Physics-Informed Deep Operator Learning for Computational Hydraulics Modeling

arXiv:2601.08086 · physics.flu-dyn · Submitted 2026-01-13 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.

Kai: We're looking at the paper: "Physics-Informed Deep Operator Learning for Computational Hydraulics Modeling".

Mira: Here's the gist of the paper. A physics-informed DeepONet framework is developed for computational hydraulics modeling that learns the solution operator of the 2D shallow water equations (SWEs) to create fast surrogate models because it enhances generalization capability for out-of-distribution cases while maintaining physical consistency.

Kai: Framework Architecture The DeepONet framework employs a two-branch neural network architecture designed to learn the solution operator of the 2D SWEs.

Mira: This architecture consists of a branch network and a trunk network.

Kai: The branch network processes discretized input functions that represent the boundary conditions, initial conditions, parameters, or other forcing terms of the SWEs.

Paper summary: Kai: So what does this framework actually claim? They claim that PI-SWE-DeepONet has much enhanced prediction capability when you look at scenarios that are outside the training data, which they call out-of-distribution cases.

Mira: That’s the key finding, isn't it? The physics constraint helps it generalize better to those situations compared to the purely data-driven SWE-DeepONet model.

Lev: It means if you run this on a real flood event that looks a bit different from what you trained it on, the physics part keeps the predictions more stable and prevents them from diverging too quickly.

Kai: They show that this physics-informed model exhibits slower error growth and larger breakdown distances when compared to the standard SWE-DeepONet.

Mira: But they also mention a trade-off, which is important for us to keep in mind: the simulated results have slightly higher errors for cases that are within the training data range.

Lev: That’s the cost of adding physics fidelity. It means you have to tune those loss weights carefully because you’re balancing fitting the data against satisfying those continuous equations, which is a tricky balancing act.

Kai: They set their ratio of data loss to PDE loss at two to one based on in-distribution test cases, but they acknowledge that this weighting is subjective because the scales are so different.

Conclusion: Mira: So looking at the paper "Physics-Informed Deep Operator Learning for Computational Hydraulics Modeling", what does it all boil down to in simple terms?

Kai: It boils down to this framework giving us a way to get fast predictions for hydraulic modeling, and you can choose your model based on what you need.

Lev: If you need reliable predictions for situations that are totally outside your training set, like an unexpected flood scenario, PI-SWE-DeepONet is the better choice because of its enhanced generalization.

Mira: And if your goal is just to replicate the results from a trusted traditional hydraulic solver within the data you’ve already seen, then SWE-DeepONet might be simpler and sufficient.

Kai: The implication for us is that this method gives us a tool where we can switch on or off that physics constraint, giving us flexibility in how we use it.

Lev: It provides an automated workflow through the HydroNet package for building these surrogate models, which cuts down on the computational time needed for those complex simulations.

Mira: The paper suggests that as long as you prioritize physical consistency or need to model uncertainty beyond your training data, this physics-informed operator learning approach is a worthwhile direction to explore in computational hydraulics.

Xiaofeng Liu, Yong G. Lai

Department of Civil and Environmental Engineering, The Pennsylvania State University · Institute of Computational and Data Sciences, The Pennsylvania State University · Technical Service Center, U.S. Bureau of Reclamation

physics.flu-dyn

Submitted: 2026-01-13

Updated: 2026-01-13

Code: https://github.com/psu-efd/HydroNet

License: http://creativecommons.org/licenses/by-nc-nd/4.0/

Importance score: 79/100

The gist: A physics-informed DeepONet framework is developed for computational hydraulics modeling that learns the solution operator of the 2D shallow water equations (SWEs) to create fast surrogate models…

Key concepts

DeepONet Framework
A neural network architecture designed to learn the solution operator of a partial differential equation (PDE). It uses two parts: a branch network for input functions (like boundary conditions) and a trunk network for spatial coordinates. These are combined via an inner product to predict the solution directly.
Physics-Informed Mode (PI-SWE-DeepONet)
A training method where the model learns not just from data, but also by minimizing errors related to the governing SWEs. This is achieved by adding a physics loss term to the total loss function. This forces the learned model to respect the actual laws of fluid dynamics during prediction.
Competing Objectives
The challenge in training is balancing two conflicting goals: accurately fitting the available training data and strictly minimizing violations of the governing physical equations (PDE residuals). Finding a right balance between these two objectives requires careful tuning of loss weights.
Out-of-Distribution Performance
This measures how well the trained model performs on situations or input data that are outside the range it was originally trained on. The physics-informed model (PI-SWE-DeepONet) shows significantly better performance and stability in these unknown cases compared to standard models.

Terminology

Summary

A physics-informed DeepONet framework is developed for computational hydraulics modeling that learns the solution operator of the 2D shallow water equations (SWEs) to create fast surrogate models because it enhances generalization capability for out-of-distribution cases while maintaining physical consistency. p. 1

Where the paper says it

manuscript submitted to Abstract Traditional 2D hydraulic models face significant computational challenges that limit their applications that are time-sensitive or require many model evaluations. This study presents a physics-informed Deep Operator Network (DeepONet) framework for computational hydraulics modeling that learns the solution operator of the 2D shallow water equations (SWEs) to create fast surrogate models.

Framework Architecture

The DeepONet framework employs a two-branch neural network architecture designed to learn the solution operator of the 2D SWEs. This architecture consists of a branch network and a trunk network. p. 5

Where the paper says it

1 Physics-Informed DeepONet Framework The DeepONet framework employs a two-branch neural network architecture designed to learn the solution operator of the 2D SWEs. As illustrated in Figure 1, the architecture consists of two parallel networks: a branch network and a trunk network.

