• Title/Summary/Keyword: 3D U-Net

Search Result 68, Processing Time 0.028 seconds

Investigating Data Preprocessing Algorithms of a Deep Learning Postprocessing Model for the Improvement of Sub-Seasonal to Seasonal Climate Predictions (계절내-계절 기후예측의 딥러닝 기반 후보정을 위한 입력자료 전처리 기법 평가)

  • Uran Chung;Jinyoung Rhee;Miae Kim;Soo-Jin Sohn
    • Korean Journal of Agricultural and Forest Meteorology
    • /
    • v.25 no.2
    • /
    • pp.80-98
    • /
    • 2023
  • This study explores the effectiveness of various data preprocessing algorithms for improving subseasonal to seasonal (S2S) climate predictions from six climate forecast models and their Multi-Model Ensemble (MME) using a deep learning-based postprocessing model. A pipeline of data transformation algorithms was constructed to convert raw S2S prediction data into the training data processed with several statistical distribution. A dimensionality reduction algorithm for selecting features through rankings of correlation coefficients between the observed and the input data. The training model in the study was designed with TimeDistributed wrapper applied to all convolutional layers of U-Net: The TimeDistributed wrapper allows a U-Net convolutional layer to be directly applied to 5-dimensional time series data while maintaining the time axis of data, but every input should be at least 3D in U-Net. We found that Robust and Standard transformation algorithms are most suitable for improving S2S predictions. The dimensionality reduction based on feature selections did not significantly improve predictions of daily precipitation for six climate models and even worsened predictions of daily maximum and minimum temperatures. While deep learning-based postprocessing was also improved MME S2S precipitation predictions, it did not have a significant effect on temperature predictions, particularly for the lead time of weeks 1 and 2. Further research is needed to develop an optimal deep learning model for improving S2S temperature predictions by testing various models and parameters.

An assessment of the applicability of multigroup cross sections generated with Monte Carlo method for fast reactor analysis

  • Lin, Ching-Sheng;Yang, Won Sik
    • Nuclear Engineering and Technology
    • /
    • v.52 no.12
    • /
    • pp.2733-2742
    • /
    • 2020
  • This paper presents an assessment of applicability of the multigroup cross sections generated with Monte Carlo tools to the fast reactor analysis based on transport calculations. 33-group cross section sets were generated for simple one- (1-D) and two-dimensional (2-D) sodium-cooled fast reactor problems using the SERPENT code and applied to deterministic steady-state and depletion calculations. Relative to the reference continuous-energy SERPENT results, with the transport corrected P0 scattering cross section, the k-eff value was overestimated by 506 and 588 pcm for 1-D and 2-D problems, respectively, since anisotropic scattering is important in fast reactors. When the scattering order was increased to P5, the 1-D and 2-D problem errors were increased to 577 and 643 pcm, respectively. A sensitivity and uncertainty analysis with the PERSENT code indicated that these large k-eff errors cannot be attributed to the statistical uncertainties of cross sections and they are likely due to the approximate anisotropic scattering matrices determined by scalar flux weighting. The anisotropic scattering cross sections were alternatively generated using the MC2-3 code and merged with the SERPENT cross sections. The mixed cross section set consistently reduced the errors in k-eff, assembly powers, and nuclide densities. For example, in the 2-D calculation with P3 scattering order, the k-eff error was reduced from 634 pcm to -223 pcm. The maximum error in assembly power was reduced from 2.8% to 0.8% and the RMS error was reduced from 1.4% to 0.4%. The maximum error in the nuclide densities at the end of 12-month depletion that occurred in 237Np was reduced from 3.4% to 1.5%. The errors of the other nuclides are also reduced consistently, for example, from 1.1% to 0.1% for 235U, from 2.2% to 0.7% for 238Pu, and from 1.6% to 0.2% for 241Pu. These results indicate that the scalar flux weighted anisotropic scattering cross sections of SERPENT may not be adequate for application to fast reactors where anisotropic scattering is important.

