Preprints
https://doi.org/10.5194/wes-2022-104
https://doi.org/10.5194/wes-2022-104
09 Nov 2022
 | 09 Nov 2022
Status: this preprint is currently under review for the journal WES.

Stochastic Gradient Descent for Wind Farm Optimization

Julian Quick, Pierre-Elouan Rethore, Mads Mølgaard Pedersen, and Rafael Valotta Rodrigues

Abstract. It is important to optimize wind turbine positions to mitigate potential wake losses. To perform this optimization, atmospheric conditions, such as the inflow speed and direction, are assigned probability distributions according to measured data, and these conditions are propagated through engineering wake models to estimate the annual energy production (AEP). This study presents stochastic gradient descent (SGD) for wind farm optimization, which is an approach that estimates the gradient of the AEP using Monte Carlo simulation, allowing for the consideration of an arbitrarily large number of atmospheric conditions. This method does not require that the atmospheric conditions be discretized, in contrast to the typical rectangular quadrature approximation of AEP. SGD is demonstrated using wind farms with square boundaries, considering cases with 25, 64, and 100 turbines, and the results are compared to a deterministic optimization approach. It is shown that SGD finds a larger optimal AEP in substantially less time than the deterministic counterpart as the number of wind turbines is increased.

Julian Quick et al.

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on wes-2022-104', Ahmad Vasel-Be-Hagh, 31 Dec 2022
  • RC2: 'Comment on wes-2022-104', Anonymous Referee #2, 03 Jan 2023
  • RC3: 'Comment on wes-2022-104', Anonymous Referee #3, 26 Jan 2023

Julian Quick et al.

Julian Quick et al.

Viewed

Total article views: 466 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
338 113 15 466 10 7
  • HTML: 338
  • PDF: 113
  • XML: 15
  • Total: 466
  • BibTeX: 10
  • EndNote: 7
Views and downloads (calculated since 09 Nov 2022)
Cumulative views and downloads (calculated since 09 Nov 2022)

Viewed (geographical distribution)

Total article views: 454 (including HTML, PDF, and XML) Thereof 454 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 25 Mar 2023
Download
Short summary
Wind turbine positions are often optimized to avoid wake losses. These losses depend on atmospheric conditions, such as the wind speed and direction. The typical optimization scheme involves discretizing the atmospheric inputs, then considering every possible set of these discretized inputs in every optimization iteration. This work presents stochastic gradient descent (SGD) as an alternative, which randomly samples the atmospheric conditions during every optimization iteration.