The branch network processes discretized input functions that represent the boundary conditions, initial conditions, parameters, or other forcing terms of the SWEs. p. 5
Where the paper says it

As illustrated in Figure 1, the architecture consists of two parallel networks: a branch network and a trunk network. The branch network processes discretized input functions that represent the boundary conditions, initial conditions, parameters, or other forcing terms of the SWEs.

The trunk network processes the spatiotemporal coordinates (x, y, t) at which the solution is to be evaluated. p. 5
Where the paper says it

These input functions are sampled at discrete points and encoded into a high-dimensional feature representation through the branch network. The trunk network, operating in parallel to the branch network, processes the spatiotemporal coordinates (x, y, t) at which the solution is to be evaluated.

The outputs of the branch and trunk networks are then combined through an inner product (or more generally, a dot product) to produce the output of the operator. p. 5
Where the paper says it

For steady state problems in the present study, the input is only the spatial coordinates (x, y). The outputs of the branch and trunk networks are then combined through an inner product (or more generally, a dot product) to produce the out- –6–

Training Signal Construction

The training of the model is carried out by minimizing the total loss function (Ltotal) consisting of four loss terms. The total loss function is expressed as a weighted sum of all loss components: Ltotal = λdataLdata + λPDEsLPDEs + λBC/IC LBC/IC + λotherLother (6). p. 8

Where the paper says it

These losses are optional and they are omitted for the examples in this work. The total loss function is then expressed as a weighted sum of all loss components: Ltotal = λdataLdata + λPDEsLPDEs + λBC/IC LBC/IC + λotherLother (6) where the weights λdata, λPDEs, λBC/IC , and λother are hyperparameters that measure the relative importance of corresponding loss components.

In the purely data-driven mode, only the data loss function Ldata is minimized, which measures the discrepancy between the predicted and the true values as a mean square error (MSE). p. 6
Where the paper says it

In the purely data-driven DeepONet mode, termed “SWE-DeepONet” in this work, only the data loss function Ldata is minimized. Data loss measures the discrepancy between the predicted and the true values, which can be expressed as the mean square error (MSE): Ldata = 1 Ndata N Xdata i=1 (qˆi − qi) 2 (1) where qˆi is the predicted value, qi is the true value, and Ndata is the number of data points.

In the physics-informed mode, termed “PI-SWE-DeepONet”, flow physics governed by the SWEs is additionally incorporated into the training process by adding the physics loss function LPDEs. p. 6
Where the paper says it

The true value qi can be generated by running physics-based numerical models such as SRH-2D or HEC-RAS 2D. In the physics-informed DeepONet mode, termed “PI-SWE-DeepONet” in this work, the flow physics governed by the SWEs is additionally incorporated into the training process by adding the physics loss function LPDEs.

The physics loss function measures the violation of the governing equations and is computed as a summation of PDE residuals at each PDE point. p. 6
Where the paper says it

In the physics-informed DeepONet mode, termed “PI-SWE-DeepONet” in this work, the flow physics governed by the SWEs is additionally incorporated into the training process by adding the physics loss function LPDEs. This loss function measures the violation of the governing equations and is computed as the summation of PDE residuals at each PDE point.

Performance Gains and Trade-offs

The PI-SWE-DeepONet model possesses much enhanced prediction capability than SWE-DeepONet when applied to out-of-distribution scenarios. The physics constraint improves the generalization capability of the model for out-of-distribution cases and is shown to exhibit slower error growth and larger breakdown distances in comparison with SWE-DeepONet. p. 1

Where the paper says it

The physics-informed model is shown to exhibit slower error growth and larger breakdown distances in comparison with SWE-DeepONet. The gain of the physics-informed training, however, comes with costs, chief among which are the simulated results have slightly higher errors for in-distribution cases.

However, this gain comes with costs, chief among which are the simulated results have slightly higher errors for in-distribution cases. p. 1
Where the paper says it

The physics-informed model is shown to exhibit slower error growth and larger breakdown distances in comparison with SWE-DeepONet. The gain of the physics-informed training, however, comes with costs, chief among which are the simulated results have slightly higher errors for in-distribution cases.

The study suggests that PI-SWE-DeepONet is preferred for out-of-distribution predictions or when physical consistency is a priority. p. 1
Where the paper says it

It reflects the existence of a tension between the two competing training objectives: fitting the results from the traditional hydraulic model and satisfying the continuous governing equations. In this study, guidelines are developed for selecting the appropriate approach based on a real-world case: PI-SWE-DeepONet is preferred for out-of-distribution predictions, uncertain training data, or when physical consistency is a priority, while SWE-DeepONet is recommended if the modeling objective is to replicate faithfully the traditional hydraulic model results within the training distribution.

Model Selection Guidelines

Guidelines are developed for selecting the appropriate approach based on a real-world case. PI-SWE-DeepONet is recommended for out-of-distribution predictions, uncertain training data, or when physical consistency is a priority. p. 1

Where the paper says it

It reflects the existence of a tension between the two competing training objectives: fitting the results from the traditional hydraulic model and satisfying the continuous governing equations. In this study, guidelines are developed for selecting the appropriate approach based on a real-world case: PI-SWE-DeepONet is preferred for out-of-distribution predictions, uncertain training data, or when physical consistency is a priority, while SWE-DeepONet is recommended if the modeling objective is to replicate faithfully the traditional hydraulic model results within the training distribution.

SWE-DeepONet is recommended if the modeling objective is to replicate faithfully the traditional hydraulic model results within the training distribution. p. 1
Where the paper says it

It reflects the existence of a tension between the two competing training objectives: fitting the results from the traditional hydraulic model and satisfying the continuous governing equations. In this study, guidelines are developed for selecting the appropriate approach based on a real-world case: PI-SWE-DeepONet is preferred for out-of-distribution predictions, uncertain training data, or when physical consistency is a priority, while SWE-DeepONet is recommended if the modeling objective is to replicate faithfully the traditional hydraulic model results within the training distribution.

