서브메뉴
검색
Computational and Data-Driven Analysis and Control of Unsteady Flows in a Pump
Computational and Data-Driven Analysis and Control of Unsteady Flows in a Pump
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105255
- ISBN
- 9798297960862
- DDC
- 620
- 저자명
- Zhong, Yonghong.
- 서명/저자
- Computational and Data-Driven Analysis and Control of Unsteady Flows in a Pump
- 발행사항
- [Sl] : University of California, Los Angeles, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 202 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-04, Section: B.
- 주기사항
- Advisor: Taira, Kunihiko.
- 학위논문주기
- Thesis (Ph.D.)--University of California, Los Angeles, 2025.
- 초록/해제
- 요약Turbomachinery is used in various industries including aerospace, bio, civil and mechanical engineering. Analysis and control of fluid flows in such systems remain a challenge to date owing to their high dimensionality and nonlinear interactions that occur across a multitude of spatial and temporal scales. There is numerous unsteadiness in turbomachinery coming from wake-body interactions, which can result in a loss of operational efficiency. For the purpose of this study, we aim to develop control strategies for unsteadiness attenuation. In support of this objective, we consider two model problems including vortex-blade interaction and flow inside a centrifugal pump. Vortex-blade interaction is a model problem that describes the influence of the vortices on the blade. Such configuration not only happens inside a pump but is also commonly seen in rotorcraft. The second model problem is a flow inside a centrifugal pump. Apart from the unsteadiness that arises from the wake-blade interaction, backward flow, and flow separation can occur near the tongue of a centrifugal pump. Over a range of operating speeds of the pump, the unsteadiness that emerges around the tongue can become significant in its magnitude to greatly affect the efficiency of the pump.While challenging, there exist opportunities to reduce the unsteadiness in turbomachinery by analyzing how the flow unsteadiness influences the overall flow around blades and tongue, and developing a flow modification approach that can attenuate the fluctuations. To this end, we consider the flow analysis, machine-learning based reconstruction, modal analysis, and flow control on the vortex-blade interaction. The flow analysis is to obtain the overall dynamical features of the unsteady flow based on sparse sensor measurements after numerical simulation. Machine learning models are developed to accurately reconstruct lift and drag forces, the pressure distribution on the surface of the blade, and the vorticity field. Once the dynamics of the flow field is captured, the optimally time-dependent (OTD) mode analysis is used to capture the key perturbation dynamics of the unsteady flow. The leading-edge vortex induced by the impingement of the vortex is highlighted as the most receptive region for perturbation amplification. Building upon on the deep understanding of the flow dynamics and perturbation dynamics of vortex-blade interactions, we develop active flow control method to mitigate the fluctuations. Although simple blowing is effective in reducing the lift fluctuation for a discrete vortex-blade interaction, applying blowing and suction is beneficial to achieve long-term reduction in lift fluctuation for continuous vortex impingement.The three-step framework including flow analysis, modal analysis, and flow control is then applied to turbulent flows inside a centrifugal pump. A model pump volute is chosen for studying the impeller-tongue interactions at off-design conditions. Starting with the large-eddy simulations, which is utilized to characterize primary flow features within the volute. The flow unsteadiness is primarily concentrated on the volute side of the tongue for the partial-load flow rate cases. On the other hand, for overload pump flows, large coherent flow structures emerge on the discharge side of the tongue. The backflow appear on the volute side of the tongue contributes to the high unsteadiness of the flow. Resolvent analysis is then applied to the overload pump flow to extract the dominant modal structures that are responsible for flow unsteadiness. Both the boundary layer and the wakes on the discharge side of the tongue are identified as receptive regions for perturbation amplification. The active flow control strategy is then developed for the overload pump flow. To disrupt the large flow structures on the discharge side of the tongue, a local actuator is placed on the tongue surface of its volute side, introducing non-harmonic disturbances that interacts with the oscillatory flow near the tongue. The large flow structures are broken into small ones that dissipate with the aim to reduce the overall fluctuations.The research demonstrates powerful computational and data-driven methodologies for enhancing pump performance and expanding its operational range.
- 일반주제명
- Fluid mechanics
- 일반주제명
- Aerospace engineering
- 일반주제명
- Mechanical engineering
- 키워드
- Turbomachinery
- 키워드
- Machine learning
- 키워드
- Rotorcraft
- 기타저자
- University of California, Los Angeles Mechanical Engineering 0330
- 기본자료저록
- Dissertations Abstracts International. 87-04B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017360043
■00520260202105255
■006m o d
■007cr#unu||||||||
■020 ▼a9798297960862
■035 ▼a(MiAaPQ)AAI32278525
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a620
■1001 ▼aZhong, Yonghong.
