Deep learning algorithm to predict friction coefficient of matching pairs at different temperature domains based on friction sound.

The tribological properties of matching pairs at different temperature domain is predicted through friction sound. Two deep learning algorithms respectively by virtue of the Nonlinear Auto-Regressive models with Exogenous Inputs and long short-term memory neural network are adopted to analyze the acoustic features of friction sound to predict tribological properties of polymer surface at five temperatures within a large temperature domain under different working conditions. The deep learning algorithm precisely fits the friction coefficient in accordance with the performance analysis. [Display omitted] • A real-time friction coefficient prediction method based on friction sound. • The mapping relationship between friction sound and friction coefficient with the same random nonlinear characteristics. • The prediction method was widely applicable to the prediction of friction coefficient under wide temperature range. • This method can be used for in-situ monitoring on tribological properties of mating pairs. [ABSTRACT FROM AUTHOR]