• Journal of Radiation Industry
• /
• v.17 no.4
• /
• pp.359-367
• /
• 2023

In this study, the function and purpose of the disposal cover, which is an engineering barrier installed to isolate the disposal vault of the near-surface disposal facility for radioactive waste from natural/man-made intrusion, and the design details of the demonstration facility for performance verification were described. The Demonstration facility was designed in a partially divided form to secure the efficiency of measurement while being the same as the actual size of the surface disposal facility to be built in the Intermediate & low-level radioactive waste disposal site of the Korea Radioactive Waste Agency (KORAD). The instruments used for measurement consist of a multi-point thermometer, FDR (Frequency Domain Reflectometry) sensor, inclinometer, acoustic sensor, flow meter, and meteorological observer. It is used as input data for the monitoring system. The 3D monitoring system was composed of 5 layers using the e-government standard framework, and was developed based on 4 components: screen, control module, service module, and DBIO(DataBase Input Output) module, and connected them to system operation. The monitoring system can provide real-time information on physical changes in the demonstration facility through the collection, analysis, storage, and visualization processes.

• International conference on construction engineering and project management
• /
• 2024.07a
• /
• pp.431-438
• /
• 2024

Monitoring unsafe activities on construction sites is challenging due to a variety of factors including the diversity of tasks and workers involved, the potential of human error and lack of real-time hazard detection. With technological advancements, several digital technologies have been proposed and applied to improve the monitoring process. Despite the potential of these technological advancements to reduce manual effort in traditional monitoring, the challenge lies in selecting and implementing the technology that best meets the specific needs of contractors. This paper aims to streamline the research of digital technologies in the construction domain by achieving three key objectives: (1) classify the types of unsafe activities that can be monitored automatically, (2) determine the specific data required for effective monitoring processes, and (3) identify the technologies that can facilitate such data collection process. We conduct a systematic literature review on cutting-edge technological studies to achieve the research aims. The findings of this research serve as a valuable resource for construction practitioners, offering insights into both the benefits and limitations of digital technologies in enhancing the monitoring process. Moreover, the study recommends preparatory elements that practitioners should undertake to integrate these technologies effectively into their monitoring frameworks. The study empowers practitioners by providing a deep understanding, enabling them to create a comprehensive safety management program aligned with the digital transformation process.

• Annual Conference of KIPS
• /
• 2024.10a
• /
• pp.678-681
• /
• 2024

Fine-tuning is critical to enhance the model's ability to operate effectively in resource-constrained environments by incorporating domain-specific data, improving reliability, fairness, and accuracy. Large language models (LLMs) traditionally prefer centralized training due to the ease of managing vast computational resources and having direct access to large, aggregated datasets, which simplifies the optimization process. However, centralized training presents significant drawbacks, including significant delay, substantial communication costs, and slow convergence, particularly when deploying models to devices with limited resources. Our proposed framework addresses these challenges by employing a federated fine-tuning strategy with Whisper-tiny model for keyword recognition system (KWR). Federated learning allows edge devices to perform local updates without the need for constant data transmission to a central server. By selecting a cluster of clients and aggregating their updates each round based on federated averaging, this strategy accelerates convergence, reduces communication overhead, and achieves higher accuracy in comparatively less time, making it more suitable than centralized approach. By the tenth round of federated updates, the fine-tuned model demonstrates notable improvements, achieving over 95.48% test accuracy. We compare the FL-finetuning method with and centralized strategy. Our framework shows significant improvement in accuracy in fewer training rounds.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.10 no.2
• /
• pp.35-46
• /
• 2008

Soil moisture is one of the important components in hydrological processes and also controls the subsurface flow mechanism at a hillslope scale. In this study, time series of soil moisture were measured at a hillslope located in Gwangneung National Arboretum, Korea using a multiplex Time Domain Reflectometry(TDR) system measuring soil moisture with bi-hour interval. The Box-Jenkins transfer function and noise model was used to estimate spatial distributions of soil moisture histories between May and September, 2007. Rainfall was used as an input parameter and soil moisture at 10 cm depth was used as an output parameter in the model. The modeling process consisted of a series of procedures(e.g., data pretreatment, model identification, parameter estimation, and diagnostic checking of selected models), and the relationship between soil moisture and rainfall was assessed. The results indicated that the patterns of soil moisture at different locations and slopes along the hillslope were similar with those of rainfall during the measurment period. However, the spatial distribution of soil moisture was not associated with the slope of the monitored location. This implies that the variability of the soil moisture was determined more by rainfall than by the slope of the site. Due to the influence of vegetation activity on soil moisture flow in spring, the soil moisture prediction in spring showed higher variability and complexity than that in early autumn did. This indicates that vegetation activity is an important factor explaining the patterns of soil moisture for an upland forested hillslope.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.1085-1089
• /
• 2015

