RW3B.4.pdf Light, Energy and the Environment © OSA 2014 Improved Tubular Receivers for Point-focus Concentrators John Pye¹, Graham Hughes², José Zapata¹, Joe Coventry¹, Charles-Alexis Asselineau¹, Ehsan Abbasi¹, Martin Kaufer¹, and Felix Venn¹ ¹ Research School of Engineering, Australian National University, Canberra, Australia ² Research School of Earth Sciences, Australian National University, Canberra, Australia Abstract: Optics, thermal emission, convection and internal flow are treated in a unified model for a tubular cavity receiver on a paraboloidal dish concentrator, and a new design is presented that shows a 40% reduction in receiver losses at the design point, compared the previous state-of-the-art receiver. Concepts to extend this work further through bladed configurations and through the use of active airflow control are also presented. OCIS codes: (350.6050) Solar energy; (220.4298) Non-imaging optics; (220.1770) Concentrators. 1. Overview Point-focus solar thermal concentrators, whether they are paraboloidal dish concentrators or heliostat-and-central-tower systems, depend on high-efficiency receivers for their cost-effective operation. High efficiency depends on losses from a cascade of different loss mechanisms being minimised, namely: blocking and shading losses, reflector optical losses, atmospheric attenuation losses, receiver spillage losses, receiver reflection losses, thermal emission losses, convection losses, conduction losses. To these first-law effects, we can add heat-transfer irreversibilities and flow irreversibilities, if a second-law or exergetic efficiency is desired. Important trade-offs are at play in these receivers: a smaller aperture reduces radiative and convective losses, but increases spillage losses, for example, and a cavity geometry will improve light-trapping but will also increase the thermal emission losses on an aperture basis. In this paper, we present the methodology and results to date from a study to analysis and improve the performance of a tubular cavity receiver for use in a dish concentrator. We then discuss how these concepts can be applied to central tower receivers to similarly improve their performance. 2. Tubular receiver model An axisymmetric receiver is considered, with tubes wound helically around the internal receiver surfaces. Optically, we allow for adiabatic regions within the receiver, where no tubes are present and the wall of the receiver has total losses (locally) equalling the total absorbed heat. Tubes are assumed to be covered with a high-absorptivity painted coating. Incident radiative flux from the sun is modelled in a Monte Carlo ray tracing model that incorporates 'sun shape', mirror surface errors and cavity receiver shape. A flux profile within the receiver is determined for a number of axisymmetric bins of arbitrary size. Thermal radiation losses are determined by a radiosity analysis, again using the same set of arbitrary bins, by single-step Monte Carlo ray tracing of diffusely-emitted rays uniformly distributed over each bin surface in turn. The ray tracing permits calculation of view factors, then a simple radiosity analysis allows calculation of the thermal radiation losses as a function of surface temperature for each bin, which is normally considered to be covered with tubes containing the receiver working fluid. Convection losses are calculated by a computational fluid dynamics model, currently treating only natural convection losses. Typically the cavity receiver shows a 'stratified' region where hot air is trapped in a region above the downward-pointing aperture; the size of this stratified region is key in determining the total losses by natural convection. The results from a full 3D simulation are converted to equivalent axisymmetric convection losses for the integrated model. Internal flow in the tubes of the receiver is modelled in a one-dimensional hydrodynamic model with steam as the working fluid, and two-phase pressure drops and convection coefficients calculated using published correlations. Convection coefficients between the working fluid and the tube wall are typically high, hence working fluid temperature profile is a strong driver for the thermal emission losses of the cavity, and hence on the optimal shape for the cavity. The integration of the above phenomena requires iteration to determine the tube external temperature profile that balances the combined modes of heat transfer from incident solar radiation as well as internal and external heat transfer. A stochastic optimisation process has been developed to screen out best-performing geometries as raytracing progresses with increasing precision and increasing numbers of rays. Finally, high-precision simulations are used to rank short-listed geometries [1].