Multiobjective optimization of a sugarcane biorefinery involving process and environmental aspects

• Multiobjective optimization studies can indicate best operating policies in biorefineries.
• NLP multiobjective optimization studies with conflicting process and environmental objectives.
• A tool that may help find a trade-off between greater profits and more sustainable process.

Process modeling and simulation are essential to predict process energetic demands, and determine possible throughputs and process emissions in biorefineries. In this paper non-linear multiobjective optimization studies with conflicting process and environmental objectives are performed in a sugarcane biorefinery. The process produces bioethanol (first and second generation), bioelectricity and concentrates vinasse in multiple-effect evaporators. Objective functions, concerning process and environmental issues, are defined and studied in five optimization problems. Decision variables are fraction of bagasse diverted to E2G production (Var1), and the fraction of vinasse that is concentrated up to 7.0°Bx in multiple-effect evaporator (Var2). The results show that Var1 cannot assume values greater than circa 0.5, due to thermal demands of the integrated process, while Var2 cannot be lower than 0.15, due to process demands for vegetal steam. Flows of concentrated vinasse are generally higher when maximization of bioethanol throughputs is imposed. When electric power generation and bioethanol throughput maximization are the objectives, the former varies 45.6%, while the latter varies 16.6%, among non-dominated solutions. The adopted approach can be used as a decision-making tool that may help to choose suitable operating conditions, in order to obtain a trade-off between greater profits and a more sustainable process.