Competing Objectives

Adding PDE residuals into DeepONet loss changes the learning target fundamentally. The network is additionally required to satisfy the continuous shallow water equations, which the numerical model (e. p. 1

Where the paper says it

This study presents a physics-informed Deep Operator Network (DeepONet) framework for computational hydraulics modeling that learns the solution operator of the 2D shallow water equations (SWEs) to create fast surrogate models. The framework can operate in two modes: a purely data-driven SWE-DeepONet that learns from numerical solver such as SRH-2D, and a physics-informed PI-SWE-DeepONet that additionally incorporates the continuous SWEs as constraints during training.

g., SRH2D) does not satisfy exactly due to differences in formulation and discretization schemes. p. 1
Where the paper says it

1 Introduction Two-dimensional (2D) depth-averaged hydraulic models solve the shallow water equations (SWEs), which are derived from depth-averaging the Navier-Stokes equations under the hydrostatic pressure assumption. They have been widely used in recently years due to the availability of models such as SRH-2D (Sedimentation and River Hydraulics 2D) (Y.

Two inherently competing objectives are at play: fit the training data versus minimize PDE residuals. p. 1
Where the paper says it

It reflects the existence of a tension between the two competing training objectives: fitting the results from the traditional hydraulic model and satisfying the continuous governing equations. In this study, guidelines are developed for selecting the appropriate approach based on a real-world case: PI-SWE-DeepONet is preferred for out-of-distribution predictions, uncertain training data, or when physical consistency is a priority, while SWE-DeepONet is recommended if the modeling objective is to replicate faithfully the traditional hydraulic model results within the training distribution.

The challenge lies in selecting the loss weights to balance data fidelity and physics constraints, as they operate on fundamentally different scales.

Conclusion

The study presents a physics-informed DeepONet framework for computational hydraulics modeling. The developed framework can switch on and off the SWEs physics constraint, resulting in PISWE-DeepONet and SWE-DeepONet models. p. 1

Where the paper says it

This study presents a physics-informed Deep Operator Network (DeepONet) framework for computational hydraulics modeling that learns the solution operator of the 2D shallow water equations (SWEs) to create fast surrogate models. The framework can operate in two modes: a purely data-driven SWE-DeepONet that learns from numerical solver such as SRH-2D, and a physics-informed PI-SWE-DeepONet that additionally incorporates the continuous SWEs as constraints during training.

PI-SWE-DeepONet significantly enhances model generalizability to out-of-distribution scenarios while having a slower error growth rate and larger breakdown distances compared to SWE-DeepONet. p. 1
Where the paper says it

Based on a real-world case, steady flows in a reach of the Sacramento River in California, it is demonstrated that PI-SWE-DeepONet possesses much enhanced prediction capability than SWE-DeepONet when applied to out-of-distribution scenarios. The physics-informed model is shown to exhibit slower error growth and larger breakdown distances in comparison with SWE-DeepONet.

The choice between models should be guided by application requirements, favoring PI-SWE-DeepONet when predictions are needed for out-of-distribution cases or physical consistency is the priority. p. 1
Where the paper says it

It reflects the existence of a tension between the two competing training objectives: fitting the results from the traditional hydraulic model and satisfying the continuous governing equations. In this study, guidelines are developed for selecting the appropriate approach based on a real-world case: PI-SWE-DeepONet is preferred for out-of-distribution predictions, uncertain training data, or when physical consistency is a priority, while SWE-DeepONet is recommended if the modeling objective is to replicate faithfully the traditional hydraulic model results within the training distribution.

The framework is implemented in the open-source HydroNet package, providing an automated workflow for surrogate modeling. p. 8
Where the paper says it

This flexibility makes the framework adaptable to diverse hydraulic modeling scenarios, from constant discharges (steady-state) to complex unsteady open channel flows with multiple inflow and exit boundaries. In this study, the algorithms and workflow presented are implemented in the opensource Python package HydroNet, which uses PyTorch for deep learning model development and training.

Open Research Section

All code and data that support the findings of this study are available from the HydroNet GitHub repository. The framework can switch on and off the SWEs physics constraint, resulting in PISWE-DeepONet and SWE-DeepONet models. p. 1

Where the paper says it

This study presents a physics-informed Deep Operator Network (DeepONet) framework for computational hydraulics modeling that learns the solution operator of the 2D shallow water equations (SWEs) to create fast surrogate models. The framework can operate in two modes: a purely data-driven SWE-DeepONet that learns from numerical solver such as SRH-2D, and a physics-informed PI-SWE-DeepONet that additionally incorporates the continuous SWEs as constraints during training.

The framework, implemented in the open-source HydroNet package, provides an automated workflow to perform a surrogate modeling. p. 8
Where the paper says it

This flexibility makes the framework adaptable to diverse hydraulic modeling scenarios, from constant discharges (steady-state) to complex unsteady open channel flows with multiple inflow and exit boundaries. In this study, the algorithms and workflow presented are implemented in the opensource Python package HydroNet, which uses PyTorch for deep learning model development and training.

Future work should explore the potential of physics-informed training to reduce data requirements and extend the framework to unsteady flows. p. 4
Where the paper says it

The physics-informed training process allows the model to learn from both available simulation data and the underlying flow physics, creating a more robust and reliable surrogate for applications. This approach has been shown to reduce the need for a large set of training data and potential to improve the predictive capability to untrained scenarios (Wang et al.

p. 4
Where the paper says it

The physics-informed training process allows the model to learn from both available simulation data and the underlying flow physics, creating a more robust and reliable surrogate for applications. This approach has been shown to reduce the need for a large set of training data and potential to improve the predictive capability to untrained scenarios (Wang et al.