■24510▼aComputational and Data-Driven Analysis and Control of Unsteady Flows in a Pump
■260 ▼a[Sl]▼bUniversity of California, Los Angeles▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a202 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-04, Section: B.
■500 ▼aAdvisor: Taira, Kunihiko.
■5021 ▼aThesis (Ph.D.)--University of California, Los Angeles, 2025.
■520 ▼aTurbomachinery is used in various industries including aerospace, bio, civil and mechanical engineering. Analysis and control of fluid flows in such systems remain a challenge to date owing to their high dimensionality and nonlinear interactions that occur across a multitude of spatial and temporal scales. There is numerous unsteadiness in turbomachinery coming from wake-body interactions, which can result in a loss of operational efficiency. For the purpose of this study, we aim to develop control strategies for unsteadiness attenuation. In support of this objective, we consider two model problems including vortex-blade interaction and flow inside a centrifugal pump. Vortex-blade interaction is a model problem that describes the influence of the vortices on the blade. Such configuration not only happens inside a pump but is also commonly seen in rotorcraft. The second model problem is a flow inside a centrifugal pump. Apart from the unsteadiness that arises from the wake-blade interaction, backward flow, and flow separation can occur near the tongue of a centrifugal pump. Over a range of operating speeds of the pump, the unsteadiness that emerges around the tongue can become significant in its magnitude to greatly affect the efficiency of the pump.While challenging, there exist opportunities to reduce the unsteadiness in turbomachinery by analyzing how the flow unsteadiness influences the overall flow around blades and tongue, and developing a flow modification approach that can attenuate the fluctuations. To this end, we consider the flow analysis, machine-learning based reconstruction, modal analysis, and flow control on the vortex-blade interaction. The flow analysis is to obtain the overall dynamical features of the unsteady flow based on sparse sensor measurements after numerical simulation. Machine learning models are developed to accurately reconstruct lift and drag forces, the pressure distribution on the surface of the blade, and the vorticity field. Once the dynamics of the flow field is captured, the optimally time-dependent (OTD) mode analysis is used to capture the key perturbation dynamics of the unsteady flow. The leading-edge vortex induced by the impingement of the vortex is highlighted as the most receptive region for perturbation amplification. Building upon on the deep understanding of the flow dynamics and perturbation dynamics of vortex-blade interactions, we develop active flow control method to mitigate the fluctuations. Although simple blowing is effective in reducing the lift fluctuation for a discrete vortex-blade interaction, applying blowing and suction is beneficial to achieve long-term reduction in lift fluctuation for continuous vortex impingement.The three-step framework including flow analysis, modal analysis, and flow control is then applied to turbulent flows inside a centrifugal pump. A model pump volute is chosen for studying the impeller-tongue interactions at off-design conditions. Starting with the large-eddy simulations, which is utilized to characterize primary flow features within the volute. The flow unsteadiness is primarily concentrated on the volute side of the tongue for the partial-load flow rate cases. On the other hand, for overload pump flows, large coherent flow structures emerge on the discharge side of the tongue. The backflow appear on the volute side of the tongue contributes to the high unsteadiness of the flow. Resolvent analysis is then applied to the overload pump flow to extract the dominant modal structures that are responsible for flow unsteadiness. Both the boundary layer and the wakes on the discharge side of the tongue are identified as receptive regions for perturbation amplification. The active flow control strategy is then developed for the overload pump flow. To disrupt the large flow structures on the discharge side of the tongue, a local actuator is placed on the tongue surface of its volute side, introducing non-harmonic disturbances that interacts with the oscillatory flow near the tongue. The large flow structures are broken into small ones that dissipate with the aim to reduce the overall fluctuations.The research demonstrates powerful computational and data-driven methodologies for enhancing pump performance and expanding its operational range.
■590 ▼aSchool code: 0031.
■650 4▼aFluid mechanics
■650 4▼aAerospace engineering
■650 4▼aMechanical engineering
■653 ▼aTurbomachinery
■653 ▼aOptimally time-dependent
■653 ▼aMachine learning
■653 ▼aRotorcraft
■653 ▼aOperational efficiency
■690 ▼a0204
■690 ▼a0548
■690 ▼a0538
■71020▼aUniversity of California, Los Angeles▼bMechanical Engineering 0330.
■7730 ▼tDissertations Abstracts International▼g87-04B.
■790 ▼a0031
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17360043▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
