This document is an excerpt from the Visualizing PI System Data Workbook v2017
In this course videos, we will work with a fictitious plant named OSIsoft Plant. You can download a copy of the database & instructions here.
This simple plant has two production lines, where each has a combination of one mixing tank and one storage tank. If you are working with your own PI System and your own data, that's even better! We just wanted to provide you context for the class videos.
This plant could be schematically shown as:
As shown here, each tank has different process variables such as Internal and External Temperatures, Flow Rate, Pressure and Level whose values are continuously collected from devices on the Plant. In the early days of PI System, these process variable were the only data items whose historical data could be stored in Data Archive.
There are some other data associated with each of these tanks such as the manufacturer, model and the installation date which are stored in the maintenance sheets available on tables in SQL Server. Moreover, all the information related to the material flowing in these tanks is kept in tables on the Plant’s SQL Servers.
Despite the fact that these tables are available on the SQL server, their information could not easily be integrated with the historical data stored in Data Archive. Hence, using AF and hierarchy becomes critical in bringing all the important data and information in one place: PI System.
At the OSIsoft Plant, predictions on the level of each mixing tank are critical in running a smooth production. This data, Level_Forecast, is stored in a “Future” point on the Data Archive and could be viewed on PI System displays or be compared to the actual value of level in any PI Applications.A collection of PI Points are built on Data Archive for storing the values of process variables. There is also a hierarchy built in AF for this Plant, bringing all the important information and data, including the process variable time series data, to one place.