Abstract
By investigating the long-term observations at Atmospheric Radiation Measurement (ARM) Southern Great Plains (SGP), we find that the routinely used Beer-Bouguer-Lambert law and the models that empirically separate direct normal irradiance (DNI) from measurements of global horizontal irradiance (GHI) have dramatic and unexpected bias in computing cloudy-sky DNI. This bias has led to tremendous uncertainty in estimating the electricity generation by solar energy conversion systems. To effectively reduce the bias, this study proposes a physical solution of all-sky DNI that computes solar radiation in the infinite-narrow beam along the sun direction and the scattered radiation falls within the circumsolar region. In sharp contrast with the other DNI models, this method uses a finite-surface integration algorithm that computes solar radiation in differential solid angles and efficiently infers its contribution to a surface perpendicular to the sun direction. The new model substantially reduces the uncertainty in DNI by a factor of 2–7.
Original language | American English |
---|---|
Article number | 100893 |
Number of pages | 18 |
Journal | iScience |
Volume | 23 |
Issue number | 3 |
DOIs | |
State | Published - 27 Mar 2020 |
Bibliographical note
Publisher Copyright:© 2020 The Author(s)
NREL Publication Number
- NREL/JA-5D00-74861
Keywords
- direct normal irradiance
- DNI
- FARMS
- FARMS-NIT
- Fast All-sky Radiation Model for Solar applications
- GHI
- global horizontal irradiance