TY - GEN
T1 - Comparative Study of Future State Predictions of Unsteady Multiphase Flows Using DMD and Deep Learning
AU - Raj, Neil Ashwin
AU - Tafti, Danesh
AU - Muralidhar, Nikhil
AU - Karpatne, Anuj
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.
PY - 2024
Y1 - 2024
N2 - Flow across an array of solid obstructions is a common phenomenon observed in many applications such as multiphase flows, heat exchangers, and environmental flows. In this work, we aim to train deep learning models and to predict the time evolution of unsteady flow fields in a domain of randomly arranged 2D cylinders at Reynolds number 50. Two different approaches are used and compared in this paper, dynamic mode decomposition (DMD) which is a dimensionality-reduction algorithm based on singular value decomposition (SVD) and long short-term memory (LSTM) neural networks. In both cases, the model is trained on the first 165 time steps and then is tested on predicting the next 300 time steps. Two flow fields with different spectral characteristics are used to compare the performance of the two techniques. The LSTM architecture owing to its ability to learn nonlinear dynamics performs better than the DMD algorithm in the case with more temporal time scales present.
AB - Flow across an array of solid obstructions is a common phenomenon observed in many applications such as multiphase flows, heat exchangers, and environmental flows. In this work, we aim to train deep learning models and to predict the time evolution of unsteady flow fields in a domain of randomly arranged 2D cylinders at Reynolds number 50. Two different approaches are used and compared in this paper, dynamic mode decomposition (DMD) which is a dimensionality-reduction algorithm based on singular value decomposition (SVD) and long short-term memory (LSTM) neural networks. In both cases, the model is trained on the first 165 time steps and then is tested on predicting the next 300 time steps. Two flow fields with different spectral characteristics are used to compare the performance of the two techniques. The LSTM architecture owing to its ability to learn nonlinear dynamics performs better than the DMD algorithm in the case with more temporal time scales present.
KW - Convolutional auto-encoder (CAE)
KW - Deep learning
KW - Dynamic mode decomposition (DMD)
KW - Long short-term memory (LSTM)
KW - Multiphase flow
KW - Unsteady flow
UR - http://www.scopus.com/inward/record.url?scp=85187775826&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85187775826&partnerID=8YFLogxK
U2 - 10.1007/978-981-99-7177-0_76
DO - 10.1007/978-981-99-7177-0_76
M3 - Conference contribution
AN - SCOPUS:85187775826
SN - 9789819971763
T3 - Lecture Notes in Mechanical Engineering
SP - 923
EP - 935
BT - Fluid Mechanics and Fluid Power, Volume 4 - Select Proceedings of FMFP 2022
A2 - Singh, Krishna Mohan
A2 - Dutta, Sushanta
A2 - Subudhi, Sudhakar
A2 - Singh, Nikhil Kumar
T2 - 9th International and 49th National Conference on Fluid Mechanics and Fluid Power, FMFP 2022
Y2 - 14 December 2022 through 16 December 2022
ER -