Neural Network Model for Examination of the Temperature Characteristics of Boundary Film on Friction Surface

A mathematical model based on BP neural network has been established to examine the temperature characteristics of the boundary film on the friction surfaces of a screw nut pair which is characterized by wear self compensation feature. The network could be used to predict the effect of the boundary film on tribological behavior. It is also capable of learning and the error is small while being trained according to L M rule. The outputs of the network are precise and in good agreement with the experimental ones. The network model could be used as an effective calculation tool for the tribological design of engineers.