Systems and methods for predicting failure of electronic systems and assessing level of degradation and remaining useful life

A method for monitoring the health-state for electronic equipment, comprising:

• measuring current and voltage at an input and an output of the electronic equipment and acquiring data therefrom;
• storing the measured data in a memory;
• using a processor having access to the measured data, calculating performance metrics for the equipment;
• separating the measured data into a plurality of data classes;
• generating performance models for at least one data class;
• extracting diagnostic features from measured data values by comparing calculated performance metrics with the performance models, wherein extracting diagnostic features includes signal processing;
• and
• identifying the source and severity of a fault based upon the diagnostic features.

A method for monitoring the health-state for electronic equipment, comprising:

• measuring current and voltage at an input and an output of the electronic equipment and acquiring data therefrom, wherein said electronic equipment is an electric power converter used to convert power from one form to at least another form;
• storing the measured data in a memory
• using a processor having access to the measured data, calculating performance metrics for the equipment;
• separating the measured data into a plurality of data classes;
• generating performance models for at least one data class;
• extracting diagnostic features from measured data values by comparing calculated performance metrics with the performance models;
• and
• identifying the source and severity of a fault based upon the diagnostic features.

A method for monitoring the health-state for electronic equipment, comprising:

• measuring current and voltage at an input and an output of the electronic equipment and acquiring data therefrom, wherein said electronic equipment includes a plurality of electronic circuit components;
• storing the measured data in a memory;
• using a processor having access to the measured data, calculating performance metrics for the equipment;
• separating the measured data into a plurality of data classes;
• generating performance models for at least one data class;
• extracting diagnostic features from measured data values by comparing calculated performance metrics with the performance models;
• and
• identifying the source and severity of a fault based upon the diagnostic features.