{"type": "FeatureCollection", "features": [{"id": "10.1016/j.renene.2021.02.003", "type": "Feature", "geometry": null, "properties": {"updated": "2026-09-22T16:16:19Z", "type": "Journal Article", "created": "2020-11-05", "title": "Virtual fatigue diagnostics of wake-affected wind turbine via Gaussian Process Regression", "description": "<p>We propose a data-driven model to predict the short-term fatigue Damage Equivalent Loads (DEL) on a wake-affected wind turbine based on wind field inflow sensors and/or loads sensors deployed on an adjacent up-wind wind turbine. Gaussian Process Regression (GPR) with Bayesian hyperparameters calibration is proposed to obtain a surrogate from input random variables to output DELs in the blades and towers of the up-wind and wake-affected wind turbines. A sensitivity analysis based on the hyperparameters of the GPR and Kullback-Leibler divergence is conducted to assess the effect of different input on the obtained DELs. We provide qualitative recommendations for a minimal set of necessary and sufficient input random variables to minimize the error in the DEL predictions on the wake-affected wind turbine. Extensive simulations are performed comprising different random variables, including wind speed, turbulence intensity, shear exponent and inflow horizontal skewness. Furthermore, we include random variables related to the blades lift and drag coefficients with direct impact on the rotor aerodynamic induction, which governs the evolution and transport of the meandering wake. In addition, different spacing between the wind turbines and W\u00f6hler exponents for calculation of DELs are considered. The maximum prediction normalized mean squared error, obtained in the tower base DELs in the fore-aft direction of the wake affected wind turbine, is less than 4%. In the case of the blade root DELs, the overall prediction error is less than 1%. The proposed scheme promotes utilization of sparse structural monitoring (loads) measurements for improving diagnostics on wake-affected turbines.</p>", "keywords": ["bepress|Physical Sciences and Mathematics|Physics|Engineering Physics", "engrXiv|Engineering|Risk Analysis", "engrXiv|Engineering|Other Engineering", "bepress|Engineering", "engrXiv|Engineering|Mechanical Engineering|Fluid Mechanics", "bepress|Engineering|Mechanical Engineering", "engrXiv|Engineering|Mechanical Engineering", "bepress|Engineering|Mechanical Engineering|Applied Mechanics", "Gaussian Process Regression", "02 engineering and technology", "7. Clean energy", "Virtual sensing", "wind turbine", "bepress|Engineering|Computational Engineering", "engrXiv|Engineering|Civil and Environmental Engineering", "0202 electrical engineering", " electronic engineering", " information engineering", "uncertainty", "Fatigue", "wake", "engrXiv|Engineering|Civil and Environmental Engineering|Structural Engineering", "Uncertainty", "engrXiv|Engineering|Mechanical Engineering|Applied Mechanics", "Bayesian Calibration", "engrXiv|Engineering|Engineering Physics", "bepress|Engineering|Risk Analysis", "engrXiv|Engineering", "bepress|Engineering|Civil and Environmental Engineering", "engrXiv|Engineering|Computational Engineering", "Wake", "bepress|Engineering|Aerospace Engineering|Aerodynamics and Fluid Mechanics", "bepress|Engineering|Civil and Environmental Engineering|Structural Engineering", "fatigue", "bepress|Engineering|Other Engineering", "Sensitivity analysis", "Wind turbine", "Bayesian Gaussian process regression"]}, "links": [{"href": "https://doi.org/10.1016/j.renene.2021.02.003"}, {"rel": "related", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/Renewable%20Energy", "name": "related record", "description": "related record", "type": "application/json"}, {"rel": "self", "type": "application/geo+json", "title": "10.1016/j.renene.2021.02.003", "name": "item", "description": "10.1016/j.renene.2021.02.003", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/10.1016/j.renene.2021.02.003"}, {"rel": "collection", "type": "application/json", "title": "Collection", "name": "collection", "description": "Collection", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main"}], "time": {"date": "2020-11-05T00:00:00Z"}}, {"id": "10.5281/zenodo.4896835", "type": "Feature", "geometry": null, "properties": {"license": "Open Access", "updated": "2026-09-22T16:21:05Z", "type": "Report", "title": "Virtual loads predictions of wake-affected wind turbines: Gaussian process regression