Friction-related failures are a frequent cause of catastrophic damage in mechanical systems, resulting in significant repair expenses and/or production delays in industry worldwide. To circumvent these failures, it is of utmost importance to have robust and cost-effective online monitoring systems able to determine the wear rate and detect events leading to failure
Research at Empa focused on developing a new online monitoring system that are not only able to measure wear rate and detect anomaly but predict the wear rate as well as the remaining in service component lifetime.
The three major achievements are (a) the possibility to detect online the different tribological states (steady-state, pre-scuffing and scuffing), (b) to following online changes of the surfaces states, even differences within the steady-state and (c) predict scuffing several minutes before failure takes place.

• Pandiyan V., Prost J., Vorlaufer G., Varga M., and Wasmer K., "Identification Of Abnormal Tribological Regimes Using a Microphone And Semi-Supervised Machine-Learning Algorithm", Friction, Vol. XX, Issue: X, Paper ID: XX, pp: 1-XX, 2021,

• Shevchik S.A., Zanoli S., Saeidi F., Meylan B., Flück G., Wasmer K., "Monitoring of Friction-Related Failures Using Diffusion Maps of Acoustic Time Series" Mechanical Systems and Signal Processing, Vol. 148, Issue February 2021, paper ID: 107172, pp: 1-14, 2021,
  https://doi.org/10.1016/j.ymssp.2020.107172

• Shevchik S.A., Saeidi F., Meylan B. and Wasmer K., “Prediction of Failure in Lubricated Surfaces Using Acoustic Time-Frequency Features and Random Forest Algorithm”, IEEE Transactions on Industrial Informatics, Vol. 13, Issue 4, pp: 1541-1553, 2017,
  http://dx.doi.org/10.1109/TII.2016.2635082

• Saeidi F., Shevchik S.A. and Wasmer K., “Automatic Detection of Scuffing Using Acoustic Emission”, Tribology International, Vol. 94, pp: 112-117, 2016, http://dx.doi.org/10.1016/j.triboint.2015.08.021

   

The efficiency of processes involving frictional contacts between surfaces is often characterized by wear rates or friction coefficients. However, the classification and forecasting of wear rates in friction related processes is a real industrial challenge that is unsolved today.
Research at Empa focused on developing an online wear monitoring system combing acoustic emission and state-of-the-art AI algorithm.
The main achievements are that we can classify with high accuracy the wear rate at specific time.

• Deshpande P., Pandiyan V., Meylan B., and  Wasmer K., "Acoustic Emission and Machine Learning Based Classification of Wear Generated Using a Pin-On-Disc Tribometer Equipped with a Digital Holographic Microscope", Wear, Vol. XXX-XX, pp: XX-XX, 2021,
  https://doi.org/10.1016/j.wear.2021.203622

   

Wear occurs in the form of adhesion, abrasion, scuffing, galling and scoring between contacts. However, the rate of the wear phenomenon depends primarily on the physical properties and the surrounding environment. Monitoring the integrity of surfaces by offline inspections leads to significant wasted machine time. Today is it mainly made using supervised AI methods. However, this method has the main disadvantages that data of abnormal regimes are difficult if not impossible to acquire.
Research at Empa focused on proposeing a semi-supervised AI algorithm that is trained only to the normal regime and classify any deviation as abnormal regimes.
The main achievements is that we could differentiate normal and abnormal regimes with a accuracy of 97% and 80%, respectively.

• Pandiyan V., Prost J., Vorlaufer G., Varga M., and Wasmer K., "Identification Of Abnormal Tribological Regimes Using a Microphone And Semi-Supervised Machine-Learning Algorithm", Friction, Vol. XX, Issue: X, Paper ID: XX, pp: 1-XX, 2021,