Data mining of unfiltered controller data

• US 7,827,122 B1
• Filed: 03/09/2006
• Issued: 11/02/2010
• Est. Priority Date: 03/09/2006
• Status: Expired due to Fees

• Current Assignee: Rockwell Automation Technologies Incorporated (Allen-Bradley Co., Inc.)
• Original Assignee: Rockwell Automation Technologies Incorporated (Allen-Bradley Co., Inc.)
• Inventors: Miller, Scott A., Sadowski, Randall P., Campbell, John T. Jr., Glavach, Mark A., Overstreet, Keith A., Sturrock, David T.
• Primary Examiner(s): Sparks; Donald
• Assistant Examiner(s): Bharadwaj; Kalpana
• Application Number:
  US11/371,433
• Time in Patent Office: 1,699 Days
• Field of Search: 706/12, 705/37, 707/1
• US Class Current: 706/12
• CPC Class Codes: G05B 17/02 electric
• CPC Class Codes: G05B 23/0232 based on qualitative trend analysis, e.g. system evolution
• CPC Class Codes: G05B 23/0264 Control of logging system, e.g. decision on which data to store; time-stamping measurements
• CPC Class Codes: G06Q 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange

Abstract

A data analysis system for utilization in an industrial environment comprises an aggregation component that aggregates data from multiple controllers on a factory floor. An analysis component performs data mining on the aggregated data to determine one of correlations between variables within the aggregated data and trends within the aggregated data. For example, the controllers can be programmable logic controllers, robotic controllers, and the like.

Claims

1. A data analysis system comprising the following computer-executable components:

an aggregation component that aggregates data from multiple controllers on a factory floor to yield aggregated data;

an analysis component that automatically performs data mining on the aggregated data and determines at least one of correlations between variables within the aggregated data or trends within the aggregated data; and

a simulation engine that simulates at least one industrial process associated with at least a subset of the multiple controllers and generates at least one prediction relating to future operation of the at least one industrial process, wherein results of the data mining undertaken by the analysis component are employed to update the simulation engine.

2. The system of claim 1, the multiple controllers comprise one or more of at least one programmable logic controller or at least one robotic controller.

3. The system of claim 1, further comprising an updating component that automatically updates a control process based at least in part upon results of the data mining.

4. The system of claim 1, the analysis component comprises a filter component that determines a time range for data to be analyzed based upon time of creation of the data, the analysis component performing data mining on data within the time range determined by the filter component.

5. The system of claim 4, wherein the filter component is associated with a set of rules defining time ranges corresponding to respective events associated with the industrial process, the rules employed to determine the time range for data to be analyzed given an event for which analysis is being performed.

6. The system of claim 1, further comprising a security component that ensures that an individual requesting access to the aggregated data is authorized for the requested access.

7. The system of claim 1, further comprising:

a data repository that retains at least a portion of the aggregated data; and

an audit component that accesses the data repository and generates an audit trail with respect to at least a portion of the aggregated data.

8. The system of claim 7, further comprising an identifier component that determines identities of controllers associated with aggregated data, the audit component configured to generate an audit trail with respect to a particular controller.

9. The system of claim 1, data received from the multiple controllers conforms to a hierarchically structured data model.

10. The system of claim 9, the hierarchically structured data model is based at least in part upon at least one of ISA S88 or ISA S95.

11. The system of claim 1, further comprising an input expander component that automatically determines data to be provided to the analysis component from data retained by the controllers upon occurrence of a predefined event.

12. The system of claim 1, the analysis component comprises a data fusion processor that aids in performing data mining on the aggregated data.

13. The system of claim 1, further comprising a forecasting component that forecasts an event based at least in part upon results of the data mining.

14. The system of claim 1, further comprising a schedule component that determines a time to schedule maintenance of a device based at least in part upon results of the data mining.

15. The system of claim 14, wherein the scheduling component combines the results of the data mining with information regarding at least one of availability of maintenance personnel, current work orders, or identity of maintenance personnel to yield combined data, and determines the time to schedule maintenance of the device based on the combined data.

16. The system of claim 1, wherein the analysis component determines actual source of an error or an alarm based upon data mining on the aggregated data from multiple controllers.

17. The system of claim 1, wherein the aggregation component determines a data storage capacity associated with each of the controllers.

18. The system of claim 17, wherein the aggregation component pulls data from a controller upon utilization of a threshold percentage of data storage capacity of a controller.

19. The system of claim 1, wherein the results of the data mining comprise at least a trend detected in the industrial process, and the simulation engine is updated in accordance with the trend.

20. A methodology for analyzing an industrial process, comprising:

configuring a data repository to receive unfiltered data from a plurality of controllers;

configuring at least one processing entity to automatically search at least a subset of the received data to locate at least one of patterns or trends in the data;

employing a simulation engine to simulate the industrial process and to generate at least one prediction relating to future operation of the industrial process;

providing the at least one of patterns or trends to the simulation; and

configuring the simulation engine to utilize the at least one of the patterns or trends in connection with updating the simulation engine.

21. The methodology of claim 20, further comprising configuring the processing entity to determine correlations between variables within the data.

22. The methodology of claim 20, wherein the configuring at least one processing entity comprises:

defining a set of rules defining time ranges corresponding to respective events associated with the industrial process; and

configuring the at least one processing entity to search a subset of the unfiltered data corresponding to a time range associated with an event being analyzed in accordance with the set of rules.

23. The methodology of claim 20, further comprising associating the processing entity with a security application, the security application configured to prohibit access to results of the data mining and data mining algorithms to authorized individuals.

24. The methodology of claim 20, further comprising generating an audit trail based upon data received from the plurality of controllers, at least one of the plurality of controllers is a programmable logic controller.

25. The methodology of claim 20, the processing entity is a data fusion processor.

26. A methodology for analyzing a control process, comprising:

simulating the control process using a simulation engine that generates at least one prediction relating to future operation of the industrial process;

aggregating raw data from a plurality of programmable logic controllers to yield aggregated data;

automatically performing data mining on the aggregated data to determine correlations between variables within the aggregated data;

determining a root cause of an alarm based upon the data mining on the aggregated data;

automatically updating the control process based at least in part upon results of the data mining;

providing results of the data mining to the simulation engine; and

updating the simulation engine in accordance with the results of the data mining.