Contextualization for historians in industrial systems

• US 7,930,639 B2
• Filed: 09/26/2007
• Issued: 04/19/2011
• Est. Priority Date: 09/26/2007
• Status: Active Grant

• Current Assignee: Rockwell Automation Technologies Incorporated (Rockwell Automation, Inc.)
• Original Assignee: Rockwell Automation Technologies Incorporated (Rockwell Automation, Inc.)
• Inventors: Palmieri, Frank Anthony Jr., Baier, John J., Campbell, John T., McGreevy, Robert J., Herbst, Robert J., Jasper, Taryl J.
• Primary Examiner(s): Nguyen; Cao (Kevin)
• Application Number:
  US11/862,183
• Publication Number:
  US 20090083649A1
• Time in Patent Office: 1,301 Days
• Field of Search: 715/700, 715/748, 715762-765, 715740-743, 715851-853, 709223-224, 706/11, 706/37
• US Class Current: 715/763
• CPC Class Codes: G05B 17/02 electric
• CPC Class Codes: Y02P 90/80 Management or planning

Abstract

Systems and methods that discover relations and correlates among disparate pieces of data, to infer possible relationships between the industrial process and historian data/events to improve industrial operations. A correlation component can employ heuristic models to capture process data/event data, and can further include an implicit correlation component and an explicit correlation component. Accordingly, relations among various parameters can be discovered (e.g., dynamically) and proper corrective adjustments supplied to the industrial process.

Claims

1. An industrial automation system, comprising:

one or more processors;

a memory communicatively coupled to the one or more processors, the memory having stored therein computer-executable instructions configured to implement the system, including;

a plurality of embedded historians configured to collect data associated with an industrial process within an organization;

a hierarchical data model representing the organization and distributed across at least two system components within the organization, wherein the hierarchical data model exposes functionality of at least a first of the plurality of embedded historians to at least one of a second of the plurality of embedded historians or one or more of the at least two system components; and

a correlation component that infers relationships between events and operation of the industrial process based on the data, the correlation component comprising at least an implicit correlation component configured to infer relationships between causes of triggering events substantially in real-time.

2. The industrial automation system of claim 1, the correlation component further comprising an explicit correlation component that employs predetermined models to infer the relationships.

3. The industrial automation system of claim 1, the correlation component further comprising heuristic models.

4. The industrial automation system of claim 1, wherein the hierarchical data model comprises nodes that represent units associated with the organization.

5. The industrial automation system of claim 1, further comprising an artificial intelligence component configured to facilitate trend verification between the data and an outcome of the industrial process.

6. The industrial automation system of claim 1, further comprising a Human Machine Interface (HMI) configured to graphically display a view of the industrial automation system.

7. The industrial system of claim 1, wherein the plurality of embedded historians are configured to employ the organizational hierarchy data model to locate data to be collected.

8. The industrial system of claim 1, wherein the embedded historians are associated with at least one of a controller, a module in a chassis, a server, a sensor, or a factory component.

9. The industrial system of claim 1, wherein the hierarchical data model is configured to employ metadata associated with the data to determine a purpose of the data.

10. The industrial system of claim 1, further comprising a publish and subscribe component configured to instructs the plurality of embedded historians to record the data upon detection of a change to the data.

11. The industrial automation system of claim 1, wherein at least one of the plurality of embedded historians is configured to store at least one message that causes transition of a data value, and the implicit correlation component analyzes the at least one message to infer at least one of the relationships.

12. A method of inferring relationships in an industrial automation system comprising:

distributing a hierarchical data model of an industrial plant among at least two system components residing on one or more hierarchical levels of the industrial plant;

identifying a data type for collection by a plurality of embedded historians distributed among the at least two system components and associated with the hierarchical data model, the hierarchical data model exposing functionality of at least a first of the plurality of embedded historians to at least one of a second of the plurality of embedded historians or one or more of the at least two system components;

locating a subset of the plurality of embedded historians that have access to data of the data type;

acquiring data of the data type using the subset of the plurality of embedded historians; and

inferring a trend for operation of the industrial automation system based on the data.

13. The method of claim 12, further comprising employing heuristics models to infer the trend.

14. The method of claim 12, further comprising employing metadata to determine a relevance or function of data collected by the plurality of embedded historians.

15. The method of claim 12, further comprising employing a directory to locate a source of at least a subset of the data.

16. The method of claim 12, wherein the collecting the data comprises collecting the data across various levels of the industrial plant.

17. The method of claim 12, further comprising initiating at least one corrective action on the industrial process based at least in part on the trend.

18. An industrial controller system comprising:

collection means for collecting data related to at least one industrial process at multiple levels of an organization;

means for representing the organization as a hierarchical data model distributed among at least two devices comprising the organization, the hierarchical data model exposing functionality of the collection means to the at least two devices;

means for locating the collection means using the hierarchical data model; and

means for inferring trends of operation associated with the industrial controller system based on the data, wherein the means for inferring comprise means for inferring relationships between causes of triggering events substantially in real-time.