Research Information System
6th International Symposium of Scientific Research and Innovative Studies, Balıkesir, Turkey, 11 - 14 March 2026, pp.520-521, (Full Text)
This study presents a statistically robust framework for the green synthesis of
biolubricants from Chlorella protothecoides oil via enzymatic esterification using Resinase HT,
a fungal lipase derived from Aspergillus oryzae and immobilized through an optimized crosslinking
method. Process modeling and optimization were performed using Artificial Neural
Networks (ANNs) trained with the Whale Optimization Algorithm (WOA). The WOAenhanced
ANN model exhibited the highest predictive performance (R² = 0.8245), effectively
capturing the non-linear behavior of the system compared to conventional gradient-based
algorithms and Response Surface Methodology (RSM). Temperature was identified as the most
influential parameter affecting FAME conversion, while enzyme loading and molar ratio
showed relatively minor effects. The integrated ANN–WOA framework successfully defined
optimal operating conditions, providing a reliable and sustainable strategy for biolubricant
production under solvent-free conditions.