Gregory Duthé

Last Name

Duthé

First Name

Gregory

ORCID

0000-0002-0895-6766

Organisational unit

03890 - Chatzi, Eleni / Chatzi, Eleni

Show all metadata

Search Results

Publications1 - 10 of 23

Flexible multi-fidelity framework for load estimation of wind farms through graph neural networks and transfer learning
Duthé, Gregory; de N Santos, Francisco; Abdallah, Imad; et al. (2024)
Data-Centric Engineering
With global wind energy capacity ramping up, accurately predicting damage equivalent loads (DELs) and fatigue across wind turbine populations is critical, not only for ensuring the longevity of existing wind farms but also for the design of new farms. However, the estimation of such quantities of interests is hampered by the inherent complexity in modeling critical underlying processes, such as the aerodynamic wake interactions between turbines that increase mechanical stress and reduce useful lifetime. While high-fidelity computational fluid dynamics and aeroelastic models can capture these effects, their computational requirements limits real-world usage. Recently, fast machine learning-based surrogates which emulate more complex simulations have emerged as a promising solution. Yet, most surrogates are task-specific and lack flexibility for varying turbine layouts and types. This study explores the use of graph neural networks (GNNs) to create a robust, generalizable flow and DEL prediction platform. By conceptualizing wind turbine populations as graphs, GNNs effectively capture farm layout-dependent relational data, allowing extrapolation to novel configurations. We train a GNN surrogate on a large database of PyWake simulations of random wind farm layouts to learn basic wake physics, then fine-Tune the model on limited data for a specific unseen layout simulated in HAWC2Farm for accurate adapted predictions. This transfer learning approach circumvents data scarcity limitations and leverages fundamental physics knowledge from the source low-resolution data. The proposed platform aims to match simulator accuracy, while enabling efficient adaptation to new higher-fidelity domains, providing a flexible blueprint for wake load forecasting across varying farm configurations.
Graph Neural Networks for Aerodynamic Flow Reconstruction from Sparse Sensing
Duthé, Gregory; Abdallah, Imad; Barber, Sarah; et al. (2023)
arXiv
Sensing the fluid flow around an arbitrary geometry entails extrapolating from the physical quantities perceived at its surface in order to reconstruct the features of the surrounding fluid. This is a challenging inverse problem, yet one that if solved could have a significant impact on many engineering applications. The exploitation of such an inverse logic has gained interest in recent years with the advent of widely available cheap but capable MEMS-based sensors. When combined with novel data-driven methods, these sensors may allow for flow reconstruction around immersed structures, benefiting applications such as unmanned airborne/underwater vehicle path planning or control and structural health monitoring of wind turbine blades. In this work, we train deep reversible Graph Neural Networks (GNNs) to perform flow sensing (flow reconstruction) around two-dimensional aerodynamic shapes: airfoils. Motivated by recent work, which has shown that GNNs can be powerful alternatives to mesh-based forward physics simulators, we implement a Message-Passing Neural Network to simultaneously reconstruct both the pressure and velocity fields surrounding simulated airfoils based on their surface pressure distributions, whilst additionally gathering useful farfield properties in the form of context vectors. We generate a unique dataset of Computational Fluid Dynamics simulations by simulating random, yet meaningful combinations of input boundary conditions and airfoil shapes. We show that despite the challenges associated with reconstructing the flow around arbitrary airfoil geometries in high Reynolds turbulent inflow conditions, our framework is able to generalize well to unseen cases.
On the potential of aerodynamic pressure measurements for structural damage detection
Franz, Philip; Abdallah, Imad; Duthé, Gregory; et al. (2025)
Wind Energy Science