The development of more robust loss balancing strategies to handle the competing objectives between data fidelity and physics constraints is also needed. p. 1
Where the paper says it

It reflects the existence of a tension between the two competing training objectives: fitting the results from the traditional hydraulic model and satisfying the continuous governing equations. In this study, guidelines are developed for selecting the appropriate approach based on a real-world case: PI-SWE-DeepONet is preferred for out-of-distribution predictions, uncertain training data, or when physical consistency is a priority, while SWE-DeepONet is recommended if the modeling objective is to replicate faithfully the traditional hydraulic model results within the training distribution.

Acknowledgments

We acknowledge the time, effort, and insights provided by the reviewers and the editor.

References

Baydin, A. G., Pearlmutter, B. A., Radul, A. A., & Siskind, J. M. (2018). Automatic differentiation in machine learning: a survey. p. 7

Where the paper says it

In this study, the non-conservative form of the 2D SWEs is adopted, versus the conservative form, because it produces simpler automatic differentiation graphs and reduces residual imbalance during training within the physics-informed DeepONet framework. Deep-learning surrogate models generally do not enforce conservation laws on a control volume by computing fluxes across control-volume boundaries.

Bihlo, A., & Popovych, R. p. 7
Where the paper says it

Instead, they enforce conservation through pointwise PDE residuals. This choice of using non-conservative form is also consistent with prior physics-informed SWE studies (Bihlo & Popovych, 2022) and is physically valid because the supervised SRH-2D data inherently enforce conservation at the solution level.

O. (2022, May). p. 3
Where the paper says it

Major operator learning architectures include DeepONet (Lu et al. , 2021, 2022), Fourier Neural Operator (FNO) (Li, Kovachki, Azizzadenesheli, liu, et al.

Physics-informed neural networks for the shallow-water equations on the sphere. p. 6
Where the paper says it

The physics-informed DeepONet framework for open channel flows governed by the 2D shallow water equations. The training of the model (weights and biases in the branch and trunk networks) is carried out by minimizing the total loss function (Ltotal) consisting of four loss terms: data loss (Ldata), physics loss (LPDEs), boundary conditions and initial conditions loss (LBC/IC ), and other losses (Lother); details are explained below.

Boule, N., & Alex, T. p. 4
Where the paper says it

, 2021; Li, Kovachki, Azizzadenesheli, Liu, et al. , 2021; Boule & Alex, 2024; Chen et al.

(2024, January). p. 3
Where the paper says it

, 2021; Kovachki et al. , 2024; Li, Kovachki, Azizzadenesheli, liu, et al.

A mathematical guide to operator learning. p. 0
Where the paper says it

manuscript submitted to Physics-Informed Deep Operator Learning for Computational Hydraulics Modeling Xiaofeng Liu1,2and Yong G. Lai3 1Department of Civil and Environmental Engineering, The Pennsylvania State University, University Park, PA, USA 2 Institute of Computational and Data Sciences, The Pennsylvania State University, University Park, PA, USA 3Technical Service Center, U.

Brunner, G. p. 1
Where the paper says it

They have been widely used in recently years due to the availability of models such as SRH-2D (Sedimentation and River Hydraulics 2D) (Y. Lai, 2010) and HEC-RAS 2D (Hydrologic Engineering Center’s River Analysis System) (Brunner, 2016).

W. (2016). Hec-ras river analysis system: User’s manual [Computer software manual]. p. 1
Where the paper says it

They have been widely used in recently years due to the availability of models such as SRH-2D (Sedimentation and River Hydraulics 2D) (Y. Lai, 2010) and HEC-RAS 2D (Hydrologic Engineering Center’s River Analysis System) (Brunner, 2016).

Chen, W., Song, J., Ren, P., Subramanian, S., Morozov, D., & Mahoney, M. W. (2024). Data-efficient operator learning via unsupervised pretraining and in-context learning. p. 0
Where the paper says it

manuscript submitted to Physics-Informed Deep Operator Learning for Computational Hydraulics Modeling Xiaofeng Liu1,2and Yong G. Lai3 1Department of Civil and Environmental Engineering, The Pennsylvania State University, University Park, PA, USA 2 Institute of Computational and Data Sciences, The Pennsylvania State University, University Park, PA, USA 3Technical Service Center, U.

Goswami, S. p. 3
Where the paper says it

, 2021; Wang et al. , 2021; Goswami et al.

, Yin, M., Yu, Y., & Karniadakis, G. E. (2022). A physics-informed variational deeponet for predicting crack path in quasi-brittle materials. p. 0
Where the paper says it

Lai3 1Department of Civil and Environmental Engineering, The Pennsylvania State University, University Park, PA, USA 2 Institute of Computational and Data Sciences, The Pennsylvania State University, University Park, PA, USA 3Technical Service Center, U. Bureau of Reclamation, Denver, CO, USA Key Points: • Physics-informed DeepONet framework is developed for computational hydraulics modeling based on the 2D shallow water equations • Physics constraint comes with the cost of higher errors for in-distribution cases because of the inconsistency between discrete and continuous governing equations • Physics constraint improves the generalization capability of the model for out-ofdistribution cases Corresponding author: Xiaofeng Liu, xzl123@psu.

Haces-Garcia, F. p. 2
Where the paper says it

Such models have been explored using the convolutional neural network (CNN) for spatial variation, long short-term memory (LSTM) network for temporal evolution, and fully connected neural network for parameterto-output mapping (Song et al. , 2023; Haces-Garcia et al.

