Artificial neural network based modeling and optimization of refined palm oil process

摘要: The content and concentration of beta-carotene, tocopherol free fatty acid is one the important parameters that affect quality edible oil. In simulation based studies for refined palm oil process, three variables are usually used as input which feed flow rate (F), column temperature (T) pressure (P). These influence output acid. this work, we develop 2 different ANN models; first model on 3 inputs (F, T, P) second (T P). Artificial neural network (ANN) models set up to describe simulation. Feed forward back propagation networks designed using architecture in MATLAB toolbox. effects numbers neurons layers examined. correlation coefficient study greater than 0.99; it good agreement during training testing models. Moreover, found can process accurately, able predict outputs very close those predicted by ASPEN HYSYS simulator process. Optimization performed maximize beta-carotene at residue distillate.