(395c) Framework for Multi-Scale Modeling and Dynamic Simulation of a Biorefinery

Biorefineries are a promising approach to produce chemicals and fuels from sustainable natural resources in order to reduce carbon dioxide emissions and the dependency on fossil raw materials. A variety of process steps, e.g. mechanical pretreatment, enzymatic hydrolysis, microbial conversion and downstream processing, is necessary to convert biomass into desired products. Techniques from computer-aided process design play an important role in increasing competitiveness of bio-based processes compared with existing conventional processes [1-5]. In that context, plant-wide dynamic models of a biorefinery help to improve the process understanding and allow model-based process operation and control. To this end, we developed a Modelica library with replaceable building blocks that allow dynamic modeling of an entire biorefinery.

As an example, we consider a biorefinery based on lignocellulosic raw material. The biomass is pretreated via the OrganoCat process [6] yielding process streams containing carbon sources for microbial conversion. For modeling the microbial conversion, we built on the dynamic flux balance analysis (DFBA) approach to formulate process models for the simulation of cellular metabolism under changing environmental conditions, typical for bioreactor-based production processes. Classical flux balance analysis (FBA) is based on the stoichiometry of all intracellular reactions within defined network boundaries and additional constraints on some fluxes to result in a linear equation system [7]. In most cases, the system of equations is underdetermined because there are more unknown reaction fluxes than metabolites such that there are infinite solutions. Formulation of a cellular optimization criterion, e.g., maximization of growth or ATP production, allows choosing among the set of solutions in order to determine one particular flux distribution. The DFBA model is obtained by coupling the FBA model to the time-dependent behavior of further process variables related to the extracellular environment of a bioreactor such as biomass, substrate and product concentration [8]. This leads to a system of differential algebraic equations with embedded optimization criteria. The DFBA approach is successfully applied to model reaction networks of different size and focus [9-12]. Our framework couples DFBA process models to entire plant simulation. The framework comprises a Modelica library with replaceable building blocks for dynamic modeling of typical biorefinery process steps, a converter to embed DFBA process models formulated in Omix [13] and an external program to solve the embedded optimization problem allowing simulation from available software, e.g. Dymola. We demonstrate the applicability of our framework by coupling a dynamic model of the OrganoCat pretreatment process [6] to a DFBA process model describing the growth of Corynebacterium glutamicum on substrates derived from lignocellulosic biomass.

References:

[1] W. Marquardt, A. Harwardt, M. Hechinger, K. Kraemer, J. Viell, and A. Voll. The biorenewables opportunity - toward next generation process and product systems. AIChE Journal, 56(9):2228–2235, 2010.

[2] A. C. Kokossis and A. Yang. On the use of systems technologies and a systematic approach for the synthesis and the design of future biorefineries. Computers & Chemical Engineering, 34(9):1397–1405, 2010.

[3] M. Martín and I. E. Grossmann. On the systematic synthesis of sustainable biorefineries. Industrial & Engineering Chemistry Research, 52(9):3044–3064, 2013.

[4] R. T. Ng and C. T. Maravelias. Design of biofuel supply chains with variable regional depot and biorefinery locations. Renewable Energy, 100:90–102, 2017.

[5] B. Bao, D. K. Ng, D. H. Tay, A. Jiménez-Gutiérrez, and M. M. El-Halwagi. A shortcut method for the preliminary synthesis of process-technology pathways: An optimization approach and application for the conceptual design of integrated biorefineries. Computers & Chemical Engineering, 35(8):1374–1383, 2011.

[6] J. Viell, A. Harwardt, J. Seiler, and W. Marquardt. Is biomass fractionation by organosolv-like processes economically viable? A conceptual design study. Bioresource Technology, 150:89–97, 2013.

[7] Palsson BØ. 2006. Systems biology: Properties of reconstructed networks. New York, NY: Cambridge University Press.

[8] R. Mahadevan, J. S. Edwards, and F. J. Doyle. Dynamic flux balance analysis of diauxic growth in escherichia coli. Biophysical Journal, 83(3):1331-1340, 2002.

[9] Hanly and M. A. Henson. Dynamic flux balance modeling of microbial co-cultures for efficient batch fermentation of glucose and xylose mixtures. Biotechnology and bioengineering, 108(2):376–385, 2011.

[10] J. L. Hjersted and M. A. Henson. Optimization of fed-batch saccharomyces cerevisiae fermentation using dynamic flux balance models. Biotechnology progress, 22(5):1239–1248, 2006.

[11] K. Höffner, S. M. Harwood, and P. I. Barton. A reliable simulator for dynamic flux balance analysis. Biotechnology and Bioengineering, 110(3):792-802, 2013.

[12] X. Zhao, S. Noack, W. Wiechert, and E. v. Lieres. Dynamic flux balance analysis with nonlinear objective function. Journal of Mathematical Biology, 147(1):761, 2017.

[13] P. Droste, S. Miebach, S. Niedenführ, W. Wiechert, and K. Nöh. Visualizing multi-omics data in metabolic networks with the software Omix: A case study. Bio Systems, 105(2):154–161, 2011.