• 한국방재학회:학술대회논문집
• /
• 한국방재학회 2008년도 정기총회 및 학술발표대회
• /
• pp.669-673
• /
• 2008

In this study, the new methodology which combines Kohonen self-organizing map(KSOM) neural networks model and the conventional neural networks models such as feedforward neural networks model and generalized neural networks model is introduced to forecast flood stage in Nakdong river, Republic of Korea. It is possible to train without output data in KSOM neural networks model. KSOM neural networks model is used to classify the input data before it combines with the conventional neural networks model. Four types of models such as SOM-FFNNM-BP, SOM-GRNNM-GA, FFNNM-BP, and GRNNM-GA are used to train and test performances respectively. From the statistical analysis for training and testing performances, SOM-GRNNM-GA shows the best results compared with the other models such as SOM-FFNNM-BP, FFNNM-BP, and GRNNM-GA and FFNNM-BP shows vice-versa. From this study, we can suggest the new methodology to forecast flood stage and construct flood warning system in river basin.

• Journal of the Korean Data and Information Science Society
• /
• 제15권2호
• /
• pp.331-337
• /
• 2004

In the analysis of massive volume data, a neural network model is a useful tool. To implement the Neural network model, it is important to select initial value. Since the initial values are generally used as random value in the neural network, the convergent performance and the prediction rate of model are not stable. To overcome the drawback a possible method use samples randomly selected from the whole data set. That is, coefficients estimated by logistic regression based on the samples are the initial values.

• 전자공학회논문지B
• /
• 제33B권2호
• /
• pp.149-158
• /
• 1996

In this paper we describe designing and implementing a digital neural chip and a parallel neural machine for simulating large scale neural netsorks. The chip is a single-chip multiprocessor which has four digiral neural processors (DNP-II) of the same architecture. Each DNP-II has program memory and data memory, and the chip operates in MIMD (multi-instruction, multi-data) parallel processor. The DNP-II has the instruction set tailored to neural computation. Which can be sed to effectively simulate various neural network models including on-chip learning. The DNP-II facilitates four-way data-driven communication supporting the extensibility of parallel systems. The parallel neural machine consists of a host computer, processor boards, a buffer board and an interface board. Each processor board consists of 8*8 array of DNP-II(equivalently 2*2 neural chips). Each processor board acn be built including linear array, 2-D mesh and 2-D torus. This flexibility supports efficiency of mapping from neural network models into parallel strucgure. The neural system accomplishes the performance of maximum 40 GCPS(giga connection per second) with 16 processor boards.

• 센서학회지
• /
• 제28권3호
• /
• pp.182-186
• /
• 2019

When a camera is employed for 3D sensing, accurate camera calibration is vital as it is a prerequisite for the subsequent steps of the sensing process. Camera calibration is usually performed by complex mathematical modeling and geometric analysis. On the other contrary, data learning using an artificial neural network can establish a transformation relation between the 3D space and the 2D camera image without explicit camera modeling. However, a neural network requires a large amount of accurate data for its learning. A significantly large amount of time and work using a precise system setup is needed to collect extensive data accurately in practice. In this study, we propose a two-step neural calibration method that is effective when only a small amount of learning data is available. In the first step, the camera projection transformation matrix is determined using the limited available data. In the second step, the transformation matrix is used for generating a large amount of synthetic data, and the neural network is trained using the generated data. Results of simulation study have shown that the proposed method as valid and effective.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.77-91
• /
• 2022

Various monitoring systems have been implemented in civil infrastructure to ensure structural safety and integrity. In long-term monitoring, these systems generate a large amount of data, where anomalies are not unusual and can pose unique challenges for structural health monitoring applications, such as system identification and damage detection. Therefore, developing efficient techniques is quite essential to recognize the anomalies in monitoring data. In this study, several machine learning techniques are explored and implemented to detect and classify various types of data anomalies. A field dataset, which consists of one month long acceleration data obtained from a long-span cable-stayed bridge in China, is employed to examine the machine learning techniques for automated data anomaly detection. These techniques include the statistic-based pattern recognition network, spectrogram-based convolutional neural network, image-based time history convolutional neural network, image-based time-frequency hybrid convolution neural network (GoogLeNet), and proposed ensemble neural network model. The ensemble model deliberately combines different machine learning models to enhance anomaly classification performance. The results show that all these techniques can successfully detect and classify six types of data anomalies (i.e., missing, minor, outlier, square, trend, drift). Moreover, both image-based time history convolutional neural network and GoogLeNet are further investigated for the capability of autonomous online anomaly classification and found to effectively classify anomalies with decent performance. As seen in comparison with accuracy, the proposed ensemble neural network model outperforms the other three machine learning techniques. This study also evaluates the proposed ensemble neural network model to a blind test dataset. As found in the results, this ensemble model is effective for data anomaly detection and applicable for the signal characteristics changing over time.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 하계학술대회 논문집 C
• /
• pp.1522-1522
• /
• 1999

