On the use of a high-performance framework for efficient model coupling in hydroinformatics

Malleron, Nicolas; Zaoui, Fabrice; Goutal, Nicole; Morel, Thierry

doi:10.1016/j.envsoft.2011.05.017

Cited by 15 publications

(10 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 5 shows the results in the case of a wet floodplain (first line: h D 0:5, second line: h D 1:0) at rest, that is, v D 0 with a bank parallel to the river axis, that is, @ x b D 0, as function of the water elevation in the river . Firstly, we noted that neither the iterative process (8) nor the approximation using Riemann invariant (7) behaved like the threshold law (6) in the case of drowned threshold, that is, when 2H max < 3H min . Otherwise, when 2 > 3h, the iterative process leads to estimates of the same order than the threshold law, starting from the convergent result when h D´and going to the previous iterations when the water level in the river rises.…”

Section: Numerical Validation Of the Transverse Discharge Estimationmentioning

confidence: 96%

A 2D reconstruction for the transverse coupling of shallow water models

Goutal

Parisot

Zaoui

2014

Numerical Methods in Fluids

View full text Add to dashboard Cite

International audienceThis work is dedicated to the modeling of exchanges between a river and its surrounding floodplains during floods overflowing its bed. Modeling flows in each subdomain, that of the river and that of the floodplain, has been the subject of numerous works for over 50 years and is now well understood. A coupling strategy is considered in order to preserve the advantages of 1D and 2D models in their respective subdomains. In comparison with other transverse coupling strategies already presented in the literature, we introduce a direct method, without overlapping the models or introducing numerical parameters. This strategy is based on the resolution of the Riemann 2D problem at the coupling interface (the bank) and requires the estimation of the transverse velocity close to the interface. We propose a model of transverse velocity using successive resolutions of Riemann problems. Then, we present a numerical resolution of the coupling system, based on a finite volume method for any Riemann solvers. Particular attention was given to the essential properties of the model (conservation of mass, positivity of water depth, and well-balanced scheme). Lastly, the precision and efficiency of the method are illustrated using examples of simulations

show abstract

Section: Numerical Validation Of the Transverse Discharge Estimationmentioning

confidence: 96%

A 2D reconstruction for the transverse coupling of shallow water models

Goutal

Parisot

Zaoui

2014

Numerical Methods in Fluids

View full text Add to dashboard Cite

show abstract

“…Coupled models are used to study climate (Eyring et al, 2015), ecosystems (Lehodey, 2005;Fulton, 2010), earth systems (Sokolov et al, 2005;Warner et al, 2008), climate mitigation (Anthoff and Tol, 2010), the water-energy-food nexus (Hermann et al, 2011), and many other realms (Larson et al, 2005;Malleron et al, 2011). Coupled natural and human dynamics are of particular interest in understanding ecological and economic sustainability (Liu et al, 2007).…”

Section: Introductionmentioning

confidence: 98%

A flexible approach to model coupling through probabilistic pooling

Rising

2017

Environmental Modelling & Software

View full text Add to dashboard Cite

“…In contrast to the monolithic approach, the partitioned approach avoids software development challenges by maintaining the modularity of all the component models. [12][13][14][15][16][17][18][19] This also allows component source codes that are developed and validated independently from one another by experts to be integrated and coupled. As a result, legacy models, in-house research models, and commercial modeling and simulation packages can potentially be utilized within the system.…”

Section: Introductionmentioning

confidence: 99%

A Purposed Model Coupling Framework for the Rapid Assembly of Complex System Simulations within an Integrated Computational Environment

McNunn

Bryden

2014

52nd Aerospace Sciences Meeting

View full text Add to dashboard Cite

Often the emergent behavior of large-scale complex engineered and natural systems is a consequence of the interaction and interconnectivity between a large numbers of component subsystems. Consequently, modeling and simulating these large-scale systems requires methods that enable large numbers of heterogeneous, distributed models and databases to be assembled into a federated architecture that supports their interoperability and information sharing. However, to accurately propagate information through a federated system of models and databases a workflow is needed that identifies any data-related interdependencies that require an implicit solution and schedule the sequence of data exchanges, model runtimes, and database queries that is needed to evaluate the overall system. Although the local data exchanges that are needed to simulate the connectivity between any two components are often well understood, identifying the necessary workflow to solve a large-scale system of component models and databases is challenging and generally not practical using a hands-on or brute-force approach. As a result, a novel method is needed that identifies the computational workflow needed to accurately propagate data through a large-scale federated system of models and databases to ensure each component receives the necessary input data from the other components in the system prior to it being solved or queried. This paper develops a methodology that utilizes Tarjan's algorithm for finding the strongly connected components of a directed graph as a mechanism for organizing and scheduling the computational workflow needed to solve federated systems of models and databases based on the local input and output data associated with each component. The methodology is applied to identify the computational workflow needed to solve a system of models and databases representing the heat transfer and thermal stress in a one-dimensional gas turbine blade wall with a thermal barrier coating, and a system of models and databases representing a hybrid gas turbine, solid oxide fuel cell energy system.

show abstract

On the use of a high-performance framework for efficient model coupling in hydroinformatics

Cited by 15 publications

References 13 publications

A 2D reconstruction for the transverse coupling of shallow water models

A 2D reconstruction for the transverse coupling of shallow water models

A flexible approach to model coupling through probabilistic pooling

A Purposed Model Coupling Framework for the Rapid Assembly of Complex System Simulations within an Integrated Computational Environment

Contact Info

Product

Resources

About