Intelligent prediction of wear location and mechanism using image identification based on improved Faster R-CNN model

Abstract

Exploring on-line wear status detection methods is of great significance to the intelligent operation and health management of mechanical equipment. At present, it is common to use high-resolution imaging equipment such as metallographic microscope and scanning electron microscope to determine the wear mechanism. However, its judgment relies on the subjective experience of the observer, and it is time-consuming and difficult to be implemented in on-line detection. In the present work, we proposed an image detection method based on improved Faster R-CNN model for wear location and wear mechanism identification. The model was trained and tested using a wear image data set produced and collected by a self-made tribometer equipped with an imaging system. The results showed that the proposed method had a detection accuracy of more than 99%, and its wear location and wear mechanism identification performance was superior to edge detection technology and Yolov3 target detection models. This work shed light on the development of a new approach for realizing on-line and intelligent wear status detection of machinery components.