2022-PVPMC-Reference-Cell-Correction-POSTER

Angular Response Correction Factors for Comparing PV Reference Cells and Thermopile Pyranometers Introduction PV reference cells and thermopile pyranometers have different spectral and angular responsivities PV performance analysts may be confused when comparing data from reference cells to pyranometers Discrepancies are often attributed to spectral effects while angular effects are under-appreciated In this work we calculate estimated angular response correction factors for comparing reference cells and pyranometers in exemplary systems Discussion Assume reference cell and pyranometer calibrated identically at normal incidence Considering only angular response, the reference cell will measure lower irradiance than the pyranometer This is not a spectral effect In winter months at high latitudes differences can reach 2-4 for single-axis tracking Differences correlate with weighted-average AOI High diffuse fraction also increases the discrepancy, because IAM diffuse averages over high AOI Reference cell angular response is similar to modules, while pyranometer angular response is flat The correction factors presented here can be used as a first estimate to explain observed behavior Michael Gostein, Atonometrics, Austin, TX Simulation Method Calculations are performed using Sandia’s PVLIB in MATLAB We use Typical Meteorological Year TMY data for 3 sites from the National Solar Radiation Database NSRDB The calculation is performed for three different geometries fixed optimal* tilt, single-axis tracking, and horizontal GHI sensor orientation Optimal* tilt taken as 0.8 x latitude simplification For each data set and geometry, we use DNI, DHI, and albedo from the TMY data to calculate POA The calculation uses the Perez model for diffuse irradiance translation and the isotropic model for ground-reflected light For the pyranometer POA sensor response is calculated assuming ideal cosine response incidence angle modifier IAM 1 For the reference cell POA sensor response is calculated using an ideal IAM from the Martin-Ruiz model, including weighted-average IAM values for diffuse and ground-reflected light 𝑃𝑃𝑃𝑃 𝑃𝑃 𝐷𝐷𝐷𝐷𝐷𝐷 � cos 𝑎𝑎 𝑎𝑎𝑎𝑎 � 𝐷𝐷 𝑃𝑃𝐼𝐼 𝑎𝑎 𝑎𝑎𝑎𝑎 𝑓𝑓 𝑃𝑃 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝐷𝐷𝐷𝐷𝐷𝐷 , 𝐷𝐷𝐷𝐷 𝐷𝐷 � 𝐷𝐷 𝑃𝑃𝐼𝐼 𝑑𝑑 𝑑𝑑 𝑑𝑑𝑑𝑑 𝐺𝐺𝐷𝐷 𝐷𝐷 � 𝛼𝛼 � 1 − 𝑐𝑐 𝑎𝑎𝑐𝑐𝑐𝑐 2 � 𝐷𝐷 𝑃𝑃𝐼𝐼 𝑔𝑔 𝑃𝑃 𝑔𝑔 𝑔𝑔𝑔𝑔 𝑑𝑑 Incidence Angle Modifier IAM Simulation Results Acknowledgements This material is based upon work supported by the U.S. Department of Energy’s Solar Energy Technologies Office under Award Number DE- SC0020831. Correction Factor 𝐶𝐶𝐶𝐶 � � 𝑃𝑃𝑃𝑃 𝑃𝑃 𝑝𝑝 𝑝𝑝 𝑃𝑃 𝑝𝑝 𝑔𝑔 𝑔𝑔 𝑝𝑝 𝑃𝑃𝑝𝑝 𝑃𝑃𝑃𝑃 � 𝑃𝑃𝑃𝑃 𝑃𝑃 𝑃𝑃 𝑃𝑃 𝑑𝑑 𝑟𝑟 𝑃𝑃𝑟𝑟 𝑟𝑟 Multiply reference cell POA by CF to get estimate for pyranometer POA Diffuse Fraction 𝑘𝑘 𝑑𝑑 𝐷𝐷𝐷𝐷 𝐷𝐷 / 𝐺𝐺 𝐷𝐷𝐷𝐷 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 1.024 1.026 1.028 1.030 1.032 1.034 Correction Factor POA, Fixed Optimal* Tilt Austin, TX Boston, MA Bakersfield, CA Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 30 35 40 Weighted Average AOI degrees Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 0.200 0.300 0.400 Diffuse Fraction Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 1.010 1.020 1.030 1.040 1.050 Correction Factor POA, Single-Axis Tracking Austin, TX Boston, MA Bakersfield, CA Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 0 50 Weighted Average AOI degrees Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 0.200 0.300 0.400 Diffuse Fraction Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 1.020 1.040 1.060 1.080 1.100 1.120 Correction Factor GHI, Sensors Horizontal Austin, TX Boston, MA Bakersfield, CA Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 40 60 Weighted Average AOI degrees Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 0.200 0.300 0.400 Diffuse Fraction 0 10 20 30 40 50 60 70 80 90 AOI degrees 0.00 0.20 0.40 0.60 0.80 1.00 Incidence Angle Modifier Reference Cell ideal Thermopile Pyranometer ideal