On optimum designs of linear Fresnel solar collectors

Abstract

This paper investigates the optimal design of linear Fresnel collectors using a two-objective optimization strategy aimed at maximizing annual energy efficiency while minimizing the specific direct cost of the solar field. The analysis then compares Pareto- optimal solutions for non-uniform designs (where geometric parameters such as width, shift, and radius vary) with solutions for variable radius configurations (constant width and shift). Additionally, the study evaluates whether the curvature radius can be effectively decoupled from other opti- mization parameters. The results in this paper show that non-uniform designs generally provide greater cost- effectiveness, particularly in certain regions of the objective space. While some areas show negligible differ- ences between configurations, others distinctly favor non-uniform designs, although the differences are not significant and are reduced as the number of mirrors in the primary field increases. Furthermore, an analysis through dimensionless parameters highlights clear trends among the Pareto solutions. The investigation also reveals that predefined curvature radius criteria, such as the sun reference or Iparraguirre’s model, yield out- comes nearly identical to optimization-based solutions for both uniform and non-uniform configurations. Consequently, curvature radius can be decoupled from other design parameters, significantly reducing compu- tational complexity without compromising performance. The insights presented are valuable for advancing cost- effective designs of linear Fresnel collectors. Nevertheless, the conclusions rely on a simplified cost model, highlighting the need for more comprehensive cost modeling. However, developing such detailed models re- mains challenging due to limited operational data stemming from the low commercial maturity of linear Fresnel technology.