Improving the Accuracy of Satellite-to-Irradiance Models

With improvements in photovoltaic performance and price drops, solar energy will soon
become cheaper than many traditional sources. Irradiance-the amount of solar energy
reaching the Earth-is key In feasibility studies of solar energy systems. Over 20 years of
archival satellite imagery enable us to generate long-term data series to better understand
climatic averages of irradiance. Turquoise’s satellite-to-irradiance model must be optimized to
increase its accuracy and performance. Two categories of parameters must be optimized.
Physical model parameters-atmospheric turbidity and albedo, direct and diffuse components
of radiation-must be estimated for the model to match observations. Parameters of the
software-the optimal range of images for cloud index calculations and parallel proceSSing
parameters-must be tuned to improve efficiency.
Increased accuracy will enable Canadian investors and governments to make better
decisions, increase stakeholder trust in the satellite-based method as complementary to
ground measurements, increase the export potential of Turquoise and help accelerate the
adoption of solar energy in Canada.

Faculty Supervisor:

Dominique Orban

Student:

Partner:

Turquoise Technology Solutions Inc

Discipline:

Engineering

Sector:

University:

École Polytechnique de Montréal

Program:

Accelerate