Apparatus and method for predicting solar irradiance variation

Apparatus comprising:

• a solar irradiance predictor processor configured to process a sequence of images, the irradiance predictor processor comprising:
• a cloud motion predictor configured to predict over a time horizon motion of at least one cloud;
• and
• an event predictor configured to predict irradiance variation by back-projecting a solar disk image based on an average speed vector computed from the predicted motion of said at least one cloud.

A power generation system comprising:

• a plurality of power generators responsive to solar irradiance, the plurality of power generators disposed over respective positions on a field;
• at least one image acquisition device arranged to acquire a sequence of images;
• an irradiance predictor processor coupled to said at least one image acquisition device to process the sequence of images, the irradiance predictor processor comprising:
• a cloud classifier module configured to classify respective pixels of an image of at least one cloud to indicate a solar irradiance-passing characteristic of at least a portion of said at least one cloud;
• a cloud motion predictor configured to predict over a time horizon motion of said at least one cloud;
• an event predictor configured to predict over the time horizon occurrence of a solar obscuration event, wherein the prediction of the solar obscuration event is based on the predicted motion of said at least one cloud, wherein the event predictor includes an irradiance variation prediction for the obscuration event based on the solar irradiance-passing characteristic of said at least one cloud;
• and
• a controller responsive to the event predictor to adjust a control strategy for at least one component and/or subsystem of the power generation system based on the predicted occurrence of the obscuration event.

A method for predicting over a time horizon solar irradiance variation, the method comprising:

• processing a sequence of images, the processing comprising:
• classifying respective pixels of an image of said at least one cloud to indicate a solar irradiance-passing characteristic of at least a portion of said at least one cloud;
• predicting over the time horizon motion of said at least one cloud;
• and
• predicting over the time horizon an occurrence of a solar obscuration event, wherein the prediction of the solar obscuration event is based on the predicted motion of said at least one cloud, wherein the prediction of the solar obscuration event includes an irradiance variation prediction for the obscuration event based on the solar irradiance-passing characteristic of said at least one cloud.