F. Cruz*, M. P. Dorado, J. M. Palomar, V. Montoro Dep. of Mechanics and Mining Engineering University of Jaen (Spain)

Corresponding author

Temperature distribution analysis over the absorber surface and the working fluid requires the application of an experimental study based on heat transfer basic principles, i. e. bi-dimensional conduction. In this sense, several researchers have decided to measure the temperatures over the plate [Chuawittayawuth et al., 2002], while others have preferred to simulate the whole process, sometimes based on commercial software with high computational capacities [Kalogirou & Papamarcou, 2000]. In our work, modelling and optimization were made by previously analyzing and programming the process. According to the general theory of heat transfer, we have proposed to develop a two-dimensional model of conduction and working fluid heat transfer [Incropera & Dewitt, 1996], together with the conventional theory of solar collectors [Duffie & Beckman, 1991], for both steady state and transient regime. Analysis were made using the following considerations [Norton et al., 2001], although in a preliminary stationary analysis some of them could be neglected: a) thermal capacity of absorber plate is considered; b) density, specific heat, viscosity, conductivity and Prandtl number of refrigerant are approximated by polynomic functions depending on temperature; c) heat transfer coefficients depend on the medium temperatures of the plate and the environment on each time; d) time variations of irradiation, ambient temperature, and inlet and outlet flow temperatures are assumed; e) radiation transmission through transparent surface varies with solar time and day of the year. Later, to validate the described models, a parameter identification procedure was formulated. This procedure was based on the gradient descendent method, related with the Taylor series expansion of an objective function. This function was defined using the comparison between real and modelled parameters [Gill et al., 1981].