System and methods for automated plant asset failure detection

• US 9,535,808 B2
• Filed: 03/17/2014
• Issued: 01/03/2017
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: Aspentech Corporation
• Original Assignee: Mtelligence Corporation (Aspentech Corporation)
• Inventors: Macnab, Scott, Bates, Alexander B., Rahilly, Paul
• Primary Examiner(s): Ko, Chae
• Application Number:
  US14/217,265
• Publication Number:
  US 20140351642A1
• Time in Patent Office: 1,023 Days
• Field of Search: 714/26, 714/47.3
• US Class Current: 1/1
• CPC Class Codes: G05B 23/0229 knowledge based, e.g. expert systems; genetic algorithms
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/2257 using expert systems
• CPC Class Codes: G06N 20/00 Machine learning

Abstract

A system for performing failure signature recognition training for at least one unit of equipment. The system includes a memory and a processor coupled to the memory. The processor is configured by computer code to receive sensor data relating to the unit of equipment and to receive failure information relating to equipment failures. The processor is further configured to analyze the sensor data in view of the failure information in order to develop at least one learning agent for performing failure signature recognition with respect to the at least one unit of equipment.

Claims

1. A system for performing failure signature recognition training for at least one unit of equipment, the system comprising:

memory;

a network interface configured to;

receive a sensor data signal including sensor data relating to the unit of equipment;

receive a failure information signal including failure information relating to equipment failures; and

one or more processors in communication with the memory and the network interface, the one or more processors being configured by computer code to analyze the sensor data in view of the failure information in order to develop at least one learning agent for performing failure signature recognition with respect to the at least one unit of equipment wherein machine learning is used during a training phase to adjust parameters of the failure agent to successfully predict failures identified by the failure information and to bias the parameters to avoid at least one of false positive errors and false negative errors wherein the one or more processors are further configured by the computer code to store the sensor data within a memory wherein the sensor data covers a plurality of intervals and is stored along with metadata to flag ones of the plurality of intervals associated with failure intervals and ones of the plurality of intervals associated with normal intervals.

2. The system of claim 1 wherein the failure information relates to prior failures of the at least one unit of equipment.

3. The system of claim 1 wherein the failure information relates to prior failures of other equipment different from the at least one unit of equipment, the other equipment being functionally similar to the at least one unit of equipment.

4. The system of claim 1 wherein the failure information includes indications types, dates and codes associated with the equipment failures.

5. The system of claim 1 wherein the one or more processors are further configured by the computer code to receive an input identifying the at least one piece of equipment.

6. The system of claim 5 wherein the input identifies only the at least one unit of equipment.

7. The system of claim 5 wherein the input identifies a group of equipment including the unit of equipment.

8. The system of claim 5 wherein the input identifies a plurality of tags respectively corresponding to a plurality of sensors associated with the unit of equipment.

9. The system of claim 8 wherein the input further identifies at least one of minimum and maximum outlier removal settings for one or more of the plurality of sensors.

10. The system of claim 1 wherein the failure information includes indications of the equipment failures.

11. The system of claim 1 wherein the failure information comprises work order information, the one or more processors being further configured to identify faults in the work order information corresponding to the equipment failures and to classify the faults.

12. The system of claim 1 wherein the one or more processors are further configured by the computer code to analyze the sensor data at times of the one or more equipment failures in order to learn one or more fault signatures.

13. The system of claim 1 wherein the network interface is further configured to receive an additional sensor data signal including additional sensor data relating to an additional unit of equipment and to receive an additional failure information signal including additional failure information relating to one or more additional equipment failures;

wherein the one or more processors are further configured by the computer code to analyze the additional sensor data in view of the one or more additional equipment failures to learn one or more additional fault signatures.

14. The system of claim 13 wherein the one or more processors are further configured by the computer code to store the one or more additional fault signatures within the memory.

15. A system for performing operations relating to anomaly detection for at least one unit of equipment, the system comprising:

a memory;

a network interface configured to;

receive a sensor data signal including sensor data relating to the unit of equipment;

receive a failure information signal including failure information relating to one or more equipment failures;

one or more processors coupled to the memory, the one or more processors being configured by computer code to;

analyze the sensor data over time periods other than periods encompassing the one or more equipment failures to determine one or more normal operating states of the at least one unit of equipment;

train an anomaly agent to detect an anomaly when a current operating state of the at least one unit of equipment is outside of the one or more normal operating states wherein the anomaly agent utilizes one of a self-organizing map and a Restricted Boltzman Machine (RBM) to model the one or more normal operating states;

activate the anomaly agent to monitor additional sensor data relating to the unit of equipment;

provide the additional sensor data to the self-organizing map;

perform a comparison of an error associated with a classification of the additional sensor data by the self-organizing map to a maximum error; and

detect an anomaly condition based upon the comparison.

16. The system of claim 15 wherein the anomaly agent utilizes Gaussian probabilities to model the one or more normal operating states.

17. A system for performing failure signature recognition training for at least one unit of equipment, the system comprising:

memory;

a network interface configured to;

receive a sensor data signal including sensor data relating to the unit of equipment

receive a failure information signal including failure information relating to equipment failures; and

one or more processors in communication with the memory and the network interface, the one or more processors being configured by computer code to analyze the sensor data in view of the failure information in order to develop at least one learning agent for performing failure signature recognition with respect to the at least one unit of equipment wherein machine learning is used during a training phase to adjust parameters of the failure agent to successfully predict failures identified by the failure information and to bias the parameters to avoid at least one of false positive errors and false negative errors wherein the one or more processors are further configured by the computer code to tune a prediction interval over which the sensor data is evaluated by evaluating failure prediction accuracy over multiple interval durations.

18. A system for performing failure signature recognition training for at least one unit of equipment, the system comprising:

memory;

a network interface configured to;

receive a sensor data signal including sensor data relating to the unit of equipment;

receive a failure information signal including failure information relating to equipment failures; and

one or more processors in communication with the memory and the network interface, the one or more processors being configured by computer code to analyze the sensor data in view of the failure information in order to develop at least one learning agent for performing failure signature recognition with respect to the at least one unit of equipment wherein machine learning is used during a training phase to adjust parameters of the failure agent to achieve a desired potential failure to failure (P-F) interval.

19. The system of claim 18 wherein the desired P-F interval is a widest P-F interval that may be obtained in view of a desired level of failure prediction accuracy.

20. The system of claim 18 wherein the one or more processors are further configured by the computer code to bias the parameters to avoid at least one of false positive errors and false negative errors.