Create an event frame generation analysis

In PI System Explorer, you can create an event frame generation analysis for an individual element.

Prerequisites: Check that an event-frame template is available for the analysis.

1. In the navigator, click Elements.

2. In the browser, select the element where you want to create the analysis.

3. In the viewer, click the Analyses tab.

   The tab lists any analyses already defined for the element.

4. Create a new analysis for the element:

5. If no analyses exist, click Create a new analysis to create the first one.

   • Otherwise, click New Analysis to create a new one.

6. Enter information to identify the analysis:

   • Name

     The name that uniquely identifies this analysis.

   • Description

     Optional text that describes the analysis.

   • Categories

     Optional category that you can assign to the analysis. Click the list and select the check box next to categories you want to assign or click New to create a new category.

   • Analysis Type

     Click Event Frame Generation.

7. From the Event Frame Template list, select the template for the event frame that this analysis generates.

   To create an event frame template, see Create event frame templates. An event frame naming pattern typically includes substitution parameters. For more information on naming pattern, see Naming patterns.

   Note: Parameters %Endtime% and %Duration% in an event frame naming pattern will be evaluated at the end of an event frame when the event frame closes. An attribute value may update at any given time. To avoid confusion, substitution parameter %@Attribute% (which resolves to a value of an attribute) in an event frame naming pattern will always be (re-)evaluated in the context of a start of an event frame. Default naming pattern (%ANALYSIS% %STARTTIME:yyyy-MM-dd HH:mm:ss.fff%) will be used in the absence of a name or if the event frame name evaluates to a null value.

8. Specify the necessary expressions for the analysis:

   1. In the StartTrigger1 row, enter the Boolean expression that specifies the condition that starts an event frame.

   2. Optional. Change the default name of the start trigger.

      Note: If you change the name of a start trigger from the default name, you must change all the start triggers and use unique names for each trigger.

   3. Optional. In the True for field, enter the amount of time that the start-trigger condition must be true before the analysis starts an event frame.

      Note: Specify a value in the True for field to reduce unwanted event frames caused by momentary fluctuations in input data.

   4. Optional. In the Severity field, enter the severity of the trigger.

      The choices for severity are: None, Information, Warning, Minor, Major, and Critical. If you have defined multiple start triggers, the one with the highest severity will be the effective start trigger; if all start triggers have the same severity, the first start trigger in the list will be the effective start trigger.

   5. Add multiple start triggers by clicking on Add a new start trigger.

      For each start trigger, enter a Boolean expression, specify a value for True for and select a value for Severity.

      Note: If you have changed the default name of a start trigger, make sure to assign unique names to all the start triggers.

   6. Optional. If a different condition ends the event frame, enter the expression in the EndTrigger row. Otherwise, the event frame will close when the start trigger condition is no longer true.

   7. Optional. Click Add a new variable to insert a new expression row where you can specify a variable and expression for use in one of the trigger expressions.

   8. Optional. Click Add a new output expression to insert a new expression to write outputs to event frame attributes.

      You can map the output as static (doesn't save history) or PI point (saves history) attribute. When mapping to PI point attribute, make sure that it is mapped back to the attribute on the asset. See Understand expression analyses for more information.

9. Optional. Generate a root cause event frame for every event frame that the analysis generates.

   A root cause event frame captures asset data, which can help you analyze conditions just before the start of the event frame.

   Note: Outputs from root cause event frames are only written to static attributes.

   1. From the Advanced Event Frame Settings window, select Generate child root cause event frame before parent event frame starts.

   2. In the Duration field, enter the length of the root cause event frame.

      The root cause event frame starts this long before the generated parent event.

      Note: The root cause event frame is created only for the first instance, not for each time that the trigger is True.

10. Optional. Map a start trigger name and start trigger expression to an event frame attribute.

    Note: We recommend that you provide meaningful names to your analysis start triggers while saving them to an event frame attribute.

    1. From the Advanced Event Frame Settings window, select Save start trigger name to event frame attribute and then select an attribute (template) or create a new attribute template on the event frame template to which the start trigger name is saved.

    2. From the Advanced Event Frame Settings window, select Save start trigger expression to event frame attribute and then select an attribute (template) or create a new attribute template on the event frame template to which the start trigger expression is saved.

       The start trigger name and start trigger expression can now be seen on generated event frames, on the selected attributes.

11. Specify scheduling for the analysis. See Understand analysis scheduling.

12. Optional. To verify the analysis, review the results produced with historical data:

    1. In the analysis list, right-click the analysis and then click Preview Results.

    2. Enter a start time and end time not later than the current time, and click Generate Results to see a list of results.

13. To apply changes and save your work locally, click on the toolbar.

    This does not run the analysis.

14. Click Check In on the toolbar.

    PI System Explorer saves and checks in your analysis, and then starts running the analysis.