• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권7호
• /
• pp.1759-1772
• /
• 2023

Chinese named entity recognition (NER) is a challenging work that seeks to find, recognize and classify various types of information elements in unstructured text. Due to the Chinese text has no natural boundary like the spaces in the English text, Chinese named entity identification is much more difficult. At present, most deep learning based NER models are developed using a bidirectional long short-term memory network (BiLSTM), yet the performance still has some space to improve. To further improve their performance in Chinese NER tasks, we propose a new NER model, IDCNN-BiLSTM-Highway, which is a combination of the BiLSTM, the iterated dilated convolutional neural network (IDCNN) and the highway network. In our model, IDCNN is used to achieve multiscale context aggregation from a long sequence of words. Highway network is used to effectively connect different layers of networks, allowing information to pass through network layers smoothly without attenuation. Finally, the global optimum tag result is obtained by introducing conditional random field (CRF). The experimental results show that compared with other popular deep learning-based NER models, our model shows superior performance on two Chinese NER data sets: Resume and Yidu-S4k, The F1-scores are 94.98 and 77.59, respectively.

• 생태와환경
• /
• 제42권4호
• /
• pp.476-486
• /
• 2009

Major objective of our study was to introduce initial researches of national long-term ecological monitoring studies on Daechung Reservoir, as one of the representative lentic reservoir ecosystems in Korea. For the long-term ecological research (LTER), we conducted preliminary field monitoring during 2008~2009 and analyzed biological parameters such as phytoplankton, zooplankton, and freshwater fish along with chemical water quality and empirical model analysis. According to phytoplankton surveys, major taxa have varied largely depending on seasons and sites sampled. Overall phytoplankton data showed that cyanophyta dominated in the summer period and diatoms dominated in the winter. In zooplankton analysis, 25 species including 20 rotifers, 3 cladocerans and 2 copepods were collected during the survey. The relative abundance of rotifers (86.5%) was always greater than that of cladocerans (6.3%) or copepods (5.1%). There were distinct spatial and inter-annual changes in the abundance of zooplankton in the reservoir, displaying similar patterns in three sites with the exception of S3 during the study. According to fish surveys, 8 families and 39 species were observed during 2008~2009. The most dominant fish was an exotic species of Lepomis macrochirus (23%), indicating an severe influence of exotic species to the ecosystem. TP averaged $17.9\;{\mu}g\;L^{-1}$ ($6{\sim}80\;{\mu}g\;L^{-1}$), which was judged as a mesotrophy, and showed a distinct longitudinal gradients. TN averaged $1.585\;{\mu}g\;L^{-1}$ during the study and judged as hypereutrophic condition. Unlike TP, TN didn't show any large seasonal and spatial variations. Under the circumstances, nitrogen limitation may not happen in this system, indicating that nitrogen control is not effective in the watershed managements. These data generated in the LTER station will provide key information on long-term biological and water quality changes in relation to global warming and some clues for efficient reservoir ecosystem managements.

• Smart Structures and Systems
• /
• 제14권2호
• /
• pp.209-224
• /
• 2014

With more and more built long-term structural health monitoring (SHM) systems, it has been considered to apply monitored data to learn the reliability of bridges. In this paper, based on a long-term SHM system, especially in which the sensors were embedded from the beginning of the construction of the bridge, a method to calculate the localized reliability around an embedded sensor is recommended and implemented. In the reliability analysis, the probability distribution of loading can be the statistics of stress transferred from the monitored strain which covered the effects of both the live and dead loads directly, and it means that the mean value and deviation of loads are fully derived from the monitored data. The probability distribution of resistance may be the statistics of strength of the material of the bridge accordingly. With five years' monitored strains, the localized reliabilities around the monitoring sensors of a bridge were computed by the method. Further, the monitored stresses are classified into two time segments in one year period to count the loading probability distribution according to the local climate conditions, which helps us to learn the reliability in different time segments and their evolvement trends. The results show that reliabilities and their evolvement trends in different parts of the bridge are different though they are all reliable yet. The method recommended in this paper is feasible to learn the localized reliabilities revealed from monitored data of a long-term SHM system of bridges, which would help bridge engineers and managers to decide a bridge inspection or maintenance strategy.

