Modeling of biodiesel production from palm oil and methanol via zeolite derived catalyst as a phase boundary catalyst: An optimization study by using artificial intelligence

In this article, generalized regression artificial neural networks have been used to modeling the production of biodiesel fuel from palm oil or soybean oil. A zeolite derived catalyst was prepared by lithium modified zeolites, which can be used as a phase boundary catalyst for the reaction of soybean oil or palm oil with short chain alcohols such as methanol or ethanol. Three variables: Reaction time, Methanol:oil and Catalyst amount are considered as independent variables and yield as dependent variable. Based on the experiments, the behavior of the system has been modeled. To maximize production efficiency, collective intelligence like particle swarm optimization algorithm has been used. The obtained results are highly accurate compared to other modeling. It is clear that the optimal value of the more accurate model is more reliable than the optimal value of the low accuracy model.

کلیدواژه های Modeling of biodiesel production from palm oil and methanol via zeolite derived catalyst as a phase boundary catalyst: An optimization study by using artificial intelligence:

particle swarm optimization algorithm ، generalized regression artificial neural networks ، mathematical modelling ، biodiesel fuel.