You are here : EUR MANUTECH SLEIGHT > SLEIGHT's research projects

QUANTRI - Research project

Quantitative identification of physical characteristics of solid interfaces correlated with the tribological properties


The project is motivated by the current lack of experimental capabilities to access to direct measurements of the solid third body (i.e. tribological interface) rheology. The work follows five steps:
  1. Generation of various third body function of contact conditions coupled with rheological measurements (friction, flows);
  2. Imaging of surface morphology & third body particles, with Scanning Electron Microscopy;
  3. Image processing with different strategies;
  4. Image analysis to extract quantitative descriptors (characteristics of particles and surface features). “Classical” geometrical descriptors (e.g. perimeter, circularity, elongation), but also more complex ones (e.g. contrast, homogeneity, entropy…), are studied in order to consider the whole diversity of third body particles layout and features;
  5. Finally an integration of relevant descriptors in machine learning algorithms will allow a better understanding of the mechanisms involved in dry contacts.

Dry friction between two bodies in contact involves a layer of particles at the sliding interface, namely the third body layer, which is fully responsible for energy dissipation. Until now, the rheology of the third body has been characterized qualitatively by a broad description of its morphology (from powdery granular media to ductile continuous media) at the contact scale. The present work proposes to enrich this characterization through advanced image processing.
First, third bodies are generated during pin-on-disk experiments. Then they are characterized by using image acquisition and analysis. Wear tracks are observed by scanning electron microscopy. The images of third body are dispatched in two classes, (1) ejected particles out of the contact, and (2) trapped particles assimilated to texture.
  1. The particles of third body, constituent of the ejection flow, are analysed to extract the geometrical, morphological and topological characteristics. A dedicated algorithm has been developed to segment automatically all the images resulting from a same test, subject to training the machine learning model with representative particles, while limiting any effects of the initial surface, thus errors, on the image analysis results. There are a multitude of descriptors for ejected particles, usually used to trace back to the wear mechanism suffered by the system. A sensitivity study is conducted to choose the most relevant ones.
  2. For the analysis of textures, the co-occurrence matrix is used as a texture study tool. This matrix counts neighboring pixel pairs in terms of intensity at a given angle and distance, one matrix per couple of angle and distance. This matrix is calculated on third body images on selected areas of interest. Then different quantities (homogeneity, energy, entropy…) can be calculated. Two types of characteristics are distinguished, those to which we can give a physical interpretation, and which can help the tribologist to interpret the phenomena present in the contact, and those more mathematical to describe the texture without providing qualitative information. Similar to the study of the ejected third body, a sensitivity study is also conducted to determine the significant parameters. The data provided by graphical analysis and sensitivity analysis make it possible to investigate the study of texture with two visions: (a) an evolutionary vision, a measure of the evolution of a parameter during the life of the contact making it possible to build a scenario; (b) a correlative view, i.e., search for which panel of characteristics extracted from the image allow to go back to the rheological parameters such as the coefficient of friction.
The diversity of third body structures and their complexity requires the search for relevant descriptors of both morphology and texture. The necessary tools have been built to study third body rheology and get quantitative results.
Once the images are properly segmented, it is possible to search for correlations between tribological data and characteristics of the third body (morphology & texture). For this it is planned to build a database grouping all the images (particles and textures) coming from different tests conducted under various conditions by attributing different features similar to those presented here, and also having hidden metadata including experimental data such as temperature, composition of the atmosphere, relative humidity, speed, roughness... Once the database is built, the use of unsupervised machine learning algorithms would allow to better understand the mechanisms involved in dry contacts.

A) Pin-on-disk tests (Production of third body), B) Image of ejected particles out of the contact and segmentation result, C) Trapped particles images (for texture analysis), D) Evolution of average friction coefficient µ, E) Friction coefficient function of entropy (a texture parameter).

  • A. Bouchot, A. Ferrieux-Paquet, G. Mollon, S. Descartes, J. Debayle, Segmentation and morphological analysis of wear track/particles images using machine learning, Journal of Electronic Imaging, 2022, 31 (5), 051605
  • Alizée Bouchot, Amandine Ferrieux, Johan Debayle, Guilhem Mollon, Sylvie Descartes, Image processing applied to tribological dry contact analysis, Wear 476 (2021) 203748

  • Alizée Bouchot, Amandine Ferrieux-Paquet, Johan Debayle, Sylvie Descartes, Guilhem Mollon, How to correlate physical characterics of tribological interfaces with rheological properties?, Wear of Materials - Banff (Canada) - 16-20 April, 2023
  • Alizée Bouchot, Amandine Ferrieux-Paquet, Johan Debayle, Guilhem Mollon, Sylvie Descartes, Dry friction analysis: A correlative study between rheology and third body morphology, World Tribology Congress WTC - Lyon (France) - 10-14 Jul. 2022
  • Alizée Bouchot, Amandine Ferrieux-Paquet, Sylvie Descartes, Guilhem Mollon and Johan Debayle, Towards a quantitative characterization of wear particles using image analysis and machine learning, Proc. SPIE 11794, Fifteenth International Conference on Quality Control by Artificial Vision, 1179416 (16 July 2021)