• Korean Journal of Computational Design and Engineering
• /
• v.10 no.2
• /
• pp.133-142
• /
• 2005

This paper proposes an object-oriented conceptual information model of mechanical assemblies, named open assembly model (OAM). The proposed assembly model primarily defines hierarchical relationships between parts and subassemblies. Together with the assembly hierarchy. the model also provides a way to represent tolerances, kinematic information, and parametric assembly constraints. Relational information such as mating conditions and degree of freedom between parts and subassemblies is captured via assembly features and relationships thereof. The information model is described using class diagrams of the Unified Modeling Language (UML), and instance diagrams are used to exemplify the proposed information model. The conceptual model presented in this paper is an integrated information model for assembly representation, which could supply necessary information for tolerance analysis and synthesis, kinematic simulation, and assembly simulation. Such a conceptual information model plays an important role for the exchange of information between modeling, analysis and planning systems. Hence, the proposed model could serve as a framework for developing data exchange standards of mechanical assemblies. The proposed model is demonstrated through a case study of a planetary gear assembly.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.9
• /
• pp.1402-1408
• /
• 2017

Recently, services for personal biometric data analysis based on real-time monitoring systems has been increasing and many of them have focused on recognition of emotions. In this paper, we propose a classification model to classify anxiety emotion using biometric data actually collected from people. We propose to deploy the support vector machine to build a classification model. In order to improve the classification accuracy, we propose two data pre-processing procedures, which are normalization and data deletion. The proposed algorithms are actually implemented based on Real-time Traffic Flow Measurement structure, which consists of data collection module, data preprocessing module, and creating classification model module. Our experiment results show that the proposed classification model can infers anxiety emotions of people with the accuracy of 65.18%. Moreover, the proposed model with the proposed pre-processing techniques shows the improved accuracy, which is 78.77%. Therefore, we can conclude that the proposed classification model based on the pre-processing process can improve the classification accuracy with lower computation complexity.

• International Journal of Highway Engineering
• /
• v.15 no.3
• /
• pp.45-52
• /
• 2013

PURPOSES : This study is to construct the regression models of drainage asphalt concrete specimens and to provide the appropriate coefficients of hydraulic conductivity prediction models. METHODS: In terms of easy calculation of the hydraulic conductivity from porosity of asphalt concrete pavement, the estimation model of hydraulic conductivity was proposed using regression analysis. 10 specimens of drainage asphalt concrete pavement were made for measurement of the hydraulic conductivity. Hydraulic conductivity model proposed in this study was calculated by empirical model based on porosity and the grain size. In this study, it shows the compared results from permeability measured test and empirical equation, and the suitability of proposed model, using regression analysis. RESULTS: As the result of the regression analysis, the hydraulic conductivity calculated from the proposal model was similar to that resulted from permeability measured test. Also result of RMSE (Root Mean Square Error) analysis, a proposed regression model is resulted in more accurate model. CONCLUSIONS: The proposed model can be used in case of estimating the hydraulic conductivity at drainage asphalt concrete pavements in fields.

• Earthquakes and Structures
• /
• v.18 no.4
• /
• pp.493-506
• /
• 2020

In this paper, a new opposition based charged system search (CSS) is proposed to be used as a parameter identification of highly nonlinear semi-active magneto-rheological damper. By replacing the opposition particles with current solutions, the mentioned strategy is used to enhance the search space and to increase the exploration of CSS. To investigate the effectiveness of the proposed method, a nonlinear modified Bouc-Wen model of MR damper is considered to find its parameters, and compare it with those achieved from experimental model of MR damper. Also, by exploiting the sensitivity analysis and using the importance vector, the less importance parameters in the Bouc-Wen model are eliminated which makes the MR damper model simpler. Results demonstrate the new proposed algorithm (OBLCSS) has a high ability to tackle highly nonlinear problems. Based on the results of the α importance vector, a simplified model is proposed and its parameters are identified by using the presented OBLCSS algorithm. The simplified proposed model also has a high capability of estimating damper responses.

• The Journal of Korea Robotics Society
• /
• v.13 no.1
• /
• pp.8-15
• /
• 2018

This paper proposes a convolutional neural network model for distinguishing areas occupied by obstacles from a LiDAR image converted from a 3D point cloud. The channels of a LiDAR image used as input consist of the distances to 3D points, the reflectivities of 3D points, and the heights of 3D points from the ground. The proposed model uses a LiDAR image as an input and outputs a result of a segmented LiDAR image. The proposed model adopts refinement modules with skip connections to segment a LiDAR image. The refinement modules with skip connections in the proposed model make it possible to construct a complex structure with a small number of parameters than a convolutional neural network model with a linear structure. Using the proposed model, it is possible to distinguish areas in a LiDAR image occupied by obstacles such as vehicles, pedestrians, and bicyclists. The proposed model can be applied to recognize surrounding obstacles and to search for safe paths.

• Computers and Concrete
• /
• v.15 no.1
• /
• pp.37-54
• /
• 2015

In general, cracks significantly deteriorate the in-situ performance of concrete members and structures, especially in urban metro tunnels that have been embedded in saturated soft soils. The microcapsule self-healing method is a newly developed healing method for repairing cracked concrete. To investigate the optimal microcapsule parameters that will have the best healing effect in concrete, a 3D analytical probability healing model is proposed; it is based on the microcapsule self-healing method's healing mechanism, and its purpose is to predict the healing efficiency and healing probability of given cracks. The proposed model comprehensively considers the radius and the volume fraction of microcapsules, the expected healing efficiency, the parameters of cracks, the broken ratio and the healing probability. Furthermore, a simplified probability healing model is proposed to facilitate the calculation. Then, a Monte Carlo test is conducted to verify the proposed 3D analytical probability healing model. Finally, the influences of microcapsules' parameters on the healing efficiency and the healing probability of the microcapsule self-healing method are examined in light of the proposed probability model.

