Historians: Read Event
This article provides a starter solution that can be used to obtain events and their time-series data from a historian.
What Does This Article Cover?
An Intelligence Hub data pipeline solution can be created to obtain Event Frame data from AVEVA PI and use the event frame start and end datetime to obtain process values for the Event Frame's assets and timespan. The payload might be formatted as a wide table and written to a data warehouse or data lake, for example. This article will focus on the AVEVA PI System, including Asset Framework, and may evolve to include other historians. The article provides a starter solution.
What is an Event?
An Asset Framework Event, often referred to as a PI Event Frame, is a feature of AVEVA PI System's Asset Framework used to capture and organize time-series data associated with specific periods of interest, such as:
- Batch processes
- Equipment shutdowns
- Alarm conditions
- Quality excursions
General Intelligence Hub design considerations for obtaining events
These considerations pertain to obtaining events.
- Define the events that are in scope for the given solution.
- Define the attribute values that may need to be obtained for the respective events.
- Determine how the in-scope attributes will be queried, perhaps based on timespan and event, for example.
- Start by querying 10s or 100s of points or tags and scale from there to determine query duration. Optimally, the duration of the query should be approximately a few seconds.
- Determine if resources need to be scaled to enable the query to be more efficient.
- Determine how the new Connection Input impacts other Inputs sharing the same Connection.
- The resulting data payload might be written to a MQTT broker, database, data lake, data warehouse, or made available by the Intelligence Hub REST Data Server.
- This solution type is conducive to a wide table format with columns per attribute and metadata values. An Intelligence Hub Model can be used to define column names and data types.
AVEVA PI System Asset Framework design considerations for obtaining Event Frames and time-series data
These considerations pertain to obtaining Event Frame data and related process data PI System Asset Framework.
- This solution uses the Intelligence Hub PI Connection Event Frame Input that can be configured to use most Event Frame queries that can be defined in PI System Explorer. For example, an Event Frame template may be used.
- A PI System Event Frame is not considered an Intelligence Hub event. Therefore, the solution most poll for new event frames.
- The solution must be optimized considering latency, polling rate, and Pipeline performance.
- The Connection Input will likely return many Event Frame instances.
- An Event Frame returns one value for Event Frame attributes. It might be beneficial to obtain multiple values for many attributes over the time span of the Event Frame.
- For example, obtain the value of all asset attributes configured for a reactor every minute during the duration of a batch phase. The starter solution addresses this scenario.
- Many other solution scenarios could be pursued, including alternatives for the time span and obtaining attribute values.
- A project file may be downloaded [here].
Summary
Please utilize the considerations provided in this article as you implement an Intelligence Hub solution.
Additional Resources