• NEWSLETTER 전기공업
• /
• no.99-23 s.240
• /
• pp.30-39
• /
• 1999

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.87-90
• /
• 2020

표 질의응답은 반-정형화된 표 데이터에서 질문에 대한 답을 찾는 문제이다. 본 연구에서는 한국어 표 질의응답을 위한 표 데이터에 적합한 TAPAS를 이용한 언어모델 사전학습 방법과 표에서 정답이 있는 셀을 예측하고 선택된 셀에서 정확한 정답의 경계를 예측하기 위한 표 질의응답 모형을 제안한다. 표 사전학습을 위해서 약 10만 개의 표 데이터를 활용했으며, 텍스트 데이터에 사전학습된 BERT 모델을 이용하여 TAPAS를 사전학습한 모델이 가장 좋은 성능을 보였다. 기계독해 모델을 적용했을 때 EM 46.8%, F1 63.8%로 텍스트 텍스트에 사전학습된 모델로 파인튜닝한 것과 비교하여 EM 6.7%, F1 12.9% 향상된 것을 보였다. 표 질의응답 모델의 경우 TAPAS를 통해 생성된 임베딩을 이용하여 행과 열의 임베딩을 추출하고 TAPAS 임베딩, 행과 열의 임베딩을 결합하여 기계독해 모델을 적용했을 때 EM 63.6%, F1 76.0%의 성능을 보였다.

• Korean Journal of Optics and Photonics
• /
• v.4 no.4
• /
• pp.434-441
• /
• 1993

Focusing geometry dependence of output Stokes energy in single pass methane Raman shifter is investigated. The experimental result shows that the threshold energy decreases as confocal parameter increases. This result can be explained by gain suppression caused by Stokes - anti-Stokes coupling. In this paper, we give simple analysis for the focusing geometry dependence of Stokes - anti-Stokes coupling and present the criterion of confocal parameter to reduce the gain suppression. Focusing geometry dependence of stimulated Brillouin scattering is measured and the result is in good agreement with theoretical prediction.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.21 no.5
• /
• pp.49-55
• /
• 2017

In the seventies, a number of researchers carried out experiments on pullout tests with prototype equipment, and the pullout test was certified as a reliable nondestructive testing(NDT) method for determining the strength of concrete. To estimate the strength of high-strength concrete, we propose a simplified pullout test that uses as a break-off bolt a standard 10mm bolt with a groove on the shaft, an insert nut, and a pullout instrument that includes a hydraulic oil pump without a load cell. To verify the advantages of the simplified pullout test(low cost, simplicity, and convenience), four wall specimens were tested with two levels of concrete strength, 30 MPa and 50 MPa, using a simplified pullout tester with a load cell. The pullout load and concrete compressive strength were measured every day until day 7, day 14, day 21 and day 28. It was found that the pullout load was very similar to the compressive strength. Therefore, we have verified that a simplified pullout test can be used to evaluate the in-place strength of high-strength concrete in structures. The prediction equation of the groove diameter of the break-off bolt(y) with the concrete strength(x) was derived as y=0.05x+3.79, with a coefficient of determination of 0.88 found through regression analysis.

• Journal of the Korea Institute of Building Construction
• /
• v.18 no.1
• /
• pp.17-24
• /
• 2018

The pullout test, a nondestructive testing(NDT), for pre-installed inserts is perhaps the most widely used technique to estimate the in-situ compressive strength of concrete. It measures the force needed to pullout a standardized metal insert embedded into concrete members. The pullout test was certified by the American Society for Testing and Materials(ASTM) and Canadian Standards Association(CSA) as a reliable method for determining the strength of concrete in concrete structures under construction. To easily estimate the strength of ultra-high-strength concrete, a simplified pullout tester, primarily composed of a standard 12mm bolt with a groove on the shaft as a break-off bolt, an insert nut, and a hydraulic oil pump without a load cell, was proposed. Four wall and two slab specimens were tested for two levels of concrete strength, 80MPa and 100MPa, using a simplified pullout tester with a load cell to verify the advantages of the pullout test and simplified pullout test. The compressive strength of concrete, pullout load, and the rupture of the break-off bolt were measured 11 times, day 1 to 7, 14, 21, 28, and 90. The correlation of the pullout load and the compressive strength of each specimen show a higher degree of reliability. Therefore, a simplified pullout test can be used to evaluate the in-place strength of ultra-high-strength concrete in structures. The prediction equation for the groove diameter of the break-off bolt(y) with the concrete strength(x) was proposed as y=0.0184x+5.4. The results described in this research confirm the simplified pullout's utility and potential for low cost, simplicity, and convenience.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.6
• /
• pp.137-143
• /
• 2015

A delay-line type time-to-digital converter (TDC) implemented in a field-programmable gate array (FPGA) is most widely owing due to its simple structure and high conversion rate. However, the delay-line type TDC suffers from nonlinearity error caused by the long delay-line because its time interval measurement range is determined by the length of the used delay line. In this study, a new TDC structure with a shorter delay line by taking a pulse train as an input is proposed for improved time accuracy and efficient use of resources. The proposed TDC utilizes a pulse-train with four transitions and a transition state detector that identifies the used transition among four transitions and prevents the meta-stable state without a synchronizer. With 72 delay cells, the measured resolution and maximum non-linearity were 20.53 ps, and 1.46 LSB, respectively, and the time interval measurement range was 5070 ps which was enhanced by approximately 343 % compared to the conventional delay-line type TDC.