Development of Pressure Estimator and Velocity Field Corrections for Particle Image Velocimetry Using Physics- Informed Neural Network

Calvin Christian Chandra; Luqman Fathurrohim; Pramudita Satria Palar; Lavi Rizki Zuhal

Calvin Christian Chandra Flow Diagnostics Laboratory, Faculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung, Jalan Ganesa 10, Bandung 40132, Indonesia
Luqman Fathurrohim Flow Diagnostics Laboratory, Faculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung, Jalan Ganesa 10, Bandung 40132, Indonesia
Pramudita Satria Palar Flow Diagnostics Laboratory, Faculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung, Jalan Ganesa 10, Bandung 40132, Indonesia
Lavi Rizki Zuhal Flow Diagnostics Laboratory, Faculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung, Jalan Ganesa 10, Bandung 40132, Indonesia

Keywords: deep learning, flow pressure estimator, flow velocity field corrector, particle image velocimetry, physics-informed neural network

Abstract

Flow diagnostics using particle image velocimetry (PIV) has always been a viable option, but errors or faults in the experiment can lead to misinterpreted data. Meanwhile, physics-informed neural network (PINN) usage has been on the rise because of its versatility. This work intends to analyze the possibilities of implementing PINN for PIV and test on a couple of flow cases to observe whether misinterpretations in PIV output can be minimized, along with providing pressure prediction in the analysis domain. This work modifies an already existing PINN program to better suit PIV applications which is then implemented on uniform flow and backstep flow. The PINN was tested on uniform flow and backstep flow which show that the PINN can produce a denser velocity prediction and predict a pressure field without any prior pressure data. Also, it is capable of filling in gaps of missing data and correct invalid velocity data. results for individual cases are satisfactory for both velocity and pressure predictions but can be improved further.

References

[1] Westerweel, J., Fundamentals of digital particle image velocimetry, Measurement Science and Technology, pp. 1379-1392, 1997.
[2] Raffel, M., Willert, C. E., Scarano, F., Kähler, C. J., Wereley, S. T. & Kompenhans, J., Particle Image Velocimetry: A Practical Guide, Cham: Springer International Publishing AG, 2018.
[3] Sciacchitano, A., Dwight, R. P. & Scarano, F., Navier–Stokes simulations in gappy PIV data, Experiments in Fluids, pp. 1421-1435, 2012.
[4] Wang, H., Liu, Y. & Wang, S., Dense velocity reconstruction from particle image velocimetry/particle tracking velocimetry using a physics-informed neural network, Physics of Fluids, 34(1), 2022.
[5] Octavianus, F. Y., Visual-Based Fluid Motion Estimator with Deep Learning, unpublished.
[6] Putra, C., On Physics-Informed Neural Network (PINN) in Solving Viscous Fluid Flow Problems, unpublished.
[7] Zhang, M. & Piggott, M. D., Unsupervised Learning of Particle Image Velocimetry, Jul. 2020.
[8] Hossain, M. A., Rahman, M. T. & Ridwan, S., Numerical Investigation of Fluid Flow Through A 2D Backward Facing Step Channel, International Journal of Engineering Research & Technology, 2(10), Oct. 2013.