Physics-Informed Data-Driven Fluid Flow Reconstruction, State Estimation, and Prediction

Dear Colleagues,

Recent advances in high-fidelity numerical simulations, advanced experimental diagnostics, and sensor technologies have generated unprecedented volumes of spatio-temporal flow data. These data provide new opportunities for understanding complex flow physics, but they also raise fundamental challenges: how can we reconstruct unobserved flow states from sparse or incomplete measurements, infer hidden flow quantities and boundary conditions, and predict the future evolution of highly nonlinear, multi-scale fluid systems?

Traditional computational fluid dynamics methods remain indispensable for resolving flow physics with high fidelity. However, their direct use can be prohibitively expensive for rapid prediction, real-time control, uncertainty quantification, inverse analysis, and digital-twin applications. At the same time, purely data-driven models often suffer from limited generalizability, insufficient physical interpretability, and poor robustness when extrapolated beyond the training regime. A promising direction is therefore to develop physics-informed and data-driven methodologies that integrate fluid–mechanical principles, reduced-order representations, data assimilation, and modern machine learning architectures.

This Special Issue, ‘Physics-Informed Data-Driven Fluid Flow Reconstruction, State Estimation, and Prediction,’ aims to provide a focused forum for recent advances at the interface of fluid mechanics, scientific machine learning, and predictive modeling. We welcome original research and review articles that address flow reconstruction, sparse sensing, state estimation, reduced-order modeling, turbulence prediction, and real-time forecasting of complex fluid systems. Particular emphasis is placed on methods that combine physical constraints with data-driven learning, improve interpretability and robustness, and demonstrate clear relevance to canonical flows, wall-bounded turbulence, compressible flows, aerodynamic applications, experimental measurements, or flow-control problems.

Topics of interest include, but are not limited to:

1. Data-driven fluid flow reconstruction, super-resolution, and sparse sensing;
2. State estimation and data assimilation for fluid systems;
3. Physics-informed neural networks and scientific machine learning;
4. Reduced-order modeling based on POD, DMD, resolvent analysis, and related methods;
5. Sparse identification of nonlinear flow dynamics;
6. Neural operators, including Fourier neural operators and DeepONet, for flow prediction;
7. Hybrid physics–AI models for turbulence modeling and closure development;
8. Machine learning for wall-bounded, compressible, and multiphase flows;
9. Prediction of extreme events and intermittent dynamics in turbulent flows;
10. Real-time flow forecasting, digital twins, and flow-control-oriented prediction;
11. Uncertainty quantification and interpretability of data-driven flow models;
12. AI-assisted aerodynamic analysis, design, and optimization.

By bringing together contributions from computational fluid dynamics, experimental fluid mechanics, data assimilation, reduced-order modeling, and artificial intelligence, this Special Issue seeks to promote robust, interpretable, and physically consistent data-driven tools for next-generation fluid mechanics research and engineering applications.

Dr. Qinmin Zheng
Dr. Jian Teng
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-anonymized peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Fluids is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• deep learning for flow forecasting
• super-resolution and flow reconstruction
• physics-informed neural networks (PINNs)
• proper orthogonal decomposition (POD)
• reduced-order modeling (ROM)
• sparse identification of nonlinear dynamics (SINDy)
• operator learning in fluid systems
• data assimilation and state estimation
• uncertainty quantification and interpretability