Remaining useful life predictions using digital-twin simulation model

A prediction system comprising:

• one or more sensors configured to generate parameter data related to a performance of an electro-mechanical element of a vehicle, wherein the parameter data is acquired during a historical period;
• and
• a processor in communication with the one or more sensors, and configured to:
• execute a digital-twin simulation model of the electro-mechanical element based on one or more estimated operating conditions of the electro-mechanical element to generate simulated behavior data that indicates expected behavior of the electro-mechanical element over time;
• generate deviation data that characterizes how the parameter data deviates from the simulated behavior data during the historical period, wherein the deviation data includes a deterministic component and a stochastic component;
• generate extrapolated deviation data by extrapolating the deterministic component and the stochastic component of the deviation data forward in time after the historical period;
• execute in the processor a deviation model to generate updated deviation data based on the extrapolated deviation data and the simulated behavior data;
• generate an updated digital-twin simulation model of the electro-mechanical element based on the updated deviation data;
• execute in the processor the updated digital-twin simulation model of the electro-mechanical element based on the one or more estimated operating conditions of the electro-mechanical element to generate updated simulated behavior data that indicates the expected behavior of the electro-mechanical element over time;
• calculate a remaining useful life of the electro-mechanical element in response to the updated simulated behavior data and the extrapolated deviation data;
• report the remaining useful life to a person associated with the vehicle;
• compare the updated simulated behavior data of the electro-mechanical element to one or more replacement thresholds;
• and
• based on the updated simulated behavior data of the electro-mechanical element exceeding at least one of the one or more replacement thresholds, output a recommendation to service or replace the electro-mechanical element on the vehicle.

A method for remaining useful life prediction comprising:

• generating parameter data related to a performance of an electro-mechanical element of a vehicle using one or more sensors, wherein the parameter data is acquired during a historical period;
• executing in a processor a digital-twin simulation model of the electro-mechanical element based on one or more estimated operating conditions of the electro-mechanical element to generate simulated behavior data that indicates expected behavior of the electro-mechanical element over time;
• generating deviation data that characterizes how the parameter data deviates from the simulated behavior data during the historical period, wherein the deviation data includes a deterministic component and a stochastic component;
• generating a plurality of extrapolated data sets by extrapolating the deterministic component and the stochastic component of the deviation data forward in time after the historical period using a plurality of extrapolation techniques;
• selecting extrapolated deviation data from one of the plurality of extrapolated data sets;
• executing in the processor a deviation model to generate updated deviation data based on the extrapolated deviation data selected from the one of the plurality of extrapolated data sets and the simulated behavior data;
• generating an updated digital-twin simulation model of the electro-mechanical element based on the updated deviation data;
• executing in the processor the updated digital-twin simulation model of the electro-mechanical element based on the one or more estimated operating conditions of the electro-mechanical element to generate updated simulated behavior data that indicates the expected behavior of the electro-mechanical element over time;
• calculating a remaining useful life of the electro-mechanical element in response to the updated simulated behavior data and the extrapolated deviation data;
• reporting the remaining useful life to a person associated with the vehicle;
• comparing the updated simulated behavior data of the electro-mechanical element to one or more replacement thresholds;
• and
• based on the updated simulated behavior data of the electro-mechanical element exceeding at least one of the one or more replacement thresholds, outputting a recommendation to service or replace the electro-mechanical element on the vehicle.