Session: 06-02: CSP Receivers and Reactors I

Paper Number: 170124

170124 - Multi-Variate Optimization of Generation 3 Csp Falling Particle Receivers Using High-Performance Computing 

Abstract: 

As the demand for efficient and alternative electricity and heat solutions grows, optimizing the design and operation of Concentrated Solar Power (CSP) systems becomes increasingly important. Generation 3, or particle based, CSP technology presents potential for improving CSP performance while also reducing overall system cost. This study presents a comprehensive numerical optimization of a Generation 3 CSP falling particle receiver (FPR), a critical technology for enhancing the efficiency and viability of Generation 3 solar thermal energy systems. 

The optimization process aims to determine the optimal configuration of the FPR for a given tower height that minimizes a pseudo-Levelized Cost of Heat (pLCOH), back cavity wall temperature, and side cavity wall temperature. The pLCOH is defined as the cost of the tower, skip-hoist, receiver, and heliostat field divided by the power absorbed by the particle working fluid over four representative days of the year (spring, summer, winter, and fall). The study parametrically varies several factors, including receiver power, receiver characteristic length, snout depth, cavity normalized offset in width and height, number of troughs, and horizontal acceptance angle, while keeping tower height, heliostat size, and curtain spacing fixed. High-performance computing resources at Sandia National Laboratories are leveraged to enable large parametric simulations in minimal clock time.  

CoPylot, the SolarPILOT Python API, is utilized to optimally establish a heliostat field layout for a fixed tower height and parametrically input receiver configuration. PySolTrace, the SolTrace Python API, is then employed to analyze the power absorbed by the heat transfer fluid and associated FPR surfaces on an hourly basis for each of the four days. This includes analyzing the rays incident on the snout and cavity surfaces. These rays are correlated to flux and temperature maps on each surface. The absorption of rays intersecting the particle curtain(s) is assessed using a curtain transmissivity function dependent on fall height. Each curtain surface is divided into ten segments to capture spatial variations in transmissivity. In addition to curtain transmissivity, advective and radiative losses are modeled using correlations established in the literature to capture an overall FPR efficiency.  

The optimization process employs multi-variate optimization for each parametric run, targeting pLCOH, back cavity wall temperature, and east/west cavity wall temperature. The top solutions from each iteration are used to generate new parametric input variables, spanning the range of the top solution set. This iterative process is repeated five times to refine the optimal solution subset. 

Initial findings of the study suggest a receiver with a 58 MWt rating, 8.6 m by 8.6 m aperture, 4 trough configuration, 160 degree acceptance angle, 2.5 m snout depth, and 1.25 normalized cavity width & height minimizes pLCOH and cavity wall temperatures for a 55 m (~180 ft) optical tower height. Modeling assumptions and inputs are under active improvement, and a full results set will be presented at the ASME ES Conference. Following the outcomes of this analysis, the optimal solution(s) will be further analyzed using Computational Fluid Dynamics (CFD) to capture the detailed physics of the FPR phenomena, providing valuable insights for the design and operation of advanced CSP systems.  

Presenting Author: Luke McLaughlin Sandia National Laboratories

Presenting Author Biography: Luke received his Ph.D. from the University of Wyoming in 2022 and has since served as a R&D Mechanical Engineer at Sandia National Laboratories. Luke's areas of research include thermal energy storage and concentrating solar thermal technologies.