Abstract
A convolutional neural network (CNN) autoencoder model has been developed to generate 3D realizations of time-averaged velocity in the wake of the wind turbines at the Sandia National Laboratories Scaled Wind Farm Technology (SWiFT) facility. Large-eddy simulations (LES) of the SWiFT site are conducted using an actuator surface model to simulate the turbine structures to produce training and validation datasets of the CNN. The simulations are validated using the SpinnerLidar measurements of turbine wakes at the SWiFT site and the instantaneous and time-averaged velocity fields from the training LES are used to train the CNN. The trained CNN is then applied to predict 3D realizations of time-averaged velocity in the wake of the SWiFT turbines under flow conditions different than those for which the CNN was trained. LES results for the validation cases are used to evaluate the performance of the CNN predictions. Comparing the validation LES results and CNN predictions, we show that the developed CNN autoencoder model holds great potential for predicting time-averaged flow fields and the power production of wind turbines while being several orders of magnitude computationally more efficient than LES.
Highlights
Luis Hernández-Callejo, In a wind farm, turbine wake interactions cause power losses and may increase fatigue loads on downwind wind turbines [1,2]
The accurate prediction of turbine wakes is an important consideration in wind farm layout optimization, which can improve the efficiency of power production and reduce the overall levelized cost of energy
We develop a convolutional neural network (CNN) autoencoder model for generating 3D realizations of time-averaged turbulent wake flow of wind turbines at the Sandia National Laboratories Scaled Wind Farm Technology (SWiFT) site in Lubbock, Texas
Summary
Luis Hernández-Callejo, In a wind farm, turbine wake interactions cause power losses and may increase fatigue loads on downwind wind turbines [1,2]. Renganathan et al [41] combined an MLP and GP with a Convolutional Neural Network (CNN) decoder to map the wind turbine operation parameters, such as inflow wind speed, turbulent intensity, turbine power generation, atmospheric-boundary layer (ABL) Richardson number, rotor angular speed, and pitch angle, to the wake flow field. Aird et al [42] developed a mask Region based Convolutional Neural Network (R-CNN) model that identifies turbine wakes in Lidar scan images with high accuracy, even with some missing data points, and is able to character wake shapes in its forming and dissipating Despite these contributions, the accuracy of the existing algorithms for velocity field predictions varies with flow conditions and wind farm layouts, limiting their application for wind farm optimization.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have