On the stabilization of process systems described by the laws of thermodynamics

In this thesis we propose a framework for modeling constraints on the dynamic behavior of systems obeying the laws of thermodynamics. These constraints induce dissipativity and we call a system obeying such constraints a process system. Passivity theory then shows that low dimensional macroscopic analysis can be used for control system design. In inputs to the process control system are process fluxes whereas the outputs are process inventories. The inputs and outputs converge to their setpoints and the state vector converges to a stationary passive state provided one exists. The proposed control system is known as an inventory control system. Several examples are developed to support the theory and an extension is proposed for systems with equipment constraints.; We incorporate the condition of e -controllability on the design process to handle those situations where constraints are present. e -controllability is derived from the application of Passivity to Thermodynamics. An algorithm is presented that allows the designer to find the optimal e -controllable design for all known disturbances in the system. Examples are presented and dynamic simulation results are shown to verify the algorithm.; Due to limitations with the inventory control approach, a storage function is constructed by exploiting the properties of the entropy function. The uncontrolled states of the system are guaranteed stable through the application of Newton's theorem for vector valued functions. Various examples are presented including a nonequilibrium flash, an adiabatic flash and an isothermal flash unit. The theory is extended to include unit operations with more than one stage and is shown to hold under certain restrictive assumptions.