Development of a Mathematical Model for Optimal Design of Integrated Utility and Hydrogen Supply Networks

Abstract

This thesis suggests a mathematical model for optimal design of integrated utility and hydrogen supply networks (IUHSNs) in a huge chemical industrial complex. Utility supply networks and hydrogen (H2) supply networks have been consistently individually studied in terms of techno-economic feasibility of large scale. A number of studies have proposed and improved mathematical models for design of each optimized supply network. However, although two different networks can coexist in a large-scale industrial complex, few studies have been conducted to develop an integrated utility and H2 supply network design in terms of a techno-economic optimization framework. This thesis designs an IUHSN which includes a number of utilities (steam, water, and electricity), and H2 sources (manufacturers) and sinks (end customers). Steam methane reforming (SMR) process is used as an intermediate linkage between both networks because 1) SMR process is the general process to produce H2 and 2) it is possible for the integrated model to use steam vents which are generated from a utility supply network. Chapter 3.I develops a mathematical model which is formulated as multi-period stochastic mixed integer linear (MPSMILP) programming model concerning uncertain raw material prices to obtain an optimal design of the IUHSN reflecting the reality. This model allows identification of a promising design strategy to minimize total supply cost, because sources and sinks can be connected to each other to transfer unused resources and products. Furthermore, an optimal location of SMR process can be decided in an IUHSN. Five different case studies are performed at Yeosu industrial complex in South Korea. In Chapter 3.II, a two-stage stochastic mixed integer linear programming (TSSMILP) model is designed for integrating various utility supply networks and global H2 supply network. The developed model not only optimizes total cost but also considers demand uncertainty about water, electricity, steam, and H2 which are mainly used in each network to improve model identification. This model also chooses SMR process as necessary process to integrate two networks. Case studies are examined in South Korea. The results of an IUHSN model can help a decision maker establish an optimal design concerning various uncertain parameters. Furthermore, cost optimization and network management can be performed by an IUHSN model.