What Is A PI System
What Is A PI System
What is a PI System?
Learning Outcomes
After completing this topic you should be able to:
Define the components of a PI System.
Draw a diagram of the architecture of a PI System.
The PI System
The PI System was originally developed by OSIsoft to collect Plant Information from PLC, DCS
and SCADA systems. The PI System collects, stores, and manages time stamped data. This
data may have timestamps in the past, current or future.
Components of a PI System are:
Computers with a PI Interface collect data (known as points or tags) from a data source.
These interface nodes get data from your data sources and send it to the Data Archive.
This data may be collected from a variety of places, such as:
o the plant, weather stations,
o IT networks,
o location data for trucks,
o Telemetry from monitoring systems.
Data is stored in the Data Archive in such a way as to make user retrieval as efficient as
possible. The data is accessible to users in different ways: directly or via tools providing
context.
Accessing the data in context is provided by linking the data points to assets defined in
an Asset Framework (AF) system.
To visualize the data collected and stored, users use tools in the Visualization Suite:
o PI Vision (browser-based graphs and symbols),
o PI Datalink (a Windows based Excel add in),
o PI ProcessBook (a Windows based application for graphical displays).
This diagram shows the 3 main categories and components of a typical PI System:
Architecture of a PI System
The architecture varies from simple to complex; some customers have only a single interface
sending data to a single Data Archive. There are many more combinations and configurations of
the PI software components, so make sure to ask your PI System administrator about how your
infrastructure is laid out.
Here is a conceptual diagram as an example:
Quick Check
Having completed this topic, are you able to:
Define the components of a PI System?
Draw a diagram of the architecture of a PI System?
Building Blocks of the PI System
LEARNING OUTCOMES
Define the terms of Asset Framework (AF) and its components: elements and attributes.
Define AF attribute types: static (none), PI Point, point array, formula, string builder,
table lookup and Analysis.
What is an AF Element/Asset?
In Asset Framework, company locations, sites, processes
and each piece of equipment is represented by an
Element. Company Assets may be defined with an AF
Element. The AF encourages organization of assets into a
structure that makes it easier to find information.
A self-explanatory element structure for assets goes a long
way to help users find the data they are seeking. With well-
defined elements showing context for the assets, data can
be located without the user needing to understand the
technical details of each piece of equipment. The AF
element structure assists in promoting a hierarchical and
logical organization of assets.
What is an AF Attribute?
Note: All relevant data about an asset is grouped together with AF Attributes on AF Elements.
This allows users to build displays and reports that maintain a complete view of the company’s
assets.
What is a PI Point?
A PI Point (sometimes referred to as a tag) is a unique storage point for data in the Data
Archive. It is a single point of measurement and has a value with a timestamp, such as a
temperature of 31.2 oC on 2019-Dec-24 23:59.
Point name
Points stored in the Data Archive each have a unique name. It is
a common practice to name the PI Points based on Control
Systems point names. Since the point is the name that identifies
the point to users, a consistent point-naming convention should
be used that is meaningful to people in your organization.
Knowing the naming convention can be helpful in searching for
points.
Try to determine what the following point may refer to:
M03_E1P1_MOTDRV1202_RUNSTAT
It refers to - Machine3 Enclosure 1 Panel 1 Motor Drive 1202
Run Status.
PI Vision Screenshot:
Searching for PI Points Is this intuitive? Probably not, unless you have spent time
memorizing the equipment’s naming conventions.
Most of the time, the PI Points themselves are not going to be easy enough for users to work
with; therefore, OSIsoft recommends developing your Asset Framework hierarchy to leverage
the very “human friendly” nature of AF Attributes on AF Assets. Building an intuitive AF structure
will make end users’ work much easier to accomplish.