This study investigates the potential of using aerodynamic pressure time series measurements to detect structural damage in elastic, aerodynamically loaded structures. Our work is motivated by the increase in the dimensions of modern wind turbine blade (WTB) designs, whose complex behavior necessitates the adoption of improved simulation and structural monitoring solutions. In refining the tracking of aerodynamic interactions and their effects on such structures, we propose to exploit aerodynamic pressure measurements, available from a novel, cost-effective, and non-intrusive sensing system, for structural damage assessment on WTBs. This proof-of-concept study is based on a series of wind tunnel experiments on an NACA 633418 airfoil. The airfoil is mounted on a vertically oscillating cantilever beam with structural damage introduced in the form of a crack by gradually sawing the cantilever beam close to its support. The pressure distribution on the airfoil is measured under diverse configurations of inflow conditions and structural states, including different angles of attack, wind velocities, heaving frequencies, and crack lengths. We further propose an algorithm, relying on convolutional neural networks (CNNs), for damage detection and rating based on the monitored signals. Analysis of the dynamics of the system using reference acceleration measurements and a finite element (FE) model and application of the suggested method on the experimental data indicate that aerodynamic pressure measurements on airfoils can indeed be used as an indirect approach for damage detection and severity classification on elastic, beam-like structures in mildly turbulent environments.
Modeling and Monitoring Erosion of the Leading Edge of Wind Turbine Blades
Duthé, Gregory; Abdallah, Imad; Barber, Sarah; et al. (2021)
Energies
Leading edge surface erosion is an emerging issue in wind turbine blade reliability, causing a reduction in power performance, aerodynamic loads imbalance, increased noise emission, and, ultimately, additional maintenance costs, and, if left untreated, it leads to the compromise of the functionality of the blade. In this work, we first propose an empirical spatio-temporal stochastic model for simulating leading edge erosion, to be used in conjunction with aeroelastic simulations, and subsequently present a deep learning model to be trained on simulated data, which aims to monitor leading edge erosion by detecting and classifying the degradation severity. This could help wind farm operators to reduce maintenance costs by planning cleaning and repair activities more efficiently. The main ingredients of the model include a damage process that progresses at random times, across multiple discrete states characterized by a non-homogeneous compound Poisson process, which is used to describe the random and time-dependent degradation of the blade surface, thus implicitly affecting its aerodynamic properties. The model allows for one, or more, zones along the span of the blades to be independently affected by erosion. The proposed model accounts for uncertainties in the local airfoil aerodynamics via parameterization of the lift and drag coefficients’ curves. The proposed model was used to generate a stochastic ensemble of degrading airfoil aerodynamic polars, for use in forward aero-servo-elastic simulations, where we computed the effect of leading edge erosion degradation on the dynamic response of a wind turbine under varying turbulent input inflow conditions. The dynamic response was chosen as a defining output as this relates to the output variable that is most commonly monitored under a structural health monitoring (SHM) regime. In this context, we further proposed an approach for spatio-temporal dependent diagnostics of leading erosion, namely, a deep learning attention-based Transformer, which we modified for classification tasks on slow degradation processes with long sequence multivariate time-series as inputs. We performed multiple sets of numerical experiments, aiming to evaluate the Transformer for diagnostics and assess its limitations. The results revealed Transformers as a potent method for diagnosis of such degradation processes. The attention-based mechanism allows the network to focus on different features at different time intervals for better prediction accuracy, especially for long time-series sequences representing a slow degradation process.
Physics-aware deep learning for wind prediction over complex terrain
Duthé, Gregory (2019)
Rapidly computing the wind flow over complex terrain features is a challenging problem with many potential fields of application. Such domains include autonomous UAV path planning, naturally ventilated building design or even wind farm layout optimization. Traditionally, this requires performing Computational Fluid Dynamics (CFD) simulations, which are computationally expensive and time consuming. One recent approach has been to use deep convolutional neural networks as a data-driven simulation alternative which leverages CFD simulation results for training. Albeit fast and relatively accurate, this method does not, however, explicitly take into account flow properties. We believe that doing so could improve prediction accuracy, especially in previously under-predicted areas which usually exhibit strong shearing flows and large velocity magnitudes. In this thesis, we investigate training deep neural networks with physics based loss functions and loss weighting methods, constructed to accentuate desirable flow features and properties. Furthermore, we explore other potential solutions, such as a physics informed adversarial approach, or using the recently proposed SPADE network which reported good spatial expressivity. Comparisons to baseline results are provided for all tested solutions, allowing us to select the combination of methods which provides the most accurate wind flow predictions. We show that training a network with a loss weighted with the velocity gradient magnitude results in a 5% decrease in maximum prediction error while retaining a comparable mean prediction error, when compared to the baseline approach.
