US 12,298,732 B2
Insight driven programming tags in an industrial automation environment
Jordan C. Reynolds, Austin, TX (US); John J. Hagerbaumer, Mequon, WI (US); Troy W. Mahr, Pleasant Prairie, WI (US); Thomas K. Jacobsen, Wake Forest, NC (US); and Giancarlo Scaturchio, Pisa (IT)
Assigned to Rockwell Automation Technologies, Inc., Mayfield Heights, OH (US)
Filed by Rockwell Automation Technologies, Inc., Mayfield Heights, OH (US)
Filed on Mar. 18, 2022, as Appl. No. 17/698,715.
Prior Publication US 2023/0297061 A1, Sep. 21, 2023
Int. Cl. G05B 19/05 (2006.01)
CPC G05B 19/056 (2013.01) [G05B 2219/13004 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system to train a machine learning model for implementation in an industrial automation environment, the system comprising:
a memory that stores executable components; and
a processor, operatively coupled to the memory, that executes the executable components, the executable components comprising:
a processing component configured to identify program tags that represent process variables in an industrial process, ingest feature vectors that represent the program tags, utilize machine learning algorithms to process the feature vectors and determine correlations between the program tags, and generate a correlation matrix that indicates the correlations between the program tags based on the utilization of the machine learning algorithms;
a design component configured to generate a control program configured for implementation by a programmable logic controller and select a program tag in the control program that represents a target variable in the industrial process, wherein the target variable comprises one of the process variables;
the processing component configured to access the correlation matrix, select ones of the program tags that represent independent process variables that are correlated with the selected program tag that represents the target variable in the industrial process based on the correlations indicated by the correlation matrix, and generate a recommendation that indicates the ones of the program tags; and
the design component configured to modify the control program using the ones of the program tags to optimize the target variable based on the correlations between the independent process variables represented by the ones of the program tags and the target variable represented by the selected program tag and transfer the control program for implementation by the programmable logic controller.