• ETRI Journal
• /
• v.41 no.2
• /
• pp.235-241
• /
• 2019

In this paper, we propose a rank-weighted reconstruction feature to improve the robustness of a feed-forward deep neural network (FFDNN)-based acoustic model. In the FFDNN-based acoustic model, an input feature is constructed by vectorizing a submatrix that is created by slicing the feature vectors of frames within a context window. In this type of feature construction, the appropriate context window size is important because it determines the amount of trivial or discriminative information, such as redundancy, or temporal context of the input features. However, we ascertained whether a single parameter is sufficiently able to control the quantity of information. Therefore, we investigated the input feature construction from the perspectives of rank and nullity, and proposed a rank-weighted reconstruction feature herein, that allows for the retention of speech information components and the reduction in trivial components. The proposed method was evaluated in the TIMIT phone recognition and Wall Street Journal (WSJ) domains. The proposed method reduced the phone error rate of the TIMIT domain from 18.4% to 18.0%, and the word error rate of the WSJ domain from 4.70% to 4.43%.

• Korean Journal of Environmental Biology
• /
• v.36 no.4
• /
• pp.558-566
• /
• 2018

The objective of the study was to determine nutrition regime and limitation in the Boryeng Reservoir where there's a water tunnel between Geum River and the reservoir. Evaluation was conducted through in situ algal bioassay experiments (in situ ABEs) using the cubitainer setting in the reservoirs. For in situ ABEs, we compared and analyzed variations in chlorophyll-a (CHL-a) and phosphorus concentrations in Boryeong Reservoir before and after the water tunnel construction. We then analyzed the nutrient effects on the reservoir. Analysis for nitrogen and phosphorus was done in the three locations of the reservoir and two locations of the ABEs. The in situ ABEs results showed that phosphorous and Nitrogen, the primary limiting nutrient regulating the algal biomass was not limited in the system. The treatments of phosphorus or simultaneous treatments of N+P showed greater algal growth than in the control of nitrate-treatments, indicating a phosphorus deficiency on the phytoplankton growth in the system. The water from the Geum River had 5 times higher total phosphorus (TP) than the water in the reservoir. Efficient management is required as pumping of the river water from Geum River may accelerate the eutrophication of the reservoir.

• Environmental Engineering Research
• /
• v.24 no.3
• /
• pp.397-403
• /
• 2019

Microcystis sp. is one of the most common harmful cyanobacteria that release toxic substances. Counting algal cells is often used for effective control of harmful algal blooms. However, Microcystis sp. is commonly observed as a colony, so counting individual cells is challenging, as it requires significant time and labor. It is urgent to develop an accurate, simple, and rapid method for counting algal cells for regulatory purposes, estimating the status of blooms, and practicing proper management of water resources. The flow cytometer and microscope (FlowCAM), which is a dynamic imaging particle analyzer, can provide a promising alternative for rapid and simple cell counting. However, there is no accurate method for counting individual cells within a Microcystis colony. Furthermore, cell counting based on two-dimensional images may yield inaccurate results and underestimate the number of algal cells in a colony. In this study, a three-dimensional cell counting approach using a novel model algorithm was developed for counting individual cells in a Microcystis colony using a FlowCAM. The developed model algorithm showed satisfactory performance for Microcystis sp. cell counting in water samples collected from two rivers, and can be used for algal management in fresh water systems.

• Journal of Drive and Control
• /
• v.16 no.1
• /
• pp.14-21
• /
• 2019

Axial piston pump is a significant part in wheeled armored vehicle, for generating hydraulic power of vehicle power system. The Axial Piston Pump is a high-performer, core functional item that is developed and applied to most of the military models in the development of military weapon systems. However, in the case of military equipment, there are conditions of limited size and weight required depending on the operating conditions and the operating environment. Under these conditions, it is required that the performance and the environmental resistance are verified to exert the required output. A unique technology is needed for the development of such equipment both in the present and in the future. Therefore, in this study, mathematical modeling of an axial piston pump is presented as a basic data for securing proprietary technology. In addition, a simulation model is designed and compared with the models of six kinds of pistons through simulation. It was established that when the number of pistons of the axial piston pump, which is the development objective, is seven, the model is suitable for the wheeled armored vehicle.

• Advances in concrete construction
• /
• v.7 no.4
• /
• pp.249-262
• /
• 2019

In this study, high strength aluminum alloys (AA) plates are proposed as a new construction material for strengthening reinforced concrete (RC) beams. The purpose of this investigation is to evaluate AA plate's suitability as externally bonded reinforcing (EBR) materials for retrofitting shear deficient beams. A total of twenty RC beams designed to fail in shear were strengthened with different spacing and orientations. The specimens were loaded with four-points loading till failure. The considered outcome parameters included load carrying capacity, deflection, strain in plates, and failure modes. The results of all tested beams showed an increase up to 37% in the load carrying capacity and also an increase in deflection compared to the control un-strengthened beams. This demonstrated the potential of adopting AA plates as EBR material. Finally, the shear contribution from the AA plates was predicted using the models available in the ACI440-08, TR55 and FIB14 design code for fiber reinforced polymer (FRP) plates. The predicted results were compared to experimental testing data with the ratio of the experimentally measured ultimate load to predicted load, range on the average, between 93% and 97%.

• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.35 no.6
• /
• pp.139-146
• /
• 2019

The modular building system is a type of prefabricated construction method, and is an industrialized building system that transports, assembles, and completes a three-dimensional module manufactured in a factory to the site. The economics of a modular building system where 50 to 80% of the entire process takes place in a modular factory is dominated by productivity of the factory manufacturing process. Since the building of the module is finished by the combination of unit parts produced by each material, it is necessary to manage the process in each module unit. However, currently marketed process control programs do not reflect the features of these modular methods. In this paper, we introduce Modular Factory Design software(MFD 2019) that can make modular unit production plan which reflects production base(modular factory) and production target(application and number of modular units). In order to verify software compatibility and reliability, two production plans with different production methods were formulated and simulated.

• Annual Conference of KIPS
• /
• 2013.11a
• /
• pp.1367-1369
• /
• 2013

Applying Application-Level Multicast technology (ALM) to Distributed Interactive Applications (DIAs) becomes more and more popular. Especially for DIAs embedded priority that the sender forwards data to receivers due to their respective priorities. The priority-based directed minimum spanning tree (PST) algorithm was designed for these DIAs. However, the PST has no efficient priority selection and filtering mechanism. The system will consume a tremendous amount of resource for reconstructing distribution tree and becomes unstable and unscalable. In this paper, First, We propose a novel priority-based application level multicast algorithm: Predict-and-Quantize for Priority with directed minimum Spanning Tree (PQPST), which can efficiently predict efficient priorities for the receivers and quantize the predicted priorities to build the multicast distribution tree. Second, we propose Priority Discrepancy Heuristic Mechanism (PDHM), which sets different thresholds of priority discrepancy within the priority discrepancy interval to control the distribution tree construction can efficiently decrease the repeated distribution tree construction, and we get the best heuristic priority discrepancy interval by PQPST. According to the simulation results, the PQPST and PDHM can efficiently improve the performance of the PST algorithm.

• Korean Journal of Comparative Education
• /
• v.24 no.3
• /
• pp.47-69
• /
• 2014

At-risk students are increasing because of developmental trait of adolescence, dissolution of family, poverty, and study stress. These risk situations influence individual student and society negatively. The purpose of this study is explore the reason of occurrence of at-risk student and draw out improvement direction of Gyeongsangbuk-do at-risk student support policy based on current situation of at-risk control and support system implementing in Gyeongsangbuk-do. The improvement ways based on the analysis of Gyeongsangbuk-do's at-risk student support strategy are as the followings. First, it is the construction of at-risk student's prevention and early discovery system. Second, it is the construction of educational safety network for guiding discovered at-risk student, Third, it is the connection with family and community.

• Journal of Drive and Control
• /
• v.18 no.1
• /
• pp.24-30
• /
• 2021

There is an increasing interest in condition-based maintenance for the prevention of economic loss due to failure. Moreover, immense research is being carried out in related technologies in the field of construction machinery. In particular, data-based failure diagnosis methods that employ AI (machine & deep learning) algorithms are in the spotlight. In this study, we have focused on the failure diagnosis and mode classification of reduction gear of excavator's travel device by using the AI algorithm. In addition, a remote monitoring system has been developed that can monitor the status of the reduction gear by using the developed diagnosis algorithm. The failure diagnosis algorithm was performed in the process of data acquisition of normal and abnormal under various operating conditions, data processing and analysis by the wavelet transformation, and learning. The developed algorithm was verified based on three-evaluation conditions. Finally, we have built a system that can check the status of the reduction gear of travel devices on the web using the Edge platform, which is embedded with the failure diagnosis algorithm and cloud.

• Advances in concrete construction
• /
• v.10 no.6
• /
• pp.489-498
• /
• 2020

This paper investigates mechanical properties of roller compacted concrete (RCC) involving reclaimed asphalt pavement (RAP). In this way, a set of 276 cylindrical RCC specimens were prepared with different RAP sizes (i.e., fine, coarse & total) at various ratios (i.e., 10%, 20%, and 40%). Results reveal that incorporation of RAP decreases unconfined compressive strength (UCS), modulus of elasticity (E), and indirect tensile (IDT) strength of RCC. For each RAP size, a regression model was used to maximize RAP content while satisfying the UCS lower limit (27.6 Mpa) mentioned by ACI as a minimum requirement for RCC used in pavement construction. Moreover, UCS of RAP incorporated mixes, dissimilar to that of control mixes, was found to be sensitive and insensitive to the testing temperature and curing time after 7 days, respectively. The results also demonstrate that the higher amounts of RAP, the more flexibility in RCC is. This issue was also proved by the results of modulus of elasticity test. In addition, the toughness index (TI) shows that increase in RAP content leads to up to 43% increase in energy absorbance capacity of RCC.