서브메뉴
검색
Parallel Optimization with Substructuring Techniques for Structural-Acoustics
Parallel Optimization with Substructuring Techniques for Structural-Acoustics
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105322
- ISBN
- 9798297667112
- DDC
- 910.916
- 서명/저자
- Parallel Optimization with Substructuring Techniques for Structural-Acoustics
- 발행사항
- [Sl] : The Pennsylvania State University, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 297 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-04, Section: B.
- 주기사항
- Advisor: Hanford, Amanda D.
- 학위논문주기
- Thesis (Ph.D.)--The Pennsylvania State University, 2025.
- 초록/해제
- 요약In order to develop a structure that meets specific noise or vibration requirements, designers often test a variety of designs using numerical analysis tools. If a set of random designs is simulated one at a time, the requirements may be met, but this process can be time-consuming and complex once constraints are introduced. Therefore, designers utilize structural-acoustic optimization (SAO) procedures to effectively optimize a structure for low noise, vibration, or other desired acoustic/structural behaviors. Although effective, traditional SAO procedures for frequency-based problems can be inefficient. Finite element analysis matrices are typically utilized to solve for the desired noise or vibration quantity, and a single global or local optimizer is used to guide the numerical solver in calculating an optimal design while considering the constraints and requirements. This process becomes computationally expensive due to the high degrees of freedom in frequency-dependent matrix solutions and lengthy searches of the optimization space using a high number of function evaluations. This research aims to reduce the costs of SAO solvers by decreasing the computation time of both the numerical solver and the optimization algorithms.The methods developed and researched to achieve lower computation times in this dissertation are categorized into three primary areas: structural-acoustic solvers, optimization algorithms, and solutions to SAO problems. The first area utilizes a dynamic substructuring technique to reduce computation time while maintaining accuracy compared to finite element analysis. Within the substructuring field, two new primary improvements are explored: spectral-Galerkin-based substructuring and polynomialbased interface reduction. In the second area, a multi-algorithm optimization approach is examined to replace single-algorithm optimization. This approach integrates various algorithms by sharing information among them to achieve reduced computation times through fewer primary function evaluations. Finally, the third area investigates a novel subproblem decomposition optimization algorithm that combines substructuring with multi-algorithm optimization techniques. The concept stems from decomposing a large SAO problem into smaller subproblems that can be solved via a parallel implementation to reduce computation time. This dissertation will utilize a variety of test problems, ranging from pure structural vibration to complete structural-acoustic sound power radiation minimization, to evaluate new methods developed within each area of research. By either combining the research efforts in these three areas or utilizing the information presented in each individual area, the impact of this research will enable designers and researchers to perform structural-acoustic optimization faster than traditional methods while maintaining an optimal design.
- 일반주제명
- Islands
- 일반주제명
- Decomposition
- 일반주제명
- Polynomials
- 일반주제명
- Acoustics
- 일반주제명
- Optimization algorithms
- 일반주제명
- Radiation
- 일반주제명
- Linear algebra
- 일반주제명
- Mathematics
- 기본자료저록
- Dissertations Abstracts International. 87-04B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017360207
■00520260202105322
■006m o d
■007cr#unu||||||||
■020 ▼a9798297667112
■035 ▼a(MiAaPQ)AAI32289584
■035 ▼a(MiAaPQ)PennState23195mbl5743
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a910.916
■1001 ▼aLuu, Matthew Bryan Chang.
■24510▼aParallel Optimization with Substructuring Techniques for Structural-Acoustics
■260 ▼a[Sl]▼bThe Pennsylvania State University▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a297 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-04, Section: B.
■500 ▼aAdvisor: Hanford, Amanda D.
■5021 ▼aThesis (Ph.D.)--The Pennsylvania State University, 2025.
■520 ▼aIn order to develop a structure that meets specific noise or vibration requirements, designers often test a variety of designs using numerical analysis tools. If a set of random designs is simulated one at a time, the requirements may be met, but this process can be time-consuming and complex once constraints are introduced. Therefore, designers utilize structural-acoustic optimization (SAO) procedures to effectively optimize a structure for low noise, vibration, or other desired acoustic/structural behaviors. Although effective, traditional SAO procedures for frequency-based problems can be inefficient. Finite element analysis matrices are typically utilized to solve for the desired noise or vibration quantity, and a single global or local optimizer is used to guide the numerical solver in calculating an optimal design while considering the constraints and requirements. This process becomes computationally expensive due to the high degrees of freedom in frequency-dependent matrix solutions and lengthy searches of the optimization space using a high number of function evaluations. This research aims to reduce the costs of SAO solvers by decreasing the computation time of both the numerical solver and the optimization algorithms.The methods developed and researched to achieve lower computation times in this dissertation are categorized into three primary areas: structural-acoustic solvers, optimization algorithms, and solutions to SAO problems. The first area utilizes a dynamic substructuring technique to reduce computation time while maintaining accuracy compared to finite element analysis. Within the substructuring field, two new primary improvements are explored: spectral-Galerkin-based substructuring and polynomialbased interface reduction. In the second area, a multi-algorithm optimization approach is examined to replace single-algorithm optimization. This approach integrates various algorithms by sharing information among them to achieve reduced computation times through fewer primary function evaluations. Finally, the third area investigates a novel subproblem decomposition optimization algorithm that combines substructuring with multi-algorithm optimization techniques. The concept stems from decomposing a large SAO problem into smaller subproblems that can be solved via a parallel implementation to reduce computation time. This dissertation will utilize a variety of test problems, ranging from pure structural vibration to complete structural-acoustic sound power radiation minimization, to evaluate new methods developed within each area of research. By either combining the research efforts in these three areas or utilizing the information presented in each individual area, the impact of this research will enable designers and researchers to perform structural-acoustic optimization faster than traditional methods while maintaining an optimal design.
■590 ▼aSchool code: 0176.
■650 4▼aIslands
■650 4▼aDecomposition
■650 4▼aPolynomials
■650 4▼aAcoustics
■650 4▼aOptimization algorithms
■650 4▼aRadiation
■650 4▼aLinear algebra
■650 4▼aMathematics
■690 ▼a0986
■690 ▼a0405
■71020▼aThe Pennsylvania State University.
■7730 ▼tDissertations Abstracts International▼g87-04B.
■790 ▼a0176
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17360207▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


