Physic-based vs data-based digital twins for bush bearing wear diagnostic

Industry is increasingly turning to predictive maintenance by using digital twins (DTs) follow and predict evolution of mechanical system. This article presents, compares discusses two DTs diagnose wear bush bearings under dynamic loads. The first DT driven a model based on data analysis statistical process control (SPC). second physical laws: Boussinesq's Archard's equations. Both are fed recorded test bench instrumented with sensors temperature, acceleration displacement. Rules for fault detection were identified, explained applied the implemented. implemented detected abnormal behaviours. data-based SPC easier implement it detects change in behaviour earlier. In contrast, physic-based has advantage being predictive, so can be used when only operating conditions system known. work contribution new diagnostic tools.