Reconstruction of 3D flow field around a building model in wind tunnel: a novel physics-informed neural network framework adopting dynamic prioritization self-adaptive loss balance strategy

Abstract

Physics-informed neural networks (PINNs) have recently emerged and attracted extensive attention as an alternative approach to computational fluid dynamics (CFD) methods, which can provide competitive solutions to a variety of forward and inverse fluid problems. In this study, we reconstruct a 3D wind field around a building model in wind tunnel test with a Reynolds number of 2.4 × 10⁴ by formulating a novel PINN framework, which is the first exploration of PINNs for building wind engineering problems. To surmount the hurdle in multi-objective optimization for PINN training, a dynamic prioritization (dp) self-adaptive loss balance strategy is proposed (termed dpPINN), which adaptively reconciles the loss terms of distinct scales to facilitate convergence in PINN training. A zero-equation turbulence model and the wind velocity data collected in near-wall regions are embedded in dpPINN training. Comparison results indicate that dpPINN predictions show good consistency with observation data, which is superior to two current PINN paradigms in prediction accuracy. Furthermore, the influence of neural network configurations, turbulence models, and the layout arrangements of training points on the dpPINN prediction is investigated. It is demonstrated that the dpPINN could be a powerful auxiliary means for airflow simulation and reconstruction in wind engineering applications.

KEYWORDS:

• Buildings
• Flow fields
• Reconstruction
• Physics-informed neural network (PINN)
• Wind tunnel test

Acknowledgement

The work described in this paper was supported by a grant from the Research Grants Council (RGC) of the Hong Kong Special Administrative Region (SAR), China (grant number PolyU 152308/22E) and a grant from The Hong Kong Polytechnic University (grant number 1-WZ0C). The authors also appreciate the funding support by the Innovation and Technology Commission of Hong Kong SAR Government to the Hong Kong Branch of National Engineering Research Center on Rail Transit Electrification and Automation (grant number K-BBY1), the National Natural Science Foundation of China (Grant No. 52202426), Start-up Fund for RAPs under the Strategic Hiring Scheme of The Hong Kong Polytechnic University (Grant No. 1-BD23), and grants from Wuyi University Hong Kong-Macao Joint R&D Fund (Grants No. 2019WGALH15, 2019WGALH17, and 2021WGALH15).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.