• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.6
• /
• pp.571-576
• /
• 2002

There have been a lot of considerations dealing with improving the performance of neuro-fuzzy system. The studies on the neuro-fuzzy modeling have largely been devoted to two approaches. First is to improve performance index of system. The other is to reduce the structure size. In spite of its satisfactory result, it should be noted that these are difficult to extend to high dimensional input or to increase the membership functions. We propose a novel neuro-fuzzy system based on the efficient clustering method for initializing the parameters of the premise part. It is a very useful method that maintains a few number of rules and improves the performance. It combine the various algorithms to improve the performance. The Expectation-Maximization algorithm of Gaussian mixture model is an efficient estimation method for unknown parameter estimation of mirture model. The obtained parameters are used for fuzzy clustering method. The proposed method satisfies these two requirements using the Gaussian mixture model and neuro-fuzzy modeling. Experimental results indicate that the proposed method is capable of giving reliable performance.

• Journal of Korea Water Resources Association
• /
• v.50 no.7
• /
• pp.489-502
• /
• 2017

The objective of this study is to propose and evaluate the BAYES-ESP, which is a dam inflow prediction method based on Ensemble Streamflow Prediction method (ESP) and Bayesian theory. ABCD rainfall-runoff model was used to predict monthly dam inflow. Monthly meteorological data collected from KMA, MOLIT and K-water and dam inflow data collected from K-water were used for the model calibration and verification. To estimate the performance of ABCD model, ESP and BAYES-ESP method, time series analysis and skill score (SS) during 1986~2015 were used. In time series analysis monthly ESP dam inflow prediction values were nearly similar for every years, particularly less accurate in wet and dry years. The proposed BAYES-ESP improved the performance of ESP, especially in wet year. The SS was used for quantitative analysis of monthly mean of observed dam inflows, predicted values from ESP and BAYES-ESP. The results indicated that the SS values of ESP were relatively high in January, February and March but negative values in the other months. It also showed that the BAYES-ESP improved ESP when the values from ESP and observation have a relatively apparent linear relationship. We concluded that the existing ESP method has a limitation to predict dam inflow in Korea due to the seasonality of precipitation pattern and the proposed BAYES-ESP is meaningful for improving dam inflow prediction accuracy of ESP.

• Journal of Navigation and Port Research
• /
• v.30 no.6 s.112
• /
• pp.439-446
• /
• 2006

In ship structures many of the structural plates have cutouts, for example, at inner bottom structure, girder, upper deck hatch, floor and dia-frame etc. In the case where a plate has a cutout it experiences reduced buckling and ultimate strength and at the same time the in-plane stress under compressive load produced by hull girder bending will be redistributed In general, actual ship structure adopted reinforcement of stiffener around the cutout in order to preventing from buckling so it need to examine a buckling and ultimate strength behaviour considering a cutout because In many ship yards used class rule for calculating buckling strength but it is difficult to evaluate perforate stiffened plate with random size. In the present paper, we investigated several kinds of perforated stiffened model from actual ship and then was performed finite element series analysis varying the cutout ratio, web height, thickness and type of cross-section using commercial FEA program(ANSYS) under compressive load.

• Korean Chemical Engineering Research
• /
• v.58 no.3
• /
• pp.346-355
• /
• 2020

This study proposed a new optimization-based decision model for an enterprise resource planning and production scheduling of a pharmaceutical intermediates manufacturing plant. To do this work, we first define the inflow and outflow information as well as the model structure, and develop an optimization model to minimize the production time (i.e., makespan) using a mixed integer linear programing (MILP). The unique feature of the proposed model is that the optimal process scheduling is established based on real-time resource logistics information using a radio frequency identification (RFID) technology, thereby theoretically requiring no material inventories. essential information for process operation, such as the required amount of raw materials and estimated arrival timing to manufacturing plant, is used as logistics constraints in the optimization model to yield the optimal manufacturing scheduling to satisfy final production demands. We illustrated the capability of the proposed decision model by applying the optimization model to two scheduling problems in a real pharmaceutical intermediates manufacturing process. As a result, the optimal production schedule and raw materials order timing were identified to minimize the makespan while satisfying all the product demands.

• The korean journal of orthodontics
• /
• v.38 no.6
• /
• pp.376-387
• /
• 2008

Objective: The purpose of this study was to compare the three dimensional changes of tooth movement using four different types of maxillary molar distalization appliances; pendulum appliance (PD), mini-implant supported pendulum appliance (MPD), stainless steel open coil spring (SP) and mini-implant supported stainless steel open coil spring (MSP). Methods: These experiments were performed using the Calorific $machine^{(R)}$ which can simulate dynamic tooth movement. Computed tomography (CT) images of the experimental model were taken before and after tooth movement in 1 mm thicknesses and reconstructed into a three dimensional model using V-works $4.0^{TM}$. These reconstructed images were superimposed using Rapidform $2004^{TM}$ and the direction and amount of tooth movement were measured. Results: The mean reciprocal anchor loss ratio at the first premolar was 17 - 19% for the PD and SP groups. The appliances using mini-implants (MPD or MSP) resulted in less anchorage loss (7 - 8%). On application of a pendulum appliance or MPD, distalization was obtained by tipping rather than by bodily movement. Furthermore, the maxillary second molar tipped distally and bucally. But on application of MSP, distalization was achieved almost by bodily movement. Conclusions: Regarding tooth movement patterns during molar distalization, stainless steel open coil spring with indirect skeletal anchorage was relatively superior to other methods.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.12
• /
• pp.209-218
• /
• 2012

