Dynamical Graph-Based Models of Brayton Cycle Systems

For many power systems, thermodynamic cycle systems are used as subsystems for energy injection or removal. For model-based control of large, interconnected systems, model fidelity and computational efficiency must be balanced. Graph-based models have proven success at accurately modeling multiple energy domains in a computationally efficient manner. This paper presents a graph-based modeling approach governed by energy conservation to produce a general graph framework for thermodynamic cycles with mass transport. To investigate model performance, case studies on a reverse Brayton cycle and open Brayton cycle system are considered. The graph-based model of these systems produces comparable model accuracy to alternative dynamic modeling techniques while providing the modularity, scalability, and computational efficiency of graph-based models. A 98% reduction in computational time is achieved by the graph-based modeling approach.