US-12625485-B2 - Dynamic industrial artificial intelligence configuration and tuning

US12625485B2US 12625485 B2US12625485 B2US 12625485B2US-12625485-B2

Abstract

Various systems and methods are presented regarding monitoring and controlling operation of a process. A visual representation of the process can be created based on a supermodel comprising models (representing one or more devices) and nodes (representing respective device variables and constraints). Further, the process can be represented by levels, wherein devices at each level can be self-aware and have onboard artificial intelligence, such that a device at any level can auto-configure itself in accordance with a requirement placed upon it. Field-level devices (IFLDs) can be smart devices which auto-configure based upon a requirement from a higher-level device. Accordingly, system awareness can be incorporated across all levels of the process enabling overall and device-specific optimization of the process. IFLDs can auto-configure to collect and transmit data in accordance with an instruction from a higher-level device, leading to efficient data collection, reduced data bandwidth/processing, and expedited system optimization.

Inventors

Bijan Sayyarrodsari
Cyril Perducat
Dan Li

Assignees

ROCKWELL AUTOMATION TECHNOLOGIES, INC.

Dates

Publication Date: 20260512
Application Date: 20230801

Claims (20)

1 . A system comprising a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory, wherein the computer executable components comprise: a configuration component configured to construct a graphical representation of an industrial process, wherein the graphical representation comprises a group of models, a first model of the group of models represents operation of a device of the industrial process according to a first configuration of the device, and the first model generates an output based on a value of output data received from the device; a visualization component configured to present the graphical representation of the process on a human-machine interface (HMI); and an artificial intelligence (AI) component configured to, in response to determining that a value of the output does not satisfy a constraint applied to the output: determine a second model, from among multiple models stored in a model repository, capable of representing the device and producing an output that satisfies the constraint, replace the first model with the second model in the graphical representation, and send configuration data defined by the second model to the device, wherein sending the configuration data to the device causes the device to operate in accordance with a second configuration associated with the second model.
2 . The system of claim 1 , wherein the output of the first model comprises a value of a parameter generated by at least one function included in the first model.
3 . The system of claim 1 , wherein the AI component is further configured to: in response to determining that the value of the output does not satisfy the constraint: generate a notification that operation of the device represented by the first model does not satisfy the constraint.
4 . The system of claim 3 , wherein the visualization component is further configured to present the notification on the HMI.
5 . The system of claim 3 , wherein the AI component is further configured to: send the configuration data to the device with an instruction to apply the configuration data to the device to control operation of the device.
6 . The system of claim 5 , wherein the configuration component is further configured to: instruct the device to transmit a current operating configuration of the device; receive the current operating configuration from the device; perform a comparison of the current operating configuration of the device with the second configuration; and confirm that the current operating configuration of the device matches the second configuration based on a result of the comparison.
7 . The system of claim 3 , wherein the device is one of a sensor, an actuator, a valve, an industrial controller, a motor drive, a telemetry device, a meter, a smart device, a device configured to monitor operation of a component/equipment included in the industrial process, or a device configured to control operation of a component/equipment included in the industrial process.
8 . The system of claim 1 , wherein the AI component is further configured to record an operation conducted on the HMI, wherein the operation is an interaction with the graphical representation, and the interaction with the graphical representation comprises at least one of adding a model to the group of models, removing a model from the group of models, associating a device configuration with a model of the group of models, adding a device, removing a device, implementing a configuration at a device included in the industrial process, responding to a notification, selecting data to store, selecting information to store, selecting a layout representation of the graphical representation, selecting a model representation of the graphical process, selecting a node associated with a model of the group of models, adjusting a constraint to a model of the group of models, connecting a first model to a second model, selecting a model type, or constructing a model.
9 . The system of claim 8 , wherein the AI component is further configured to replicate the interaction with the graphical representation based on an interpretation of the operation.
10 . The system of claim 8 , wherein the graphical representation of the industrial process comprises at least one of: a real-time version presenting a real-time graphical representation of the industrial process; or an offline version presenting a graphical representation of the industrial process based on at least one of real-time data received from one or more devices operating in the industrial process or historical data captured from a prior operation of the one or more devices operating in the industrial process.
11 . The system of claim 10 , wherein user interaction with the graphical representation is via the offline version.
12 . The system of claim 11 , wherein the configuration component is further configured to review an authorization to determine whether authorization exists prior to implementing, to the real-time version, a model or configuration represented on the offline version.
13 . The system of claim 1 , wherein at least one of the first model or the second model comprises a parametric model, a parametric hybrid model, a linear model, a non-linear model, a kinetic model, a first principles reasoning model, a solver, a historical data model, a cost function analysis model, a regression cost function model, a binary classification cost function model, a multi-class classification cost function model, a mixed-integer non-liner program model, a deep learning-based model, a backpropagation model, a static backpropagation model, a recurrent backpropagation model, a gradient computation model, a chain rule model, an error determination model, or a mathematical model configured to represent operation of a component in the process, and the component is a device, a group of devices, or a component block.
14 . The system of claim 1 , wherein the output generated by the first model is provided to a third model of the group of models as an input to the third model, and the constraint applied to the output is defined by the third model.
15 . A computer-implemented method for visualizing an industrial process, comprising: constructing a graphical representation of an industrial process, wherein the graphical representation includes a first model representing operation of a device of the industrial process according to a first configuration of the device, and the first model generates an output based on a value of output data generated by the device; monitoring the value of the output data received from the device represented by the first model; and in response to determining that the value of the output data does not satisfy a constraint applied to the output: identifying a second model, from among multiple models stored in a model repository, capable of representing the device and producing an output that satisfies the constraint; replacing the first model with the second model in the graphical representation; and sending configuration data defined by the second model to the device, wherein the sending of the configuration data to the device configures the device to operate in accordance with a second configuration associated with the second model.
16 . The computer-implemented method of claim 15 , further comprising presenting, on a human-machine interface (HMI), the graphical representation.
17 . The computer-implemented method of claim 15 , wherein at least one of the first model or the second model is one of a parametric model, a parametric hybrid model, a linear model, a non-linear model, a kinetic model, a first principles reasoning model, a solver, a historical data model, a cost function analysis model, a regression cost function model, a binary classification cost function model, a multi-class classification cost function model, a mixed-integer non-liner program model, a deep learning-based model, a backpropagation model, a static backpropagation model, a recurrent backpropagation model, a gradient computation model, a chain rule model, an error determination model, or a mathematical model configured to represent operation of a component in the industrial process, and the component is a device, a group of devices, or a component block.
18 . A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a process to cause the processor to: construct a graphical representation of an industrial process, wherein the graphical representation includes a first model representing operation of a device of the industrial process according to a first configuration of the device; monitor the value of the output data received from the device represented by the first model; and in response to determining that the value of the output data does not satisfy a constraint applied to the output: identify a second model, from among multiple models stored in a model repository, capable of representing the device and producing an output that satisfies the constraint; replace the first model with the second model in the graphical representation; and send configuration data defined by the second model to the device, wherein sending the configuration data to the device configures the device to operate in accordance with a second configuration associated with the second model.
19 . The computer program product of claim 18 , wherein the program instructions executable by the processor further cause the processor to: in response to the determining that the value of the output data does not satisfy a constraint applied to the output: generate a notification that operation of the device represented by the first model does not satisfy the constraint; and present, on a human machine interface, the graphical representation and the notification.
20 . The computer program product of claim 19 , wherein the program instructions executable by the processor further cause the processor to, in response to determining that the value of the output data does not satisfy the constraint: transmit an instruction for the device to apply the configuration data to the device to control operation of the device.