, Ross, N., Glennie, C. L., Rifai, H. S., Hoskere, V., & Ekhtari, N. (2025). Rapid 2D hydrodynamic flood modeling using deep learning surrogates. p. 2
Where the paper says it

Surrogate models learn the input-output relationship of the original model from a limited set of training simulations, then provide rapid predictions for new inputs without having to solve the full partial differential equation (PDE) system. It is widely accepted that the most promising surrogate models are based on deep machine learning techniques.

Kantorovich, L. V. (1958). On the translocation of mass. Keppler, R. p. 4
Where the paper says it

, 2024; Rivera-Casillas et al. , 2025; Keppler et al.

, Fensholt, R., & Koch, J. (2025, October). p. 4
Where the paper says it

, 2024; Rivera-Casillas et al. , 2025; Keppler et al.

Learning 2D shallow water equations with physics-informed neural operator networks. p. 6
Where the paper says it

The physics-informed DeepONet framework for open channel flows governed by the 2D shallow water equations. The training of the model (weights and biases in the branch and trunk networks) is carried out by minimizing the total loss function (Ltotal) consisting of four loss terms: data loss (Ldata), physics loss (LPDEs), boundary conditions and initial conditions loss (LBC/IC ), and other losses (Lother); details are explained below.

Kramer, B., Peherstorfer, B., & Willcox, K. E. (2024). Learning nonlinear reduced models from data with operator inference. p. 1
Where the paper says it

This study presents a physics-informed Deep Operator Network (DeepONet) framework for computational hydraulics modeling that learns the solution operator of the 2D shallow water equations (SWEs) to create fast surrogate models. The framework can operate in two modes: a purely data-driven SWE-DeepONet that learns from numerical solver such as SRH-2D, and a physics-informed PI-SWE-DeepONet that additionally incorporates the continuous SWEs as constraints during training.

Lai, Y., Ben, A., & Xiaofeng, L. p. 0
Where the paper says it

manuscript submitted to Physics-Informed Deep Operator Learning for Computational Hydraulics Modeling Xiaofeng Liu1,2and Yong G. Lai3 1Department of Civil and Environmental Engineering, The Pennsylvania State University, University Park, PA, USA 2 Institute of Computational and Data Sciences, The Pennsylvania State University, University Park, PA, USA 3Technical Service Center, U.

(2025). Time for large-scale modeling? p. 2
Where the paper says it

Despite their widespread adoption and proven reliability, traditional 2D hydraulic models face computational challenges that have placed a limit on their applicability range. High-fidelity simulations of large-scale flood events or complex river systems can require hours to days of computational time, even on modern high-performance computing systems.

results of srh-2d parallelization effort. p. 8
Where the paper says it

, 2017, 2019). The framework also leverages pyHMT2D, a Python interface for the SRH-2D hydraulic model, to process simulation results, extract flow field data from result files, and generate collocation points for physics-informed training.

Lai, Y. G. (2010). Two-dimensional depth-averaged flow modeling with an unstructured hybrid mesh. p. 1
Where the paper says it

Other challenges are also discussed, such as the loss weighting approach. 1 Introduction Two-dimensional (2D) depth-averaged hydraulic models solve the shallow water equations (SWEs), which are derived from depth-averaging the Navier-Stokes equations under the hydrostatic pressure assumption.

Lai, Y. G., Smith, D. L., & Israel, J. (2016). 2D and 3D Flow Modeling of the Sacramento River Fremont Weir Section. p. 9
Where the paper says it

The Sacramento River is one of the major rivers in California, USA, and has historically been subject to complex flow dynamics due to its confluence with the San Joaquin River and the intricate channel geometry in the delta region. In this study, a section of the river near the Fremont Weir is simulated under a constant-discharge flow condition.

Li, Z., Kovachki, N. p. 3
Where the paper says it

Operator learning represents a transformative approach in scientific machine learning (SciML) that addresses the fundamental challenge of learning solution operators for parametric partial differential equations (Lu et al. , 2021; Kovachki et al.

p. 3
Where the paper says it

Operator learning represents a transformative approach in scientific machine learning (SciML) that addresses the fundamental challenge of learning solution operators for parametric partial differential equations (Lu et al. , 2021; Kovachki et al.

p. 3
Where the paper says it

Operator learning represents a transformative approach in scientific machine learning (SciML) that addresses the fundamental challenge of learning solution operators for parametric partial differential equations (Lu et al. , 2021; Kovachki et al.

p. 3
Where the paper says it

Operator learning represents a transformative approach in scientific machine learning (SciML) that addresses the fundamental challenge of learning solution operators for parametric partial differential equations (Lu et al. , 2021; Kovachki et al.

, Azizzadenesheli, K. p. 3
Where the paper says it

, 2021; Kovachki et al. , 2024; Li, Kovachki, Azizzadenesheli, liu, et al.

p. 3
Where the paper says it

, 2021; Kovachki et al. , 2024; Li, Kovachki, Azizzadenesheli, liu, et al.

p. 3
Where the paper says it

, 2021; Kovachki et al. , 2024; Li, Kovachki, Azizzadenesheli, liu, et al.

, Liu, B., Bhattacharya, K., Stuart, A., & Anandkumar, A. (2021). Neural operator: Learning maps between function spaces with applications to PDEs. p. 3
Where the paper says it

manuscript submitted to mappings between infinite-dimensional function spaces rather than finite-dimensional vectors. Operator learning represents a transformative approach in scientific machine learning (SciML) that addresses the fundamental challenge of learning solution operators for parametric partial differential equations (Lu et al.

Li, Z., Kovachki, N. B., Azizzadenesheli, K., liu, B., Bhattacharya, K., Stuart, A., & Anandkumar, A. (2021). Fourier neural operator for parametric partial differential equations. p. 3
Where the paper says it

manuscript submitted to mappings between infinite-dimensional function spaces rather than finite-dimensional vectors. Operator learning represents a transformative approach in scientific machine learning (SciML) that addresses the fundamental challenge of learning solution operators for parametric partial differential equations (Lu et al.

