Overview of the proposed methodology.
Spectral analysis of the original data for the coastal, mid-island, and offshore location. In parenthesis the number of years of data included.
A sine-curve fit to the monthly mean wind speed of the given number of years of data, separately for each location.
The distribution of the five day types at the three locations based on the original wind speed data.
The reference days fitted to the empirical hourly mean wind speeds of the days for each day type.
Distibution of day types as a function of daily mean wind speed, per location. The day types, numbered one through five, correspond to the day types “No peak,” “00:00-06:00,” “06:00-12:00,” “12:00-18:00,” “18:00-24:00,” respectively.
Spectral analysis comparison of original data with the reference method and the new method: the spectral information of the original data is reasonably well maintained by both methods for synthetic wind speed data.
The distribution of the different day types in the test year of original data, the data generated by the new method and the data from the reference method.
Measured wind speed data information.
Seasonal mean wind speeds in m/s per location.
Lag-terms and standard deviations of the determined AR-models for variation of the daily mean wind speed from the reference value given by location.
Lag-terms and standard deviations of the AR-models for variation of hourly mean wind speed from the reference value given by day type and location.
Seasonal variation factors for the test year, in relation to reference values calculated from the available years of historical data.
Quantitative comparison of statistical characteristics of the test year data set, the synthetic data generated by the reference method and the new method. λ and κ refer to the Weibull scale and shape parameters respectively.
Number of “off-days” in comparison to the test year for the average results of the scenarios from the new method and the results from the reference method for all three locations.
Article metrics loading...
Full text loading...