Models for characterizing Darrieus turbines and synthesizing wind speed scenarios

Ashwath Mukesh Bhat
DOI: https://doi.org/10.48550/arXiv.2112.01661
2021-12-03
Abstract:With the evolving global energy dynamics, new and innovative ways to integrate renewable energy technologies into residential and commercial complexes are being adopted. Techno-economic models such as NREL-SAM and HOMER are handy for evaluating such integration of renewables. However, the unavailability of libraries for vertical axis wind turbines (VAWT) and wind speed data limitations have been identified in the two platforms (NREL-SAM and HOMER). Accordingly, this project looks into developing models for evaluating power curve tables for Darrieus turbines and synthesizing wind speed scenarios. Analytical and numerical models for Darrieus turbines have been presented, and the ability to evaluate power curve tables from both models has been demonstrated. Double-Multiple Streamtube model was implemented for the analytical modelling, and 2-D CFD simulations on OpenFOAM were developed for the numerical modelling. The analytical model was found to fail for small-sized turbines, likely due to poor lift and drag coefficient values at high local angles of attack, but matched sufficiently with experimental results for large-sized turbines. Numerical simulations for Darrieus turbines were of two types - torque evaluation at constant wind speed and angular velocity, and starting characteristics for constant wind speed. Auto-regressive moving average models were implemented to characterize historical wind speed measurements, thus synthesizing wind speed scenarios. For this, the wind speeds or wind speed residuals (following detrending) were first normalized using the cumulative distribution functions, following which an ARMA model was fit.
Fluid Dynamics
What problem does this paper attempt to address?