Measurement Accuracy of Twin-Sensor Rotating Shadowband Irradiometers (RSI)

Kurzfassung

The annual amount and temporal distribution of solar irradiation is a key parameter for a proper design of solar power applications. Project developers of solar power plants therefore need to procure extensively irradiation data for their projected sites. This is particularly crucial for Concentrating Solar Power (CSP) plants using direct solar beam irradiation. Reasons are a higher spatial and temporal variation of direct solar beam irradiation compared to global irradiation; moreover, it requires a higher effort and thus higher costs to gather accurate good-quality direct beam data and subsequently direct solar beam data is even less available for suitable regions. Best practice is to combine actual and accurate measurements on ground with historical data derived from satellites (avalable yet since the early 1980´s) to get a valid and reliable data set. Typically used high-precision instruments like pyrheliometers and pyranometers as specified in ISO norms are severly affected by soiling. They are often not appropriate for the use at remote locations (where many CSP plants are projected to be built) because they need daily maintenance due to their higher affinity for soiling, require a higher power supply and cause higher costs. Instead Rotating Shadowbank Irradiometers (RSI) are wídely used and show significant advantages over the typical precision equipment under measurement conditions at remote sites due to low power requirement, their robustness and low sensibiltiy to soiling. RSIs typically use photodiodes as irradiance sensors with immediate time response instead of thermopiles. This enables the determination of global but also diffuse and direct irradiance with one sensing element by a brief drop of the signal during the rotation of a shadowbank over the sensor. Although the primordial accuracy of photodiodes is mediocre compared to highly precise sensors, their direct beam response usually remains within 2 to 3% RMSD compred to precise sensors when functional corrections and proper calibration are applied on their systematic deviations. Main causes for the systematic deviations are a limited spectral sensitivity and temperature effects; potentially insufficient calibration at the manufacturer or a drift of the sensitivity may increase them. However, the detailed dependencies vary among individual photodiodes and ened to be adjusted individually for each sensor. In a new updated instrument version. the RSI is assembled with a twin sensor of two independent photodiodes for enhanced performance and accuracy.