Description

TECHNICAL FIELD This application relates to techniques facilitating at least one of representation, monitoring, or controlling operation of a process. BACKGROUND Traditional approaches for optimizing an enterprise-wide system typically involves complex, and potentially tedious, mathematical programming that can become daunting, if not impracticable, as the scale/scope of the enterprise system (e.g., a manufacturing plant) increases. Hence, the ability for a conventional enterprise system to meet one or more objectives and/or requirements can be problematic in itself, even to the point of being unachievable. A problem with conventional enterprise systems (e.g., an automation system) is the enterprise system can be highly dependent on complex metadata models utilizing data from disparate sources that oftentimes result in a non-explainable highly complex model(s). Further, the sheer volume of data being generated, transmitted, and/or processed across an automated process can be unwieldy (e.g., the volume of sensor data generated during operation of a die-casting plant comprising furnace/melting equipment, molten metal delivery equipment, die cast machine, molten metal injection system, die-clamping equipment, casting ejection and extraction equipment, die spraying equipment, and the like) and can place a significant operational burden on the devices/systems operating in the higher levels of the enterprise model (e.g., programmable logic controllers (PLCs) responding to the received sensor data). Further, in a conventional system, the plethora of data is largely useless until it has been processed by the higher-level devices. Furthermore, the data may not include data of interest to a higher-level component in the enterprise system, the data may not be in a format for use by a higher-level system, and suchlike. Accordingly, the structure and complexity of conventional enterprise systems render a response to a simple requirement (e.g., reduce scrap) to be potentially unachievable by the very model designed and configured to achieve the requirement. The above-described background is merely intended to provide a contextual overview of some current issues and is not intended to be exhaustive. Other contextual information may become further apparent upon review of the following detailed description. SUMMARY The following presents a summary to provide a basic understanding of one or more embodiments described herein. This summary is not intended to identify key or critical elements, or delineate any scope of the different embodiments and/or any scope of the claims. The sole purpose of the summary is to present some concepts in a simplified form as a prelude to the more detailed description presented herein. In one or more embodiments described herein, systems, devices, computer-implemented methods, methods, apparatus and/or computer program products are presented that facilitate monitoring and/or controlling operation of equipment in a process, wherein the equipment is monitored and/or controlled by a device that is configured with on-board intelligence to enable real-time adjustment of the device and/or the equipment being monitored in accordance with changing operating conditions at the process. The system can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise a configuration component configured to construct a graphical representation of an industrial process, wherein the graphical representation comprises a group of models. The computer executable components can further comprise a visualization component configured to present the graphical representation of the process on a human-machine interface (HMI). In an embodiment, the group of models can comprise a first model representing operation of a first sub-group of devices in the industrial process. In an embodiment, the first model can comprise a first input and a first output, wherein the first output comprises a parameter generated by at least one function included in the first model, wherein a constraint has been applied to the parameter. The computer executable components can further comprise an artificial intelligence (AI) component configured to monitor output data received from a device represented by the first model and determine if a value of the output data associated with the parameter does not match the constraint. The AI component can be further configured to, in response to determining the output data does not match the constraint, modifying the first model to generate at least one of a parameter or a constraint that matches the value of the output data, or generating a notification that operation of the device represented by the first model does not match the constraint. In an embodiment, the visualization component can be further configured to present the notification on the HMI. In a further embodiment,