• 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.

• 한국정보과학회논문지:시스템및이론
• /
• 제29권11호
• /
• pp.589-600
• /
• 2002

변형 모델은 볼륨영상으로부터 관심 대상 객체의 3차원적 경계면 구조 추출을 위해 효과적인 접근 방법을 제공한다. 그러나, 기존 변형 모델은 초기 조건에 민감하고, 심한 함몰 및 돌출 부위를 가지는 복잡한 경계면을 잘 표현하지 못하면, 모델 내 구성 요소들 간에 자기교차를 일으킬 수 있는 세가지 주요 제한점이 있다. 본 논문에서는 기존 변형 모델이 갖는 이러한 제한점을 개선함으로써 복잡한 기하학적 표면 형태를 가지는 객체의 경계면 추출에 효과적인 변형 모델을 제안한다. 첫째, 제안 변형 모델은 다해상도 볼륨영상 피라미드를 기반으로 모델구성 요소들을 계층적으로 리샘플링한다. 이 접근은 객체의 경계면을 멀티스케일 방식으로 추출함으로써 초기화에의 의존성을 극복할 뿐 아니라, 모델 구성 요소들의 크기를 복셀 크기에 따라 항상 균일하게 유지함으로써 모델이 영상의 복잡한 특성 정보에 따라 유동적으로 변형될 수 있게 한다. 둘째, 제안 변형 모델은 기존 모델에서 가지는 내력과 외력 외에 자기교차방지력을 포함한다. 자기교차방지력은 제한 거리 이내로 근접한 비인접 모델구성 요소들간에 척력을 적용함으로써 자기교차를 사전에 방지한 수 있게 한다. 본 논문에서는 다양한 합성 볼륨영상 및 뇌 MR 볼륨영상에 대한 실험을 통해서 제안 모델이 초기화 위치에 의존하지 않고 자기교차 없이 복잡한 함몰 및 돌출 경계면 구조를 성공적으로 추출한 결과를 보인다.

• Asian Journal of Atmospheric Environment
• /
• 제4권3호
• /
• pp.150-156
• /
• 2010

Inter-comparison of chemical transport models (CTMs) was conducted among four modeling research groups. Model performance of the ensemble approach to $O_3$ and $PM_{2.5}$ simulation was evaluated by using observational data with a time resolution of 1 or 6 hours at four sites in the Kanto area, Japan, in summer 2007. All groups applied the Community Multiscale Air Quality model. The ensemble average of the four CTMs reproduced well the temporal variation of $O_3$ (r=0.65-0.85) and the daily maximum $O_3$ concentration within a factor of 1.3. By contrast, it underestimated $PM_{2.5}$ concentrations by a factor of 1.4-2, and did not reproduce the $PM_{2.5}$ temporal variation at two suburban sites (r=~0.2). The ensemble average improved the simulation of ${SO_4}^{2-}$, ${NO_3}^-$, and ${NH_4}^+$, whose production pathways are well known. In particular, the ensemble approach effectively simulated ${NO_3}^-$, despite the large variability among CTMs (up to a factor of 10). However, the ensemble average did not improve the simulation of organic aerosols (OAs), underestimating their concentrations by a factor of 5. The contribution of OAs to $PM_{2.5}$ (36-39%) was large, so improvement of the OA simulation model is essential to improve the $PM_{2.5}$ simulation.

• Asian Journal of Atmospheric Environment
• /
• 제12권2호
• /
• pp.139-152
• /
• 2018

The urban model inter-comparison study (UMICS) was conducted in order to improve the performance of air quality models (AQMs) for simulating fine particulate matter ($PM_{2.5}$) in the Greater Tokyo Area of Japan. UMICS consists of three phases: the first phase focusing on elemental carbon (UMICS1), the second phase focusing on sulfate, nitrate and ammonium (UMICS2), and the third phase focusing on organic aerosol (OA) (UMICS 3). In UMICS2/3, all the participating AQMs were the Community Multiscale Air Quality modeling system (CMAQ) with different configurations, and they similarly overestimated $PM_{2.5}$ nitrate concentration and underestimated $PM_{2.5}$ OA concentration. Various sensitivity analyses on CMAQ configurations, emissions and boundary concentrations, and meteorological fields were conducted in order to seek pathways for improvement of $PM_{2.5}$ simulation. The sensitivity analyses revealed that $PM_{2.5}$ nitrate concentration was highly sensitive to emissions of ammonia ($NH_3$) and dry deposition of nitric acid ($HNO_3$) and $NH_3$, and $PM_{2.5}$ OA concentration was highly sensitive to emissions of condensable organic compounds (COC). It was found that $PM_{2.5}$ simulation was substantially improved by using modified monthly profile of $NH_3$ emissions, larger dry deposition velocities of $HNO_3$ and $NH_3$, and additionally estimated COC emissions. Moreover, variability in $PM_{2.5}$ simulation was estimated from the results of all the sensitivity analyses. The variabilities on CMAQ configurations, chemical inputs (emissions and boundary concentrations), and meteorological fields were 6.1-6.5, 9.7-10.9, and 10.3-12.3%, respectively.

