Modelling, synthesis and analysis of biorefinery networks

Abstract

The quest for satisfying the needs of a growing population along with the urge to address environmental concerns require the development of novel and more sustainable production processes that utilize renewable resources, such as biomass. Biorefineries have gained interest, as integrated facilities for the conversion of biomass into chemicals, fuels and energy, because they have the potential to maximize biomass value while reducing emissions. The design of biorefinery networks is a complex decisionmaking problem that involves the selection of feedstocks, processing technologies, products, geographical locations, and operating conditions, among others. Unlike petroleumbased processing networks, biorefineries rely on feedstocks that are nonhomogeneous across geographical areas in terms of their availability, type and properties. For this reason, the performance of biorefinery networks depends on their geographical distribution and surrounding markets. Moreover, biorefinery research is ongoing and these processes are not well established, which means that a large number of potential technologies are continuously developed and need to be evaluated. This corresponds to large amounts of data being available yet not consolidated, systematized or ready to use. In addition, the political and social contexts change rapidly, which requires means for fast assessments given each specific context. It is thus clear that methods and tools to address some of the design challenges are required, for example, the synthesis of reactions to convert available biomassbased feedstocks into desired products, the selection of processing routes and technologies from a large set of alternatives, or the generation of hybrid technologies through process intensification. Systematic process synthesis and design methods have been developed for traditional chemical processes. However, although many concepts are still valuable, these methods are not directly applicable to biorefinery networks and need to be further extended. The main contribution of this work is the development of a systematic framework for synthesis of biorefinery networks that integrates the necessary models, methods and tools for the problem formulation and solution. The developed framework is generic, hence useful within a large range of applications, being able to manage the complexity of problems including synthesis of chemical process, bioprocess and biorefineries. It is also flexible, namely easily adaptable to different problem types. Moreover, an ontology-based data structure for data management and the implementation of the framework in a computer-aided tool, Super-O, make the solution procedure faster, more efficient and easily accessible by non-expert users. The framework has been applied to three relevant biorefinery synthesis examples in order to test its application range and highlight its key features.