and deep neural networks", "description": "Load analysis of wind turbines may be performed either via physics-based models or via direct measurement. On the first case, loads are calculated with aero-elastic models, based on significant assumptions on the mechanical and aeroelastic properties of the structure and the acting forces (wind, wave and control). Otherwise, loads can be directly measured based on a sensor network, which entails increased costs due to installation, maintenance and calibration of sensors and IT infrastructure. These costs can be manageable for a single wind turbine but become substantial on densely instrumented wind farms. In turn, the increased costs negatively affect the levelized cost of energy. This is, in fact, the main reason why stakeholders shy away from applying monitoring technologies in wind farms. To overcome the above challenges, we need an alternative way of estimating the loads which would involve a reduced number of sensors while replicating the actual load measurement scenario. To this end, we propose data-driven models to predict the loads acting on different components of a wind turbine. These models use SCADA, wind inflow and other variables to predict loads in components of interest of a wind turbine. We have already successfully demonstrated this concept in the past on simulated wind turbine Damage Equivalent Loads (DELs) based on Gaussian Process regression [1,2], and on real wind turbine data [3]. In this work, we validate this approach on actual wind turbine data from the Alpha Ventus Wind Farm obtained within the framework of the Research Alpha Ventus (RAVE) project. Two Senvion turbines are selected for this study. One of the wind turbines is used to train and validate a regression model to predict the tower base DELs based on SCADA, wind inflow and other environmental variables. Afterwards, the trained model is used to predict the loads in the second wind turbine. Load prediction is attained with two machine learning methods, the first one based on Gaussian Process Regression (GPR) and the second one based on Artificial Neural Networks (ANN). For the first one, a decision tree is used to separate the different operating modes of the wind turbine (idling, operating and transitioning). The decision tree is built on simple heuristics on a subset of SCADA variables (mean and standard deviation of the rotor RPM and blade pitch angle). Subsequently, a GPR is built for each one of the operating modes. In the second method, the SCADA variables are fed to the ANN after undergoing an initial transformation for data compression and collinearity reduction.", "keywords": ["machine learning", "structural health monitoring", "13. Climate action", "wind turbines", "virtual sensing", "7. Clean energy"], "contacts": [{"organization": "Avenda\u00f1o-Valencia, Luis David, Abdallah, Imad, Venu, Anish, Chatzi, Eleni,", "roles": ["creator"]}]}, "links": [{"href": "https://doi.org/10.5281/zenodo.4896835"}, {"rel": "self", "type": "application/geo+json", "title": "10.5281/zenodo.4896835", "name": "item", "description": "10.5281/zenodo.4896835", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/10.5281/zenodo.4896835"}, {"rel": "collection", "type": "application/json", "title": "Collection", "name": "collection", "description": "Collection", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main"}], "time": {"date": "2019-01-01T00:00:00Z"}}], "links": [{"rel": "self", "type": "application/geo+json", "title": "This document as GeoJSON", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items?keywords=Virtual+sensing&f=json", "hreflang": "en-US"}, {"rel": "alternate", "type": "text/html", "title": "This document as HTML", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items?keywords=Virtual+sensing&f=html", "hreflang": "en-US"}, {"rel": "collection", "type": "application/json", "title": "Collection URL", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main", "hreflang": "en-US"}, {"type": "application/geo+json", "rel": "first", "title": "items (first)", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items?keywords=Virtual+sensing&", "hreflang": "en-US"}, {"rel": "last", "type": "application/geo+json", "title": "items (last)", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items?keywords=Virtual+sensing&offset=2", "hreflang": "en-US"}], "numberMatched": 2, "numberReturned": 2, "distributedFeatures": [], "timeStamp": "2026-09-22T18:52:23.252789Z"}