최소 단어 이상 선택하여야 합니다.
최대 10 단어까지만 선택 가능합니다.
다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
NTIS 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
DataON 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
Edison 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
Kafe 바로가기주관연구기관 | 한국과학기술정보연구원 Korea Institute of Science and Technology Information |
---|---|
연구책임자 | 이종숙 |
참여연구자 | 구기범 , 권예진 , 김남규 , 김민아 , 김재성 , 김한기 , 노현지 , 류기명 , 서정현 , 신정훈 , 안선일 , 온누리 , 유석종 , 이세훈 , 이정철 , 이준 , 이준학 , 이중연 , 전인호 , 채희승 , 허영주 , 황규현 , 김한슬 , 황순욱 , 이식 , 최윤근 , 허태상 , 고명주 , 육진희 , 최장원 , 정용환 , 김후성 , 장한빛나래 , 김현섭 , 김덕수 |
보고서유형 | 2단계보고서 |
발행국가 | 대한민국 |
언어 | 한국어 |
발행년월 | 2019-12 |
과제시작연도 | 2019 |
주관부처 | 과학기술정보통신부 Ministry of Science and ICT |
등록번호 | TRKO202000006301 |
과제고유번호 | 1711097338 |
사업명 | 한국과학기술정보연구원연구운영비지원(R&D)(주요사업비) |
DB 구축일자 | 2020-07-29 |
키워드 | 계산과학공학.R&D 효율화.인공지능.플랫폼.과학적 가시화.computational science and engineering.R&D efficiency.AI.platform.Scientific.Visualization. |
ㅇ 계산과학 데이터 플랫폼 기술수준
- 계산과학 데이터 자동분류 방안 연구
- 인공지능 기반 데이터 분석 모델 공유 및 재활용 방안 연구
- 슈퍼컴퓨터 5호기 연계 기술 개발
ㅇ 계산과학공학 플랫폼을 위한 가시화 기술 개발
- 계산과학공학 플랫폼 적용을 위한 4개 핵심 기술 개발
- 세계최고기술수준 대비 70% 수준 달성
- 볼륨렌더링 가시화 성능 기존 대비 최대 16.97배 향상
ㅇ 플랫폼 활용 커뮤니티 협력체계 활성화
- PRAGMA 국제행사
- 계산과학공학 학술모임
<
ㅇ 계산과학 데이터 플랫폼 기술수준
- 계산과학 데이터 자동분류 방안 연구
- 인공지능 기반 데이터 분석 모델 공유 및 재활용 방안 연구
- 슈퍼컴퓨터 5호기 연계 기술 개발
ㅇ 계산과학공학 플랫폼을 위한 가시화 기술 개발
- 계산과학공학 플랫폼 적용을 위한 4개 핵심 기술 개발
- 세계최고기술수준 대비 70% 수준 달성
- 볼륨렌더링 가시화 성능 기존 대비 최대 16.97배 향상
ㅇ 플랫폼 활용 커뮤니티 협력체계 활성화
- PRAGMA 국제행사
- 계산과학공학 학술모임
(출처 : 초록 3p)
IV. Results of the study
□ Development and acquisition of element technology managing and processing data of computational science engineering
- Development of a prediction model of automatic classification of computational science engineering data with high performance(mean F-1 0.995)
◦ A
IV. Results of the study
□ Development and acquisition of element technology managing and processing data of computational science engineering
- Development of a prediction model of automatic classification of computational science engineering data with high performance(mean F-1 0.995)
◦ A study of computational science engineering data characteristics for automatic classification of computational science engineering data
◦ Development of a module extracting more than 1,200 types of characteristics from raw data which includes data redundancy, validationtest, valid keyword, and frequency extraction
◦ Achievement of the performance of a high-level prediction model (mean F-1:0.995, precision:0.996, recall:0.995)
- Development of an environment for solving problems of computational science engineering, data, AI convergence in a way of crowd sourcing method
◦ Support of user-to-user sharing of whole process of AI analysis including computational science engineering raw files, feature data, feature extraction codes, AI analysis code, AI models
◦ Development of extraction environment of large HPC-based AI feature data
◦ Provision of formulaic framework for developing feature data extraction codes and AI models and support development of automated AI model services by recycling the codes
- Acquisition of AI as a Computational Simulation Service element technologies by linking computational science engineering, data, and AI
◦ Development of an easy and convenient way of producing AI based simulation service and executing environment with less programming efforts
◦ Development of data and AI model based service execution environment which can be linked to simulation execution of computational science engineering platform
◦ Management and execution of data and AI model based service that is equipped with and execute diverse AI based service independent ofsimulation execution
- Development of pilot service and AI model based on computational science engineering data for validation of feasibility and utilization of developed platform element technology including error prediction of computational science engineering data
◦ Development of an artificial intelligence model and service predicting energy value of Quantum Espresso material (R2 performance: 0.982)
◦ Development of AI model and service predicting simulation time of Quantum Espresso Relaxation (R2 Performance: 0.845)
◦ Development of simulation error prediction model and service of Quantum Espresso Relaxation (Man F-1 Performance: 0.991)
◦ Development of based pilot AI model and service based on computational science engineering data such as GAMESS, KFLOW
□ The development of visualization technology for computational science engineering platform
- Achievement of 70 percent of visualization technology development level of computational science engineering platform by developing 3 core technology
◦ Completion of developing web socket bidirectional communication technology for platform/visualization server connection
◦ Completion of developing an interactive visualization environment of simulation data in web environment by producing web user interface
◦ Achievement of 70 percent of technology level compared to ParaViewWeb Visualizer, the world's highest level technology
- Realization of real-time visualization by improving performance of detailed algorithms of light-weight server and rendering technology
◦ Achievement of 16.97 times higher real-time volume rendering performance of light-weight visualization server for image-based rendering real-timecontrol
◦ Improvement of web rendering performance by developing rendering technology optimized in web client environment
□ Establishment and activation of platform-based computational science engineering community cooperation system
- Expansion of platform utilization through establishment of domestic and overseas cooperation system
◦ Computational science engineering and domestic and overseas academic events held 5 times
- Activation of computational science engineering community through platform promotion
◦ Won ASOCIO ICT award in Digital Government
◦ KISTI-KOFAC, KISTI-Pohang Technopark-Handong University MOU
(출처 : SUMMARY 10p)
과제명(ProjectTitle) : | - |
---|---|
연구책임자(Manager) : | - |
과제기간(DetailSeriesProject) : | - |
총연구비 (DetailSeriesProject) : | - |
키워드(keyword) : | - |
과제수행기간(LeadAgency) : | - |
연구목표(Goal) : | - |
연구내용(Abstract) : | - |
기대효과(Effect) : | - |
Copyright KISTI. All Rights Reserved.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.