Computationally efficient CFD prediction of bubbly flow using physics-guided deep learning. (October 2020)

• Record Type: Journal Article
• Authors: Bao, Han; Feng, Jinyong; Dinh, Nam; Zhang, Hongbin
• Is Part Of: International journal of multiphase flow. Volume 131(2020)
• Publication Date: 2020-10
• Subjects: Deep learning -- Two-phase bubbly flow -- Coarse-mesh CFD -- Physical feature -- Data similarity; Multiphase flow -- Periodicals; Écoulement polyphasique -- Périodiques; Multiphase flow; Periodicals; 620.1064
• DOI: 10.1016/j.ijmultiphaseflow.2020.103378
• Languages: English
• ISSNs: 0301-9322

Abstract

Highlights: Physics-guided deep learning is applied to enable computationally efficient computational fluid dynamics (CFD) simulation. Simulation errors of velocities and void fraction are predicted to improve coarse-mesh CFD simulation of two-phase flow. Computational costs of fine-mesh CFD simulation and the physics-guided data-driven approach are compared. Data similarity measurement is introduced to construct the training database of deep neural networks. Abstract: To realize efficient computational fluid dynamics (CFD) prediction of two-phase flow, a multi-scale framework was proposed in this paper by applying a physics-guided data-driven approach. Instrumental to this framework, Feature Similarity Measurement (FSM) technique was developed for error estimation in two-phase flow simulation using coarse-mesh CFD, to achieve a comparable accuracy as fine-mesh simulations with fast-running feature. By defining physics-guided parameters and variable gradients as physical features, FSM has the capability to capture the underlying local patterns in the coarse-mesh CFD simulation. Massive low-fidelity data and respective high-fidelity data are used to explore the underlying information relevant to the main simulation errors and the effects of phenomenological scaling. By learning from previous simulation data, a surrogate model using deep feedforward neural network (DFNN) can be developed and trained to estimate the simulation error of coarse-mesh CFD. In a demonstration case of two-phase bubbly flow, the DFNN model well captured and corrected the unphysical "peaks" in the velocity and void fraction profiles near the wall in the coarse-mesh configuration, even for extrapolative predictions. The research documented supports the feasibility of the physics-guided deep learning methods for coarse mesh CFD simulations which has a potential for the efficient industrial design.