Method for providing a physically explainable fault information of a bearing by a fault detection model

A fault detection apparatus for providing physically explainable fault information of a bearing built in a machine by a fault detection model, comprising at least one processor configured to perform the steps:

• obtaining sensor data measured at the bearing as input data relating to an input data domain and a fault detection model which is trained on sensor data related to the input data domain to output a predicted failure value of the bearing by processing the obtained sensor data;
• mapping the measured sensor data from the input data domain to a selected data domain and resulting in an augmented fault detection model which outputs augmented predicted failure value related to the selected data domain, wherein the selected data domain has a physical meaning to a fault of the bearing;
• performing a feature attribution on the augmented fault detection model for the obtained sensor data, wherein a quantified importance of at least one individual feature of the input data to the augmented predicted failure value in the selected data domain is identified via said performing the feature attribution;
• displaying the at least one individual feature and the respective quantified importance in the selected data domain at a user interface;
• and
• in response to one quantified importance of the at least one individual feature being detected at a predefined frequency that is related to a root cause of the fault of the bearing, automatically outputting an alarm to a user interface, followed by taking measures comprising changing settings of the machine, stopping the machine, performing maintenance work on the machine, or combinations thereof.