The fundamental challenge in the synthesis/design optimization of energy systems is the definition of system configuration and design parameters. The traditional way to operate is to follow the previous experience, starting from the existing design solutions. A more advanced strategy consists in the preliminary identification of a superstructure that should include all the possible solutions to the synthesis/design optimization problem and in the selection of the system configuration starting from this superstructure through a design parameter optimization. This top–down approach cannot guarantee that all possible configurations could be predicted in advance and that all the configurations derived from the superstructure are feasible. To solve the general problem of the synthesis/design of complex energy systems, a new bottom–up methodology has been recently proposed by the authors, based on the original idea that the fundamental nucleus in the construction of any energy system configuration is the elementary thermodynamic cycle, composed only by the compression, heat transfer with hot and cold sources and expansion processes. So, any configuration can be built by generating, according to a rigorous set of rules, all the combinations of the elementary thermodynamic cycles operated by different working fluids that can be identified within the system, and selecting the best resulting configuration through an optimization procedure. In this paper, the main concepts and features of the methodology are deeply investigated to show, through different applications, how an artificial intelligence can generate system configurations of various complexity using preset logical rules without any “ad hoc” expertise.
A nonlinear inverse model for airborne wind energy system analysis, control, and design optimization
