Hourly day-ahead wind power forecasting at two wind farms in northeast Brazil using WRF model. (1st September 2021)

• Record Type: Journal Article
• Authors: Duarte Jacondino, William; Nascimento, Ana Lucia da Silva; Calvetti, Leonardo; Fisch, Gilberto; Augustus Assis Beneti, Cesar; da Paz, Sheila Radman
• Is Part Of: Energy. Volume 230(2021)
• Publication Date: 2021-09-01
• Subjects: WRF -- Onshore -- Forecast -- Wind power -- Brazil; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2021.120841
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: Wind energy is rapidly growing industry in Brazil. Wind speed forecasting is necessary in the planning, controlling, and monitoring for the reliable and efficient operation of the wind power systems. Thus, this study focuses on the impact of different physics parameterization in forecasting wind speed in two onshore wind farms using the Weather and Research Forecasting (WRF) model. The wind farms are located in Parazinho, in the northeast of Brazil, a region with high wind resource. Hindcasts are performed for a high (i.e., July 2017) and low (i.e., April 2017) wind speed regimes using different forecast lead-times (i.e., 24–48 h). The best performing setup consists of Thompson microphysics, Bougeault-Lacarrere PBL, Betts-Miller cumulus, New Goddard Longwave/Shortwave radiation, and Pleim-Xiu Land Surface schemes. Our findings also suggest that the model forecast setting with the TKE closure scheme, namely BouLac, performed better than that setting with first-order closure ACM2. The best mean monthly error (MAE) obtained is 1.1 m s −1 for wind and 12.6% for wind power. Highlights: Hourly day-ahead wind speed forecast at 95 m AGL for two wind farms in northeast Brazil using WRF model is examined. Several sensitivity experiments using different physical option were perfomed. The local closure PBL BouLAC scheme had the greatest impact in wind speed forecast. The results can provide valuable insights for wind power forecasting investigation.