Device control using processed sensor data corresponding to unexpected operations

1. A non-transitory computer-readable medium comprising computer-executable instructions that, when executed, are configured to cause a processor to perform operations comprising:

receiving a first dataset from a first industrial automation component, wherein the first dataset corresponds to a first processed result of a first raw sensor data acquired by a first sensor;

receiving a second dataset from a second industrial automation component, wherein the second dataset corresponds to a second processed result of a second raw sensor data acquired by a second sensor;

receiving data indicating an expected operation related to one or more operations of an industrial automation system including the first industrial automation component monitored by the first sensor and the second industrial automation component monitored by the second sensor;

identifying a signature based on the first dataset, the second dataset, or both, wherein the signature indicates an unexpected operation as compared to the expected operation;

performing a root cause analysis using the signature to determine a relationship indicating a first set of changes of the first dataset corresponding to a second set of changes in the data indicating the expected operation; and

sending a command to a device associated with the first sensor, the second sensor, or both based on the determined relationship, wherein the command is configured to cause the device to modify an operation.

2. The non-transitory computer-readable medium of claim 1 , wherein the first processed result, the second processed result, or both comprise a classification of one or more first portions of the first raw sensor data acquired by the first sensor, one or more second portions of the second raw sensor data acquired by the second sensor, or both.

3. The non-transitory computer-readable medium of claim 1 , wherein the device is configured to implement control logic to control the operation of the device based on a decision tree; and

wherein the instructions that, when executed, are configured to cause the processor to retrain one or more portions of the decision tree based on the modified operation.

4. The non-transitory computer-readable medium of claim 3 , wherein the instructions that, when executed, are configured to cause the processor to retrain the one or more portions of the decision tree comprises:

identifying one or more branches of the decision associated with the signature; and

updating one or more logic operations within the one or more branches based on the modified operation.

5. The non-transitory computer-readable medium of claim 1 , wherein the unexpected operation comprises a quality target corresponding to a defect of one or more products produced by the industrial automation system.

6. The non-transitory computer-readable medium of claim 1 , wherein the device is distinct from the first industrial automation component and the second industrial automation component.

7. The non-transitory computer-readable medium of claim 1 , wherein the device is configured to implement control logic to control the operation of the device based on a neural network; and

wherein the instructions that, when executed, are configured to cause the processor to retrain one or more portions of the neural network based on the modified operation.

8. The non-transitory computer-readable medium of claim 7 , wherein the instructions that, when executed, are configured to cause the processor to retrain the one or more portions of the neural network comprises:

identifying one or more property layers of associated with the signature; and

updating the identified one or more property layers based on the modified operation.

9. The non-transitory computer-readable medium of claim 1 , wherein the instructions that, when executed, are configured to cause the processor to perform operations comprising:

determining a machine component associated with the raw sensor data acquired by the first sensor;

receiving contextual information identifying a machine including the machine component;

determining a hierarchy group of the machine component based on the contextual information; and

transmitting a control signal to the sensor that causes the sensor to transmit data associated with the raw sensor data to an additional device configured to process the data, the raw sensor data, or both, based on the hierarchy group.

10. A method, comprising:

receiving, via a processor, a first dataset from a first industrial automation component, wherein the first dataset corresponds to a first processed result of a first raw sensor data acquired by a first sensor;

receiving, via the processor, a second dataset from a second industrial automation component, wherein the second dataset corresponds to a second processed result of a second raw sensor data acquired by a second sensor;

receiving, via the processor, data indicating an expected operation related to one or more operations of an industrial automation system including the first industrial automation component monitored by the first sensor and the second industrial automation component monitored by the second sensor;

identifying, via the processor, a signature based on the first dataset, the second dataset, or both, and the data indicating the expected operation, wherein the signature indicates an unexpected operation as compared to the expected operation;

performing, via the processor, a root cause analysis using the signature to determine a relationship indicating a first set of changes of the first dataset corresponding to a second set of changes in the data indicating the expected operation; and

sending, via the processor, a command to a device associated with the first sensor, the second sensor, or both based on the determined relationship, wherein the command is configured to cause the device to modify an operation.

11. The method of claim 10 , wherein the signature comprises a data profile of the first dataset, the second dataset, or both.

12. The method of claim 10 , comprising:

receiving, via the processor, a subsequent indication of the signature associated with the unexpected operation;

determining, via the processor, a portion of a model implemented by the device based on the signature; and

retraining, via the processor, the portion of the model.

13. The method of claim 10 , wherein the data indicating an expected operation comprises a model indicating an expected combination of the first dataset, the second dataset, or both, and a quality target associated with a production of one or more products of the industrial automation system.

14. The method of claim 10 , comprising determining, via the processor, that the signature corresponds to a new operating condition;

labeling the signature based on the new operating condition; and

generating the command based on the new operating condition prior to sending the command.

15. The method of claim 10 , wherein the unexpected operation comprises an image of a completed product of one or more products of the industrial automation system.

16. A non-transitory computer-readable medium comprising computer-executable instructions that, when executed, are configured to cause a processor to perform operations comprising:

receiving a first dataset from a first industrial automation component, wherein the first dataset corresponds to a first processed result of a first raw sensor data acquired by a first sensor;

receiving a second dataset from a second industrial automation component, wherein the second dataset corresponds to a second processed result of a second raw sensor data acquired by a second sensor;

receiving data indicating an expected operation related to one or more operations of an industrial automation system including the first industrial automation component monitored by the first sensor and the second industrial automation component monitored by the second sensor;

identifying a signature based on the first dataset, the second dataset, or both, and the data indicating the expected operation, wherein the signature indicates an unexpected operation as compared to the expected operation;

performing a root cause analysis using the signature to determine a relationship indicating a first set of changes of the first dataset corresponding to a second set of changes in the data indicating the expected operation;

sending a command to a device associated with the first sensor, the second sensor, or both based on the determined relationship, wherein the command is configured to cause the device to modify an operation;

retrieving a model, wherein the device associated with the first sensor, the second sensor, or both, is configured to implement control logical to control operation of the device based on the model; and

updating the model based on the result of the root cause analysis.

17. The non-transitory computer-readable medium of claim 16 , wherein the relationship indicates a direct correlation, an indirect correlation, a linear correlation, or a nonlinear correlation between the unexpected operation and the first dataset, the second dataset, or both.

18. The non-transitory computer-readable medium of claim 16 , wherein the first dataset, the second dataset, or both comprise a classification of the sensor data acquired by the first sensor, the sensor data acquired by the second sensor, or both.

19. The non-transitory computer-readable medium of claim 16 , wherein the first industrial automation component comprises a smart device distinct from the device.

20. The non-transitory computer-readable medium of claim 16 , wherein the first dataset is a label indicating an operational state of the device.