Monitoring machine operation with different sensor types to identify typical operation for derivation of a signature

1. A method comprising:

receiving sensor information from a primary sensor, the primary sensor positioned to receive information from a portion of an industrial operation;

receiving sensor information from one or more secondary sensors, the one or more secondary sensors are arranged to provide additional information about the portion of the industrial operation monitored by the primary sensor indicative of current operating conditions of the portion of the industrial operation;

using the sensor information from the one or more secondary sensors and a machine learning algorithm to determine if the portion of the industrial operation is operating in a normal or abnormal condition;

using the sensor information from the one or more secondary sensors to identify a transient condition and excluding sensor information of the primary sensor during the transient condition; and

in response to determining that the portion of the industrial operation is operating in a normal condition, using sensor information from the primary sensor during the determined normal operating condition to derive a primary sensor signature for the sensor information from the primary sensor.

2. The method of claim 1 , further comprising:

receiving additional sensor information from one of:

the primary sensor positioned to receive information from the portion of the industrial operation used to derive the primary sensor signature; and

another primary sensor positioned to receive information from a portion of an industrial operation similar to the portion of the industrial operation used to derive the primary sensor signature;

determining if the additional sensor information exceeds the primary sensor signature by a threshold amount; and

sending an alert in response to determining that the additional sensor information exceeds the primary sensor signature by the threshold amount.

3. The method of claim 2 , wherein the primary sensor signature comprises frequency information and determining if the additional sensor information exceeds the primary sensor signature by a threshold amount comprises determining if the additional sensor information at a frequency exceeds the primary sensor signature at the frequency by the threshold amount.

4. The method of claim 2 , wherein the portion of the industrial operation at least similar to the portion of the industrial operation used to derive the primary sensor signature is the portion of the industrial operation used to derive the primary sensor signature and the primary sensor used to receive the additional sensor information is the primary sensor used to derive the primary sensor signature.

5. The method of claim 1 , wherein the information from the primary sensor comprises a repeating pattern.

6. The method of claim 1 , further comprising:

receiving sensor information from the one or more secondary sensors indicative of a change to operation of the portion of the industrial operation monitored by the primary sensor wherein the change to the operation comprises an additional normal operating condition;

receiving additional sensor information from the primary sensor during the additional normal operating condition; and

using the additional sensor information from the primary sensor during the determined additional normal operating condition and using the machine learning algorithm to derive an additional primary sensor signature for the sensor information from the primary sensor without sending an alert of an abnormal operating condition.

7. The method of claim 1 , wherein the machine learning algorithm updates the primary sensor signature in response to:

receiving additional sensor information from the primary sensor and receiving additional sensor information from the one or more secondary sensors;

and/or receiving sensor information from a primary sensor and receiving sensor information from one or more secondary sensors positioned to receive information from a portion of an industrial operation at least similar to the portion of the industrial operation used to derive the primary sensor signature.

8. The method of claim 1 , wherein the primary sensor comprises one of a vibration sensor, an acoustic sensor, a pressure sensor and a flow sensor.

9. The method of claim 1 , wherein:

a secondary sensor of the one or more secondary sensors comprises one of a temperature sensor, a current sensor, a voltage sensor, a camera, a motion sensor, a vibration sensor, an acoustic sensor, a pressure sensor, a flow sensor and information from a programmable logic controller (“PLC”);

one or more of the secondary sensors are of a different type than the primary sensor;

one or more of the secondary sensors perform a different function than the primary sensor; and/or

one or more of the secondary sensors comprise a virtual sensor comprising a sensing function based on sensor information from two or more other sensors.

10. A component comprising:

a primary sensor module configured to receive sensor information from a primary sensor, the primary sensor positioned to receive information from a portion of an industrial operation;

a secondary sensor module configured to receive sensor information from one or more secondary sensors, the one or more secondary sensors are arranged to provide additional information about the portion of the industrial operation monitored by the primary sensor indicative of current operating conditions of the portion of the industrial operation;

a conditions module configured to use the sensor information from the one or more secondary sensors and a machine learning algorithm to determine if the portion of the industrial operation is operating in a normal or abnormal condition;

a transient exclusion module configured to use the sensor information from the one or more secondary sensors to identify a transient condition and wherein the signature module is configured to exclude sensor information of the primary sensor during the transient condition; and

a signature module configured to, in response to determining that the portion of the industrial operation is operating in a normal condition, use sensor information from the primary sensor during the determined normal operating condition to derive a primary sensor signature for the sensor information from the primary sensor,

wherein said modules comprise one or more of hardware circuits, a programmable hardware device, and executable code, the executable code stored on one or more computer readable storage media.

