Wind farm layout optimization is crucial for maximizing energy production in renewable energy systems. This video introduces a novel layout optimization method for offshore wind farms that aims to minimize power deficits caused by the wake effect, without restrictions on the number of turbines, their shape, or the extent of the concession area.

The core of this algorithm is a gradient descent method, which progressively and randomly incorporates new turbines into the concession area. As the optimization process unfolds, the turbines expand outward, seeking areas with less perturbation while pushing previous ones aside. To avoid local maxima at the end of the optimization process, the algorithm suppresses locations that contribute the most to potential power deficits.

A comprehensive map of potential for the entire area is then generated, and a greedy algorithm places new turbines to finalize the layout. Remarkably, this entire process is completed in just 25 seconds.

To significantly accelerate the search process and the creation of the potential map, a validated simplification is introduced: for turbines affected by multiple wakes, the power output is calculated using a linear aggregation of power deficits rather than the traditional linear or quadratic aggregation of speed deficits. This approach draws an analogy between power deficit and repulsive non-isotropic electrostatic potential energy, allowing the utilization of conservative field properties.

As a result, this method is 20 times faster than any other layout optimization algorithm currently available in the literature. Additionally, irregular concession areas are effectively managed using Stokes’ theorem to identify outer points.