Stochastic modeling of lift and drag dynamics under turbulent wind inflow conditions
Article first published online: 20 JAN 2014
Copyright © 2014 John Wiley & Sons, Ltd.
How to Cite
Luhur, M. R., Peinke, J., Schneemann, J. and Wächter, M. (2014), Stochastic modeling of lift and drag dynamics under turbulent wind inflow conditions. Wind Energ.. doi: 10.1002/we.1699
- Article first published online: 20 JAN 2014
- Manuscript Accepted: 22 NOV 2013
- Manuscript Revised: 11 OCT 2013
- Manuscript Received: 28 APR 2013
- wind tunnel measurements;
- fractal square grid;
- turbulent inflow;
- stochastic model;
- lift dynamics;
- drag dynamics
We present a new approach to model the complex dynamics of aerodynamic forces on an airfoil in turbulent inflow conditions. Our ansatz is based on stochastic differential equations and aims at replacing traditional look-up table methods used in wind turbine simulation systems by the effective response dynamics of lift and drag forces. The parameters of the model are derived directly from empirical data. Measurements were taken in the closed loop wind tunnel of the University of Oldenburg for an airfoil FX 79-W-151A. The turbulent inflow was generated using a fractal square grid as it is possible to generate in this way wind speed fluctuations with similar statistics as observed in nature. Forces were measured using two strain gauge force sensors at two end points of the vertically installed airfoil. The modeling is performed by applying a stochastic approach on the measured data. By estimating the first two Kramers–Moyal coefficients, a first-order stochastic differential equation called the Langevin equation is obtained. The stochastic model achieved through this approach is extended to account for oscillation effects contained in lift and drag dynamics that probably stem from unsteady aerodynamic effects. The results are optimized by applying a χ2 test on the probability density functions (PDFs) of model and measurements. With the knowledge of the Langevin equation, synthetic time series are generated. Their stationary PDFs as well as conditional PDFs show good agreement with the actual measurements. A comparison of classical averaging and the stochastic approach shows that stochastic analysis achieves additional insight into the local dynamics of lift and drag forces. Copyright © 2014 John Wiley & Sons, Ltd.