Thermochemical conversion of biomass, coal, and natural gas to produce liquid fuels often includes the production of Fischer-Tropsch (FT) liquids utilizing a synthetic gas (syngas) intermediate [1]. Specifically, coal gasification, biomass gasification or natural gas reforming is used to create the syngas which is then converted to liquid hydrocarbons in the FT reactors [2,3]. The raw liquid product is subsequently upgraded to produce the necessary hydrocarbon fuels. Current studies generally focus on a small set of process designs with specific operating conditions to evaluate the cost of crude oil that would make the design competitive with petroleum derived fuels. However, it remains unclear whether any one of the designs represents the topology of a refinery that can produce transportation fuels at the lowest cost. Additionally, even if the process topology is correct, the selection of process operating conditions in the given process design may be suboptimal, leading to a higher cost of liquid fuels. To address both of these issues, a process synthesis approach can be used which can assess a large feasible set of process flowsheets to find the one(s) that yield a minimal liquid fuels cost [4]. A mixed-integer nonlinear non-convex optimization based mathematical model (MINLP) that seeks to minimize the cost of liquid fuels production would allow for analysis of the distinct topologies and operating conditions.
To ensure a high-quality solution to the MINLP, a global optimization approach is proposed to find the CBGTL refinery system topology that provides the lowest levelized cost of transportation fuel production. Continuous variables are used to model stream flow rates, process operating conditions, and thermodynamic properties while binary variables are used to model the logical use of a process unit or a process stream. The MINLP model includes simultaneous heat and power integration utilizing heat engines to recover electricity from the process waste heat. A branch-and-bound scheme is introduced that utilizes piecewise linear underestimators to determine tight lower bounds on the optimal solution. The capability of the method is demonstrated using two large case studies that feature 2,266 nonconvex terms, 21 binary variables, 15,232 continuous variables, and 15,533 constraints.
[1] National Academy of Sciences, National Academy of Engineering, and National Research Council. Liquid Transportation Fuels from Coal and Biomass: Technological Status, Costs, and Environmental Issues. Prepublication. Washington, D. C., EPA, 2009.
[2] R. C. Baliban, J. A. Elia, and C. A. Floudas. Toward Novel Biomass, Coal, and Natural Gas Processes for Satisfying Current Transportation Fuel Demands, 1: Process Alternatives, Gasification Modeling, Process Simulation, and Economic Analysis. Ind. Eng. Chem. Res., 49:7343-7370, 2010.
[3] J. A. Elia, R. C. Baliban, and C. A. Floudas. Toward Novel Biomass, Coal, and Natural Gas Processes for Satisfying Current Transportation Fuel Demands, 2: Simultaneous Heat and Power Integration. Ind. Eng. Chem. Res., 49:7371-7388, 2010.
[4] R. C. Baliban, J. A. Elia, and C. A. Floudas. Optimization Framework for the Simultaneous Process Synthesis, Heat and Power Integration of a Thermochemical Hybrid Biomass, Coal, and Natural Gas Facility. Comp. Chem. Eng., doi:10.1016/j.compchemeng.2011.01.041,2011.
See more of this Group/Topical: Computing and Systems Technology Division