Study on Electronic Structures and Properties in High $T_c\;YBa_2Cu_O_{7-x}\;and\;YBa_2Cu_4O_8$ Superconductors (고온 초전도체 YBa$_2 Cu_3O_{7-x}$와 YBa$_2Cu_4O_8$의 전자구조와 성질에 관한 연구)

  • Son Man-Shick;Ha Hyun-Shick;Paek U-Hyon;Lee Kee-Hag
    • Journal of the Korean Chemical Society
    • /
    • v.35 no.4
    • /
    • pp.316-323
    • /
    • 1991
  • We calculated a difference between the YBa$_2Cu _3O_{7-x}$ superconductor (123 system) of critical temperature, 95 K and the YBa$_2Cu_4 O_8$ superconductors (124 system) of critical temperature, 80 K in Y-system superconductors using Extended Huckel Theory (EHT). The valence electron population (VEP), reduced overlap population (ROP) and net charge for the charged cluster models relating to the layer and the chain in 123 and 124 systems were compared. The VEPs of Cu atom in the layer of 123 and 124 systems populated d$_{z^2}$ orbital more than d$_{x^2-y^2}$ orbital, and in the chain of 123 and 124 systems populated d$_{y^2-z^2}$ orbital more than d$_{z^2}$ orbital. The ROP of the Cu(1)-O(1) in the layer of 123 system was larger than the value of the Cu(1)-O(2), but the ROP of the Cu(1)-O(2) in the layer of 124 system was larger than the value of the Cu(1)-O(1). The ROP of Cu(2)-O(4) in the chain of 123 and 124 systems were larger than the value of the Cu(2)-O(3). In 123 system the net charge values of the Cu in the layer was larger than the value of the Cu in the chain. However, in 124 system the net charge value of the Cu in the chain was larger than the value in the layer.

  • PDF

독서자원과 그 협력방안을 위한 시도적 연구-경상북도를 중심으로

  • 김남석
    • Journal of Korean Library and Information Science Society
    • /
    • v.3
    • /
    • pp.1-31
    • /
    • 1976
  • If Y'ational developments are depend on the dex-elopment of regional >ociety and de~.elopment of regional society are rely on intellcctual delrelopment of members of regional society, going abreas! oi economic capacity. Estnhlis!l:r,ent of libraries as a treasure-house of mental resources and its Jet-elopment are inevit-able and important. But in every respect :f modern libraries are not equipped with contemporary devclij;!~i:c:\ulcornert of society. One c~i the resolving problems, like this, opening the door of the liI~-,\ulcornerr:~fa-c ilities, as one object, which located in every regional areas, u.9ng the existing facilities, the width of the library services mus: I)c opened. I-' ;- I n,> study is aimed at like the following: 1. is analysizing the situation ~vhich the library facilities of one ti:u~:sa!~d five hundred and fifty nine located here and there ir, Ii.;~-ui:gij~opkr ovince in Korea and s i s t ~t.l lousand seven hundred aid tliirl:~ six village libraries, for connecting as a library networks. 2. Tl\ulcornerc ivhole libraries in Kyungbook province and dividing into four ::arts as a Daegu, Phohang, Ahndong and Kimchun area as a transportation area, school area and resources area which exhibiting the view-point of helping-net~vorksl ike the following,

  • PDF

NEUTRONICS MODELING AND SIMULATION OF SHARP FOR FAST REACTOR ANALYSIS

  • Yang, W.S.;Smith, M.A.;Lee, C.H.;Wollaber, A.;Kaushik, D.;Mohamed, A.S.
    • Nuclear Engineering and Technology
    • /
    • v.42 no.5
    • /
    • pp.520-545
    • /
    • 2010
  • This paper presents the neutronics modeling capabilities of the fast reactor simulation system SHARP, which ANL is developing as part of the U.S. DOE's NEAMS program. We discuss the three transport solvers (PN2ND, SN2ND, and MOCFE) implemented in the UNIC code along with the multigroup cross section generation code $MC^2$-3. We describe the solution methods and modeling capabilities, and discuss the improvement needs for each solver, focusing on massively parallel computation. We present the performance test results against various benchmark problems and ZPR-6 and ZPPR critical experiments. We also discuss weak and strong scalability results for the SN2ND solver on the ZPR-6 critical assembly benchmarks.

