Abstract
Probabilistic methods are commonly adopted to estimate both ultimate and fatigue loads in wind turbines, and multiple papers have presented a variety of approaches. In contrast, system-level, integrated design optimization methods for wind turbines have so far relied on deterministic methods to estimate loads and deflections. This work aims at addressing this gap by adopting importance sampling to estimate ultimate blade deflection for use within a rotor design optimization. A comparison between the results of this approach against traditional approaches using deterministic load evaluations show that the probabilistic approach is more conservative and that the solutions found with deterministic approaches violate the probabilistic constraint on tip deflection between 13% and 17%. This highlights the importance of accurate load and deflection estimates and demonstrates that probabilistic approaches can be successfully integrated within a wind turbine design process despite the higher computational costs.
Original language | American English |
---|---|
Number of pages | 14 |
State | Published - 2020 |
Event | AIAA SciTech Forum and Exposition - Orlando, Florida Duration: 6 Jan 2020 → 10 Jan 2020 |
Conference
Conference | AIAA SciTech Forum and Exposition |
---|---|
City | Orlando, Florida |
Period | 6/01/20 → 10/01/20 |
NREL Publication Number
- NREL/CP-2C00-75555
Keywords
- importance sampling
- wind turbine optimization