Conference Year
2018
Keywords
ORC system, positive displacement, CFD, 3D modeling, real gas
Abstract
Over the last decade, environmental and economic concerns have pushed the researchers to find new solutions in the track of a more responsible use of energy. Particularly, small scale organic Rankine cycles (ORCs) have been regarded as candidate for a better employment of waste energy. In order to increase the performance of these cycles and to extend their operating range, attention has been drawn on the behavior of the different components both experimentally as well as by means of computational fluid dynamics (CFD) simulations. The numerical approach has been increasingly used in the study of the machines that compose the cycle, avoiding the problems that typically affect the experimental analyses, e.g. compatibility of refrigerant with sealing systems, and allowing for the preliminary test of new machines to be added to the cycle. In this work, the numerical analysis will be focused not only on the single components considered as a stand-alone, but rather extended on their reciprocal interaction and on the system integration of the different machines. A Whole ORC Model (WOM), can thus be built and employed as a virtual test bench. Such a virtual model can be of paramount importance in predicting the behavior of the cycle in off-design conditions or in gathering information about fluid stagnation locations. The analysis can be even extended by coupling the WOM with the external world. Specifically, the grid demand and the heat flux at the evaporator can vary: such change is translated in a variation in the boundary conditions. The response of the cycle to the external variation can be therefore monitored and studied. A full three-dimensional, transient analysis and the framework in which the WOM is developed are presented in this work. The numerical strategies employed are described, with particular attention to the fashion in which the real gas effect of the working fluid and the motion inside the positive displacement machines are treated. The performance variation in response to an external change is reported to show the capability of the virtual test bench in helping both the system conductor as well as the designer.