Server and system for automatic selection of tags for modeling and anomaly detection

A server system comprising:

• at least one processor configured to be coupled to an industrial process;
• and
• a non-transitory computer-readable storage medium tangibly storing thereon program logic for execution by the processor, the program logic comprising:
• logic executable by the at least one processor for receiving at least one data communication from at least one industrial control system of the industrial process, the at least one data communication comprising a plurality of signals comprising a plurality of data values and associated plurality of tags received over time from the industrial process;
• and
• logic executable by the at least one processor that automatically associates tags defining a subsystem within the industrial process, wherein the automatic association is operated by at least a portion of the program logic executable by the at least one processor by:
• providing a curating service configured to generate a report of a general score and a per-user score based on correlated tags and display the report via at least one graphical user interface rendered on at least one digital display;
• calculating a time-weighted average over a specific time period for each of the plurality of signals;
• selecting a specific number of different day periods sampled from the plurality of tags and associated plurality of signals;
• performing a linear regression of plurality of signals over each period of the specific number of different day periods for each tag of the plurality of tags;
• calculating a slope and intercept of the linear regression for each period of the specific number of different day periods for each tag of the plurality of tags;
• for each of the plurality of tags, calculating a residual value of each data value of the plurality of signals over each period of the specific number of different day periods by subtracting an expected value calculated from the linear regression from an actual value of each data value of the plurality of signals over each period of the specific number of different day periods;
• for each of the plurality of tags, calculating a normalized value of the residual value for each data value of the plurality of signals over each period of the specific number of different day periods by subtracting a mean value of the residual values over each period of the specific number of different day periods from the residual value for each data value of the plurality of signals over each period of the specific number of different day periods and dividing the residual value for each data value of the plurality of signals over each period of the specific number of different day periods by its residual standard deviation;
• for each of the plurality of tags, calculating an absolute value of a dot product of the normalized residual values and the residual values for each tag of the remainder of a following number of tags of the plurality of tags;
• assigning each of the absolute values of the dot products calculated for the specific number of different day periods as a score;
• and
• defining the subsystem within the industrial process by recording a top number of scores and associated tag names for each tag of the plurality of tags so that the tag is associated with the top number of most closely associated tags and scores;
• wherein the server system is configured to use the top number of scores for each tag of the plurality of tags to automatically identify physical system relationships;
• wherein the server system is configured to use the physical system relationships to automatically select tags for modeling and anomaly detection;
• and
• wherein when an intelligent review of a report by the curating service reveals that the report concerns a component in a subsystem, the curating service ranks the report higher with respect to a report about a component in a different subsystem.