• Tribology and Lubricants
• /
• v.16 no.6
• /
• pp.461-468
• /
• 2000

In this study, a neural network based diagnostic system of a rotating spindle system supported by ball bearings was introduced. In order to create actual failure situations, two exemplary abnormal status were made. Out of several possible data source locations, ten measurement spots were chosen. In order to discriminate multiple abnormal status, a neural network system was introduced using back propagation algorithm updating connecting weight between each nodes. In order to find the optimal structure of the neural network system reducing the information sources, magnitude of the weight of the network was referred. Hinton diagram was used to visually inspect the least sensitive weight connecting between input and hidden layers. Number of input node was reduced from 10 to 7 and prediction rate was increased to 100%.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.2521-2525
• /
• 2005

In this paper, we presented a predictive control technique, which is based on wavelet neural network (WNN), for the control of chaotic systems whose precise mathematical models are not available. The WNN is motivated by both the multilayer feedforward neural network definition and wavelet decomposition. The wavelet theory improves the convergence of neural network. In order to design predictive controller effectively, the WNN is used as the predictor whose parameters are tuned by error between the output of actual plant and the output of WNN. Also the training method for the finding a good WNN model is the Extended Kalman algorithm which updates network parameters to converge to the reference signal during a few iterations. The benefit of EKF training method is that the WNN model can have better accuracy for the unknown plant. Finally, through computer simulations, we confirmed the performance of the proposed control method.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.26 no.10
• /
• pp.1518-1527
• /
• 1989

We proposed a technique to design a multimicrocomputer system for real-time parallel processing with an interconnection network which has good network latency time. In order to simplify the performance evaluation and the design procedure under the hard real-time constraints we defined network latency time which takes into account the queueing delays of the networks. We designed a dynamic interconnection network following the proposed technique, and the simulation results show that we can easily estimate the multimicrocomputer system's approximate performance using the defined network latency time before the actual design, so this definition can help the efficient design of the real-time parallel processing systems.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.6
• /
• pp.540-545
• /
• 2012

In this paper, we propose a fault diagnosis method using artificial neural network and SVM (Support Vector Machine) to detect and isolate faults in the nonlinear systems. The proposed algorithm consists of two main parts: fault detection through threshold testing using a artificial neural network and fault isolation by SVM fault classifier. In the proposed method a fault is detected when the errors between the actual system output and the artificial neural network nominal system output cross a predetermined threshold. Once a fault in the nonlinear system is detected the SVM fault classifier isolates the fault. The computer simulation results demonstrate the effectiveness of the proposed SVM and artificial neural network based fault diagnosis method.

• Journal of the Korean Society for Precision Engineering
• /
• v.19 no.4
• /
• pp.73-78
• /
• 2002

This paper presents a neural network approach to estimate the surface roughness by considering the relationship between the polishing operation parameters and the surface roughness. The neural network model predicts the post-machining surface roughness by using several factors such as pre-machining surface roughness, pressure, feed rate, spindle speed, and the number of polishing as inputs. In this paper, the several neural network models are implemented to estimate the surface roughness by using actual experimental data. The experimental results show that the neural network approach is more appropriate to represent the polishing characteristics of mold and die compared with the results obtained by the approach using exponential function.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.05a
• /
• pp.129-133
• /
• 2003

The work described in this paper deals with the proposed application of an Artificial Neural Network Model for the Advanced Pressurized Water Reactor APR-1400 transient identification. The approach adopted for testing the network take note of the expectation which should be fulfilled by a network for real-time application, like testing with data in on-line mode and use of actual or real-life patterns for training. The recall test performed demonstrates that use of neural network for transient identification is indeed an attractive preposition.

• Korean System Dynamics Review
• /
• v.1 no.2
• /
• pp.61-89
• /
• 2000

The purpose of this study is first, to find out what factors affect the cooperation and trust within the functions in the policy implementation network and in what mechanism these factors interact, second to investigate the whys to manage trust and cooperation successfully in the dynamic situation such as the network setting. For these purpose, this study reviews the concept and characteristics of policy implementation organizations, second, extracts the various factors affecting trust and cooperation in the network situation, third applies and analyzes the relationship among factors to system dynamics model based on the game theory. The results of this study could be summarized as follows: It was found that the utility change within the participants by persuasion ＆ mutual understanding and change of rule would be leading to success in policy implementation network. Also bureaucratic management such as power enforcement does not have any good impact in the managing network. In this study, system simulation method tried to analyze the hypothesis. Quantitative and case analyses were not accompanied and analysis was limited to two-person game theory. So there is some doubt this results could be generalized to actual situation which is N-person game.

• Transactions of the Society of Information Storage Systems
• /
• v.6 no.1
• /
• pp.18-24
• /
• 2010

The performance of the speech recognition mainly depends on uncertain factors such as speaker's conditions and environmental effects. The present study deals with the speech recognition of a number of multi-syllable isolated Korean words using soft computing techniques such as back-propagation neural network, fuzzy inference system, and fuzzy neural network. Feature patterns for the speech recognition are analyzed with 12th order thirty frames that are normalized by the linear predictive coding and Cepstrums. Using four models of speech recognizer, actual experiments for both single-speakers and multiple-speakers are conducted. Through this study, the recognizers of combined fuzzy logic and back-propagation neural network and fuzzy neural network show the better performance in identifying the speech recognition.

• Proceedings of the KIEE Conference
• /
• 1998.07g
• /
• pp.2293-2295
• /
• 1998

In this paper, an improved on-line neural network model is suggested. This neural network is designed to store and recall sequence of key strokes in on-line. The network stores incoming patterns as weight connections between series of layers. The layer has a 2-dimensionally distributed neurons where the location of neurons are relevant to the actual location of computer keyboard. To store longer patterns, the network has circulating layer connections and different patterns can be superposed on the same layer. Also, when the patterns are stored over the layers, the starting layer is not fixed but changed by the characteristics of Patterns to increases network capability. The ways how to choose the starting layer during the store and recall process are investigated.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.11
• /
• pp.4483-4501
• /
• 2020

Recent findings identified the scatter and shape of MFD (macroscopic fundamental diagram) is heavily influenced by the spatial distribution of link density in a road network. This implies that the concept of MFD can be utilized to divide a heterogeneous road network with different degrees of congestion into multiple homogeneous subnetworks. Considering the actual traffic data is usually incomplete and inaccurate while most traffic partition algorithms rely on the completeness of the data, we proposed a three-step partitioned algorithm called Iso-MB (Isoperimetric algorithm - Merging - Boundary adjustment) permitting of incompletely input data in this paper. The proposed algorithm was implemented and verified in a simulated urban transportation network. The existence of well-defined MFD in each subnetwork was revealed and discussed and the selection of stop parameter in the isoperimetric algorithm was explained and dissected. The effectiveness of the approach to the missing input data was also demonstrated and elaborated.