Optical efficiency measurement of receivers for parabolic trough solar thermal power plants in solar simulators

Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2018; Aachen 1 Online-Ressource (xii, 181 Seiten) : Illustrationen (2018). = Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2018
Two linear focus solar simulator test benches were built for the calorimetric measurement of the optical efficiency of parabolic trough receivers. The test benches, called ElliRec and OptiRec, consist of mirrors of elliptical cylinder geometry, flat end mirrors, and metal-Halide lamps. Water at ambient temperature is directed on a spiralling path at the inner surface of the absorber. Absorbed power of the receiver is calculated from temperature increase and flow rate of the water and amounts to 6.5 kW or 4.8 kW and is compared to that by a reference to yield the relative optical efficiency. Expanded uncertainty at k = 2 of optical efficiency in the solar simulator of a receiver relative to that of a reference receiver is 0.6 % for the ElliRec and 0.34 % for the OptiRec; expanded uncertainty at k = 2 of the absolute power measurement is 1.2 % for ElliRec and 1.6 % for OptiRec. Several aspects of the transfer of laboratory measurement to the application in the field are investigated. Assuming idealized spectra for absorber and glass systematic deviations of optical efficiency between solar simulator and field of typically up to 4 % were calculated. For comparison of relative measurements, as typically performed, deviations between solar simulator and field are < 0.2 % for measured spectra for both test benches. An experimental comparison of the ElliRec solar simulator with two spectrophotometric test benches showed a disagreement of < 0.6 % standard deviation for five out of six compared receivers. Differences between the ElliRec and OptiRec for seven receivers were measured to < 1.1 %. A 1-d model is used to calculate changes from room temperature to operating temperature yielding corrections for the hot receiver of up to 0.7 % depending on bellow design. Detailed 3-d-ray-tracing models yield a spatial non-uniformity in longitudinal direction calculated for - 2.03 m to + 2.03 m is 1.6 % for ElliRec and 2.3 % for OptiRec. Neglecting source volume and slope deviations of the mirror it is shown, that in contrast to the field, in the solar simulator angle and position of incidence on the absorber are linked. The investigation of the impact of focus deviation based on the Fresnel Equations shows a decreasing effective absorptance with increasing collector angle of incidence and mean focus deviation. For example, assuming Gaussian radiation distribution of focus deviation of half of the absorber radius and a collector angle of incidence of 0° results in an additional loss in effective absorptance of 1 % due to reduced absorptance. The optical efficiency measurement in the linear focus solar simulator is a fast measurement with high repeatability. The influence of the effects in transfer to the field are small enough to consider the measurement a meaningful assessment of the quality of a receiver and to resolve differences between products. Due to the simple optical geometry the method has the potential to be easily reproduced by other laboratories.
Published by Aachen