Recently, the Field Programmable Stateful Logic Array (FPSLA) was proposed as one of the most promising system integration technologies which will extend the life of the Moore's law. This work is the first proposal of the FPSLA design automation flow, and the approaches to logic synthesis, synchronization, physical mapping, and automatic placement of the FPSLA designs. The synchronization at each gate for pipelining determines the x-coordinates of cells, and reduces the placement to 1-dimensional problems. The objective function and its gradients for the non-linear optimization of the net length and placement density have been remodeled for the reduced global placement problem. Also, a recursive algorithm has been proposed to legalize the placement by relaxing the density overflow of bipartite bin groups in a top-down hierarchical fashion. The proposed model and algorithm are implemented, and validated by applying them to the ACM/SIGDA benchmark designs. The output state of a gate in an FPSLA needs to be duplicated so that each fanout gate can be connected to a dedicated copy. This property has been taken into account by merging the duplicated nets into a hyperedge, and then, splitting the hyperedge into edges as the optimization progresses. This yields additional 18.4% of the cell count reduction in the most dense logic stage. The practicality of the FPSLA can be further enhanced primarily by incorporating into the logic synthesis the constraint to avoid the concentrated fains of gates on some logic stages. In addition, an efficient algorithm needs to be devised for the routing problem which is based on a complicated graph. The graph models the nanowire crossbar which is trimmed to be embedded into the FPSLA fabric, and therefore, asymmetric. These CAD tools can be used to evaluate the fabric efficiency during the architecture enhancement as well as automate the design.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.8
• /
• pp.96-105
• /
• 1998

The aim of this paper is to propose the method of WDHMM(Weighted DHMM), using the MLP-VQ for the improvement of speaker-independent connect-digit recognition system. MLP neural-network output distribution shows a probability distribution that presents the degree of similarity between each pattern by the non-linear mapping among the input patterns and learning patterns. MLP-VQ is proposed in this paper. It generates codewords by using the output node index which can reach the highest level within MLP neural-network output distribution. Different from the old VQ, the true characteristics of this new MLP-VQ lie in that the degree of similarity between present input patterns and each learned class pattern could be reflected for the recognition model. WDHMM is also proposed. It can use the MLP neural-network output distribution as the way of weighing the symbol generation probability of DHMMs. This newly-suggested method could shorten the time of HMM parameter estimation and recognition. The reason is that it is not necessary to regard symbol generation probability as multi-dimensional normal distribution, as opposed to the old SCHMM. This could also improve the recognition ability by 14.7% higher than DHMM, owing to the increase of small caculation amount. Because it can reflect phone class relations to the recognition model. The result of my research shows that speaker-independent connected-digit recognition, using MLP-VQ and WDHMM, is 84.22%.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.29 no.5
• /
• pp.397-403
• /
• 2016

During the rapid economic growth in Korea since the 1970s, many underground facilities were constructed such as under passes and railways. Seismic design has been mandated in 1988, but the structures built before 1988 were not reflected on the seismic design. Accordingly, these underground structures require effective seismic reinforcing methods to ensure safety when the earthquake happens. By these reasons, in this study, using the proposed pre-flexed members, RC box structure was analyzed for seismic reinforcement of the corner. This method is based on a principle that enlarging the resistance against the external force by installing the pre-flexed member to the box structure corner. To evaluate validity, a newly developed member with CornerSafe was compared with traditional type reinforcement using experiments and finite element analysis. In finite element mode, nonlinearity of steel was modeled based on J2 plasticity model and concrete was based on CEB FIP MODEL CODE 1990. Also, composite ratios of box and pre-flexed member were computed for design application. The reinforcement and box structure were analyzed under the bond condition completely attached by the tie, and the results of experiment and finite element analysis were same in the force-displacement curve.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2019.11a
• /
• pp.150-152
• /
• 2019

In this study, the effects of the wave field changes due to the coastal structure on the hydrodynamic performance of the OWC wave energy, converter are analyzed using a three-dimensional numerical wave tank technique (NWT). The OWC device is simulated numerically by introducing a linear pressure drop model, considering the coupling effect between the turbine and the OWC chamber in the time domain. The flow distribution around the chamber is different due to the change of reflection characteristics depending on the consideration of the breakwater model. The wave energy captured from the breakwater is spatially distributed on the plane of the front of the breakwater, and the converted pneumatic power increased when concentrated in front of the chamber. The change of the standing wave distribution is repeated according to the relationship between the incident wavelength and the length of the breakwater, and the difference in energy conversion performance of the OWC was confirmed.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.6
• /
• pp.533-538
• /
• 2013

The damage caused by Recent frequently occurring locality torrential rains is increasing rapidly. In case of densely populated metropolitan area, casualties and property damage is a serious due to landslides and debris flows and floods. Therefore, the importance of predictions about the torrential is increasing. Precipitation characteristic of the bad weather in Korea is divided into typhoons and torrential rains. This seems to vary depending on the duration and area. Rainfall is difficult to predict because regional precipitation is large volatility and nonlinear. In this paper, Very short-term precipitation forecasting pattern model is implemented using KLAPS data used by Korea Meteorological Administration. we designed very short term precipitation forecasting pattern model using GA-based RBFNNs. the structural and parametric values such as the number of Inputs, polynomial type,number of fcm cluster, and fuzzification coefficient are optimized by GA optimization algorithm.