCN and the mapping capability of BPN. The ability and effectiveness of identifying a ninlinear dynamic system using the proposed architecture will be demonstrated by computer simulation.

• Journal of Korean Institute of Industrial Engineers
• /
• v.25 no.1
• /
• pp.111-124
• /
• 1999

This evapaper is toluate the forecasting performance of three neural network(NN) approaches against ARIMA model using the famous time series analysis competition data. The first NN approach is to analyze the second Makridakis (M2) Competition Data using Multilayer Perceptron (MLP) that has been the most popular NN model in time series analysis. Since it is recently known that MLP suffers from bias/variance dilemma, two approaches are suggested in this study. The second approach adopts Cascade Correlation Network (CCN) that was suggested by Fahlman & Lebiere as an alternative to MLP. In the third approach, a time series is separated into two series using Noise Filtering Network (NFN) that utilizes autoassociative memory function of neural network. The forecasts in the decomposition analysis are the sum of two prediction values obtained from modeling each decomposed series, respectively. Among the three NN approaches, Decomposition Analysis shows the best forecasting performance on the M2 Competition Data, and is expected to be a promising tool in analyzing socio-economic time series data because it reduces the effect of noise or outliers that is an impediment to modeling the time series generating process.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.59 no.7
• /
• pp.1320-1326
• /
• 2010

In this paper, we develop a design methodology of granular-based neurocomputing networks realized with the aid of the clustering techniques. The objective of this paper is modeling and evaluation of approximation and generalization capability of the Linear-Type Superconducting Power Supply (LTSPS). In contrast with the plethora of existing approaches, here we promote a development strategy in which a topology of the network is predominantly based upon a collection of information granules formed on a basis of available experimental data. The underlying design tool guiding the development of the granular-based neurocomputing networks revolves around the Fuzzy C-Means (FCM) clustering method and the Radial Basis Function (RBF) neural network. In contrast to "standard" Radial Basis Function neural networks, the output neuron of the network exhibits a certain functional nature as its connections are realized as local linear whose location is determined by the membership values of the input space with the aid of FCM clustering. To modeling and evaluation of performance of the linear-type superconducting power supply using the proposed network, we describe a detailed characteristic of the proposed model using a well-known NASA software project data.

• Geophysics and Geophysical Exploration
• /
• v.26 no.3
• /
• pp.114-125
• /
• 2023

The physics-informed neural network (PINN) has been proposed to overcome the limitations of various numerical methods used to solve partial differential equations (PDEs) and the drawbacks of purely data-driven machine learning. The PINN directly applies PDEs to the construction of the loss function, introducing physical constraints to machine learning training. This technique can also be applied to wave equation modeling. However, to solve the wave equation using the PINN, second-order differentiations with respect to input data must be performed during neural network training, and the resulting wavefields contain complex dynamical phenomena, requiring careful strategies. This tutorial elucidates the fundamental concepts of the PINN and discusses considerations for wave equation modeling using the PINN approach. These considerations include spatial coordinate normalization, the selection of activation functions, and strategies for incorporating physics loss. Our experimental results demonstrated that normalizing the spatial coordinates of the training data leads to a more accurate reflection of initial conditions in neural network training for wave equation modeling. Furthermore, the characteristics of various functions were compared to select an appropriate activation function for wavefield prediction using neural networks. These comparisons focused on their differentiation with respect to input data and their convergence properties. Finally, the results of two scenarios for incorporating physics loss into the loss function during neural network training were compared. Through numerical experiments, a curriculum-based learning strategy, applying physics loss after the initial training steps, was more effective than utilizing physics loss from the early training steps. In addition, the effectiveness of the PINN technique was confirmed by comparing these results with those of training without any use of physics loss.

• Nuclear Engineering and Technology
• /
• v.55 no.6
• /
• pp.2305-2314
• /
• 2023

The governing equations of atmospheric dispersion most often taking the form of a second-order partial differential equation (PDE). Currently, typical computational codes for predicting atmospheric dispersion use the Gaussian plume model that is an analytic solution. A Gaussian model is simple and enables rapid simulations, but it can be difficult to apply to situations with complex model parameters. Recently, a method of solving PDEs using artificial neural networks called physics-informed neural network (PINN) has been proposed. The PINN assumes the latent (hidden) solution of a PDE as an arbitrary neural network model and approximates the solution by optimizing the model. Unlike a Gaussian model, the PINN is intuitive in that it does not require special assumptions and uses the original equation without modifications. In this paper, we describe an approach to atmospheric dispersion modeling using the PINN and show its applicability through simple case studies. The results are compared with analytic and fundamental numerical methods to assess the accuracy and other features. The proposed PINN approximates the solution with reasonable accuracy. Considering that its procedure is divided into training and prediction steps, the PINN also offers the advantage of rapid simulations once the training is over.

