Thermodynamics of Multi-Domain Power Networks: Principles for Optimization and Control with Applications to Turboelectric Systems

Cleveland State University, Cleveland OH

Investigators

Abstract

This grant will support research leading to advancing the understanding and application of thermodynamic principles that govern power transfer in interconnected energy systems with electric power storage, conversion, and utilization. Environmental sustainability, economy and strategic competitiveness continue to fuel the quest for more efficient energy management methods. Such methods encompass, in addition to traditional industrial processes involving conversions between heat and work, forefront technologies such as microgrids, hybrid vehicles and electrified aircraft propulsion. Therefore, there is a need to develop theoretical foundations of interconnected thermodynamic systems that go beyond the classical setting. This project introduces innovative ideas allowing meaningful interpretations of entropy and its connection to efficiency in extended physical domains. It develops methodologies for optimal design and control in application to electrified turbine engines. Educational and outreach activities include undergraduate student mentoring and collaboration with the MathCorps Cleveland summer program that targets students in grades 6 to 12 who are underrepresented in science and technology. Entropy is connected to the ability to perform work efficiently and is the basis for the optimal design and operation of systems deriving work from heat. Entropy cannot be defined for Hamiltonian systems due to their lack of a preferent direction of power transmission, unlike the diffusive power transfer patterns found in thermofluid systems. This project adopts a systems-theoretic approach to thermodynamics to define entropy generation and exergy efficiency by cyclic averages. The objective is to provide counterparts to key notions of irreversible thermodynamics, along with methods for model-free parametric optimization and feedback control design. The objective will be achieved with tasks organized along three thrusts: i) Formalizing the property of energy cyclo-directionality, applicable to port-Hamiltonian systems on graphs and providing definitions for average entropy generation and exergy efficiency; ii) Solving parametric optimization problems considering second law principles. This involves online spectral estimation techniques to enable model-free, self-optimization methods with tools such as extremum-seeking and formulating multivariable, frequency-domain optimal control methodologies under entropy-related objectives; and iii) Demonstrating and assessing the practical validity of this framework using a turboelectric propulsion system where an engine simulation interacts in real-time with a physical electromechanical system in the laboratory. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.