Advances in Wind Farm Layout Optimization

Dear Colleagues,

Wind farm layout optimization (WFLO) is a vivid field of research dealing with the surprisingly difficult question of how to optimally arrange a set of wind turbines inside a local area (wind farm). As soon as the problem consists of placing two or more turbines in the farm, these turbines possibly represent wake effects causing wind obstacles for one another, dependent on the wind direction. In mathematics, this problem is typically seen as a constraint optimization (i.e., maximization or minimization) task. Researchers maximize power output or output efficiency or minimize some type of cost function. However, as the problem is so complex, an overall algorithm finding a global optimum in general settings, furthermore at low-computation time consumption, has not been found yet. Approaches to solving the problem are usually classified a) according to the wake model used, b) according to the class of optimization approaches (gradient-based approaches and gradient-free algorithms), or c) according to the target function class. Many contributions so far make assumptions, e.g., constant wind speeds across the entire area, interpolating the wind speed measurement point raster enough to assume that wind speed is differentiable, defining possible turbine locations over a rather coarse raster (discrete computational domain), thus strongly reducing the possible locations, or others. While many of these assumptions are fair, they make a comparison of these approaches difficult.

This Special Issue aims at providing original research in WFLO contributions and making it as comparable as possible. Authors are invited (but not obligated) to use the free software package “wflo”, available for the software R from the CRAN repository (see https://CRAN.R-project.org/package=wflo). wflo provides a quality data set as well as a standardized workflow and tool chain for WFLO researchers to focus on their actual contribution: the optimization approach. It also serves as a unified benchmark which allows for comparison of approaches across the entire WFLO research branch.

Guest Editors

Prof. Dr. Georg Stadtmann

Dr. Carsten Croonenbroeck