Assessment of the optimal curvature radius of linear Fresnel primary mirrors

Abstract

The curvature radius design of linear Fresnel primary mirrors relates to two possible configurations: non- uniform, each mirror has a particular radius value, and uniform, all mirrors have the same value. Different non-uniform designs have been proposed but never compared to understand their differences or whether these differences exist. Moreover, it is not clear how much performance loss would result from using a simpler uniform curvature design, particularly concerning the tolerance to optical errors. This study aims to fill these gaps by comparing the annual average optical efficiency and acceptance angle of different curvature radius design propositions for two study cases. The results reveal that among the non-uniform designs the sun reference is the fair option since it offers annual efficiency with deviations less than 0.4% to the maximum but also show the higher tolerance to optical errors, outperforming other options in terms of acceptance angle. In contrast, uniform curvature designs, while simpler in terms of manufacturing, operation, and maintenance, exhibit lower overall optical performance. Compared to the optimum non-uniform design, the best uniform configuration shows a reduction in annual efficiency by up to 1.9% and a higher drop in acceptance, up to 15%. Notably, these opti- mum designs (uniform and non-uniform) allow for a simplified determination of mirror radii without the need for time-intensive performance evaluations such as ray-tracing. Nevertheless, further exploitation of these results in design optimization requires a due accounting of the related manufacturing and maintenance costs impacting the foreseeable specific costs of each design method.