Identifying interaction patterns to speed-up the AEP evaluation in large offshore wind farms with uniformly distributed turbines and exclusion zones

The evaluation of annual energy production (AEP) is the key step for estimating wind farm income and assessing project viability through metrics like the levelized cost of energy. However, calculating AEP is highly time-intensive, particularly in wind farm layout optimization algorithms, which must evaluate AEP for numerous potential layouts. Even with simplified models, evaluating a single layout for medium or large farms can take several seconds, limiting the exploration of feasible solutions. In uniformly distributed offshore wind farms, repeated interaction patterns between turbines lead to redundant calculations of wind speed deficits in traditional methods. This work addresses the issue by identifying geometric interaction patterns and their frequency within a given layout. By leveraging these patterns, the method drastically reduces AEP computation times, achieving up to a 3000-fold improvement for a 16 × 16 turbine farm. The approach is especially effective for uniformly spaced, rhomboid-shaped offshore wind farms, reducing computation times to less than 20 ms regardless of farm size. For irregularly shaped farms or those with exclusion zones, the method still achieves significant time reductions, performing over 20 times faster than traditional methods. This innovation enables deeper exploration of the solution space in optimization processes, greatly enhancing efficiency in wind farm design. • A completely new formulation to aggregate power deficits due to wakes. • By far, the fastest method to evaluate the annual energy production of wind farms. • To be applied in offshore wind farms with uniformly distributed turbines. • Valid for rhomboidal layouts, arbitrary contours and exclusion zones. • Computation times in the order of 10 ms almost regardless of the wind farm size.