Estimating flow fields with reduced order models

Sommer, Kamil David; Reineking, Lucas; Ravichandran, Yogesh Parry; Skoda, Romuald; Mönnigmann, Martin

doi:10.1016/j.heliyon.2023.e20930

Cited by 4 publications

(3 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a result, a strategy to overcome temporal error propagation needs to be created in order to enable long‐term prediction for non‐periodic dynamics. In the future, we aim to improve the temporal model and apply the presented framework to predict the flow field in highly packed bed configurations 4…”

Section: Discussionmentioning

confidence: 99%

“…The transformation from the system's physical coordinates to the modal coordinates is typically accomplished using a Galerkin projection (see, e.g., refs. [3, 4], for applications to engineering problems). The intrusive nature of the underlying POD‐Galerkin approach requires access to the governing equations for the creation of the reduced‐order model.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Reduced‐order modeling framework using two‐level neural networks

Mjalled,

Torres,

Mönnigmann

2023

Proc Appl Math and Mech

Self Cite

View full text Add to dashboard Cite

Established reduced‐order modeling (ROM) methods, for instance, Galerkin‐projection, approximate the solution by linearly projecting high‐dimensional spaces to a lower‐dimensional space spanned by the reduced basis. However, the accuracy of these methods may be insufficient for complex and multiscale simulations due to the restriction to a linear space. Alternatively, autoencoders (AEs) can be used for nonlinear dimensionality reduction. We combine nonlinear dimensionality reduction techniques with time series prediction to build data‐driven ROMs. The presented framework consists of two‐level neural networks. The high‐dimensional space is nonlinearly compressed in the first level using the encoder function of the AE. Subsequently, the temporal evolution of the latent vector is predicted in the second level. The original solution can be easily reconstructed using the decoder function of the AE. In comparison with the projection‐based ROMs, for example, proper orthogonal decomposition (POD) with Galerkin projection, this framework allows naturally to include parameters in the prediction without nonlinear interpolation of the linear basis. We demonstrate the framework on a two‐dimensional flow field simulation around circular bodies parameterized with the inlet fluid velocity.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Reduced‐order modeling framework using two‐level neural networks

Mjalled,

Torres,

Mönnigmann

2023

Proc Appl Math and Mech

Self Cite

View full text Add to dashboard Cite

show abstract

“…The accuracy of the model was improved, and the optimized entropy efficiency reached 58.73%. Sommer K D et al [20] introduced greedy algorithm to evaluate the fluid velocity and pressure field of centrifugal pump, which improved the robust observability of the system and the fluid flow characteristics in the pump. Omer Incebaya et al [21] optimized the hydraulic performance data of a centrifugal blood pump using the bee algorithm to obtain higher hydraulic efficiency and minimum pump wall shear stress.…”

Section: Introductionmentioning

confidence: 99%

Multi-Objective Optimization Design of Low Specific Speed Centrifugal Pumps Based on Genetic Algorithm

Wang,

Zhou,

Zheng

2023

IEEE Access

View full text Add to dashboard Cite

Aiming at the problems of low working efficiency and high energy loss in the operation of centrifugal pump, an intelligent algorithm for centrifugal pump design was proposed. On the basis of hydraulic design, a multi-objective optimization mathematical model of centrifugal pump was established. A Genetic algorithm (GA) was designed to optimize the major structural parameters of the pump and the optimal combination of parameters was obtained. Calculation results suggested that the optimized pump efficiency was significantly augmented and the energy loss was significantly lower. The external characteristics test platform of centrifugal pump was established. The test results indicated that the head, efficiency and energy loss of the pump had been effectively improved after optimization. Under the rated working conditions, the centrifugal pump based on GA algorithm had increased the head by 26.92%, boosted the efficiency by 32.28%, and diminished the energy loss by 14.38%, which achieved the expected goal. Obviously, the GA algorithm is significantly better than the Particle Swarm Optimization (PSO) algorithm in improving the performance of centrifugal pump. The accuracy and superiority of the designed GA algorithm were verified. This served a reference for the future design of pumps with high efficiency and low energy consumption.

show abstract

Automated optimal experimental design strategy for reduced order modeling of aerodynamic flow fields

Wang,

Martins,

2024

Aerospace Science and Technology

View full text Add to dashboard Cite

Estimating flow fields with reduced order models

Cited by 4 publications

References 44 publications

Reduced‐order modeling framework using two‐level neural networks

Reduced‐order modeling framework using two‐level neural networks

Multi-Objective Optimization Design of Low Specific Speed Centrifugal Pumps Based on Genetic Algorithm

Automated optimal experimental design strategy for reduced order modeling of aerodynamic flow fields

Contact Info

Product

Resources

About