Text-mining technique in the past had difficulty in realizing the analysis algorithm due to text complexity and degree of freedom that variables in the text have. Although the algorithm demanded lots of effort to get meaningful result, mechanical text analysis took more time than human text analysis. However, along with the development of hardware and analysis algorithm, big data technology has appeared. Thanks to big data technology, all the previously mentioned problems have been solved while analysis through text-mining is recognized to be valuable as well. However, applying text-mining to Korean text is still at the initial stage due to the linguistic domain characteristics that the Korean language has. If not only the data searching but also the analysis through text-mining is possible, saving the cost of human and material resources required for text analysis will lead efficient resource utilization in numerous public work fields. Thus, in this paper, we compare and evaluate the public document classification by handwork to public document classification where word frequency(TF-IDF) in a text-mining-based text and Cosine similarity between each document have been utilized in big data environment.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.49 no.8
• /
• pp.25-32
• /
• 2012

In this paper, we propose a method for improving the performance of low complexity LS channel estimation for OFDM in fast time varying channels. The CE-BEM channel model used for the low complexity LS channel estimation has a problem on its own and deteriorates channel estimation performance. In this paper, we first use time domain windowing in order to remove the effect of ICI caused by data symbols. Then samples are taken from the results of the LS channel estimation and the effects of the windowing are removed from them. For resolving the defect of CE-BEM, the channel responses are recovered by interpolating the resultant samples with DPSS employed as basis functions the characteristics of which is well matched to the time variation of the channel. Computer simulations show that the proposed channel estimation method gives rise to performance improvement over conventional methods especially when channel variation is very fast and confirm that not only which type of functions is selected for the basis but how many functions are used for the basis is another key factor to performance improvement.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.2E
• /
• pp.26-31
• /
• 1994

A new feature extraction technique utilizing strange attractor and artificial neural network for speaker recognition is presented. Since many signals change their characteristics over long periods of time, simple time-domain processing techniques should e capable of providing useful information of signal features. In many cases, normal time series can be viewed as a dynamical system with a low-dimensional attractor that can be reconstructed from the time series using time delay. The reconstruction of strange attractor is described. In the technique, the raw signal will be reproduced into a geometric three dimensional attractor. Classification decision for speaker recognition is based upon the processing or sets of feature vectors that are derived from the attractor. Three different methods for feature extraction will be discussed. The methods include box-counting dimension, natural measure with regular hexahedron and plank-type box. An artificial neural network is designed for training the feature data generated by the method. The recognition rates are about 82%-96% depending on the extraction method.

• Geophysics and Geophysical Exploration
• /
• v.19 no.3
• /
• pp.131-135
• /
• 2016

Coarray is a parallel processing technique introduced in the Fortran 2008 standard. Coarray can implement parallel processing using simple syntax. In this research, we examined applicability of Coarray to seismic parallel processing by comparing performance of seismic data processing programs using Coarray and MPI. We compared calculation time using seismic wave propagation modeling and one to one communication time using domain decomposition technique. We also compared performance of parallel reverse-time migration programs using Coarray and MPI. Test results show that the computing speed of Coarray method is similar to that of MPI. On the other hand, MPI has superior communication speed to that of Coarray.

• Wind and Structures
• /
• v.34 no.1
• /
• pp.59-72
• /
• 2022

In this work the numerical results of the flow around a 5:1 rectangular cylinder at Reynolds numbers 3 000 and 40 000, zero angle of attack and smooth incoming flow condition are presented. Implicit Large Eddy Simulations (ILES) have been performed with a high-order accurate spatial scheme and an implicit high-order accurate time integration method. The spatial approximation is based on a discontinuous Galerkin (dG) method, while the time integration exploits a linearly-implicit Rosenbrock-type Runge-Kutta scheme. The aim of this work is to show the feasibility of high-fidelity flow simulations with a moderate number of DOFs and large time step sizes. Moreover, the effect of different parameters, i.e., dimension of the computational domain, mesh type, grid resolution, boundary conditions, time step size and polynomial approximation, on the results accuracy is investigated. Our best dG result at Re=3 000 perfectly agrees with a reference DNS obtained using Nek5000 and about 40 times more degrees of freedom. The Re=40 000 computations, which are strongly under-resolved, show a reasonable correspondence with the experimental data of Mannini et al. (2017) and the LES of Zhang and Xu (2020).

• Journal of the Korean GEO-environmental Society
• /
• v.17 no.7
• /
• pp.5-13
• /
• 2016

This research is focused on the seepage behavior of embankment which had the weak zone with big permeability. The distributed TDR (Time Domain Reflectometer) and point sensors such as settlement gauge, pore water pressuremeter, vertical total stressmeter, and FDR (Frequency Domain Reflectometer) sensor were used to measure the seepage characteristics and embankment behavior. Also, the measured data were compared to the data of 2-D and 3-D numerical analysis. The dimension of model embankment was 7 m length, 5 m width and 1.5 m height, which is composed of fine-grained sands and the water level of embankment was 1.3 m height. The seepage behavior of measuring and numerical analysis were very similar, it means that the proper sensing system can monitor the real-time safety of embankment. The result by 2-D and 3-D numerical analysis showed similar saturation processing, however in case of weak zone, the phreatic lines of 2-D showed faster movement than that of 3-D analysis, and finally they converged.