• 한국대기환경학회지
• /
• 제34권5호
• /
• pp.687-696
• /
• 2018

In order to develop effective emission abatement strategies for coal-fired power plants, we analyzed the shutdown effects of coal-fired power plants on $PM_{2.5}$ concentration in June by employing air quality model for the period from 2013 to 2016. WRF (Weather Research and Forecast) and CMAQ(Community Multiscale Air Quality) models were used to quantify the impact of emission reductions on the averaged $PM_{2.5}$ concentrations in June over Chungcheongnam-do area in Korea. The resultant shutdown effects showed that the averaged $PM_{2.5}$ concentration in June decreased by 1.2% in Chungcheongnam-do area and decreased by 2.3% in the area where the surface air pollution measuring stations were located. As a result of this study, it was confirmed that it is possible to analyze policy effects considering the change of meteorology and emission and it is possible to quantitatively estimate the influence at the maximum impact region by utilizing the air quality model. The results of this study are expected to be useful as a basic data for analyzing the effect of $PM_{2.5}$ concentration change according to future emission changes.

• Interaction and multiscale mechanics
• /
• 제2권1호
• /
• pp.45-68
• /
• 2009

An aim of the study is to develop an efficient numerical simulation technique that can handle the two-scale analysis of fluid permeation filters fabricated by the partial sintering technique of small spherical ceramics. A solid-fluid mixture homogenization method is introduced to predict the mechanical characters such as rigidity and permeability of the porous ceramic filters from the micro-scale geometry and configuration of partially-sintered particles. An extended finite element (X-FE) discretization technique based on the enriched interpolations of respective characteristic functions at fluid-solid interfaces is proposed for the non-interface-fitted mesh solution of the micro-scale analysis that needs non-slip condition at the interface between solid and fluid phases of the unit cell. The homogenization and localization performances of the proposed method are shown in a typical two-dimensional benchmark problem whose model has a hole in center. Three-dimensional applications to the body-centered cubic (BCC) and face-centered cubic (FCC) unit cell models are also shown in the paper. The 3D application is prepared toward the computer-aided optimal design of ceramic filters. The accuracy and stability of the X-FEM based method are comparable to those of the standard interface-fitted FEM, and are superior to those of the voxel type FEM that is often used in such complex micro geometry cases.

• Nuclear Engineering and Technology
• /
• 제47권5호
• /
• pp.523-533
• /
• 2015

Three-dimensional flow phenomena in a wire-wrapped 37-pin fuel assembly mock-up of a Japanese loop-type sodium-cooled fast reactor, Monju, were investigated with a numerical analysis using a general-purpose commercial computational fluid dynamics code, CFX. Complicated and vortical flow phenomena in the wire-wrapped 37-pin fuel assembly were captured by a Reynolds-averaged Navier-Stokes flow simulation using a shear stress transport turbulence model. The main purpose of the current study is to understand the three-dimensional complex flow phenomena in a wire-wrapped fuel assembly to support the license issue for the core design. Computational fluid dynamics results show good agreement with friction factor correlation models. The secondary flow in the corner and edge subchannels is much stronger than that in an interior subchannel. The axial velocity averaged in the corner and edge subchannels is higher than that averaged in the interior subchannels. Three-dimensional multiscale vortex structures start to be formed by an interaction between secondary flows around each wire-wrapped pin. Behavior of the large-scale vortex structures in the corner and edge subchannels is closely related to the relative position between the hexagonal duct wall and the helically wrapped wire spacer. The small-scale vortex is axially developed in the interior subchannels. Furthermore, a driving force on each wire spacer surface is closely related to the relative position between the hexagonal duct wall and the wire spacer.

