Thesis Archive
AN EVALUATION OF OPTIMAL AND SUB-OPTIMAL SOURCE-SINK CONSERVATION NETWORKS GENERATED BY P-GRAPH STUDIO USING THE MONTE CARLO SIMULATION METHOD
John Ismael A. Arogo
Antonio Louis O. Coronel
Christian Marco J. Janairo
Abstract:
The development of resource conservation networks is a strategy to reduce the consumption of fresh resources and the generation of wastes such that its optimal design maximizes the utilization of resources. This work focuses on the development of a methodology for generating design options for various types of resource conservation networks. Process-Graph (P-Graph) theory was initially developed for process network synthesis and designed for the cost optimization of processes. However, a novel application of P-Graph is presented in this work by translating the generic resource conservation network problem into a P-Graph model. By understanding the underlying algorithms of P-Graph as it is implemented in the software P-Graph studio, another field of application is discovered. A general structure for resource conservation networks enables the generation of all combinatorially feasible network structures and the identification of optimal and sub-optimal networks. In addition, the robustness of selected network structures to variations in system parameter and performance were tested by performing Monte Carlo Simulations. This work presents three (3) case studies to demonstrate the capability of the model. The first case study looks at the minimization of freshwater usage and subsequent creation of a wastewater recycle network in an Acrylonitrile plant. The second case study deals with the minimization of fresh hydrogen gas consumption in refinery plants through the creation of a hydrogen fuel recycle network. Constraints such as resource availability and demand, accompanied with quality considerations and parameter variations were accounted for in the generation of these resource conservation networks. Results show that even if there are only a handful of processes in the resource conservation network, thousands of feasible network designs exist (more than 24,000 networks identified for the hydrogen fuel recycle network), there may be several degenerate solutions and that variations in system parameters can influence the robustness of network configurations.
Adviser:
Dr. Kathleen B. Aviso
Dr. Raymond R. Tan
[email protected]
[email protected]