Computer-implemented system and method for generating a risk-adjusted probabilistic forecast of renewable power production for a fleet

摘要

Probabilistic forecasts of the expected power production of renewable power sources, such as solar and wind, are generally provided with a degree of uncertainty. The expected power production for a fleet can be projected as a time series of power production estimates over a time period ahead of the current time. The uncertainty of each power production estimate can be combined with the costs and risks associated with power generation forecasting errors, and displayed or visually graphed as a single, deterministic result to assist power grid operators (or planners) in deciding whether to rely on the renewable power source.

主权项

1. A computer-implemented system for generating a risk-adjusted probabilistic forecast of renewable power production for a fleet, comprising:
a database comprising:
irradiance observations measured at regular intervals for the location of a renewable power production plant; a computer comprising a processor and memory within which code for execution by the processor is stored, comprising:
a clearness index module configured to build a set of clearness indexes comprising the irradiance observations divided by clear sky irradiance, and further configured to order the set of clearness indexes into a time series successively spaced at input time intervals;a clearness index correlation coefficient module configured to find a clearness correlation coefficient between the location of the renewable power production plant and the location of one other renewable power production plant in the same power production fleet as a function of physical distance between the two plants in proportion to overhead cloud speed;a weighted clearness index module configured to weight the set of clearness indexes by the clearness correlation coefficient, and further configured to order the set of weighted clearness indexes into a time series at the input time intervals;a power estimation module configured to estimate expected power production statistics from the time series of the clearness indexes and the time series of the weighted clearness indexes;a power forecasting module configured to forecast expected power production as a function of the expected power production statistics and overall power ratings of the two renewable power production plants, and further configured to apply a time lag correlation coefficient to the expected power production to build a time series at output time intervals;an error module configured to determine error associated with the expected power production for at least one of the output time intervals of the time series of the expected power production, and to identify risk associated with the error; and a risk-adjustment module configured to adjust the expected power production based on the risk.