Industrial automation edge as a service
A cloud-based edge-as-as-service (EaaS) system allows edge gateways to be easily configured and deployed on the cloud for collection, contextualization, and egress of industrial data to downstream applications, including analytic applications, work order management systems, or visualization systems. The EaaS system uses predefined device profiles to automatically discover relevant data items on plant floor devices and present these data items to a user. Model configuration interfaces served by the EaaS system allow the user to map selected data items to predefined models for organizing or contextualizing the selected data, and for egressing the contextualized data to the target applications. These model configurations can be deployed on the cloud platform as edge gateways that use the resulting information models to collect and contextualize relevant items of device data during runtime and to export the modeled data to the target application.
1 . A 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 user interface component configured to
render configuration interfaces on a client device and to receive, via interaction with the configuration interfaces, a selection of a device profile, from a library of device profiles, corresponding to an industrial asset and defining data tags available on the industrial asset that are relevant to an analytic objective,
render, based on the device profile, a list of the data tags, and
receive, via interaction with the configuration interfaces, mapping input that binds a selected subset of the data tags from the list to an analytic model selected from a library of analytic models, wherein the analytic model corresponds to a target application and defines data mappings to the target application; and
a modeling component configured to generate an information model based on the mapping input, wherein the information model defines a contextualization of the selected subset of the data tags and a mapping of the selected subset of the data tags to the target application based on the data mappings defined by the analytic model,
wherein the system is configured to instantiate an edge gateway comprising the information model on a cloud platform, the edge gateway comprising:
an edge gateway component configured to collect device data from the selected subset of the data tags specified by the information model, and
an egress component configured to contextualize the device data based on the information model to yield modeled data, and to send the modeled data to the target application based on the mapping defined by the information model.
2 . The system of claim 1 , wherein the edge gateway executes on the cloud platform as a virtual machine.
3 . The system of claim 1 , wherein
binding of the selected subset of the data tags to the analytic model maps the selected subset of the data tags to the target application in accordance with the data mappings defined by the analytic model.
4 . The system of claim 1 , wherein
the analytic model defines an organization of data, and
the egress component is configured to model the device data in accordance with the organization defined by the analytic model to yield the modeled data.
5 . The system of claim 1 , wherein
the device profile defines a modeling of the device data collected from the selected subset of the data tags, and
the modeling component configured to generate the information model further based on the modeling defined by the device profile.
6 . The system of claim 1 , wherein the target application is at least one of an analytic system that applies analytics to the modeled data, a predictive maintenance application that predicts industrial asset failures or performance issues based on analysis of the modeled data, a work order management system that generates maintenance work orders in response to detection of a performance issue based on analysis of the modeled data, or a visualization system that displays the modeled data.
7 . The system of claim 1 , wherein
the edge gateway further comprises an analytics component configured to apply a machine learning model to the device data or to the modeled data to yield an analytic result, and
the egress component is configured to send the analytic result to the target application.
8 . The system of claim 7 , wherein the machine learning model is a predictive maintenance model designed to predict a failure of an industrial assets based on learned trends in the device data or the modeled data.
9 . The system of claim 1 , wherein contextualization of the device data based on the information model comprises at least one of
organization of the device data based on a hierarchical organization of industrial assets defined by the information model, or
addition of contextual metadata to the device data that defines functional or mathematical relationships between the data tags.
10 . A method, comprising:
rendering, by a system comprising a processor and executing on a cloud platform, a configuration interface on a client device;
receiving, by the system via interaction with the configuration interface, a selection of a device profile, from a library of device profiles, corresponding to an industrial asset, wherein the device profile defines data tags available on the industrial asset that are relevant to an analytic objective;
rendering, by the system based on the device profile, a list of the data tags on the user interface;
receiving, by the system via interaction with the configuration interface, mapping input that selects a subset of the data tags from the list to be bound to an analytic model selected from a library of analytic models, wherein the analytic model corresponds to a target application and defines data mappings to the target application;
generating, by the system, an information model based on the mapping input, wherein the information model defines a contextualization of the subset of the data tags and a mapping of the subset of the data tags to the target application based on the data mappings defined by the analytic model;
deploying, by the system, an edge gateway comprising the information model on a cloud platform, wherein the deploying comprises:
collecting, by the edge gateway, device data from the subset of the data tags in accordance with the information model,
contextualizing, by the edge gateway, the device data based on the information model to yield modeled data, and
sending, by the edge gateway, the modeled data to the target application based on the mapping of the subset of the data tags to the target application defined by the information model.
11 . The method of claim 10 , wherein the deploying further comprises executing the edge gateway as a virtual machine on the cloud platform.
12 . The method of claim 10 , wherein
the rendering comprises
discovering, by the system, the data tags on the industrial asset based on the device profile.
13 . The method of claim 10 , wherein
the receiving of hem aping input maps the subset of the data tags to the target application in accordance with the data mappings defined by the analytic model.
14 . The method of claim 10 , wherein
the analytic model defines an organization of data, and
the deploying of the edge gateway further comprises modeling the device data in accordance with the organization defined by the analytic model to yield the modeled data.
15 . The method of claim 10 , wherein
the device profile defines a modeling of the device data collected from the subset of the data tags, and
the deploying further comprises generating the information model further based on the modeling defined by the device profile.
16 . The method of claim 10 , wherein sending comprises sending the modeled data to at least one of an analytic system that applies analytics to the modeled data, a predictive maintenance application that predicts industrial asset failures or performance issues based on analysis of the modeled data, a work order management system that generates maintenance work orders in response to detection of a performance issue based on analysis of the modeled data, or a visualization system that displays the modeled data.
17 . The method of claim 10 , wherein the deploying further comprises:
applying, by the edge gateway, a machine learning model to the device data or to the modeled data to yield an analytic result, and
sending the analytic result to the target application.
18 . A non-transitory computer-readable medium having stored thereon instructions that, in response to execution, cause a system comprising a processor and executing on a cloud platform to perform operations, the operations comprising:
rendering a configuration interfaces on a client device;
receiving, via interaction with the configuration interface, a selection of a device profile, from a library of device profiles, corresponding to an industrial device, wherein the device profile defines data tags available on the industrial device that are relevant to an analytic objective;
rendering, based on the device profile, a list of the data tags on the user interface;
receiving, via interaction with the configuration interface, mapping input that selects a subset of the data tags from the list to be bound to an analytic model selected from a library of analytic models, wherein the analytic model corresponds to a target application and defines data mappings to the target application;
generating an information model based on the mapping input, wherein the information model defines a contextualization of the subset of the data tags and a mapping of the subset of the data tags to the target application based on the data mappings defined by the analytic model;
deploying an edge gateway comprising the information model on a cloud platform, wherein the deploying comprises:
collecting, by the edge gateway, device data from the subset of the data tags of the industrial device in accordance with the information model,
contextualizing, by the edge gateway, the device data based on the information model to yield modeled data, and
sending, by the edge gateway, the modeled data to the target application based on the mappings of the subset of the data tags to the target application defined by the information model.
19 . The non-transitory computer-readable medium of claim 18 , wherein the deploying further comprises executing the edge gateway as a virtual machine on the cloud platform.
20 . The non-transitory computer-readable medium of claim 18 , wherein
the analytic model defines an organization of data, and
The deploying further comprises modeling the device data in accordance with the organization defined by the analytic model to yield the modeled data.