An artificial neural network approach to bifurcating phenomena in computational fluid dynamics

Pichi, Federico; Ballarin, Francesco; Rozza, Gianluigi; Hesthaven, Jan S.

doi:10.1016/j.compfluid.2023.105813

Cited by 28 publications

(3 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Common choices for the interpolatory map are radial basis functions, 40 Gaussian process, 41 , 42 cubic splines, 43 , 44 or artificial neural networks. 45 , 46 We are going to show how to compute an efficient approximation of such a map by exploiting only the directions of maximal variations in a multi‐fidelity setting, without the need to perform additional simulations.…”

Section: Reduced Order Modelsmentioning

confidence: 99%

A multifidelity approach coupling parameter space reduction and nonintrusive POD with application to structural optimization of passenger ship hulls

Tezzele

Fabris

Sidari

et al. 2022

Numerical Meth Engineering

View full text Add to dashboard Cite

Nowadays, the shipbuilding industry is facing a radical change toward solutions with a smaller environmental impact. This can be achieved with low emissions engines, optimized shape designs with lower wave resistance and noise generation, and by reducing the metal raw materials used during the manufacturing. This work focuses on the last aspect by presenting a complete structural optimization pipeline for modern passenger ship hulls which exploits advanced model order reduction techniques to reduce the dimensionality of both input parameters and outputs of interest. We introduce a novel approach which incorporates parameter space reduction through active subspaces into the proper orthogonal decomposition with interpolation method. This is done in a multi‐fidelity setting. We test the whole framework on a simplified model of a midship section and on the full model of a passenger ship, controlled by 20 and 16 parameters, respectively. We present a comprehensive error analysis and show the capabilities and usefulness of the methods especially during the preliminary design phase, finding new unconsidered designs while handling high dimensional parameterizations.

show abstract

Section: Reduced Order Modelsmentioning

confidence: 99%

A multifidelity approach coupling parameter space reduction and nonintrusive POD with application to structural optimization of passenger ship hulls

Tezzele

Fabris

Sidari

et al. 2022

Numerical Meth Engineering

View full text Add to dashboard Cite

show abstract

“…To demonstrate the network architecture associated with the LSTM model, we first introduce a simple Artificial Neural Network (ANN), followed by its extension to a Recurrent Neural Network (RNN) and finally to an LSTM network. ANN is a modelling technique able to formulate a nonlinear functional relationship between input and output data: see, e.g., [52,53,54,55,56]. In our case the input-output data are the pairs…”

Section: Long Short-term Memory Networkmentioning

confidence: 99%

A non-intrusive data-driven reduced order model for parametrized CFD-DEM numerical simulations

Hajisharifi

Romanò

Girfoglio

et al. 2023

Journal of Computational Physics

View full text Add to dashboard Cite

“…Researchers have recently discussed the predicted Artificial Neural Network (ANN) approaches to explain the different models. Pichi et al [24] utilized the concept of ANN to present the model representing the triangular cavity flow. ANN was used by Alsaiari et al [25] to estimate the water productivity of diverse strategies of solar stills.…”

Section: Introductionmentioning

confidence: 99%

Stochastic Levenberg–Marquardt Neural Network Implementation for Analyzing the Convective Heat Transfer in a Wavy Fin

et al. 2023

View full text Add to dashboard Cite

The present research examines the steady, one-dimensional thermal distribution and heat transfer of a wavy fin. This heat transfer analysis considers convective effects as well as temperature-dependent thermal conductivity. Furthermore, a novel implementation of a neural network with backpropagated Levenberg–Marquardt algorithm (NN-BLMA)-based machine learning intelligent strategies is provided to interpret the heat transfer analysis of a convective wavy fin. The non-linear ordinary differential equation (ODE) of the study problem is converted into its non-dimensional form using the similarity transformation technique. The dimensionless equation obtained is then numerically explored via the Runge–Kutta–Fehlberg scheme. A data set for varying the pertinent parameters is generated, and an artificial neural network model is designed to estimate the heat transfer behavior of the wavy fin. The effectiveness of the proposed NN-BLMA is subsequently endorsed by analyses using a regression model, mean square error, and histograms. The findings of comprehensive computational parametric studies illustrate that the presented technique, NN-BLMA is an effective convergent stochastic numerical solver employed for the heat transfer model of the convective wavy fin. The wavy fin’s temperature dispersion optimizes as the thermal conductivity parameter rises. Heat transfer rate is higher in wavy fin compared to rectangular fin.

show abstract

An artificial neural network approach to bifurcating phenomena in computational fluid dynamics

Cited by 28 publications

References 40 publications

A multifidelity approach coupling parameter space reduction and nonintrusive POD with application to structural optimization of passenger ship hulls

A multifidelity approach coupling parameter space reduction and nonintrusive POD with application to structural optimization of passenger ship hulls

A non-intrusive data-driven reduced order model for parametrized CFD-DEM numerical simulations

Stochastic Levenberg–Marquardt Neural Network Implementation for Analyzing the Convective Heat Transfer in a Wavy Fin

Contact Info

Product

Resources

About