• Wind and Structures
• /
• v.32 no.4
• /
• pp.371-391
• /
• 2021

Tornadoes are the most devastating meteorological natural hazards. Many empirical and theoretical numerical models of tornado vortex have been proposed, because it is difficult to carry out direct measurements of tornado velocity components. However, most of existing numerical models fail to explain the physical structure of tornado vortices. The present paper proposes a new empirical numerical model for a tornado vortex, and its load effects on a low-rise and a tall building are calculated and compared with those for existing numerical models. The velocity components of the proposed model show clear variations with radius and height, showing good agreement with the results of field measurements, wind tunnel experiments and computational fluid dynamics. Normal stresses in the columns of a low-rise building obtained from the proposed model show intermediate values when compared with those obtained from existing numerical models. Local forces on a tall building show clear variation with height and the largest local forces show similar values to most existing numerical models. Local forces increase with increasing turbulence intensity and are found to depend mainly on reference velocity Uref and moving velocity Umov. However, they collapse to one curve for the same normalized velocity Uref / Umov. The effects of reference radius and reference height are found to be small. Resultant fluctuating force of generalized forces obtained from the modified Rankine model is considered to be larger than those obtained from the proposed model. Fluctuating force increases as the integral length scale increases for the modified Rankine model, while they remain almost constant regardless of the integral length scale for the proposed model.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.2
• /
• pp.51-58
• /
• 2023

This paper proposes a deep learning method for estimating the heart rate from facial videos. Our proposed method estimates remote photoplethysmography (rPPG) signals to predict the heart rate. Although there have been proposed several methods for estimating rPPG signals, most previous methods can not be utilized in low-power single board computers due to their computational complexity. To address this problem, we construct a lightweight student model and employ a knowledge distillation technique to reduce the performance degradation of a deeper network model. The teacher model consists of 795k parameters, whereas the student model only contains 24k parameters, and therefore, the inference time was reduced with the factor of 10. By distilling the knowledge of the intermediate feature maps of the teacher model, we improved the accuracy of the student model for estimating the heart rate. Experiments were conducted on the UBFC-rPPG dataset to demonstrate the effectiveness of the proposed method. Moreover, we collected our own dataset to verify the accuracy and processing time of the proposed method on a real-world dataset. Experimental results on a NVIDIA Jetson Nano board demonstrate that our proposed method can infer the heart rate in real time with the mean absolute error of 2.5183 bpm.

• The Journal of Information Systems
• /
• v.26 no.3
• /
• pp.63-90
• /
• 2017

Purpose This study proposes a novel system trading model using case-based reasoning (CBR) based on absolute similarity threshold. The proposed model is designed to optimize the absolute similarity threshold, feature selection, and instance selection of CBR by using genetic algorithm (GA). With these mechanisms, it enables us to yield higher returns from stock market trading. Design/Methodology/Approach The proposed CBR model uses the absolute similarity threshold varying from 0 to 1, which serves as a criterion for selecting appropriate neighbors in the nearest neighbor (NN) algorithm. Since it determines the nearest neighbors on an absolute basis, it fails to select the appropriate neighbors from time to time. In system trading, it is interpreted as the signal of 'hold'. That is, the system trading model proposed in this study makes trading decisions such as 'buy' or 'sell' only if the model produces a clear signal for stock market prediction. Also, in order to improve the prediction accuracy and the rate of return, the proposed model adopts optimal feature selection and instance selection, which are known to be very effective in enhancing the performance of CBR. To validate the usefulness of the proposed model, we applied it to the index trading of KOSPI200 from 2009 to 2016. Findings Experimental results showed that the proposed model with optimal feature or instance selection could yield higher returns compared to the benchmark as well as the various comparison models (including logistic regression, multiple discriminant analysis, artificial neural network, support vector machine, and traditional CBR). In particular, the proposed model with optimal instance selection showed the best rate of return among all the models. This implies that the application of CBR with the absolute similarity threshold as well as the optimal instance selection may be effective in system trading from the perspective of returns.

• Korean Journal of Computational Design and Engineering
• /
• v.5 no.2
• /
• pp.175-183
• /
• 2000

Proposed in this paper is a model-bated AFA (automatic feedrate-adjustment) method for maintaining smooth cutting-loads (i.e., cutting-force) during 2D-profile milling. Before the cutting-force model was established, some assumptions were verified through a series of preliminary cutting experiments (The results found that the curving-force was independent of the cutting speed and the cutting action at the cutter bosom). From the data obtained during the main cutting experiments, a “chip-load/cutting-force model”representing the cutting-force as a function of the chip-load (i.e., effective cutting-depth) and a feedrate is proposed. Based on the model. an AFA scheme for maintaining smooth cutting-force by adjusting the feedrate (i.e., F-code) according to the changes in chip-load was proposed. To check the validity of the proposed AFA scheme. another set of cutting experiments was conducted by using feedrate-adjusted NC-data while monitoring the actual machining processes using an accelerometer. The experimental results showed that the proposed AFA-scheme was quite effective.