Mechanical Wear Life Prediction Based on Abrasive Debris Generation

Abstract

Assessment and prediction of mechanical wear life are imperative for ensuring the effective application, reducing maintenance cost and minimizing material waste, especially for the machines whose main failure modes are related to wear. For an assembled product, the prediction of wear life is difficult because there is no an appropriate method for direct measurement. The direct product of wear is believed to contain key information of the wear status of the contact surfaces and has been suggested as an indicator for mechanical failures. Currently, several online oil debris monitoring methods are proposed with high precision for accurate debris detection, but there is still a lack of a model to describe the relationship between the wear status and the debris. This paper presents a physics-based model to predict the generation of abrasive debris of contact pairs with a certain roughness, so as to predict the wear life of mechanical components. The probabilistic model is given in a numerical way, based on the boundary element method and atomic attrition mechanism. The proposed method is able to provide predictive features of wear debris including the amount, the distribution of sizes and morphological information. Combining the online oil debris detection approaches, the method is applied to the prediction of an aviation hydraulic pump. The experimental result indicates that the method is effective in predicting the remaining useful life of mechanical components.