Graph Transformers for inverse physics: reconstructing flows around arbitrary 2D airfoils
Duthé, Gregory; Abdallah, Imad; Chatzi, Eleni (2025)
arXiv
We introduce a Graph Transformer framework that serves as a general inverse physics engine on meshes, demonstrated through the challenging task of reconstructing aerodynamic flow fields from sparse surface measurements. While deep learning has shown promising results in forward physics simulation, inverse problems remain particularly challenging due to their ill-posed nature and the difficulty of propagating information from limited boundary observations. Our approach addresses these challenges by combining the geometric expressiveness of message-passing neural networks with the global reasoning of Transformers, enabling efficient learning of inverse mappings from boundary conditions to complete states. We evaluate this framework on a comprehensive dataset of steady-state RANS simulations around diverse airfoil geometries, where the task is to reconstruct full pressure and velocity fields from surface pressure measurements alone. The architecture achieves high reconstruction accuracy while maintaining fast inference times. We conduct experiments and provide insights into the relative importance of local geometric processing and global attention mechanisms in mesh-based inverse problems. We also find that the framework is robust to reduced sensor coverage. These results suggest that Graph Transformers can serve as effective inverse physics engines across a broader range of applications where complete system states must be reconstructed from limited boundary observations.
Hybrid Model for Inflow Conditions Inference on Airfoils Under Uncertainty
Marykovskiy, Yuriy; Deparday, Julien; Abdallah, Imad; et al. (2023)
AIAA Journal
Estimation of inflow conditions, such as wind speed and angle of attack, is vital for assessing aerodynamic performance of a lifting profile. This task is particularly challenging in the field due to the inherent stochasticity of the inflow variables. In practice, the field installation of a measurement system exacerbates the measurement uncertainty. Here, we present a hybrid model to infer the inflow conditions on a wind turbine blade along with a process to quantify the involved uncertainty. The model combines potential flow theory and conformal mapping with pressure measurements from a novel monitoring system, which eliminates the need for external reference pressure measurements. Stagnation point location and wind speed are formulated as outputs of an optimization problem, in which pressure differences along the surface of an airfoil are connected to the potential flow solution through the Bernoulli equation. The proposed scheme is experimentally validated. The hybrid model offers a practical and robust solution for inflow condition estimation, suitable for field deployment on wind turbine or aircraft. The uncertainty quantification process provides valuable insights for improving monitoring system design and quantifying the accuracy of the predictive scheme before actual field installation.
Local flow and loads estimation on wake-affected wind turbines using graph neural networks and PyWake
Duthé, Gregory; de Nolasco Santos, Francisco; Abdallah, Imad; et al. (2023)
Journal of Physics: Conference Series
The estimation of the flow conditions throughout a modern wind farm is a relevant albeit complex problem involving several different modelling scales. In this context, recent years have seen the rise of the demand - normally industry-driven - of quick, computationally inexpensive, wind farm flow field simulators upon which further calculations might be undertaken (e.g. power output, fatigue loads, etc.). Naturally, the trustworthiness of such models hinges upon their ability to accurately estimate wakes, under the evident trade-off between computational time and accuracy. One such simulator that has been recently gaining traction within the wind community is PyWake, a Python-based flow solver, with several possible add-ons and modularity options. Based on runs from PyWake, this