arXiv subjects

M. A. Mendez

Publications and source records attributed to M. A. Mendez.

5 recordsLinked to original sources

arXiv2024

Challenges and Opportunities for Machine Learning in Fluid Mechanics

Big data and machine learning are driving comprehensive economic and social transformations and are rapidly re-shaping the toolbox and the methodologies of applied scientists. Machine learning tools are designed to learn functions from data with little to no need of prior knowledge. As continuous developments in experimental and numerical methods improve our ability to collect high-quality data, machine learning tools become increasingly viable and promising also in disciplines rooted in physical principles. These notes explore how machine learning can be integrated and combined with more classic methods in fluid dynamics. After a brief review of the machine learning landscape, we show how many problems in fluid mechanics can be framed as machine learning problems and we explore challenges and opportunities. We consider several relevant applications: aeroacoustic noise prediction, turbulence modelling, reduced-order modelling and forecasting, meshless integration of (partial) differential equations, super-resolution and flow control. While this list is by no means exhaustive, the presentation will provide enough concrete examples to offer perspectives on how machine learning might impact the way we do research and learn from data.

physics.flu-dyn↗

arXiv2022

Control of a Wind-Turbine via Machine Learning techniques

This article presents two model-free controllers for wind-turbine torque and pitch control. These controllers are based on reinforcement learning (RL) and Bayesian optimization (BO) and do not rely on any mathematical model of the wind-turbine dynamics, in contrast to classical approaches designed on linearized models. The model-free controllers were benchmarked against a proportional-integral-derivative (PID) regulator in a numerical environment using Blade Element Momentum theory for computing the aerodynamic torque and the blade loads. The results showed that the model-free approaches could increase power harvesting while reducing wind turbine loads.

physics.flu-dyn↗

arXiv2022

Artificial Neural Networks Modelling of Wall Pressure Spectra Beneath Turbulent Boundary Layers

We analyse and compare various empirical models of wall pressure spectra beneath turbulent boundary layers and propose an alternative machine learning approach using Artificial Neural Networks (ANN). The analysis and the training of the ANN are performed on data from experiments and high-fidelity simulations by various authors, covering a wide range of flow conditions. We present a methodology to extract all the turbulent boundary layer parameters required by these models, also considering flows experiencing strong adverse pressure gradients. Moreover, the database is explored to unveil important dependencies within the boundary layer parameters and to propose a possible set of features from which the ANN should predict the wall pressure spectra. The results show that the ANN outperforms traditional models in adverse pressure gradients, and its predictive capabilities generalise better over the range of investigated conditions. The analysis is completed with a deep ensemble approach for quantifying the uncertainties in the model prediction and integrated gradient analysis of the model sensitivity to its inputs. Uncertainties and sensitivities allow for identifying the regions where new training data would be most beneficial to the model's accuracy, thus opening the path towards a self-calibrating modelling approach.

physics.flu-dyn↗

arXiv2020

Dynamics of the Jet Wiping Process via Integral Models

The jet wiping process is a cost-effective coating technique that uses impinging gas jets to control the thickness of a liquid layer dragged along a moving strip. This process is fundamental in various coating industries (mainly in hot-dip galvanizing) and is characterized by an unstable interaction between the gas jet and the liquid film that results in wavy final coating films. To understand the dynamics of the wave formation, we extend classic laminar boundary layer models for falling films to the jet wiping problem, including the self-similar integral boundary layer (IBL) and the weighted integral boundary layer (WIBL) models. Moreover, we propose a transition and turbulence model (TTBL) to explore modelling extensions to larger Reynolds numbers and to analyze the impact of the modelling strategy on the liquid film dynamics. The validity of the long-wave formulation was first analyzed on a simpler problem, consisting of a liquid film falling over an upward-moving wall, using Volume Of Fluid (VOF) simulations. This validation proved the robustness of the integral formulation in conditions that are well outside their theoretical limits of validity. Finally, the three models were used to study the response of the liquid coat to harmonic and non-harmonic oscillations and pulsations in the impinging jet. The impact of these disturbances on the average coating thickness and wave amplitude is analyzed, and the range of dimensionless frequencies yielding maximum disturbance amplification is presented.

physics.flu-dyn↗

arXiv2019

Multi-Scale Proper Orthogonal Decomposition of Complex Fluid Flows

Data-driven decompositions are becoming essential tools in fluid dynamics, allowing for tracking the evolution of coherent patterns in large datasets, and for constructing low order models of complex phenomena. In this work, we analyze the main limits of two popular decompositions, namely the Proper Orthogonal Decomposition (POD) and the Dynamic Mode Decomposition (DMD), and we propose a novel decomposition which allows for enhanced feature detection capabilities. This novel decomposition is referred to as Multiscale Proper Orthogonal Decomposition (mPOD) and combines Multiresolution Analysis (MRA) with a standard POD. Using MRA, the mPOD splits the correlation matrix into the contribution of different scales, retaining non-overlapping portions of the correlation spectra; using the standard POD, the mPOD extracts the optimal basis from each scale. After introducing a matrix factorization framework for data-driven decompositions, the MRA is formulated via 1D and 2D filter banks for the dataset and the correlation matrix respectively. The validation of the mPOD, and a comparison with the Discrete Fourier Transform (DFT), DMD and POD are provided in three test cases. These include a synthetic test case, a numerical simulation of a nonlinear advection-diffusion problem, and an experimental dataset obtained by the Time-Resolved Particle Image Velocimetry (TR-PIV) of an impinging gas jet. For each of these examples, the decompositions are compared in terms of convergence, feature detection capabilities, and time-frequency localization.

physics.flu-dyn↗