• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2004.04a
• /
• pp.3-8
• /
• 2004

When most manufacturing company produce grinding product operators decide grinding condition by experience and subjective judgment. The study on grinding manufacture have been developed to get the grinding condition with the same result when non-experienced or experienced worker work. The objective of this study is to develope the grinding condition and predict the result of grinding by neural network. Several discussions were made in following areas as; getting MRR with image processing, the architecture optimization of neural network with experiment design, analysis of the input neurons using sensitivity approach. The results showed that the developed approach was the best method in predicting grinding condition with respect to surface finish quality.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.24 no.2
• /
• pp.81-94
• /
• 1999

ATM network design procedure is different from the current circuit or packet networks design procedure because of the variety of the offered services and the variability of requested bandwidth for each connection of ATM network. A number of optimization models for the link dimensioning of ATM network design have been proposed in the literature. However, most of the literature did not consider the modularity of resources allocated to a transmission path and the non-bifurcation of a VP link over the more than one TP, which are standardized in recent ITU-T Recommendations. In thIs paper, we propose a mathematical model for link dimensioning of ATM networks, based on the network synthesis method and a generalized bin-packing problem. The suggested model satisfies the constraints mentioned in the ITU-T Recommendations. We also propose efficient and practical algorithms for the suggested model. Computational experiment shows that the suggested algorithm gives efficient solutions even for moderate and large-sized networks within reasonable time.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2001.12a
• /
• pp.221-224
• /
• 2001

In this paper, we propose the initial optimized structure of the Radial Basis Function Network which is more simple in the part on the structure and converges more faster than Neural Network with the analysis method using Time-Frequency Localization and genetic algorithm. When we construct the hidden node with the Radial Basis Function whose localization is similar with an approximation target function in the plane of the Time and Frequency, we have initial structure of RBFN, After that, we evaluate the parameters of RBF in the network and the parameters needed for the network is more a few. Finally, we make a good decision of the initial structure having an ability of approximation.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.36 no.2
• /
• pp.1-14
• /
• 2011

In this paper, we deal with an access network design problem of fiber-to-the-home passive optical network (FTTH-PON). The FTTH-PON network design problem seeks to minimize the total cost of optical splitters and cables that provide optical connectivity between central office and subscribers. We develop a flow-based mixed integer programming (MIP) model with nonlinear link cost. By developing valid inequalities and preprocessing rules, we enhance the strength of the proposed MIP model in generating tight lower bounds for the problem. We develop an effective Tabu Search (TS) heuristic algorithm that provides good quality feasible solutions to the problem. Computational results demonstrate that the valid inequalities and preprocessing rules are effective for improving the LP-relaxation lower bound and TS algorithm finds good quality solutions within reasonable time bounds.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.54 no.1
• /
• pp.20-29
• /
• 2005

This paper presents an implementation of efficiency optimization of reluctance synchronous motor (RSM) using a neural network (NN) with a direct torque control (DTC). The equipment circuit considered with iron losses in RSM is analyzed theoretically, and the optimal current ratio between torque current and exiting current component are derived analytically. For the RSM driver, torque dynamic can be maintained with DTC using TMS320F2812 DSP Controller even with controlling the flux level because a torque is directly proportional to the stator current unlike induction motor. In order to drive RSM at maximum efficiency and good dynamics response, the Backpropagation Neural Network is adapted. The experimental results are presented to validate the applicability of the proposed method. The developed control system show high efficiency and good dynamic response features with 1.0 [kW] RSM having 2.57 inductance ratio of d/q.

• Proceedings of the KIEE Conference
• /
• 2006.10c
• /
• pp.266-268
• /
• 2006

In this paper, we design a Linear Quadratic Gaussian(LQG) controller for active suspensions. We can improve the inherent suspension problem, trade-off between the ride quality and the suspension travel by selecting appropriate weights in the LQ-objective function. Using an optimization-algorithm, Evolution Strategy(ES), we find the proper control gains for selected frequencies, which have major effects on the vibrations of the vehicle's state variables. The frequencies and proper control gains are used for the neural network data. During a vehicle running, the trained on-line neural network is activated and provides the proper gains for non-trained frequencies.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1530-1533
• /
• 2005

In this paper, we design a Linear Quadratic Gaussian controller for the active suspension. We can improve the inherent suspension problem, trade-off between the ride quality and the suspension travel by selecting appropriate weights in the LQ-objective function. Because any definite rules for selecting weights do not exist, we use an optimization-algorithm, Evolution Strategy (ES) to find the proper control gains for selected frequencies, which have major effects on the vibrations of the vehicle's state variables. The frequencies and proper control gains are used for the neural network data. During a vehicle running, the trained on-line neural network is activated and provides the proper gains for non-trained frequencies. For the full-state feedback control, Kalman filter observes the full states and Fourier transform is used to detect the frequency of the road.

• 한국정보통신설비학회:학술대회논문집
• /
• 2009.08a
• /
• pp.203-207
• /
• 2009

WiBro(Mobile WiMAX) has gained momentum as a top candidate to deliver the dream of full mobile wireless internet. To save cost and time in WiBro network design, simulation tool has to deploy powerful and useful analysis functions. If path loss model is more accurate, the reliability of analysis result of simulation tool will be much improved. So we emphasize on the importance of pathloss model in WiBro network design in this paper. For this, we introduce to three kinds of pathloss models(SUI, SCM, SCM-E) supposed properly models in WiBro RF (Radio Frequency) environment. Also we treat from basic theory to practical substance on the pathloss model to adopt WiBro network design/optimization. Finally, we describe about wireless network analysis tool named 'CellPLAN(R)' and techniques possible to improvethe accuracy of pathloss model.

• Journal of Institute of Control, Robotics and Systems
• /
• v.20 no.7
• /
• pp.750-755
• /
• 2014

Product shortage which causes backordering and/or lost sales cost is very popular in chemical industries, especially in commodity polymer business. This study deals with backordering cost in the supply chain optimization model under the framework of process-inventory network. Classical economic order quantity model with backordering cost suggested optimal time delay and lot size of the final product delivery. Backordering can be compensated by advancing production/transportation of it or purchasing substitute product from third party as well as product delivery delay in supply chain network. Optimal solutions considering all means to recover shortage are more complicated than the classical one. We found three different solutions depending on parametric range and variable bounds. Optimal capacity of production/transportation processes associated with the product in backordering can be different from that when the product is not in backordering. The product shipping cycle time computed in this study was smaller than that optimized by the classical EOQ model.

• Journal of Digital Contents Society
• /
• v.5 no.2
• /
• pp.151-156
• /
• 2004

This paper presents process model to accomplish traffic engineering in Internet. The process model consists of 4 stages. The first stage is the formulation of a control policy dominated network operation. The second stage is the observation of the network state through a set or monitoring functions. The third stage is the characterization or traffic and analysis or the network state. The final stage is the optimization of network performance. the four stages of the process model defined above are iterated.