• Geomechanics and Engineering
• /
• v.1 no.1
• /
• pp.53-74
• /
• 2009

In recent years, several computer-aided pattern recognition and data mining techniques have been developed for modeling of soil behavior. The main idea behind a pattern recognition system is that it learns adaptively from experience and is able to provide predictions for new cases. Artificial neural networks are the most widely used pattern recognition methods that have been utilized to model soil behavior. Recently, the authors have pioneered the application of genetic programming (GP) and evolutionary polynomial regression (EPR) techniques for modeling of soils and a number of other geotechnical applications. The paper reviews applications of pattern recognition and data mining systems in geotechnical engineering with particular reference to constitutive modeling of soils. It covers applications of artificial neural network, genetic programming and evolutionary programming approaches for soil modeling. It is suggested that these systems could be developed as efficient tools for modeling of soils and analysis of geotechnical engineering problems, especially for cases where the behavior is too complex and conventional models are unable to effectively describe various aspects of the behavior. It is also recognized that these techniques are complementary to conventional soil models rather than a substitute to them.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.49 no.3
• /
• pp.21-31
• /
• 2007

This study is aimed at the development of a runoff forecasting model by using the Fuzzy inference system and Neural Network model to solve the uncertainties occurring in the process of rainfall-runoff modeling and improve the modeling accuracy of the stream runoff forecasting. The Neuro-Fuzzy (NF) model were used in this study. The NF model, recently received a great deal of attention, improve the existing Neural Networks by the aid of the Fuzzy theory applied to each node. The study area is the downstreams of Naeseung-chun. Therefore, time-dependent data was obtained from the Wolpo water level gauging station. 11 and 2 out of total 13 flood events were selected for the training and testing set of model respectively. The schematic diagram method and the statistical analysis are conducted to evaluate the feasibility of rainfall-runoff modeling. The model accuracy was rapidly decreased as the forecasting time became longer. The NF model can give accurate runoff forecasts up to 4 hours ahead in standard above the Determination coefficient $(R^2)$ 0.7. In the comparison of the runoff forecasting using the NF and TANK models, characteristics of peak runoff in the TANK model was higher than ones in the NF models, but peak values of hydrograph in the NF models were similar.

• Journal of the Korean Society for Precision Engineering
• /
• v.12 no.11
• /
• pp.118-124
• /
• 1995

In the paper, we have proposed a new technique to determine the initial billet for the forged products using a function approximation in neural network. A three-layer neural network is used and a back propagation algorithm is employed to train the network. An optimal billet which satisfied the forming limitation, minimum of incomplete filling in the die cavity, load and energy as well as more uniform distribution of effective strain, is determined by applying the ability of function approximation of the neural network. The amount of incomplete filling in the die, load and forming energy as well as effective strain are measured by the rigid-plastic finite element method. This new technique is applied to find the optimal billet size for the axisymmetric rib-web product in hot forging. This would reduce the number of finite element simulation for determining the optimal billet of forging products, further it is usefully adopted to physical modeling for the forging design

• Smart Structures and Systems
• /
• v.15 no.4
• /
• pp.1103-1120
• /
• 2015

A self-sensing magnetorheological (MR) damper with embedded piezoelectric force sensor has recently been devised to facilitate real-time close-looped control of structural vibration in a simple and reliable manner. The development and characterization of the self-sensing MR damper are presented based on experimental work, which demonstrates its reliable force sensing and controllable damping capabilities. With the use of experimental data acquired under harmonic loading, a nonparametric dynamic model is formulated to portray the nonlinear behaviors of the self-sensing MR damper based on NARX modeling and neural network techniques. The Bayesian regularization is adopted in the network training procedure to eschew overfitting problem and enhance generalization. Verification results indicate that the developed NARX network model accurately describes the forward dynamics of the self-sensing MR damper and has superior prediction performance and generalization capability over a Bouc-Wen parametric model.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1996.11a
• /
• pp.384-388
• /
• 1996

The closed-loop system modeling of an Active/semiactive suspension system has been accomplished through an artificial neural Networks. The 7DOF full model as the system equation of motion has been derived and the output feedback linear quadratic regulator has been designed for the control purpose. For the neural networks training set of a sample data has been obtained through the computer simulation. A 7DOF full model with LQR controller simulated under the several road conditions such as sinusoidal bumps and the rectangular bumps. A general multilayer perceptron neural network is used for the dynamic modeling and the target outputs are feedback to the input layer. The Backpropagation method is used as the training algorithm. The modeling of system and the model validation have been shown through computer simulations.