Li, Z., Kovachki, N. B., Choy, C., Li, B., Kossaifi, J., Otta, S. P.,... Anandkumar, A. (2023). Geometry-informed neural operator for large-scale 3d PDEs. p. 2
Where the paper says it

Despite their widespread adoption and proven reliability, traditional 2D hydraulic models face computational challenges that have placed a limit on their applicability range. High-fidelity simulations of large-scale flood events or complex river systems can require hours to days of computational time, even on modern high-performance computing systems.

Li, Z., Zheng, H. p. 3
Where the paper says it

, 2021; Li, Kovachki, Azizzadenesheli, Liu, et al. , 2021), PhysicsInformed Neural Operators (PINO) (Li, Zheng, et al.

, Kovachki, N., Jin, D., Chen, H., Liu, B.,... Anandkumar, A. (2021). Physics-informed neural operator for learning partial differential equations. p. 3
Where the paper says it

manuscript submitted to mappings between infinite-dimensional function spaces rather than finite-dimensional vectors. Operator learning represents a transformative approach in scientific machine learning (SciML) that addresses the fundamental challenge of learning solution operators for parametric partial differential equations (Lu et al.

Lu, L., Jin, P., Pang, G., Zhang, Z., & Karniadakis, G. E. (2021). Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators. p. 3
Where the paper says it

(2021), was among the first to demonstrate the feasibility of learning solution operators for PDEs (Lu et al. DeepONet employs a two-branch architecture inspired by the universal approximation theorem for operators: a branch network processes discretized input functions (e.

Lu, L., Meng, X., Cai, S., Mao, Z., Goswami, S. p. 3
Where the paper says it

, 2021; Wang et al. , 2021; Goswami et al.

, Zhang, Z., & Karniadakis, G. E. (2022). A comprehensive and fair comparison of two neural operators (with practical extensions) based on FAIR data. p. 1
Where the paper says it

The framework can operate in two modes: a purely data-driven SWE-DeepONet that learns from numerical solver such as SRH-2D, and a physics-informed PI-SWE-DeepONet that additionally incorporates the continuous SWEs as constraints during training. Based on a real-world case, steady flows in a reach of the Sacramento River in California, it is demonstrated that PI-SWE-DeepONet possesses much enhanced prediction capability than SWE-DeepONet when applied to out-of-distribution scenarios.

Paszke, A., Gross, S., Chintala, S., Chanan, G., Yang, E., DeVito, Z.,... Lerer, A. (2017). Automatic differentiation in pytorch. p. 8
Where the paper says it

In this study, the algorithms and workflow presented are implemented in the opensource Python package HydroNet, which uses PyTorch for deep learning model development and training. The neural network architectures, including both the branch and trunk networks of DeepONet, are built using PyTorch’s neural network modules, enabling automatic differentiation for computing PDE residuals in physics-informed training (Paszke et al.

Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G.,... Chintala, S. (2019). Pytorch: an imperative style, high-performance deep learning library. p. 8
Where the paper says it

This flexibility makes the framework adaptable to diverse hydraulic modeling scenarios, from constant discharges (steady-state) to complex unsteady open channel flows with multiple inflow and exit boundaries. In this study, the algorithms and workflow presented are implemented in the opensource Python package HydroNet, which uses PyTorch for deep learning model development and training.

Rahman, M. A., Ross, Z. E., & Azizzadenesheli, K. p. 3
Where the paper says it

, 2021; Kovachki et al. , 2024; Li, Kovachki, Azizzadenesheli, liu, et al.

(2023). U-no: U-shaped neural operators. p. 3
Where the paper says it

, 2021; Li, Kovachki, Azizzadenesheli, Liu, et al. , 2021), PhysicsInformed Neural Operators (PINO) (Li, Zheng, et al.

Rivera-Casillas, P. p. 4
Where the paper says it

, 2024; Kramer et al. , 2024; Rivera-Casillas et al.

, Dutta, S., Cai, S., Loveland, M., Nath, K., Shukla, K.,... Dawson, C. (2601.08086v1>. Song, Y., Shen, C., & Liu, X. p. 0
Where the paper says it

Lai3 1Department of Civil and Environmental Engineering, The Pennsylvania State University, University Park, PA, USA 2 Institute of Computational and Data Sciences, The Pennsylvania State University, University Park, PA, USA 3Technical Service Center, U. Bureau of Reclamation, Denver, CO, USA Key Points: • Physics-informed DeepONet framework is developed for computational hydraulics modeling based on the 2D shallow water equations • Physics constraint comes with the cost of higher errors for in-distribution cases because of the inconsistency between discrete and continuous governing equations • Physics constraint improves the generalization capability of the model for out-ofdistribution cases Corresponding author: Xiaofeng Liu, xzl123@psu.

(2023). A Surrogate Model for Shallow Water Equations Solvers with Deep Learning. p. 1
Where the paper says it

manuscript submitted to Abstract Traditional 2D hydraulic models face significant computational challenges that limit their applications that are time-sensitive or require many model evaluations. This study presents a physics-informed Deep Operator Network (DeepONet) framework for computational hydraulics modeling that learns the solution operator of the 2D shallow water equations (SWEs) to create fast surrogate models.

Sun, A., Jiang, P., Shuai, P., & Chen, X. p. 4
Where the paper says it

, 2021; Li, Kovachki, Azizzadenesheli, Liu, et al. , 2021; Boule & Alex, 2024; Chen et al.

(2024). Bridging hydrological ensemble simulation and learning using deep neural operators. p. 2
Where the paper says it

It is widely accepted that the most promising surrogate models are based on deep machine learning techniques. Such models have been explored using the convolutional neural network (CNN) for spatial variation, long short-term memory (LSTM) network for temporal evolution, and fully connected neural network for parameterto-output mapping (Song et al.

