Systems and methods of operational anomaly detection

A non-transitory, tangible, computer-readable medium storing instructions that, when executed by processing circuitry, cause the processing circuitry to perform operations comprising:

• receiving an indication of an industrial automation device, wherein the industrial automation device corresponds to an asset that is targeted for operational baseline device modeling;
• determining an application of the industrial automation device;
• determining a configuration of the asset based on the application;
• determining an operating range of the asset;
• generating a plurality of device models corresponding to the asset based on the operating range, the application, and the configuration;
• storing the plurality of device models in memory in association with a plurality of indications of a plurality of process states, a plurality of operational parameters associated with the operating range, and a plurality of training status indications, wherein the industrial automation device is configured to operate in different process states of the plurality of process states;
• receiving acquired data from the industrial automation device, wherein the acquired data indicates a current operation of the asset;
• identifying a device model of the plurality of device models based on the acquired data, wherein the device model corresponds to a process state of the plurality of process states, an operational parameter of the plurality of operational parameters, and a training status indication of the plurality of training status indications;
• determining that the training status indication indicates that training of the device model has yet to be completed;
• and
• training the device model based on at least some of the acquired data based on that the training of the device model has yet to be completed.

A non-transitory, tangible, computer-readable medium storing instructions that, when executed by processing circuitry, cause the processing circuitry to perform operations comprising:

• receiving acquired data from an industrial automation device, wherein the acquired data indicates a current operation of an asset, wherein the asset is configured to operate in different process states of a plurality of process states while handling respective physical materials of a plurality of physical materials;
• accessing a plurality of device models associated with a plurality of indications of the plurality of process states, a plurality of indications of a plurality of speeds, and a plurality of training status indications, wherein each device model of the plurality of device models corresponds to a respective process state of the plurality of process states, a respective speed of the plurality of speeds, and a respective training status indication of the plurality of training status indications;
• determining a process state of the plurality of process states based on the acquired data;
• identifying a device model of the plurality of device models based on the process state and the acquired data, wherein the device model corresponds to a speed range of the plurality of speeds and a training status indication of the plurality of training status indications;
• determining that the device model is trained based on the training status indication, wherein the device model corresponds to an indication of a baseline operation while trained;
• comparing at least some of the acquired data to the baseline operation to determine a deviation between the baseline operation and the current operation;
• and
• generating, based on the deviation, one or more control signals to implement an adjustment to the industrial automation device.