Stochastic simulation and forecasting of hourly average wind speed sequences in Jamaica

Journal Article · Tue Jan 01 04:00:00 EST 1991 · Solar Energy (Journal of Solar Energy Science and Engineering); (USA)

DOI:https://doi.org/10.1016/0038-092X(91)90101-2· OSTI ID:5897292

Daniel, A R; Chen, A A ^[1]

---

1. Univ. of the West Indies, Kingston (Jamaica)

Stochastic simulation and forecast models of hourly average wind speeds are presented in this paper. The models take into account several basic features of wind speed data including autocorrelation, non-Gaussian distribution, and diurnal nonstationarity. The positive correlation between consecutive wind speed observations is taken into account by fitting an autoregressive moving average (ARMA) process to wind speed data which were transformed to make their distribution approximately Gaussian and standardized to remove diurnal nonstationarity. Based on three years of data, separate models were obtained for the months of June, July, and August. Tests were performed to verify the adequacy of the models and comparisons were made between generated and real series and between forecasted and real series to check if the wind speed behavior was faithfully reproduced. Good results were obtained.

• OSTI ID: 5897292
• Journal Information: Solar Energy (Journal of Solar Energy Science and Engineering); (USA), Journal Name: Solar Energy (Journal of Solar Energy Science and Engineering); (USA) Vol. 46:1; ISSN SRENA; ISSN 0038-092X
• Country of Publication: United States
• Language: English

Related Subjects

17 WIND ENERGY
170100* -- Wind Energy-- Resources & Availability (Climatology)
AIR FLOW
CORRELATIONS
DAILY VARIATIONS
DATA
DEVELOPING COUNTRIES
FLOW RATE
FLUID FLOW
FORECASTING
GAS FLOW
GREATER ANTILLES
HOURLY VARIATIONS
INFORMATION
ISLANDS
JAMAICA
MACHINERY
MATHEMATICAL MODELS
MONITORING
RESOURCE POTENTIAL
SIMULATION
TESTING
TURBINES
TURBOMACHINERY
VALIDATION
VARIATIONS
WEST INDIES
WIND
WIND TURBINES