Villani, C. (2009). Optimal transport: Old and new. p. 2
Where the paper says it

In recent years, researchers and practitioners have increasingly turned to surrogate modeling approaches that can approximate the behavior of traditional hydraulic models at a fraction of the computational cost. Surrogate models learn the input-output relationship of the original model from a limited set of training simulations, then provide rapid predictions for new inputs without having to solve the full partial differential equation (PDE) system.

Wang, S., Wang, H., & Perdikaris, P. (2021). Learning the solution operator of parametric partial differential equations with physics-informed deeponets. p. 3
Where the paper says it

manuscript submitted to mappings between infinite-dimensional function spaces rather than finite-dimensional vectors. Operator learning represents a transformative approach in scientific machine learning (SciML) that addresses the fundamental challenge of learning solution operators for parametric partial differential equations (Lu et al.

Wen, G., Li, Z., Azizzadenesheli, K., Anandkumar, A., & Benson, S. M. (2022). U-FNO-an enhanced fourier neural operator-based deep learning model for multiphase flow. p. 3
Where the paper says it

Major operator learning architectures include DeepONet (Lu et al. , 2021, 2022), Fourier Neural Operator (FNO) (Li, Kovachki, Azizzadenesheli, liu, et al.

Out-of-Distribution Performance

The performance on out-of-distribution cases is now presented to show the generalizability of trained surrogate models when scenarios are beyond the parameter range of the training data. PI-SWE-DeepONet shows a superior generalization capability in comparison with SWE-DeepONet and increased advantages are observed with increasing Wasserstein distance. p. 1

Where the paper says it

Based on a real-world case, steady flows in a reach of the Sacramento River in California, it is demonstrated that PI-SWE-DeepONet possesses much enhanced prediction capability than SWE-DeepONet when applied to out-of-distribution scenarios. The physics-informed model is shown to exhibit slower error growth and larger breakdown distances in comparison with SWE-DeepONet.

The error ratio r between PI-SWE-DeepONet and SWE-DeepONet further quantifies the performance of the physics-informed model. p. 1
Where the paper says it

Based on a real-world case, steady flows in a reach of the Sacramento River in California, it is demonstrated that PI-SWE-DeepONet possesses much enhanced prediction capability than SWE-DeepONet when applied to out-of-distribution scenarios. The physics-informed model is shown to exhibit slower error growth and larger breakdown distances in comparison with SWE-DeepONet.

For water depth h, the majority of samples fall well below unity, with error ratios typically between 0. p. 1
Where the paper says it

Other challenges are also discussed, such as the loss weighting approach. 1 Introduction Two-dimensional (2D) depth-averaged hydraulic models solve the shallow water equations (SWEs), which are derived from depth-averaging the Navier-Stokes equations under the hydrostatic pressure assumption.

7 and 0.9. The benefit due to the physics constraint tends to increase with distance (r decreases with distance). p. 0
Where the paper says it

Lai3 1Department of Civil and Environmental Engineering, The Pennsylvania State University, University Park, PA, USA 2 Institute of Computational and Data Sciences, The Pennsylvania State University, University Park, PA, USA 3Technical Service Center, U. Bureau of Reclamation, Denver, CO, USA Key Points: • Physics-informed DeepONet framework is developed for computational hydraulics modeling based on the 2D shallow water equations • Physics constraint comes with the cost of higher errors for in-distribution cases because of the inconsistency between discrete and continuous governing equations • Physics constraint improves the generalization capability of the model for out-ofdistribution cases Corresponding author: Xiaofeng Liu, xzl123@psu.

The breakdown distances, using a threshold of normalized RMSE ¿ 0. p. 1
Where the paper says it

Based on a real-world case, steady flows in a reach of the Sacramento River in California, it is demonstrated that PI-SWE-DeepONet possesses much enhanced prediction capability than SWE-DeepONet when applied to out-of-distribution scenarios. The physics-informed model is shown to exhibit slower error growth and larger breakdown distances in comparison with SWE-DeepONet.

5, further highlight improved stability by physics constraint. p. 0
Where the paper says it

Lai3 1Department of Civil and Environmental Engineering, The Pennsylvania State University, University Park, PA, USA 2 Institute of Computational and Data Sciences, The Pennsylvania State University, University Park, PA, USA 3Technical Service Center, U. Bureau of Reclamation, Denver, CO, USA Key Points: • Physics-informed DeepONet framework is developed for computational hydraulics modeling based on the 2D shallow water equations • Physics constraint comes with the cost of higher errors for in-distribution cases because of the inconsistency between discrete and continuous governing equations • Physics constraint improves the generalization capability of the model for out-ofdistribution cases Corresponding author: Xiaofeng Liu, xzl123@psu.

Discussion on Objectives

Adding PDE residuals into DeepONet loss changes the learning target fundamentally. The competing objectives are fit the training data versus minimize PDE residuals. p. 1

Where the paper says it

It reflects the existence of a tension between the two competing training objectives: fitting the results from the traditional hydraulic model and satisfying the continuous governing equations. In this study, guidelines are developed for selecting the appropriate approach based on a real-world case: PI-SWE-DeepONet is preferred for out-of-distribution predictions, uncertain training data, or when physical consistency is a priority, while SWE-DeepONet is recommended if the modeling objective is to replicate faithfully the traditional hydraulic model results within the training distribution.

The result is that no network can satisfy both competing objectives, necessitating a compromise between the two. p. 1
Where the paper says it

The gain of the physics-informed training, however, comes with costs, chief among which are the simulated results have slightly higher errors for in-distribution cases. It reflects the existence of a tension between the two competing training objectives: fitting the results from the traditional hydraulic model and satisfying the continuous governing equations.