11. The component of claim 10 , further comprising:

an operations module configured to receive additional sensor information from one of:

the primary sensor positioned to receive information from the portion of the industrial operation used to derive the primary sensor signature; and

another primary sensor positioned to receive information from a portion of an industrial operation similar to the portion of the industrial operation used to derive the primary sensor signature;

a violation module configured to determine if the additional sensor information exceeds the primary sensor signature by a threshold amount; and

an alert module configured to send an alert in response to determining that the additional sensor information exceeds the primary sensor signature by the threshold amount.

12. The component of claim 10 , wherein the information from the primary sensor comprises a repeating pattern.

13. The component of claim 10 , wherein:

the secondary sensor module is further configured to receive sensor information from the one or more secondary sensors indicative of a change to operation of the portion of the industrial operation monitored by the primary sensor wherein the change to the operation comprises an additional normal operating condition;

the primary sensor module is configured to receive additional sensor information from the primary sensor during the additional normal operating condition; and

a new signature module is configured to use the additional sensor information from the primary sensor during the determined additional normal operating condition and use the machine learning algorithm to derive an additional primary sensor signature for the sensor information from the primary sensor without sending an alert of an abnormal operating condition.

14. The component of claim 10 , wherein the primary sensor comprises one of a vibration sensor, an acoustic sensor, a pressure sensor and a flow sensor and wherein:

a secondary sensor of the one or more secondary sensors comprises one of a temperature sensor, a current sensor, a voltage sensor, a camera, a motion sensor, a vibration sensor, an acoustic sensor, a pressure sensor, a flow sensor and information from a programmable logic controller (“PLC”);

one or more of the secondary sensors are of a different type than the primary sensor;

one or more of the secondary sensors perform a different function than the primary sensor; and/or

one or more of the secondary sensors comprise a virtual sensor comprising a sensing function based on sensor information from two or more other sensors.

15. A computer program product comprising a computer readable storage medium having program code embodied therein, the program code executable by a processor to:

receive sensor information from a primary sensor, the primary sensor positioned to receive information from a portion of an industrial operation;

receive sensor information from one or more secondary sensors, the one or more secondary sensors are arranged to provide additional information about the portion of the industrial operation monitored by the primary sensor indicative of current operating conditions of the portion of the industrial operation;

use the sensor information from the one or more secondary sensors and a machine learning algorithm to determine if the portion of the industrial operation is operating in a normal or abnormal condition;

using the sensor information from the one or more secondary sensors to identify a transient condition and excluding sensor information of the primary sensor during the transient condition; and

in response to determining that the portion of the industrial operation is operating in a normal condition, use sensor information from the primary sensor during the determined normal operating condition to derive a primary sensor signature for the sensor information from the primary sensor.

16. The computer program product of claim 15 , wherein the program code is further configured to:

receive additional sensor information from one of:

the primary sensor positioned to receive information from the portion of the industrial operation used to derive the primary sensor signature; and

another primary sensor positioned to receive information from a portion of an industrial operation similar to the portion of the industrial operation used to derive the primary sensor signature;

determine if the additional sensor information exceeds the primary sensor signature by a threshold amount; and

send an alert in response to determining that the additional sensor information exceeds the primary sensor signature by the threshold amount.

17. The computer program product of claim 15 , wherein the information from the primary sensor comprises a repeating pattern.

18. The computer program product of claim 15 , wherein the program code is further configured to:

receive information from the one or more secondary sensors indicative of a change to operation of the portion of the industrial operation during wherein the change to the operation comprises an additional normal operating condition;

receive additional sensor information from the primary sensor during the additional normal operating condition; and

use the additional sensor information from the primary sensor during the determined additional normal operating condition and use the machine learning algorithm to derive an additional primary sensor signature for the sensor information from the primary sensor without sending an alert of an abnormal operating condition.