Experimental and numerical development of a hybrid solar/gas heating system

Abstract

The main purpose of this thesis is to understand and formulate the relationship between the hybrid system performance and design parameters, in order to integrate hybrid solar systems in the design of energy efficient buildings. The combined use of numerical simulations, Design of Experiments (DoE) technique and an optimization approach is thus adopted for this aim. Moreover, the combined use of numerical simulations and DoE technique allows the development of meta-modeling relationships between response variable, here solar fraction (SF), primary energy savings ratio (PESR), levelized cost of heat (LCOH) and design parameters. These meta-models are used, first, to perform a sensitivity analysis, and second, to optimize the design of hybrid solar systems. The desirability function approach and genetic algorithms are then considered in order to simultaneously optimize all of SF, PESR and LCOH. The proposed method is applied in a real case study, an experimental platform that has been developed for the purpose of this study. An optimal design was obtained 78.46%, 44.87% and 7.7 DA/kWh for the SF, PESR and LCOH respectively. The obtained results indicated that optimal design is achievable using the proposed approach which represents a simple and fast method to optimize hybrid solar design parameters.