Identification of technical condition of valve clearance compensators using vibration measurement and machine learning

The dynamic development of internal combustion engine design and requirements for high reliability generates the need to apply a strategy of their operation based on the current technical condition. The paper concerns vibration diagnostics of automatic compensators of valve lash of a combustion engine. It presents the course of an active experiment conducted in order to develop a methodology for identifying the state of compensators based on measures of vibration signals measured at the engine head. Based on the results of the experiment, a classifier was developed in the form of a decision tree, which with high accuracy identified the technical condition of the compensators. The set of simple rules obtained thanks to the built up trees allows for easy implementation of the diagnostic system in practice.

---

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2020).