International Journal of Heat and Mass Transfer, Vol.115, 682-693, 2017
Forecast of thermal harvesting performance under multi-parameter interaction with response surface methodology
Benefitting from transformation optics, the design of thermal harvesting devices by manipulating heat flux has aroused considerable interest owing to its potential application in heat storage and the fabrication of highly efficient and reliable solar cells and thermal collectorS. However, studies on harvesting performance under different conditions achieved by the use of different design parameters are lacking. In this study, we focus on forecasting the harvesting performances of 2D thermal concentrating schemes with variational design parameters, including the ratios of conductivities and radii, wedge angles, and temperature differences. In order to evaluate the multi-parametric harvesting performance, an expression of efficient harvesting entropy was proposed based on the entropy generation distributions of thermal harvesting schemes. Response surface methodology (RSM) was employed to achieve the regression model of efficient harvesting entropy under different conditions. The accuracy of the forecast was verified by analysis of variance (ANOVA) and five confirmation tests. Moreover, the effects of interaction influences on efficient harvesting entropy of design parameters were described by 3D response surface. In general, this study indicates that a harvesting scheme with a higher ratio of conductivities and a smaller wedge angle and ratio of radii under a specified temperature gradient can provide a better concentrating performance. With the theoretical forecast of the harvesting performance, a satisfactory thermal harvesting device can be built under homologous conditions. (C) 2017 Elsevier Ltd. All rights reserved.