Add an Asset Management Model

20 min

A model is a template that contains static and dynamic attributes, data transformations, and schemas. You create Digital Twin models and users then create instances in Belden Horizon Data Operations from the models. When you create a model, you:

Create a new model manually or from an import file.

Deploy the model from Belden Horizon Data Manager to your Belden Horizon Data Operations instance.

Configure the model by defining static and dynamic attributes, data transformations, and schemas.

You create Digital Twins models by navigating to Asset Management > Models for a company.

See Use Digital Twins to Contextualize Data for information on how to leverage Digital Twins.

Note: Digital Twins requires a Scale or Growth license.

Viewer and Administrator Roles

Note: These options are available for Belden Horizon Data Manager version 2.31.0 and later.

Visibility of edge devices is limited to the companies that you have access to. The actions that you can perform also depends on your asset management role (defined in Belden Horizon Data Manager Admin Console). You are also assigned one of the following roles:

The Viewer role allows view access to Digital Twins.

The Administrator role allows full access to Digital Twins.

Step 1: Create the Model

You have two options for creating the model. You can create a new model or import a model as a JSON file.

You can add a Asset Management model from the Asset Management Models pane.

Note: You must view the Asset Management models as tiles to add a model. See Switch between Table view/Module view of Manage Asset Management Models.

Create a New Model

To create a new model:

From the companes list, navigate to Asset Management > Models.

From the Models pane, click the Create a model tile.

From the Asset drop-down, select a model type.

Enter the name of the model in the Name field.

Optionally, enter a short description of the model in the Description field.

Click Add. A new model tile appears in the Models pane.

Import an Existing Model

You can import a model as a JSON file.

To import an existing model, click the Import Model icon, select the <model name>.json file, and click Open.

Model Versions

When you create a model, the model version starts at v.0. The version number increments when you save the model. The version number displays in the:

Model tile.

Model and instance configuration page.

Step 2: Configure the Model

After creating the model and defining a name, you can configure it with static and dynamic attributes and schema to represent your assets or data structures.

Model Image

Click the Upload image icon to select an image for the model. The image displays in the model tile on the Models pane.

Static Attributes

When you create static attributes, you define default attribute values as placeholders. Later, you change the default values in model instances. For example, you might define Manufacturer , Device Description, Model and Plant attributes and change the static attribute values later in instances.

Manage the metadata for the model from the Static Attributes tab.

Add Static Attributes

If you need to filter existing attributes, you can use the search field.

To add static attributes:

From the Static Attributes tab, click Add new static attribute.

In the Attribute Name field, either select an attribute from the drop-down list or enter a new attribute name.

In the Attribute Value field, enter a value for the attribute.

Import Static Attributes

Before importing, create a CSV file with the following headers, then add rows with attribute names and values.

Required Headers

ModelName

Value

Optional Headers

InstanceName

After creating the file:

Navigate to Static Attributes, and select Import Attributes.

In the dialog your browser opens, navigate to and open the CSV file.

Export Static Attributes

You can export all attributes to a CSV file, modify them offline, and import them again.

To export a CSV file with existing static attributes:

Navigate to Static Attributes, and select Export Attributes. Attributes export to a CSV file.

Dynamic Attributes

Next, define dynamic attributes that represent real-time data for your asset. Dynamic attributes can be JSON, boolean, array, string, or number data types. Each dynamic attribute can have an associated transformation.

Note: Dynamic attribute data type values are stored and returned in lower case.

Add Dynamic Attributes

To add a dynamic attribute:

Click the Add item icon.

In the Name field, enter a name for the attribute.

Optionally, enter a value for Unit and select a Data Type. JSON applies if you do not specify a data type.

The Skip Key option is only available when the data type is JSON. Select Skip Key to exclude a specific attribute from the payload. Note that the value persists in the payload, for example, as value='20'.

Optionally, define a topic expression consisting of static string values and parameters by surrounding the expression in curly brackets ({}). For example, devicehub.alias.{device}.tagname.

Optionally, if you've created data transformation schemas, select one from the SchemaID field.

If you need to filter existing attributes, you can use the search field.

Import and Export Dynamic Attributes

To import dynamic attributes, download the sample CSV file and populate it with your dynamic attributes. After creating the file, click the model tile, navigate to Dynamic Attributes, and click the Import Attributes icon.

