Bioenergy II: Modeling and multi-objective optimization of different biodiesel production processes

One of the most promising renewable fuels proposed as an alternative to fossil fuels is biodiesel. The competitive potential of biodiesel is limited by the price of vegetable oils, which strongly influences the final price of biofuels. An appropriate planning and design of the whole production process, from the seed to the biodiesel end product, is essential in order to contain the fallout of energy inefficiencies in the high price of the end product. This study focuses on the characteristics of the production process currently used to produce biodiesel. Refined vegetable oil can be converted into biodiesel by means of a great variety of techniques and technologies, many of which are still not suitable for application on an industrial scale. The solution of greatest interest is homogeneous alkaline transesterification with KOH and methanol. Even when dealing with this type of conversion, it is impossible to establish a universal pattern to describe the conversion or purification stages because there are various possible solutions that make each system different from another. When we look more closely at the state of the art in industrial biodiesel production plants, we also encounter the potential problems introduced by the type and characteristics of the raw materials. Comparing some of the reference solutions that have inspired numerous installations, an optimization analysis was conducted using ASPENPLUS 2006, for the modeling of the process, and modeFRONTIER 4.1 for the optimization procedure. The optimization analysis was carried out using a multi-objective genetic algorithm optimization in order to define the configuration of the main parameters that guarantee the best trade-off between the maximization of the purity of some important compounds and the minimization of energy requirements in the process. The results of this analysis were Pareto frontiers that identify a family of configurations which define the best trade-off between the objectives. Using the Pareto frontiers we then selected the configuration that requires the minimum energy consumption. Among these optimal configurations there is one which guarantees the lowest specific energy consumption while all the optimal configurations obtained respected the requirements of EN 14214, in terms of biodiesel quality. Copyright © 2010 The Berkeley Electronic Press.