서브메뉴
검색
Two-Step Reactor Analysis Cross Section Models and Optimal Construction Techniques
Two-Step Reactor Analysis Cross Section Models and Optimal Construction Techniques
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20250211153008
- ISBN
- 9798384044611
- DDC
- 539.76
- 저자명
- Folk, Thomas.
- 서명/저자
- Two-Step Reactor Analysis Cross Section Models and Optimal Construction Techniques
- 발행사항
- [Sl] : University of Michigan, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 259 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-03, Section: B.
- 주기사항
- Advisor: Kochunas, Brendan.
- 학위논문주기
- Thesis (Ph.D.)--University of Michigan, 2024.
- 초록/해제
- 요약This dissertation develops methods to improve the accuracy of cross section models within two-step reactor analysis codes and optimize the case matrices that are used in these procedures. We first examine conventional methods for interpolating cross section functions with respect to instantaneous state variables that capture the immediate effect of the reactor state condition. For each of the three models examined: full table, multiple tables, and partial derivatives, we derive point-wise error expressions and subsequently bound the error in the infinite lattice multiplication factor for a variety of PWR reactor state spaces. These contributions provide a mathematical framework for minimizing the interpolation error in case matrices that has traditionally been achieved through heuristic studies.In the second half of the dissertation, we provide two novel contributions involving the proper calculation of the history variable in the macro-depletion model and a data-driven approach to function approximation using sparse sensing and reconstruction techniques. Previously unknown, we derive an analytic expression for the weight factor used in the burnup-weighted average of the instantaneous reactor state function to account for historical effects due to off-nominal depletion conditions. Using the newly developed linear history expansion method to compute weights specific to the lattice design, state variable, and burnup dependence, we show exceptional accuracy in the computed history variables for several state variables and improve upon the static weights that are conventionally used.We then turn back to the instantaneous effects of the cross section functions and propose a sparse sensing and reconstruction method to optimally generate case matrices requiring fewer transport calculations and providing greater accuracy to the previously studied methodologies. Using sets of basis functions extracted from a Proper Orthogonal Decomposition of a data matrix constructed of cross section functions over a variety of lattice designs, we optimize the instantaneous branches of a case matrix using D-optimality criteria from optimal experimental design. These case matrices along with the reconstruction method are proven to yield superior accuracy to conventional interpolation methods.Together, these techniques provide reactor developers with improved methods to preserve underlying cross section functions - from their generation by the high-fidelity transport code to their use in the low-order core simulator.
- 일반주제명
- Nuclear engineering
- 일반주제명
- Nuclear physics
- 키워드
- Macro-depletion
- 키워드
- Multiple tables
- 기타저자
- University of Michigan Nuclear Engineering & Radiological Sciences
- 기본자료저록
- Dissertations Abstracts International. 86-03B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017164483
■00520250211153008
■006m o d
■007cr#unu||||||||
■020 ▼a9798384044611
■035 ▼a(MiAaPQ)AAI31631417
■035 ▼a(MiAaPQ)umichrackham005835
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a539.76
■1001 ▼aFolk, Thomas.
■24510▼aTwo-Step Reactor Analysis Cross Section Models and Optimal Construction Techniques
■260 ▼a[Sl]▼bUniversity of Michigan▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a259 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-03, Section: B.
■500 ▼aAdvisor: Kochunas, Brendan.
■5021 ▼aThesis (Ph.D.)--University of Michigan, 2024.
■520 ▼aThis dissertation develops methods to improve the accuracy of cross section models within two-step reactor analysis codes and optimize the case matrices that are used in these procedures. We first examine conventional methods for interpolating cross section functions with respect to instantaneous state variables that capture the immediate effect of the reactor state condition. For each of the three models examined: full table, multiple tables, and partial derivatives, we derive point-wise error expressions and subsequently bound the error in the infinite lattice multiplication factor for a variety of PWR reactor state spaces. These contributions provide a mathematical framework for minimizing the interpolation error in case matrices that has traditionally been achieved through heuristic studies.In the second half of the dissertation, we provide two novel contributions involving the proper calculation of the history variable in the macro-depletion model and a data-driven approach to function approximation using sparse sensing and reconstruction techniques. Previously unknown, we derive an analytic expression for the weight factor used in the burnup-weighted average of the instantaneous reactor state function to account for historical effects due to off-nominal depletion conditions. Using the newly developed linear history expansion method to compute weights specific to the lattice design, state variable, and burnup dependence, we show exceptional accuracy in the computed history variables for several state variables and improve upon the static weights that are conventionally used.We then turn back to the instantaneous effects of the cross section functions and propose a sparse sensing and reconstruction method to optimally generate case matrices requiring fewer transport calculations and providing greater accuracy to the previously studied methodologies. Using sets of basis functions extracted from a Proper Orthogonal Decomposition of a data matrix constructed of cross section functions over a variety of lattice designs, we optimize the instantaneous branches of a case matrix using D-optimality criteria from optimal experimental design. These case matrices along with the reconstruction method are proven to yield superior accuracy to conventional interpolation methods.Together, these techniques provide reactor developers with improved methods to preserve underlying cross section functions - from their generation by the high-fidelity transport code to their use in the low-order core simulator.
■590 ▼aSchool code: 0127.
■650 4▼aNuclear engineering
■650 4▼aNuclear physics
■653 ▼aTwo-step methodology
■653 ▼aHomogenized cross section models
■653 ▼aMacro-depletion
■653 ▼aLinear history expansion
■653 ▼aProper Orthogonal Decomposition
■653 ▼aTable interpolation
■653 ▼aMultiple tables
■653 ▼aPartial derivatives
■690 ▼a0552
■690 ▼a0756
■71020▼aUniversity of Michigan▼bNuclear Engineering & Radiological Sciences.
■7730 ▼tDissertations Abstracts International▼g86-03B.
■790 ▼a0127
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17164483▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
