Xiangnan Yu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xiangnan Yu has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- HongGuang Sun
State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing 210098, China
6 shared publications Yong Zhang
6 shared publicationsHongGuang Sun
3 shared publications- Yuntian Chen
Ningbo Key Laboratory of Advanced Manufacturing Simulation, Eastern Institute of Technology, Ningbo, China
3 shared publications Zhibo Chen
3 shared publications- Chunmiao Zheng
Southern University of Science and Technology
2 shared publications
Latest Publications
- 15 October 2026
- Published by Elsevier in Advances in Water Resources
- Vol. 218
A double power-law convolution model for anomalous transport driven by diffusion-immobile domains
- 15 October 2026
- Published by Elsevier in Advances in Water Resources
- Vol. 218
A double power-law convolution model for anomalous transport driven by diffusion-immobile domains
- 28 September 2026
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 463
A fast discovery framework of fractional transport equations through fractional-order operator linearization
- 28 September 2026
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 463
A fast discovery framework of fractional transport equations through fractional-order operator linearization
- 5 September 2026
- Published by Elsevier in Journal of Hydrology
- Vol. 679
Physics-guided symbolic regression of nonlocal transport memory from sparse observations in aquatic systems
- 5 September 2026
- Published by Elsevier in Journal of Hydrology
- Vol. 679
Physics-guided symbolic regression of nonlocal transport memory from sparse observations in aquatic systems
- Open Access
- 12 March 2026
- Published by Elsevier in SSRN Electronic Journal
Physics-Guided Symbolic Regression of Nonlocal Transport Memory from Sparse Observations in Aquatic Systems
Predicting non-Fickian transport in aquatic systems from sparse observations is challenging because historical effects and spatial heterogeneity often control present behavior. We introduce a physics-guided, sequential hybrid framework that couples nonlocal transport theory with deep-learning-assisted symbolic regression to infer transport memory directly from one-dimensional observations, such as breakthrough curves (BTCs) or spatial concentration snapshots. The physics module maps observations to a nonparametric memory representation derived from the statistics of microscopic particle dynamics, including temporal memory associated with mass-transfer-controlled transport and spatial (kinematic) memory represented by the streamtube velocity probability density function (PDF) for conductivity-driven spreading, thereby unifying temporal and spatial nonlocality within a single construct. Neural reconstruction, coupled with a stable numerical Laplace transform, produces denoised, high-resolution spectral representations, from which symbolic regression identifies compact, closed-form expressions for transport memory that are embedded back into the nonlocal model. Validation spans synthetic and experimental systems: semi-analytical fractional diffusion and a Cr(VI) column experiment (power-law kernels); a Red Cedar River tracer test (reach-specific, exponentially tempered kernels); random-walk particle tracking (Gamma-family velocity laws); groundwater tracer transport at the MADE field site (independently referred inverse-Gamma velocity PDFs for MADE-1 and MADE-2 with out-of-sample predictive skill); and sediment transport in the North Loup River (a symmetric heavy-tailed velocity PDF capturing both backward and forward scaling). Across all cases, the inferred models reproduce observed transport behavior and accurately predict withheld profiles while remaining interpretable and parsimonious, without requiring prior characterization of aquatic system properties beyond geometry and boundary conditions. Compared to a purely data-driven sparse-regression approach on the same MADE-2 dataset, which yields only a near-integer-order advection equation (\alpha = 0.99) with no dispersion, our physics-guided framework captures the anomalous spreading and predicts withheld concentration snapshots, demonstrating enhanced robustness under sparse, noisy conditions. By bridging data-driven inference with nonlocal transport physics and expanding the concept of transport memory to encompass both temporal and spatial nonlocality, the proposed framework provides a practical and interpretable pathway for modeling anomalous transport under observational constraints common in laboratory and field hydrologic systems. This capability enables practical inference of dominant transport mechanisms and plume evolution in field systems where only sparse monitoring data are available.