To export a CSV file of existing dynamic attributes, click Export Attributes.

Transformations

Use transformations to convert the raw JSON data published on topics in the message broker to a desired format. Transformations apply to specific dynamic attributes and the resulting transformation is included in the full Digital Twins and Publish Attributes payloads. Transformations clean your data and make it easier for northbound systems to consume. For example, create a transformation that eliminates duplicate timestamps and register IDs in your payload.

Transformations parse and transform the payload and specify a payload structure as the model inherits data. See Use Digital Twins to Contextualize Data for information on formatting transformations.

With transformations, you can:

Create a JSON transformation that filters specific key-value pairs . For example, this transformation only includes the following attributes in the payload.

}
  “value“: “$value“,
  “timestamp“: “$timestamp“:
}

Create a JSONata transformation to perform dynamic filtering. JSONata is a query language that allows logical conditions and computed outputs. For example, use JSONata to create an alert when a temperature tag exceeds a high or low threshold value.

{
  "asset": deviceName,
  "area": metadata.area,
  "line": metadata.line,
  "quality": success,
  "metric": "metric",
  "timestamp": $fromMillis(1510067557121, '[M01]/[D01]/[Y0001] [h#1]:[m01][P]'),
  "value": value
}

Use a playground to test out your transformations before adding them to a model.

Add the following system variables: "$EDGE_DT_SYSTEM_TIME", "$EDGE_DT_MODEL_VERSION", and "$EDGE_DT_MODEL_NAME".

Configure Transformations

To add a transformation schema:

Click the Transformation tab.

Click the Add a Schema icon.

Enter a name for the schema.

Select from JSON or JSONata from Schema Type drop-down menu.

In the code window, copy and paste the JSON-based or JSONata schema.

(Optional) You can use the System Variables in the code drawer.

Experiment in a playground.

Click Add.

To edit or delete a schema, click the Action menu for the schema and select an option.

Apply the transformation to specific dynamic attributes.

Configure Parameters

Note: Parameters are available for Belden Horizon Data Manager version 2.31.0 and later.

Parameters are variables that you can insert into topic expressions. Parameters help you save time when defining instances that have many attributes.

For example, define a device parameter and add it as a topic expression for a dynamic variable with the following syntax: devicehub.alias.{device}.tagname. Then modify the value for the device parameter for each instance on the Instance Parameters configuration page.

Note: Parameter names and values are case sensitive and cannot contain whitespaces. Allowed characters:

A-Z in lowercase and uppercase

Hyphen (-) and underscore (_) characters.

To create a parameter:

Navigate to the Parameters tab and click Add Item.

Enter a parameter name and a placeholder value.

Click Save Row. The new parameter saves.

Add the parameter to a topic expression. In this example, devicehub.alias.{device}.Temperature is added to the Temperature attribute.

Select an instance of the model and click Open Parameters.

In the Instance Parameters dialog, edit the value for the device.

Navigate to the Instance Dynamic Attributes page to see the topic defined for the dynamic attribute, based on your topic expression and parameter values.

Note: If a dynamic attribute has a topic expression defined, the topic for instances cannot be edited. To change the dynamic attribute topic, you must edit the model or parameter values.

Schema

Use the Schema tab to create a custom data structure for the model.

To create a category for the schema

Click the Action menu icon for root and select Add Node.

To add attributes to the schema, do one of the following:

Drag and drop attributes from the left-hand list to the canvas under the appropriate category.

Click the Action menu for a category and select Add item to Node. Then, select the attribute to add.

When you have configured the schema, click Save Model.

Export Schemas

To export an existing schema to a CSV file, click the Export Schema icon.

Sync Attributes

To sync the existing schema

Click the Sync Attributes icon. All attributes are appended to the current hierarchy. If they exist, the static_attributes and dynamic_attributes nodes are recreated.

To confirm that you want to proceed, click Yes.

Clear Schema

To clear the schema builder tree, click Clear Schema. Changes apply when you save the model.

Apply Model Updates to Instances

When you update your model, you can apply the changes to model instances.

To apply model updates:

Click Upgrade Instances.

Select the instances to upgrade and click Upgrade.