Optimization of the sugar hydrothermal extraction process from olive cake using neuro-fuzzy models.

The optimization of the hydrothermal extraction step in the biorefinery scheme in order to obtain sugars and antioxidants from olive cake was carried out. This process using water and low temperatures for the olive cake has not been previously studied by other authors. Central Composite Design of experiments was carried out. Four variables and three levels for each variable were tested: temperature (50, 70 and 90 °C) time (60, 120, 180 min), average particle diameter (0.42, 0.96 and 1.5 mm) and solid/liquid ratio (0.15, 0.20, 0.25). The main aim was to understand the relationship between the process variables and the yield of extraction of glucose, xylose, polyphenols and oligomers. The variables which most influenced the process were particle diameter and solid/liquid ratio. The experimental values were adjusted to a classical polynomial model and to a neuro-fuzzy system. The neuro-fuzzy demonstrated to be much more accurate when predicting the experimental values.