Determination of Nutrient Contents and In vitro Gas Production Values of Some Legume Forages Grown in the Harran Plain Saline Soils

  • Boga, M.;Yurtseven, S.;Kilic, U.;Aydemir, S.;Polat, T.
    • Asian-Australasian Journal of Animal Sciences
    • /
    • v.27 no.6
    • /
    • pp.825-831
    • /
    • 2014
  • The aim of this study was to determine the nutritive value of some legume species in salt-affected soils of South-East Anatolian region using chemical composition and in vitro gas production kinetics. In this study, Lotus corniculatus, Trifolium alexandrinum, Medicago sativa were sown and tested in four different locations. A 3 by 4 factorial design with 3 legume species and 4 salt levels (non salty electrical conductivity (EC)<4 dS/m; low salt: 4 dS/m>EC<8 dS/m, medium saline: 8 dS/m>EC<16 dS/m and high salt: 16 dS/m>EC) was used in the study. Results indicated that salinity and plants had no significant effect on ash and ether extract. Dry matter (DM), acid detergent fiber, digestible dry matter, dry matter intake (DMI) were affected by plant, salinity and plant${\times}$salinity interaction. On the other hand neutral detergent fiber, relative feed value (RFV), and DMI were affected by salinity and plant${\times}$salinity interaction. Mineral contents were affected by plant species, salinity and salinity${\times}$plants interactions. In vitro gas production, their kinetics and estimated parameters such as were not affected by salinity whereas the gas production up to 48 h, organic matter digestibility, metabolizable energy (ME), and net energy lactation ($NE_L$) were affected by plant and plant${\times}$salt interaction. Generally RFVs of all species ranged from 120 to 210 and were quite satisfactory in salty conditions. Current results show that the feed value of Medicago sativa is higher compared to Lotus corniculatus and Trifolium alexandrinum.

Traditional Chinese Medicine in the Treatment of Mental Retardation;A Review Study (정신지체에 대한 한의학 치료와 연구 동향;2003-2007년 중국 임상 논문에서)

  • Im, Jeong-Gyun;Kim, Lak-Hyung
    • The Journal of Pediatrics of Korean Medicine
    • /
    • v.22 no.1
    • /
    • pp.13-24
    • /
    • 2008
  • Objectives This study was designed to analyze the clinical studies on Mental retardation(MR) in traditional Chinese medicine(TCM). Methods For this study, we searched the clinical studies on MR, which had been published from 2003 to 2007, through web site CNKI(中國知識基礎設施工 http://www.cnki.net). There were 17 clinical studies and we focused on those studies. Results 1. In those Chinese studies, they used following words to describe Mental retardation; 小人弱智($xi\check{a}o\acute{e}rru\grave{o}zh\grave{i}$), 智能發育不全($zh\grave{i}n\acute{e}ngf\bar{a}y\grave{u}buqu\acute{a}n$), 智力低下($zh\grave{i}l\grave{i}d\bar{i}xi\grave{a}$), 精神發育遲滯($j\bar{i}ngsh\acute{e}f\bar{a}y\grave{u}ch\acute{i}zh\grave{i}$), 智能落后($zh\grave{i}n\acute{e}nglu\grave{o}h\grave{o}u$), 智能落后($zh\grave{i}n\acute{e}ngch\acute{i}hu\check{a}n$), 失天愚型患人($xi\bar{a}nti\bar{a}ny\acute{u}x\acute{i}nghu\grave{a}n'\acute{e}r$). 2. There were many kinds of TCM treatment methods for MR, such as herbal medicine, acupuncture, electroacupunture, acupoint injection, Chuna therapy, and special education. And those TCM treatments methods showed higher efficacies in the treatment of MR compared with Western medicine. 3. Mental retardation was related with the deficiency of heart, kidney, liver, spleen(心虛, 腎虛, 肝虛, 脾虛) and the pathological mechanism of Phlegm(絹) and Blood stasis(慫沂)in the studies about the Bian Zheng-the types of differential diagnosis- of MR. 4. Most of studies used Intelligence Quotient(IQ) to assess the efficacy of TCM treatment of MR. And the duration of treatment, the degree of illness, the age of patient, and the cause of illness affected the prognosis of MR. Conclusions These results suggest that traditional medicine could be one of the useful treatments on MR. And these results could be used in the clinical practices and studies on MR in Korea.

  • PDF

Application case for phase III of UAM-LWR benchmark: Uncertainty propagation of thermal-hydraulic macroscopic parameters

