Subscale Tribofilm Tribological Modelling in Boundary Lubrication Using Multivariate Analysis

Abstract

A subscale tribofilm tribological performance model has been developed to investigate the sensitivity of the model to lubricant chemistries and tribological factors in boundary lubrication conditions. Tribological modelling of tribofilm performance in boundary lubrication deals with a large number of variables. Thus, multivariate techniques were used to obtain the subscale tribofilm friction and wear models in terms of additive concentration in the oil and tribological factors. Soft Independent Modelling of Class Analogy classification techniques such as principal component analysis and partial least squares regression were used to correlate the outputs friction, wear, and lambda ratio with additive concentration in a model oil and tribological factors. The results show that boundary lubrication can be divided into three ‘subscale’ lubrication regimes which can be defined as (a) severe, (b) moderate, and (c) mild boundary lubrication conditions. The predictions from the friction and wear models showed varied dependence on the variables within the subscale lubrication regimes.