Predicting fatigue of an asset that heals
The example embodiments are directed to a system and method which can predict a degradation in the health of an asset that heals based on data sensed from a machine or equipment operated by the asset that heals and in consideration of healing of the asset. In one example, a method may include one or more of storing time-series data of an operation of a machine, predicting a fatigue value of an operator of the machine via a predictive model that comprises a Rainflow counting algorithm that determines a degradation of the operator based on a changing attribute in the time-series data and a healing function that determines a healing component of the operator based on rest of the operator, and outputting information about the predicted fatigue value via a user interface.
1. A computing system comprising:
a storage configured to store time-series data of an operation of a machine; and
a processor configured to
predict a fatigue value of an operator of the machine via a predictive model that comprises a Rainflow counting algorithm that determines a degradation of the operator based on a changing attribute in the time-series data and a healing function that determines a healing component of the operator based on rest of the operator,
wherein the processor is configured to predict the fatigue value by applying a decay function representing healing of the operator to the degradation of the operator determined by the Rainflow counting algorithm, and the processor is further configured to output information about the predicted fatigue value via a user interface.
2. The computing system of claim 1 , wherein the predictive model is executed based on a rolling window of time-series data.
3. The computing system of claim 1 , wherein the machine comprises one of an aircraft, a locomotive, an automotive vehicle, a submersible, a ship, an equipment for use in drilling or mining, and a healthcare machine.
4. The computing system of claim 1 , wherein the time-series data comprises one or more of temperature, pressure, altitude, velocity, humidity, and noise, which are captured of the machine during operation.
5. The computing system of claim 1 , wherein the healing component comprises a reduction in the degradation of the operator based on the non-operating time of the machine.
6. The computing system of claim 1 , wherein the processor configured to output a graphical pattern of the predicted fatigue value over time which identifies the degradation and the healing component of the operator.
7. The computing system of claim 1 , wherein the processor is configured to determine the degradation of the operator based on magnitudes of oscillations in the time-series data of the machine.
8. A method comprising:
storing time-series data of an operation of a machine;
predicting a fatigue value of an operator of the machine via a predictive model that comprises a Rainflow counting algorithm that determines a degradation of the operator based on a changing attribute in the time-series data and a healing function that determines a healing component of the operator based on rest of the operator, wherein the predicting the fatigue value comprises applying a decay function representing healing of the operator to the degradation of the operator determined by the Rainflow counting algorithm; and
outputting information about the predicted fatigue value via a user interface.
9. The method of claim 8 , wherein the predictive model is executed based on a rolling window of time-series data.
10. The method of claim 8 , wherein the machine comprises one of an aircraft, a locomotive, an automotive vehicle, a submersible, a ship, an equipment for use in drilling or mining, and a healthcare machine.
11. The method of claim 8 , wherein the time-series data comprises one or more of temperature, pressure, altitude, velocity, humidity, and noise, which are captured of the machine during operation.
12. The method of claim 8 , wherein the healing component comprises a reduction in the degradation of the operator based on the non-operating time of the machine.
13. The method of claim 8 , wherein the outputting comprises outputting a graphical pattern of the predicted fatigue value over time which identifies the degradation and the healing component of the operator.
14. The method of claim 8 , wherein the predicting comprises determining the degradation of the operator based on magnitudes of oscillations in the time-series data of the machine.
15. A non-transitory computer-readable medium comprising instructions which when executed are configured to cause a computer to perform a method comprising:
storing time-series data of an operation of a machine;
predicting a fatigue value of an operator of the machine via a predictive model that comprises a Rainflow counting algorithm that determines a degradation of the operator based on a changing attribute in the time-series data and a healing function that determines a healing component of the operator based on rest of the operator, wherein the predicting the fatigue value comprises applying a decay function representing healing of the operator to the degradation of the operator determined by the Rainflow counting algorithm; and
outputting information about the predicted fatigue value via a user interface.
16. The non-transitory computer-readable medium of claim 15 , wherein the predictive model is executed based on a rolling window of time-series data.
17. The non-transitory computer-readable medium of claim 15 , wherein the outputting comprises outputting a graphical pattern of the predicted fatigue value over time which identifies the degradation and the healing component of the operator.
18. The non-transitory computer-readable medium of claim 15 , wherein the predicting comprises determining the degradation of the operator based on magnitudes of oscillations in the time-series data of the machine.