A Probabilistic Learning Approach Applied to the Optimization of Wake Steering in Wind Farms

Abstract

The wake steering control in wind farms has gained significant attention in the last years. This control strategy has shown promise to reduce energy losses due to wake effects and increase the energy production in a wind farm. However, wind conditions are variable in wind farms, and the measurements are uncertain what should be considered in the design of wake steering control strategies. This paper proposes using the Probabilistic Learning on Manifold (PLoM), which can be viewed as a supervised machine learning method, to enable the wake steering optimization under uncertainty. The expected power generation is estimated considering uncertainties in wind speed and direction with good accuracy and reduced computational cost for two wind farm layouts, which expand the application of machine learning models in wake steering. Furthermore, the analysis shows the potential gain with the application of wake steering control.

Surrogate modeling of structural seismic response using probabilistic learning on manifolds

“…[41][42][43][44][45] This methodology is just coming to the attention of engineering researchers, with only a handful of papers utilizing the method outside of the research groups of the originators of the PLoM methodology. [46][47][48][49][50] In this paper, we show the utility of PLoM for estimating the joint probability distribution of EDPs under earthquake ground motions. In particular, PLoM is proposed for generating realizations of a random vector with values in a finite-dimensional Euclidean space that preserve the correlation structure of the original data and can comply with user-defined statistical constraints.…”

“…To address these challenges, this paper explores the use of Soize and Ghanem's recently developed machine learning methodology, probabilistic learning on manifolds (PLoM) 41–45 . This methodology is just coming to the attention of engineering researchers, with only a handful of papers utilizing the method outside of the research groups of the originators of the PLoM methodology 46–50 . In this paper, we show the utility of PLoM for estimating the joint probability distribution of EDPs under earthquake ground motions.…”

Polynomial-chaos-based conditional statistics for probabilistic learning with heterogeneous data applied to atomic collisions of Helium on graphite substrate

“…These problems could not be solved without these learning methods because the use of the usual methods would require computer resources, which are not available. Regarding these learning methods, let us cite, for example, learning with kernels [94,95,96], probabilistic and statistical learning [97,98,99,100], learning on the manifolds [101,102,103,104,105,106,107,108,109,110,111,112,113], and probabilistic physics-based learning [114,115,116,117,118].…”

Updating Nonlinear Stochastic Dynamics of an Uncertain Nozzle Model Using Probabilistic Learning With Partial Observability and Incomplete Dataset

“…The PLoM algorithm is particularly well-suited for scenarios involving small training datasets, and its efficiency has been demonstrated across various domains. Examples include non-convex optimization under uncertainty [41,42,43,44], model-form uncertainties using random bases [45], and the updating, design, and control of dynamical systems [46,47,48]. The statistical surrogate model will be based on conditional statistics for given control parameter, using the learned realizations obtained from PLoM under the constraints defined by the target.…”