Method for determining service event of machine from sensor data
A system and method for data collection and frequency analysis with self-organization functionality includes analyzing with a processor a plurality of sensor inputs, sampling with the processor data received from at least one of the plurality of sensor inputs at a first frequency, and self-organizing with the processor a selection operation of the plurality of sensor inputs.
1. A computer-implemented method comprising:
collecting, from at least one sensor associated with a group of industrial machines, sensor data indicating at least one current health state indicator associated with at least one industrial machine of the group of industrial machines, wherein the at least one current health state indicator includes a fault condition of the at least one industrial machine;
providing the sensor data as an input to a neural network of a machine learning system, the neural network trained to determine the at least one current health state indicator based on patterns in the sensor data;
receiving the at least one current health state indicator as an output from the neural network of the machine learning system;
receiving, from the machine learning system, an indication of an additional sensor associated with the group of industrial machines from which to collect sensor data in order to diagnose the at least one current health state indicator; and
based on the at least one current health state indicator, determining a schedule of a service event, wherein the service event is associated with the at least one industrial machine, the determining the schedule of the service event including:
providing the at least one current health state indicator to the machine learning system, and receiving, from the machine learning system, the schedule of the service event,
wherein, based on iterative feedback, the machine learning system considers additional signals from at least one of the at least one sensor, the additional sensor, or another sensor to increase confidence in the fault condition.
2. The computer-implemented method of claim 1 , wherein the at least one current health state indicator is based on at least one of:
operational data associated with at least one industrial machine of the group of industrial machines,
failure data associated with at least one industrial machine of the group of industrial machines, or
a condition of at least one industrial machine that was detected in association with a maintenance activity associated with the group of industrial machines.
3. The computer-implemented method of claim 1 , wherein the iterative feedback includes a measure of success in predicting or anticipating fault states, and the neural network is trained on the iterative feedback.
4. The computer-implemented method of claim 1 , wherein the machine learning system tests different signals with different sensors of the at least one sensor until the fault condition is positively diagnosed before determining the schedule of the service event.
5. The computer-implemented method of claim 4 , wherein the iterative feedback includes a measure of success in predicting or anticipating fault states, and the neural network is trained on the iterative feedback.
6. The computer-implemented method of claim 1 , wherein the determining the schedule of the service event further comprises:
based on the at least one current health state indicator, determining a prediction of a future health state of the at least one industrial machine of the group of industrial machines, and
determining the schedule of the service event based on a predictive maintenance task, wherein the predictive maintenance task is based on the prediction of the future health state of the at least one industrial machine.
7. The computer-implemented method of claim 6 , wherein the prediction is generated by a predictive maintenance knowledge system that is configured to generate predictions of future health states of the group of industrial machines based on current health state indicators of the group of industrial machines.
8. A computer-implemented method comprising:
receiving at least one service event that is associated with at least one industrial machine of a group of industrial machines;
collecting, from at least one sensors associated with the group of industrial machines, sensor data indicating at least one current health state indicator associated with the at least one industrial machine of the group of industrial machines, wherein the at least one current health state indicator includes a fault condition of the at least one industrial machine;
training a neural network of a machine learning circuit to determine the at least one current health state indicator based on patterns in the sensor data;
determining, by the neural network of the machine learning circuit based on the patterns recognized in the sensor data, the at least one current health state indicator;
based on the at least one current health state indicator, determining a schedule of the at least one service event; and
based on iterative feedback, considering the sensor data with the machine learning circuit to improve confidence in the fault condition or the schedule of the at least one service event.
9. The computer-implemented method of claim 8 , wherein the determining the schedule of the at least one service event further comprises: receiving an updated schedule from the machine learning circuit, which has been trained to generate updated schedules for the group of industrial machines.
10. The computer-implemented method of claim 8 , wherein the determining the schedule of the at least one service event includes at least one of:
determining a performance of at least one task associated with the at least one service event,
determining a resource associated with the at least one service event,
determining a source of a resource associated with the at least one service event,
procuring a resource associated with the at least one service event,
arranging a delivery of a resource associated with the at least one service event, or
rating a performance of at least one task associated with the at least one service event.
