Machine health monitoring of rotating machinery

An auto-diagnosis method for executing a volume of machine health monitoring of one or more machine parts by an analysis of measurements of the one or more machine parts to detect defects therein as early as possible, the auto-diagnosis method implemented by an auto-diagnosis system comprising at least one sensor in communication with the one or more machine parts, at least one processor, and at least one memory storing thereon software for an auto-diagnosis alarm handling of the one or more machine parts, the software comprising at least an alarm level module being executable by the at least one processor to cause the auto-diagnosis system to implement the auto-diagnosis method comprising:

• setting, by the alarm level module, alarm levels for each measurement point of the one or more machine parts based on a mean of a group of condition indicator values and a predetermined number of standard deviations above the mean,
• detecting, by the alarm level module, an alarm with respect to the alarms levels and the measurements of the one or more machine parts provided by the at least one sensor,
• automatically adjusting, by the alarm level module, an alarm level corresponding to the alarm, and
• automatically adjusting, by the alarm level module, the alarm level to a new state based on the measurements respective to the alarm.