Determining an appropriate weighting is non-trivial and subjective, as the normalization schemes differ between the training data and the PDE residuals. p. 1
Where the paper says it

In this study, guidelines are developed for selecting the appropriate approach based on a real-world case: PI-SWE-DeepONet is preferred for out-of-distribution predictions, uncertain training data, or when physical consistency is a priority, while SWE-DeepONet is recommended if the modeling objective is to replicate faithfully the traditional hydraulic model results within the training distribution. Other challenges are also discussed, such as the loss weighting approach.

The ratio λdata/λPDEs was set to 2:1 based on performance on in-distribution test cases. p. 8
Where the paper says it

These losses are optional and they are omitted for the examples in this work. The total loss function is then expressed as a weighted sum of all loss components: Ltotal = λdataLdata + λPDEsLPDEs + λBC/IC LBC/IC + λotherLother (6) where the weights λdata, λPDEs, λBC/IC , and λother are hyperparameters that measure the relative importance of corresponding loss components.

The choice between SWE-DeepONet and PI-SWE-DeepONet models should be guided by the application requirements. PI-SWE-DeepONet is preferred when predictions are needed for out-of-distribution cases, training data contains relatively large uncertainties, or physical consistency is the priority. p. 1

Where the paper says it

It reflects the existence of a tension between the two competing training objectives: fitting the results from the traditional hydraulic model and satisfying the continuous governing equations. In this study, guidelines are developed for selecting the appropriate approach based on a real-world case: PI-SWE-DeepONet is preferred for out-of-distribution predictions, uncertain training data, or when physical consistency is a priority, while SWE-DeepONet is recommended if the modeling objective is to replicate faithfully the traditional hydraulic model results within the training distribution.

SWE-DeepONet is the choice when the goal is to replicate faithfully the result of a trusted physics-based solver within the training distribution. p. 1
Where the paper says it

It reflects the existence of a tension between the two competing training objectives: fitting the results from the traditional hydraulic model and satisfying the continuous governing equations. In this study, guidelines are developed for selecting the appropriate approach based on a real-world case: PI-SWE-DeepONet is preferred for out-of-distribution predictions, uncertain training data, or when physical consistency is a priority, while SWE-DeepONet is recommended if the modeling objective is to replicate faithfully the traditional hydraulic model results within the training distribution.

The framework, implemented in the open-source HydroNet package, provides an automated workflow for surrogate modeling. p. 8
Where the paper says it

This flexibility makes the framework adaptable to diverse hydraulic modeling scenarios, from constant discharges (steady-state) to complex unsteady open channel flows with multiple inflow and exit boundaries. In this study, the algorithms and workflow presented are implemented in the opensource Python package HydroNet, which uses PyTorch for deep learning model development and training.

Future work should explore the potential of physics-informed training to reduce data requirements and extend the framework to unsteady flows. The development of more robust loss balancing strategies is also needed. p. 9
Where the paper says it

A 2D mesh is generated for use by SRH-2D and shown in Figure 3(a); it has 5,036 mixed triangle-quad cells. For the DeepONet surrogate models, collocation points are additionally needed on which the PDE residuals and loss functions due to boundary conditions are computed.

The paper presents a physics-informed DeepONet framework for computational hydraulics modeling. p. 1
Where the paper says it

manuscript submitted to Abstract Traditional 2D hydraulic models face significant computational challenges that limit their applications that are time-sensitive or require many model evaluations. This study presents a physics-informed Deep Operator Network (DeepONet) framework for computational hydraulics modeling that learns the solution operator of the 2D shallow water equations (SWEs) to create fast surrogate models.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Physics informed DeepONet framework can be designed to facilitate practical adoption in real-world engineering applications by developing an automated process for all four stages of modeling: generating training data, designing the operator learning architecture, training the model, and testing performance.

  2. The improved AI system can perform rapid scenario analysis, uncertainty quantification, risk analyses, and real-time forecasting while maintaining physical consistency required for engineering decision-making.

  3. The PI-SWE-DeepONet model can be used for out of distribution predictions or when physical consistency is a priority, as it exhibit[s] slower error growth and larger breakdown distances in comparison with SWE-DeepONet.

Abstract

Traditional 2D hydraulic models face significant computational challenges that limit their applications that are time-sensitive or require many model evaluations. This study presents a physics-informed Deep Operator Network (DeepONet) framework for computational hydraulics modeling that learns the solution operator of the 2D shallow water equations (SWEs) to create fast surrogate models. The framework can operate in two modes: a purely data-driven SWE-DeepONet that learns from numerical solver such as SRH-2D, and a physics-informed PI-SWE-DeepONet that additionally incorporates the continuous SWEs as constraints during training. Based on a real-world case, steady flows in a reach of the Sacramento River in California, it is demonstrated that PI-SWE-DeepONet possesses much enhanced prediction capability than SWE-DeepONet when applied to out-of-distribution scenarios. The physics-informed model is shown to exhibit slower error growth and larger breakdown distances in comparison with SWE-DeepONet. The gain of the physics-informed training, however, comes with costs, chief among which are the simulated results have slightly higher errors for in-distribution cases. It reflects the existence of a tension between the two competing training objectives: fitting the results from the traditional hydraulic model and satisfying the continuous governing equations. In this study, guidelines are developed for selecting the appropriate approach based on a real-world case: PI-SWE-DeepONet is preferred for out-of-distribution predictions, uncertain training data, or when physical consistency is a priority, while SWE-DeepONet is recommended if the modeling objective is to replicate faithfully the traditional hydraulic model results within the training distribution. Other challenges are also discussed, such as the loss weighting approach.

Sources

Related papers