11. The computer-implemented method of claim 8 , wherein the determining the schedule of the at least one service event further comprises:
based on the at least one current health state indicator, determining a prediction of a future health state of the at least one industrial machine associated with the group of industrial machines, and
determining the schedule of the at least one service event based on a predictive maintenance task, wherein the predictive maintenance task is based on the prediction of the future health state of the at least one industrial machine.
12. The computer-implemented method of claim 11 , wherein the prediction is generated by a predictive maintenance knowledge system that is configured to generate predictions of future health states of the group of industrial machines based on current health state indicators of the group of industrial machines.
13. The computer-implemented method of claim 12 , wherein the predictive maintenance knowledge system is configured to generate the predictions of the future health states of the group of industrial machines based at least one of:
a maintenance task associated with the at least one industrial machine,
a request to perform a service event associated with the at least one industrial machine, or
a recommendation for a service event associated with the at least one industrial machine.
14. The computer-implemented method of claim 8 , further comprising:
determining a service provider to perform the at least one service event; and
initiating a performance of the at least one service event by the service provider.
15. The computer-implemented method of claim 8 , further comprising:
determining a part of the at least one industrial machine that is associated with the at least one service event;
determining a part provider of the part; and
initiating a request for the part from the part provider.
16. The computer-implemented method of claim 8 , further comprising: presenting a user interface that includes an indicator of the at least one industrial machine of the group of industrial machines and an indicator of the schedule of the at least one service event.
17. The computer-implemented method of claim 8 , further comprising: presenting a user interface that includes a recommendation based on at least one of:
a preventive maintenance task associated with the at least one industrial machine of the group of industrial machines,
a reactive maintenance task associated with the at least one industrial machine of the group of industrial machines, or
a repair task associated with the at least one industrial machine of the group of industrial machines.
18. A computer-implemented method comprising:
for at least one industrial machine of a group of industrial machines,
collecting, from sensors associated with the group of industrial machines, at least one current health state indicator of the at least one industrial machine, wherein the at least one current health state indicator includes a fault condition of the at least one industrial machine; and
based on the at least one current health state indicator, determining a schedule of at least one service event for the at least one industrial machine, the determining the schedule of the at least one service event including providing the at least one current health state indicator of the at least one industrial machine to a machine learning circuit including a neural network; and
based on iterative feedback, using the machine learning circuit to consider the fault condition and sensor data from the sensors associated with the group of industrial machines before determining the schedule of the at least one service event; and
receiving, from the machine learning circuit, the schedule of the at least one service event.
19. The computer-implemented method of claim 18 , wherein the determining the schedule of the at least one service event further comprises: determining the schedule of the at least one service event associated with the at least one industrial machine based on a selected service event, wherein the selected service event is associated with another industrial machine of the group of industrial machines, and the another industrial machine is different than the at least one industrial machine.
20. The computer-implemented method of claim 18 , wherein the determining the schedule of the at least one service event further comprises:
providing the at least one current health state indicator of the at least one industrial machine to the machine learning circuit, which has been trained to determine the schedule of the at least one service event associated with the at least one industrial machine, and
receiving, from the machine learning circuit, the schedule of the at least one service event.
21. The computer-implemented method of claim 20 , further comprising: receiving, from the machine learning circuit, at least one of:
a preventive maintenance task associated with the at least one industrial machine,
a reactive maintenance task associated with the at least one industrial machine,
a repair task associated with the at least one industrial machine,
a request to perform a service event associated with the at least one industrial machine, or
a recommendation for a service event associated with the at least one industrial machine.
22. The computer-implemented method of claim 20 , further comprising:
receiving, from the machine learning circuit, a determination of a service provider to perform the at least one service event; and
initiating a performance of the at least one service event by the service provider.
23. The computer-implemented method of claim 20 , further comprising:
receiving, from the machine learning circuit, a determination of a part of the at least one industrial machine that is associated with the at least one service event;
determining a part provider of the part; and
initiating a request for the part from the part provider.
24. The computer-implemented method of claim 20 , further comprising: presenting a user interface that includes an indicator of the at least one industrial machine of the group of industrial machines and an indicator of the schedule of the at least one service event.
25. The computer-implemented method of claim 18 , wherein the iterative feedback includes a measure of success in predicting or anticipating fault states, and the neural network is trained on the iterative feedback.
26. The computer-implemented method of claim 18 , wherein the machine learning circuit is trained to determine the schedule of the at least one service event associated with the at least one industrial machine.