- 12 March 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessPhysics-Guided Symbolic Regression of Nonlocal Transport Memory from Sparse Observations in Aquatic Systems
Predicting non-Fickian transport in aquatic systems from sparse observations is challenging because historical effects and spatial heterogeneity often control present behavior. We introduce a physics-guided, sequential hybrid framework that couples nonlocal transport theory with deep-learning-assisted symbolic regression to infer transport memory directly from one-dimensional observations, such as breakthrough curves (BTCs) or spatial concentration snapshots. The physics module maps observations to a nonparametric memory representation derived from the statistics of microscopic particle dynamics, including temporal memory associated with mass-transfer-controlled transport and spatial (kinematic) memory represented by the streamtube velocity probability density function (PDF) for conductivity-driven spreading, thereby unifying temporal and spatial nonlocality within a single construct. Neural reconstruction, coupled with a stable numerical Laplace transform, produces denoised, high-resolution spectral representations, from which symbolic regression identifies compact, closed-form expressions for transport memory that are embedded back into the nonlocal model. Validation spans synthetic and experimental systems: semi-analytical fractional diffusion and a Cr(VI) column experiment (power-law kernels); a Red Cedar River tracer test (reach-specific, exponentially tempered kernels); random-walk particle tracking (Gamma-family velocity laws); groundwater tracer transport at the MADE field site (independently referred inverse-Gamma velocity PDFs for MADE-1 and MADE-2 with out-of-sample predictive skill); and sediment transport in the North Loup River (a symmetric heavy-tailed velocity PDF capturing both backward and forward scaling). Across all cases, the inferred models reproduce observed transport behavior and accurately predict withheld profiles while remaining interpretable and parsimonious, without requiring prior characterization of aquatic system properties beyond geometry and boundary conditions. Compared to a purely data-driven sparse-regression approach on the same MADE-2 dataset, which yields only a near-integer-order advection equation (\alpha = 0.99) with no dispersion, our physics-guided framework captures the anomalous spreading and predicts withheld concentration snapshots, demonstrating enhanced robustness under sparse, noisy conditions. By bridging data-driven inference with nonlocal transport physics and expanding the concept of transport memory to encompass both temporal and spatial nonlocality, the proposed framework provides a practical and interpretable pathway for modeling anomalous transport under observational constraints common in laboratory and field hydrologic systems. This capability enables practical inference of dominant transport mechanisms and plume evolution in field systems where only sparse monitoring data are available.
- 27 September 2025
- Published by Elsevier in Physica A: Statistical Mechanics and its Applications
- Vol. 679
Physical mechanism analysis of anomalous diffusion characterized by scaling law
- 27 September 2025
- Published by Elsevier in Physica A: Statistical Mechanics and its Applications
- Vol. 679
Physical mechanism analysis of anomalous diffusion characterized by scaling law
- 13 March 2024
- Published by American Physical Society (APS) in Physical Review Fluids
- Vol. 9 (3) , 034502
Impact of microscale physics in continuous time random walks for hydrodynamic dispersion in disordered media
The continuous time random walk (CTRW) approach has been widely applied to model large-scale non-Fickian transport in the flow through disordered media. Often the underlying microscopic transport mechanisms and disorder characteristics are not known, and their effect on large-scale solute dispersion is encoded by a heavy-tailed transition time distribution. Here we study how the microscale physics manifests in the CTRW framework and how it affects solute dispersion. To this end, we consider transport in disordered media with random sorption and random flow properties. Both disorder mechanisms can give rise to anomalous particle transport. We present the CTRW models corresponding to each of these physical scenarios to discuss the different manifestations of microscale heterogeneity on large-scale dispersion depending on the particle injection modes. The combined impact of random sorption and advection is studied with a CTRW model that explicitly represents both microscale disorder mechanisms. While random advection and sorption may show similar large-scale transport behaviors, they can be clearly distinguished in their response to uniform injection conditions and, in general, to initial particle distributions that are not flux weighted. These findings highlight the importance of the microscale physics for the interpretation and prediction of anomalous dispersion phenomena in disordered media.
- 13 March 2024
- Published by American Physical Society (APS) in Physical Review Fluids
- Vol. 9 (3) , 034502
Impact of microscale physics in continuous time random walks for hydrodynamic dispersion in disordered media
The continuous time random walk (CTRW) approach has been widely applied to model large-scale non-Fickian transport in the flow through disordered media. Often the underlying microscopic transport mechanisms and disorder characteristics are not known, and their effect on large-scale solute dispersion is encoded by a heavy-tailed transition time distribution. Here we study how the microscale physics manifests in the CTRW framework and how it affects solute dispersion. To this end, we consider transport in disordered media with random sorption and random flow properties. Both disorder mechanisms can give rise to anomalous particle transport. We present the CTRW models corresponding to each of these physical scenarios to discuss the different manifestations of microscale heterogeneity on large-scale dispersion depending on the particle injection modes. The combined impact of random sorption and advection is studied with a CTRW model that explicitly represents both microscale disorder mechanisms. While random advection and sorption may show similar large-scale transport behaviors, they can be clearly distinguished in their response to uniform injection conditions and, in general, to initial particle distributions that are not flux weighted. These findings highlight the importance of the microscale physics for the interpretation and prediction of anomalous dispersion phenomena in disordered media.
- 4 June 2020
- Published by Elsevier in Chaos, Solitons & Fractals
- Vol. 138, 109959
Applicability of time fractional derivative models for simulating the dynamics and mitigation scenarios of COVID-19
- 4 June 2020
- Published by Elsevier in Chaos, Solitons & Fractals
- Vol. 138, 109959
Applicability of time fractional derivative models for simulating the dynamics and mitigation scenarios of COVID-19
- 1 June 2019
- Published by Elsevier in Advances in Water Resources
- Vol. 128, 129-144
Impact of absorbing and reflective boundaries on fractional derivative models: Quantification, evaluation and application
- 1 June 2019
- Published by Elsevier in Advances in Water Resources
- Vol. 128, 129-144
Impact of absorbing and reflective boundaries on fractional derivative models: Quantification, evaluation and application