• 한국환경과학회지
• /
• 제26권10호
• /
• pp.1167-1180
• /
• 2017

Numerical Weather Prediction (NWP) models such as the Weather Research and Forecasting (WRF) model are essential for forecasting one-day-ahead solar irradiance. In order to evaluate the performance of the WRF in forecasting solar irradiance over the Korean Peninsula, we compared WRF prediction data from 2008 to 2010 corresponding to weather observation data (OBS) from the Korean Meteorological Administration (KMA). The WRF model showed poor performance at polluted regions such as Seoul and Suwon where the relative Root Mean Square Error (rRMSE) is over 30%. Predictions by the WRF model alone had a large amount of potential error because of the lack of actual aerosol radiative feedbacks. For the purpose of reducing this error induced by atmospheric particles, i.e., aerosols, the WRF model was coupled with the Community Multiscale Air Quality (CMAQ) model. The coupled system makes it possible to estimate the radiative feedbacks of aerosols on the solar irradiance. As a result, the solar irradiance estimated by the coupled system showed a strong dependence on both the aerosol spatial distributions and the associated optical properties. In the NF (No Feedback) case, which refers to the WRF-only stimulated system without aerosol feedbacks, the GHI was overestimated by $50-200W\;m^{-2}$ compared with OBS derived values at each site. In the YF (Yes Feedback) case, in contrast, which refers to the WRF-CMAQ two-way coupled system, the rRMSE was significantly improved by 3.1-3.7% at Suwon and Seoul where the Particulate Matter (PM) concentrations, specifically, those related to the $PM_{10}$ size fraction, were over $100{\mu}g\;m^{-3}$. Thus, the coupled system showed promise for acquiring more accurate solar irradiance forecasts.

• 대기
• /
• 제32권4호
• /
• pp.341-351
• /
• 2022

Recently, interest in the possibility of a washout effect using artificial rain enhancement technology to reduce high-concentration fine dust is growing. Therefore, in this study, the reduction rate of PM₁₀ concentration according to the amount of artificial rain enhancement was calculated during Asian Dust event which occurred over the Korean Peninsula on March 29, 2021 using air quality model [i.e., Community Multiscale Air Quality (CMAQ)] combined with the mesoscale model for artificial rain enhancement (i.e., WRF-MMS). According to WRF-MMS, the washout effect lasted 5 hours, and the maximum precipitation rate was calculated to be 1.5 mm hr^-1. According the CMAQ results, the PM₁₀ reduction rate was up to 22%, and the affected area was calculated to be 6.4 times greater than that of the artificial rain enhancement area. Even if the maximum amount of precipitation per hour is lowered to 0.8 mm hr^-1 (about 50% level), the PM₁₀ reduction rate appears to be up to 16%. In other words, it is believed that this technique can be used as a direct method for reducing high-concentration fine dust even when the artificial rain enhancement effect is weak.

• Journal of Radiation Protection and Research
• /
• 제48권1호
• /
• pp.28-43
• /
• 2023

Background: High-fidelity meteorological data is a prerequisite for the realistic simulation of atmospheric dispersion of radioactive materials near nuclear power plants (NPPs). However, many meteorological models frequently overestimate near-surface wind speeds, failing to represent local meteorological conditions near NPPs. This study presents a new high-resolution (approximately 1 km) meteorological downscaling method for modeling short-range (< 100 km) atmospheric dispersion of accidental NPP plumes. Materials and Methods: Six considerations from literature reviews have been suggested for a new dynamic downscaling method. The dynamic downscaling method is developed based on the Weather Research and Forecasting (WRF) model version 3.6.1, applying high-resolution land-use and topography data. In addition, a new subgrid-scale topographic drag parameterization has been implemented for a realistic representation of the atmospheric surface-layer momentum transfer. Finally, a year-long simulation for the Kori and Wolsong NPPs, located in southeastern coastal areas, has been made for 2016 and evaluated against operational surface meteorological measurements and the NPPs' on-site weather stations. Results and Discussion: The new dynamic downscaling method can represent multiscale atmospheric motions from the synoptic to the boundary-layer scales and produce three-dimensional local meteorological fields near the NPPs with a 1.2 km grid resolution. Comparing the year-long simulation against the measurements showed a salient improvement in simulating near-surface wind fields by reducing the root mean square error of approximately 1 m/s. Furthermore, the improved wind field simulation led to a better agreement in the Eulerian estimate of the local atmospheric dispersion. The new subgrid-scale topographic drag parameterization was essential for improved performance, suggesting the importance of the subgrid-scale momentum interactions in the atmospheric surface layer. Conclusion: A new dynamic downscaling method has been developed to produce high-resolution local meteorological fields around the Kori and Wolsong NPPs, which can be used in short-range atmospheric dispersion modeling near the NPPs.