• 한국지반공학회논문집
• /
• 제37권8호
• /
• pp.25-35
• /
• 2021

현재 국내에서는 비탈면 및 구조물 안정성 확보를 목적으로 네일, 록볼트 등과 함께 그라운드 앵커 공법을 사용하고 있다. 이 중 장기목적으로 사용되는 영구앵커의 경우 사용기간 중 지지력과 내구성이 확보되어야 하나, 최근 연구 결과에 따르면 장기거동에 따른 잔존 긴장력 감소와 비탈면 변형 등의 현상이 보고되고 있다. 이와 같은 잔존 긴장력 감소 문제는 앞으로 지속적으로 증가될 것으로 전망되며, 이로 인한 유지관리 비용 증가 등의 문제가 불가피할 것으로 보인다. 이에 본 연구에서는 국내·외 문헌연구를 통해 영구앵커의 긴장력에 영향을 미치는 요인들을 파악하였으며, 과거 수행된 하중계 모니터링 자료를 분석한 선행연구들을 조사하였다. 이후, 이를 기초자료로 활용하여 실제 현장에서 수집한 하중계 계측자료를 분석하여 앵커의 긴장력 감소현황을 파악하였고, 그 장기하중감소특성을 분석하였다. 마지막으로 앞의 내용을 종합하여, 설치 직후 100일 부근의 단기 데이터를 통해 영구앵커의 장기하중감소특성을 예측하는 기법을 제안하였다.

• Smart Structures and Systems
• /
• 제20권2호
• /
• pp.197-205
• /
• 2017

The dynamic performance of railway bridges under high-speed trains draws the attention of bridge engineers. The vibration issue for long-span bridges under high-speed trains is still not well understood due to lack of validations through structural health monitoring (SHM) data. This paper investigates the correlation between bridge acceleration and train speed based on structural dynamics theory and SHM system from three foci. Firstly, the calculated formula of acceleration response under a series of moving load is deduced for the situation that train length is near the length of the bridge span, the correlation between train speed and acceleration amplitude is analyzed. Secondly, the correlation scatterplots of the speed-acceleration is presented and discussed based on the transverse and vertical acceleration response data of Dashengguan Yangtze River Bridge SHM system. Thirdly, the warning indexes of the bridge performance for correlation scatterplots of speed-acceleration are established. The main conclusions are: (1) The resonance between trains and the bridge is unlikely to happen for long-span bridge, but a multimodal correlation curve between train speed and acceleration amplitude exists after the resonance speed; (2) Based on SHM data, multimodal correlation scatterplots of speed-acceleration exist and they have similar trends with the calculated formula; (3) An envelope line of polylines can be used as early warning indicators of the changes of bridge performance due to the changes of slope of envelope line and peak speed of amplitude. This work also gives several suggestions which lay a foundation for the better design, maintenance and long-term monitoring of a long-span high-speed bridge.

• Journal of Power Electronics
• /
• 제8권4호
• /
• pp.325-331
• /
• 2008

Power system perturbations are due to many reasons; one of the most common perturbation causes is switching off/on the power capacitors. This paper monitors and discusses the overvoltages which appear on local and remote capacitor connected buses in power systems. Using the Fast Fourier Transfer (FFT), the total harmonic content of voltages and currents waveforms is also estimated at all buses. The power factor during different cases of switching modes "off/on" is monitored. The monitoring technique tackles not only the longitudinal long distance mutual effects of switching power capacitors between different buses but also evaluates the overvoltage durations. A relative long term monitoring is implemented using the Matlab/Simulink environment to show severity assessments in different switching modes on the transformers' voltages and currents' waveforms.

• 한국지하수토양환경학회지:지하수토양환경
• /
• 제21권4호
• /
• pp.30-41
• /
• 2016

The amount of groundwater related data is drastically increasing domestically from various sources since 2000. To justify the more expansive continuation of the data acquisition and to derive valuable implications from the data, continued employments of sophisticated and state-of-the-arts statistical tools in the analyses and predictions are important issue. In the present study, we employed a well established machine learning technique of Gaussian Process Regression (GPR) model in the trend analyses of groundwater level for the long-term change. The major benefit of GPR model is that the model provide not only the future predictions but also the associated uncertainty. In the study, the long-term predictions of groundwater level from the stations of National Groundwater Monitoring Network located within Han River Basin were exemplified as prediction cases based on the GPR model. In addition, a few types of groundwater change patterns were delineated (i.e., increasing, decreasing, and no trend) on the basis of the statistics acquired from GPR analyses. From the study, it was found that the majority of the monitoring stations has decreasing trend while small portion shows increasing or no trend. To further analyze the causes of the trend, the corresponding precipitation data were jointly analyzed by the same method (i.e., GPR). Based on the analyses, the major cause of decreasing trend of groundwater level is attributed to reduction of precipitation rate whereas a few of the stations show weak relationship between the pattern of groundwater level changes and precipitation.