This paper proposes the methods of short-term load forecasting using Kohonen neural networks and back-propagation neural networks. Historical load data is divided into 5 patterns for the each seasonal data using Kohonen neural networks and using these results, load forecasting neural network is used for next day hourly load forecasting. Normal days and holidays are forecasted. For load forecasting in summer, max-, and min-temperature data are included in neural networks for a better forecasting accuracy. To show the possibility of the proposed method, it was tested with hourly load data of Korea Electric Power Corporation. (1993-1997)

• Journal of Platform Technology
• /
• 제6권1호
• /
• pp.3-8
• /
• 2018

This paper proposes a technique for recognizing online handwritten cursive data obtained by tracing a motion trajectory while a user is in the 3D space based on a convolution neural network (CNN) algorithm. There is a difficulty in recognizing the virtual character input by the user in the 3D space because it includes both the character stroke and the movement stroke. In this paper, we divide syllable into consonant and vowel units by using labeling technique in addition to the result of localizing letter stroke and movement stroke in the previous study. The coordinate information of the separated consonants and vowels are converted into image data, and Korean handwriting recognition was performed using a convolutional neural network. After learning the neural network using 1,680 syllables written by five hand writers, the accuracy is calculated by using the new hand writers who did not participate in the writing of training data. The accuracy of phoneme-based recognition is 98.9% based on convolutional neural network. The proposed method has the advantage of drastically reducing learning data compared to syllable-based learning.

• 설비공학논문집
• /
• 제15권3호
• /
• pp.159-165
• /
• 2003

The fault detection and diagnosis technology may be applied in order to decrease the energy consumption and the maintenance cost of the air-conditioning system. To detect partial faults of the air-conditioning system, a neural network algorithm may be used. In this study, the neural network algorithm using normalized input data by the standard deviation was applied. And the [7$\times$10$\times$10$\times$1] neural network structure was selected. Test results showed that the neural network algorithm using normalized input data was very effective to detect the condenser fouling and the evaporator fan fault of an air-conditioning system.

• 한국정밀공학회지
• /
• 제10권4호
• /
• pp.170-179
• /
• 1993

In this study, self-organized neural network is used to solve the vorrespondence problem of the axial stereo image. Edge points are extracted from a pair of stereo images and then the edge points of rear image are assined to the output nodes of neural network. In the matching process, the two input nodes of neural networks are supplied with the coordi- nates of the edge point selected randomly from the front image. This input data activate optimal output node and its neighbor nodes whose coordinates are thought to be correspondence point for the present input data, and then their weights are allowed to updated. After several iterations of updating, the weights whose coordinates represent rear edge point are converged to the coordinates of the correspondence points in the front image. Because of the feature map properties of self-organized neural network, noise-free and smoothed depth data can be achieved.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.631-634
• /
• 2004

In this paper, the Bayesian recurrent neural network (BRNN) is proposed to predict time series data. Among the various traditional prediction methodologies, a neural network method is considered to be more effective in case of non-linear and non-stationary time series data. A neural network predictor requests proper learning strategy to adjust the network weights, and one need to prepare for non-linear and non-stationary evolution of network weights. The Bayesian neural network in this paper estimates not the single set of weights but the probability distributions of weights. In other words, we sets the weight vector as a state vector of state space method, and estimates its probability distributions in accordance with the Bayesian inference. This approach makes it possible to obtain more exact estimation of the weights. Moreover, in the aspect of network architecture, it is known that the recurrent feedback structure is superior to the feedforward structure for the problem of time series prediction. Therefore, the recurrent network with Bayesian inference, what we call BRNN, is expected to show higher performance than the normal neural network. To verify the performance of the proposed method, the time series data are numerically generated and a neural network predictor is applied on it. As a result, BRNN is proved to show better prediction result than common feedforward Bayesian neural network.