Start
Jun 14, 2011 - 09:30
End
Jun 14, 2011 - 11:30
Venue
Dept Library (Skype)
Event Type
Speaker
Dr. Yogendra Shastri Visiting Research Assistant Professor Energy Biosciences Institute & Department of Agricultural and Biological Engineering University of Illinois at Urbana-Champaign
Title
Techno-economic feasibility analysis of bioenergy feedstock production using optimization and simulation models
Biomass feedstock production is an important component of the biomass based energy sector. Low energy and bulk densities seasonal availability and distributed supply create unique challenges that must be e®ectively addressed. The determination of the techno-economic feasibility of this system in the presence of existing and novel technologies is non-trivial. Moreover it must be ensured that the development and functioning of the feedstock production system is sustainable and resilient. This work takes a systems theory based approach to address these challenges through the development of optimization and agent-based simulation models. An optimization based decision making framework named BioFeed has been developed to optimize the local/regional feedstock production system. A unique feature of this mixed integer linear programming (MILP) model is the integration of the design as well as operational decisions in a single framework. The model also identi¯es system bottlenecks and quanti¯es the systemic impacts of technology improvement. The model was successfully used to study the production of switchgrass and Miscanthus in Illinois (US). The results indicated that the choice of biomass pre-processing technology signi¯cantly a®ected the total cost which varied between 45-65$/Mg. This translated to about 50% of the targeted ethanol production cost of 0.32$/l. Biomass pre-processing such as grinding and storage were the major cost components. Sensitivity analysis showed that 25% increase in pre-processing throughput rate led to 10% reduction in the total cost and energy consumption. A novel computational scheme combining mathematical programming and agent-based modeling was developed which resulted in an order of magnitude reduction in the simulation time. Moreover the optimal solution was within 5% of the rigorous MILP solution. Large scale production of ligno-cellulosic feedstock is not yet established and there is a signi¯cant lack of understanding about the development and functioning of such as system. It is a complex system with multiple scales uncertainties and emergent properties. An agent-based model was therefore developed using the theory of complex adaptive systems to study its dynamics. Important stakeholders such as biore¯nery and farmers were modeled as independent agents using an object-oriented approach. Simulation results for Miscanthus production showed that up to 15 years might be required to reach steady regional biomass productivity which has implications on biore¯nery capacity planning. The contract price could be up to 40% higher than the optimized production cost. The model will be used in future to study the resiliency of this system in the presence of uncertainties and disturbances in the future.