• 대한토목학회논문집
• /
• 제32권2A호
• /
• pp.97-108
• /
• 2012

재킷식 해양구조물인 울돌목 시험조류발전소에 대하여 장기 모니터링을 통하여 구조물 동적 응답을 계측하였으며, 계측된 동적 응답 중 저주파수 거동을 정밀하게 계측할 수 있는 동적 경사 응답을 이용하여 구조물의 고유주파수 및 모드감쇠비를 추정하고, 이와 같은 동특성이 조위와 조류 유속 등 외부 환경에 의하여 어떤 영향을 받는지를 분석하였다. 제한된 수의 응답 계측 자료로부터 구조물의 고유주파수 및 모드감쇠비를 정밀하게 추정하기 위하여 개선된 실험모드해석 방법인 LS-FDD 방법을 제안하였으며, 제안된 실험모드해석 기법을 이용하여 울돌목 시험조류발전소의 동적 경사 응답을 분석하여, 주요 3차모드의 고유주파수와 모드감쇠비를 정밀하게 추정하였다. 추정된 동특성은 시간에 따라 크게 변동하며, 이러한 변동은 조석의 영향을 지배적으로 받고 있음을 시계열 분석 및 주파수 분석을 통하여 알 수 있었다. 또한 울돌목 시험조류발전소에서 관측한 일정 기간의 조위 및 조류 유속 자료를 이용하여, 구조물의 동특성과 조류 자료 사이의 상관관계를 분석하였고, 조위 및 유속 자료만으로 구조물의 동특성을 예측할 수 있는 모델식을 결정하였다.

• Advances in concrete construction
• /
• 제10권6호
• /
• pp.547-557
• /
• 2020

This paper presents one of the demonstration projects undertaken during the FP7 EU-funded Anagennisi project (Innovative reuse of all tyre components in concrete-2014-2017) on a full-scale (30 m×40 m, thickness: 0.2 m) Steel Fibre Reinforced Concrete (SFRC) slab-on-grade using a blend of manufactured steel fibres (MSF) and Recycled Tyre Steel Fibres (RTSF). The aim of the project was to assess the use of RTSF in everyday construction practice. The Anagennisi partners, Dulex Ltd in collaboration with Gradmont-Gradacac Ltd and University of Zagreb, designed, cast and monitored the long-term shrinkage deformations of the indoor slab-on-grade slab at Gradmont's precast concrete factory in Gradacac, Bosnia and Herzegovina. A hybrid RTSF mix (20 kg/㎥ of MSF+10 kg/㎥ of RTSF) was used to comply with the design criteria which included a maximum load capacity of 20 kN/㎡. The slab was monitored for one year using surveying equipment and visual inspection of cracks. During the monitoring period, the slab exhibited reasonable deformations (a maximum displacement of 3.3 mm for both, horizontal and vertical displacements) whilst after five years in use, the owners did not report any issues and were satisfied with the construction methodology and materials used. This work confirms that RSTF is a viable and sustainable solution for slab-on-grade applications.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2006년도 정기 학술대회 논문집
• /
• pp.248-255
• /
• 2006

Among a variety of influencing components, time-variant seepage and long-term underground motion are important to understand the abnormal behavior of tunnels. Excessiveness of these two components could be the direct cause of severe damage on tunnels. however, it is not easy to quantify the effect of these on the behavior of tunnels. These parameters can be estimated by using inverse methods once the appropriate relationship between inputs and results are clarified. Various inverse methods or parameter estimation techniques such as artificial neural network and least square method can be used depending on the characteristics of given problems. Numerical analyses, experiments, or monitoring results are frequently used to prepare a set of inputs and results to establish the back analysis models. In this study, a back analysis method has been developed to estimate geotechnically hard-to-known parameters such as permeability of tunnel filter, underground water table, long-term rock mass load, size of damaged zone associated with seepage and long-term underground motion. The artificial neural network technique is adopted and the numerical models developed in the firstpart are used to prepare a set of data for learning process. Tunnel behavior especially the displacements of the lining has been exclusively investigated for the back analysis.