  • Mesado, C.;Miro, R.;Verdu, G.
    • Nuclear Engineering and Technology
    • /
    • v.52 no.8
    • /
    • pp.1626-1637
    • /
    • 2020
  • This work covers an important point of the benchmark released by the expert group on Uncertainty Analysis in Modeling of Light Water Reactors. This ambitious benchmark aims to determine the uncertainty in light water reactors systems and processes in all stages of calculation, with emphasis on multi-physics (coupled) and multi-scale simulations. The Gesellschaft für Anlagen und Reaktorsicherheit methodology is used to propagate the thermal-hydraulic uncertainty of macroscopic parameters through TRACE5.0p3/PARCSv3.0 coupled code. The main innovative points achieved in this work are i) a new thermal-hydraulic model is developed with a highly-accurate 3D core discretization plus an iterative process is presented to adjust the 3D bypass flow, ii) a control rod insertion occurrence -which data is obtained from a real PWR test- is used as a transient simulation, iii) two approaches are used for the propagation process: maximum response where the uncertainty and sensitivity analysis is performed for the maximum absolute response and index dependent where the uncertainty and sensitivity analysis is performed at each time step, and iv) RESTING MATLAB code is developed to automate the model generation process and, then, propagate the thermal-hydraulic uncertainty. The input uncertainty information is found in related literature or, if not found, defined based on expert judgment. This paper, first, presents the Gesellschaft für Anlagen und Reaktorsicherheit methodology to propagate the uncertainty in thermal-hydraulic macroscopic parameters and, then, shows the results when the methodology is applied to a PWR reactor.

Bionomics of the Green Peach Aphid(Myzus persicae $S\H{u}lzer$) Adults on Chinese cabbage(Brassica campestris) (복숭아혹진딧물(Myzus persicae $S\H{u}lzer$) 성충의 수명과 생명표)

  • Kim, Ji-Soo;Kim, Tae-Heung;Lee, Sang-Guei
    • Korean journal of applied entomology
    • /
    • v.44 no.3 s.140
    • /
    • pp.213-217
    • /
    • 2005
  • Adult development and fecundity of the green peach aphid, Myzus persicae $S\H{u}lzer$, were studied at $15{\sim}32.5^{\circ}C$ with $60{\sim}70%$ RH under 16L:8D and the results were put together to build a life table. The longevity shortened as temperatures rose at and below $25^{\circ}C$ whereas it remained relatively constant at $27.5^{\circ}C$. Total fecundity was not significantly different at all temperatures except at $30^{\circ}C$. Daily fecundity gradually increased from $15\;to\;22.5^{\circ}C$. It was 5.1 at $25^{\circ}C$ and as temperatures either went down or up, it decreased to 2.8 at both $15^{\circ}C\;and\;30^{\circ}C$. Fecundity model built from total fecundity at various temperatures suggests that $18.3^{\circ}C$ was the optimum temperature for the maximum progeny of 51. Net reproduction rate RO was highest of 36.5 at $27.5^{\circ}C$. The intrinsic rate of increase per day $r_m$ and the finite rate of increase per day $\lambda$ were highest of 0.4 and 1.6, respectively and the doubling time Dt was shortest of 1.6. Constructed life table suggests that the optimum temperature for adult development for the green peach aphid was $27.5^{\circ}C$ at which the mean generation time was 8.1.

Improvement for Transmission Speed of G3 FAX (G3 팩스 전송속도 성능 향상에 관한 연구)

  • Kim, K.T.;Bae, J.I.;Jo, B.K.;Kim, J.B.
    • Proceedings of the KIEE Conference
    • /
    • 1999.07g
    • /
    • pp.3204-3205
    • /
    • 1999
  • G3 팩스가 G1 팩스나 G2 팩스와 구별되는 점은 문서 이미지를 스캔하여 얻은 아날로그 신호(전압)를 디지털 신호로 바꿔주는 A/D 변환과 여기서 얻어진 디지털 신호를 전송을 위해 다시 아날로그 신호(주파수, 위상 진폭)로 바꿔주는 모듈레이션 과정이 중간에 신호를 압축하여 표현하는 코딩 단계가 있다는 것이다. 따라서 G1 팩스나 G2 팩스에 비해 s훨씬 적은 신호를 가지고 문서 이미지를 표현할 수 있으며, 결과적으로 전송속도가 빨라지는 것이 된다. 표준 A4용지는 수평 방향으로 1728개의 픽셀, 수직방향으로 1145라인의 픽셀로 구성된다. 따라서 층 1,978,560($1728{\times}1145$) 비트의 데이터가 A4용지 한 장에 포함될 수 있는 것이다. 팩스가 이런한 표준 A4용지를 스캔하게될 경우 수캔되는 매 라인에는 1728개의 검은색(글자 및 그림) 및 흰색 (여백)픽 셀들이 존재한다. G3 팩스는 서로 다른 Run Length에 비교적 짧은 특정 코드를 미리 할당해 놓고 실제 픽셀들 대신 이러한 코드들을 모듈레이션해서 전송한다. 따라서 전송속도는 단축된다.

  • PDF