A detailed 1D model of a parabolic trough solar receiver with a double-validation approach

Abstract

This study proposes a detailed Parabolic Trough Collector (PTC) 1D model offering precise optical and thermal analysis. Our motivations were accuracy while ensuring fast and inexpensive computation. The proposed model differs from existing models by considering a more extensive set of possible optical and thermal losses, which to our knowledge have not previously been grouped together in a single model. Testing of the model based on experiments carried out by Sandia National Laboratories showed near-similar accuracy to much more complex models, with a mean standard deviation of 0.891% and 5.99 W/m2 for thermal efficiency and thermal losses respectively. These tests were carried out using two types of selective coating: cermet and black chrome, with two operating modes, namely vacuum and pressure annulus. To expand its scope of validity, the developed model was also tested using data from a selection of PTCs of commercial lengths (70, 106, 143, and 150 m) under real operating conditions. Additional validation of the model was carried out with available results from different heat transfer fluids (HTF), including two nanofluids. We highlighted that the deviation from the results of the various PTCs, although very satisfying in general, becomes more critical with increasing length and temperature, reaching a maximum outlet temperature deviation of around 2.14 °C. On the other hand, when switching from water to the nanofluid Water/CuO (5 %), the results show an increase in outlet fluid temperature from 325.91 to 334.73 °C. Experimental validation was also carried out using a 25 kW PTC test loop located at Ben Guerir. The average relative error perceived between simulation results and experimental measurements is estimated at 0.99 % for the steady-state test (based on the HTF temperature at the PTC outlet). An additional validation based on the external temperature of the glass envelope was also considered. This approach, which is rarely used in the literature, showed a relative error of around 4.39 % between simulations and experimental data, which remains within the uncertainty range of the thermal camera used.