BIOSTEAM: THE OPEN-SOURCE BIOPROCESS SIMULATION AND TECHNO- ECONOMIC ANALYSIS MODULES IN PYTHON BY YOEL RENE CORTES-PENA THESIS Submitted in partial fulfillment of the requirements for the degree of Master of Science in Environmental Engineering in Civil Engineering in the Graduate College of the University of Illinois at Urbana-Champaign, 2019 Urbana, Illinois Adviser: Assistant Professor Jeremy Guest
ABSTRACT Preliminary techno-economic analysis (TEA) of a chemical process provides critical information on the economic feasibility, technological bottlenecks, and venture risks due to process uncertainties. Current TEA methods generally rely on proprietary software to evaluate two to three design configurations with single point sensitivity analysis. Such classical methods neglect the effect of process uncertainties and fail to evaluate the complete landscape of design decisions. The limited scope of current literature obscures comparisons between process evaluations and makes it difficult to predict how possible technological developments can impact the sustainability of a process. The main difficulties in adding rigorous uncertainty analysis is high computational time and low rate of successful convergence. Additionally, relying on proprietary software presents both an economic and intellectual barrier for evaluating emerging and conceptual processes. The Bioprocess Simulation and Techno-Economic Analysis Modules, BioSTEAM, is an open-source steady state process simulation package in python for preliminary TEA that will enable rigorous uncertainty analysis through its fast and flexible platform. BioSTEAM presents an intuitive toolset of objects that handle thermodynamic properties, material flows, unit operations, recycle systems, and process specifications. The applicability of BioSTEAM is demonstrated here in the context of a design for the co-production of biodiesel and ethanol from lipid cane. The evaluation of the lipid cane biorefinery in BioSTEAM closely matches a previous evaluation of the design using SuperPro Designer (a proprietary process simulation software). BioSTEAM is well documented and readily available at the Python Package Index, a repository for published Python packages. Although BioSTEAM has not yet incorporated many of the unit operation models presented in proprietary process simulators, its extendable and transparent architecture offers users the power to build new unit operation models and share their designs without any barriers. BioSTEAM may help foster a new open- source community that can accelerate advancements in the field of process design and simulation. ii
ACKNOWLEDGEMENTS I would like to thank Professor Jeremy Guest for his encouragement, guidance, and trust throughout the project. When Professor Guest presented the idea of this project to me, I knew that building a process simulation platform would be a challenging endeavor. Not only would I need to thoroughly revisit classical chemical engineering foundations, but also develop strong programming skills. Throughout the project, Professor Guest was confident in my skills and gave me the freedom and the trust I needed to learn about software development and begin laying out the framework for BioSTEAM. My meetings with Professor Guest helped solidify the project tremendously through our philosophical discussions of the platform's future. I could have not asked for a better advisor than Professor Guest, who's dedication and clear communication allowed this project to grow under our shared