Anomaly detection using root cause analysis in a probabilistic multi-component calibrated model

A machine-readable tangible medium embodying information indicative of instructions for causing one or more machines to perform operations comprising:

• providing, on a computer system, a digital model of an industrial plant, the digital model having numerical parameters associated with components of the industrial plant, each numerical parameter describing a relationships between an input and an output of a component, each numerical parameter having a nominal range;
• selecting a set of values for a first parameter of the numerical parameters, the set of values extending below, within, and above the nominal range of the first parameter, a frequency of selected values peaking at values within the nominal range of the first parameter;
• executing the digital model using each of the selected set of values, the executing occurring with nominal plant inputs and while all numerical parameters other than the first parameter are within respective nominal ranges;
• associating each of the selected set of values with a related output of the executed digital model;
• distinguishing i) related outputs associated with values within the nominal range of the first parameter from ii) related outputs associated with values extending below or above the nominal range of the first parameter;
• and
• saving the associations in a database.

A system for troubleshooting a process in an industrial plant, the system comprising:

• a memory;
• and
• at least one processor operatively coupled with the memory and executing program code from the memory for:
• providing, on a computer system, a digital model of an industrial plant, the digital model having numerical parameters associated with components of the industrial plant, each numerical parameter describing a relationships between an input and an output of a component, each numerical parameter having a nominal range;
• selecting a set of values for a first parameter of the numerical parameters, the set of values extending below, within, and above the nominal range of the first parameter, a frequency of selected values peaking at values within the nominal range of the first parameter;
• executing the digital model using each of the selected set of values, the executing occurring with nominal plant inputs and while all numerical parameters other than the first parameter are within respective nominal ranges;
• associating each of the selected set of values with a related output of the executed digital model;
• distinguishing i) related outputs associated with values within the nominal range of the first parameter from ii) related outputs associated with values extending below or above the nominal range of the first parameter;
• and
• saving the associations in a database.