contribution studies the applicability of graph neural networks (GNN) as a simultaneous wind turbine fleet, wind inflow and loads surrogate. As graphs are discrete mathematical sets of dependent objects, one can see how the layout-dependent interaction between wind turbines' aerodynamics, from which wake arises, lends itself to be modelled as a graph. Thus, we introduce the use of GNNs for layout-agnostic joint modelling of rotor-averaged wind speed, turbulence intensity, power production and damage equivalent loads (DEL) on individual turbines of wind farms. To this end, probabilistic samples of inflow conditions from a Weibull distribution for wind speed, uniform wind direction and conditional normal distributions of wind shear and nacelle yaw angles are generated as main inflow properties in the numerical simulations. Additionally, arbitrary wind farm layouts are created based on varied geometric shapes with random parametrization (varying orientations, length/width ratio). Both the arbitrary layouts and the random inflow conditions are then used as inputs for PyWake. PyWake's output, namely, rotor averaged wind speed, turbulence intensity, power production and DELs of each individual wind turbine in the wind farms, are used to train a general GNN model. We elect to implement an Encode-Process-Decode GNN as a graph learning model. We compare the performance of various approaches of the relational dependency representation (edge-forming techniques) amongst the nodes (wind turbines) on the graph, such as Delaunay triangulation, KNN, radius-based methods and fully connected scheme. In our analysis, we evaluate the accuracy of GNNs and its ability to generalize their joint predictions for unseen layouts and varying inflow conditions.
Development of a freely distributable CFD tool for approximate and detailed simulations of the flow around a complex of buildings
Decaix, Jean; Jaboyedoff, Pierre; Duthé, Gregory; et al. (2021)
Journal of Physics: Conference Series
Indo-Swiss Building Energy Efficiency Project (BEEP) is a cooperation project between the Ministry of Power, Government of India, and the Federal Department of Foreign Affairs of the Swiss Confederation. Started in 2011, the project's central focus is to help India mainstream Energy-Efficient and Thermally Comfortable (EETC) Building Design. BEEP works with building industry, policy makers, and building owners to catalyse adoption of EETC building design and technologies. India wants to avoid or reduce the use of air conditioning by improving natural ventilation at night, which requires numerical simulations to compute the flow around the buildings. However, the simulations of fluid flows are time consuming and are not used at the beginning of a project when the locations of the buildings are set. To improve the situation, a freely distributable environment based on the OpenFOAM toolbox has been developed providing two levels of resolution: an approximate level computing the flow in few minutes and a RANS level of simulation. The user inputs are limited to the geometry and the velocity direction and magnitude. The mesh and the numerical set up are automated. The accuracy of the two levels of resolution have been checked by computing test cases from the CEDVAL database.
Towards unsupervised fault detection for offshore wind turbine cable protection systems using contrastive learning
Duthé, Gregory; L'Homme, Yan; Abdallah, Imad; et al. (2024)
Journal of Physics: Conference Series ~ Dynamics, control, and monitoring
As offshore wind power expands globally, it is essential to ensure the reliable operation of components of such critical infrastructures. A less explored instance of such components, which are though essential in terms of operation, is found in subsea turbine cables and their protection systems, whose failure can incur prolonged shutdown periods and costly repairs. We propose a novel unsupervised machine learning approach exploiting use of Distributed Acoustic Sensing (DAS) data and contrastive learning for monitoring offshore wind turbine Cable Protection Systems (CPSs). A Transformer neural network adapted for time-series ingests the high-frequency, noisy DAS CPS time-series measurements, and is trained to learn a coherent representation of the data using a contrastive learning scheme that enforces temporal and positional consistency in the latent space. This latent representation can then be used to perform anomaly detection in an unsupervised manner, alleviating the need for costly labeled offshore anomaly data. We demonstrate that a coherent representation of the data is learnt by the model, which we then use to detect synthetic anomalies and an actual CPS stabilization event.

Publications1 - 10 of 23

Gregory Duthé

Last Name

First Name

ORCID

Organisational unit

Filters

Search Results