Artificial intelligence based design of multiple friction modifiers dispersed castor oil and evaluating its tribological properties

This present work deals with designing a new lubricant with multiple friction modifiers (FM) having better tribological properties using artificial neural network (ANN) and genetic algorithm (GA). The input variables considered are load, speed and concentration of the FMs, and coefficient of friction (CoF) is the output. Experimental data generated in pin-on-disc tribometer is used for the ANN model, whereas GA is used for optimization using the ANN models as the objective function. The experimental trials of the computationally designed lubricants achieving the minimum CoF were conducted. It has been observed that the coefficient of friction by the new lubricant containing multiple friction modifiers is 45-50% less and wear scar diameter is 87.5% less compared to commercial mineral oil samples.