Introduction (1 minute)
"Happy evening everyone! Now, we’re going to explore how to handle one
of the biggest challenges in the industrial world !! managing massive
amounts of sensor data. We’ll dive into the Plant Information
Management System (PIMS), and how it simplifies the ‘data jungle’
that many of us face in the age of Industry 4.0.
Let me ask you all—how many of you work with data from sensors or have
experience in environments that rely on real-time data from equipment?
Can anyone share some challenges you face when trying to manage or
analyze that data?"
(Allow for a few responses and engage with the answers.)
"Great! We’ll talk today about how the PI System can help address these
challenges and make sense of all that data."
The Data Jungle (2 minutes)
"Now, let’s talk about the Data Jungle. Imagine You’re working in an
industrial environment where thousands of sensors are sending data
every second. That’s an overwhelming amount of data, right? How would
you keep track of all that information? How would you know what data is
important and what is just noise?
In the context of Industry 4.0, data is flooding in from all directions—
sensors, IoT devices, machines, and so on. But here’s the catch: data
overload can easily cause chaos. Without the right system, we will be left
struggling to figure out what to do with it all.
How many of you have ever faced a situation where you had too much
data and not enough insight from it? (Pause for audience interaction.)"
"Exactly! It’s like being lost in a jungle without a map. This is where the PI
System comes in. It’s designed to organize tons of data and manage the
data into models.
AVEVA PI System Introduction (2 minutes)
"So, what exactly is this PI System? To put it simply, it’s a powerful tool
that helps to acquire, store, manage, and analyze data from sensors,
all in real-time. Imagine it as a bridge between the sensors and the users
or systems that need the data.
How many of you have worked with systems where data is spread out
across multiple platforms or devices? (Pause for responses.)
The PI System solves this by centralizing the data, ensuring that no
matter where the data comes from—whether from a pump, valve, or
temperature sensor, reactors, compressors, towers—it’s collected, stored,
and made accessible at one place.
Think of it as a digital control center that handles everything from data
collection to delivering it to the right person or application."
Step 1: Data Capture – Collecting with Ease (2 minutes)
"Let’s dive into how the PI System captures data. How difficult do you
think it is to collect data from industrial sensors? (Pause for responses.)
You’re probably aware that industrial data often comes from various
sources and uses different communication standards, right? So how do we
gather this data without endless custom coding or chasing after outdated
systems?
The answer: PI System interfaces. With over 450+ interfaces, the PI
System supports almost any industrial asset or communication standard.
This means that no matter how diverse the data sources are, the system
can collect the data seamlessly.
To ensure the data collecting is accurate and free from errors these
interfaces come with built-in intelligence—features like data buffering,
automatic filtering of bad data, and even auto-discovery of new data
sources. the data we collect is not just fast, but also clean."
PI Connectors and PI Adapters are used to facilitate the connection
between local servers (like on-site systems or assets) and global
servers (such as central PI servers or cloud environments). PI Connectors
are used to connect various data sources like, PLCs, SCADA systems and
push it to the PI Server in real time. PI Adapters, If we are integrating data
from an external cloud-based system that uses REST APIs to push data, PI
Adapter would be used to retrieve the data and send it to the central PI
Server.
Step 2: Storing and Preparing Data for Analysis (2 minutes)
"Once the data is captured, it needs to be stored in a way that it can be
easily accessed and analyzed. Now, here’s an interesting question for you:
traditional databases don’t typically handle the volume and speed of
industrial data well. Isn’t it? How would you store and manage data
coming in at high frequencies? (pause)
The PI System is built specifically for real-time performance, so it can
store and deliver data instantly as it arrives, whether it’s 1 Hz or 100 Hz.
This makes it perfect for industries with high-frequency sensor data.
For example, if we need calculations, unit conversions, or event tagging,
the PI System takes care of it.
Just imagine being able to retrieve historical data from years ago, just
like that. How long do you think it would take to pull up 10-20 years’ worth
of data from traditional systems? (Pause for answers.)"
"The PI System can retrieve this historical data in seconds, making it
incredibly fast and efficient to get both current and past data with no
complicated archiving involved."
Step 3: Adding Context to Data – Making it Understandable (2
minutes)
"Now, we all know that data alone isn’t always helpful. For instance, we
might see a sensor label like TI37.109-CP-TK9PV. Does anyone here
know what that refers to? (Pause for answers.) It’s pretty cryptic, right?
So, how do we make sense of this data?
The PI System attaches sensor data to real-world assets, like pumps,
valves, or motors. This way, instead of looking at a bunch of technical
data labels, we can see what the data represents in a business context.
Asset Analytics are used to create calculations that transform raw sensor
data into meaningful insights to measure efficiency, performance, and
asset health. By leveraging the PI Asset Framework (AF), we can
standardize the calculations across similar assets, enabling real-time and
historical analysis. It also allows for setting up alerts based on calculated
values, helping with proactive maintenance and optimization.
Step 4: Event Frames – Capturing the Story Behind the Data (1
minute)
"Next, let’s talk about Event Frames. When analyzing data, it’s not just
about raw numbers—it’s about understanding the events that drive the
data. For example, tracking batch durations, downtimes, or startups.
Why is it important to capture events? (Pause for responses.)
With the PI System, the events are automatically tagged, which means
we can easily analyze specific occurrences like equipment failures or
production batches. It turns raw data into actionable insights. Now,
comparing events , analyzing downtime causes—is quick and efficient."
PI Notifications are used to deliver critical data and alerts through various
channels, such as HTTP requests, email, and SMS.
The Last Mile: Delivering Data Where It’s Needed (1 minute)
"Once the data is captured, stored, and understood, it’s time to deliver it
to the right destination..)
Cloud Services: Uses PI Cloud Connect to integrate with cloud
platforms.
PI Integrators: Connects with platforms like Microsoft Azure,
Google Cloud, and Amazon Web Services for business analytics.
PI System Access: Provides various interfaces such as PI WebAPI,
PI OLEDB, and PI SQL Client for accessing data.
The PI System has powerful visualization tools, and it integrates
seamlessly with business applications.
Visualization: Tools like PI Vision (Web), PI DataLink, and PI
ProcessBook for visualizing and analyzing data.
PI DataLink is an Excel add-in that allows users to access and analyze PI
System data directly within Excel, making it easy to create reports and
perform detailed analysis. There’s another application PI Builder which is
used to streamline the creation and management of Asset Framework (AF)
elements, simplifying the setup of tags, templates, and attributes. PI
Vision provides a web-based platform for visualizing real-time and
historical data through interactive dashboards.
