Skip navigation
All Places > Academic Users Group > Blog
1 2 Previous Next

Academic Users Group

19 posts

2018 PI World Barcelona Recap

Posted by imudron Employee Oct 24, 2018

Academic Symposium (Monday, 13 – 17:00, 24 September) 

Welcome & Wrap-up summary speech and 6 presentations at

  1. Welcome and Overview of the Day (Nicolas Peels & Ivan Mudron)

         Welcome and Overview of the Day.png

     2. SIRIUS: Collaboration across the digital divides in the oil and gas supply chain (David Cameron)


       3. A successful study case of collaboration through PI system: University of Granada and Abbott Lab (Jose Miguel Gutierrez Guerrero)          


          4. OSIsoft Cloud Offering : Transforming Student Education with the Academic Hub Service (Erica Trump)


          5. Intelligent Buildings Remote Monitoring Using PI System at the VSB - TU of Ostrava Campus (Jan Vanus)


          6. Using the PI System in Chemical Engineering Education and Research (Nenad Bolf & Zeljka Ujevic Andrijic)



          7. Smart Industrial Concept! - Design and Operational Optimization (Rene Hofmann) 


          8. Wrap-up summary speech of Customer Innovation & Academia Director (John Matranga) 



Innovation and IoT Conference track presentation:

Title: Exploring the Future of Digital Twins with OSIsoft and SIRIUS (Evgeny Kharlamov – University of Oxford, Brandon Perry – OSIsoft, David Cameron – Unioversity of Oslo)

The ongoing population explosion of users and data is making all of us re-think how people, applications, and data come together.
One approach is Digital Twins: virtual representations of physical assets, sophisticated enough to drive complex analytics and simulations without expensive custom integration.
The promise of Digital Twins drives us at OSIsoft to address some vast prerequisite challenges: contextualizing the data, personalizing the system, enabling collaboration, and integrating seamlessly with deep learning and artificial intelligence.
OSIsoft Research is working in collaboration with SIRIUS, a research consortium for scalable data access in the Oil & Gas domain.



Industrial IT track

Title: Improving Situational Awareness for Utilities Operators and Energy Managers (Mary-Ann Ibeziako – University of Maryland College park)

The University of Maryland partnered with NIST National Cybersecurity Center for Excellence and OsiSoft PI as a beta test site by providing the Campus Co-generation,
Substation and 6,300 ton chiller plant as the real data source for the Situational Awareness Project and Energy Sector Asset Management Project.
The project was a demonstration on how to secure utility network infrastructure



Manufacturing, Supply Chain, Transportation

Title: Real Time Optimization of Gasoil Consumption for Railway Transportation (Ernesto Bedrina Ramirez – Grupo Ceteck, Pablo Martinez Fernandez – Polytechnic University of Valencia)

Real time optimization of gas & oil consumption for railway transportation with genetics algorithms.
The main goal of the project is to reach the best time/gasoil consumption rate maintaining the scheduled time stops



Analytics Track

Title: Introduction to Time-Series Analysis with PI System Data (Brian Davison - Lehigh University)

Fundamental understanding of time-based data and how to extract insights from it. With some real-world examples we will demonstrate some simple analytics using time-based data in R.
A focus will be on underlying assumptions behind time series, and how to sample and generate statistics and predictive models on such data.





30+ new contacts (Media, Existing universities, Customers, New universities - most of them from host country #Spain)
Main upcoming trend: Customers are wiling to know more about the program and some are pushing us to collaboration with universities (Johnson&Johnson, Lonza, Umicore)


Academia Interaction Dinner

- building community

- building trust
-have discussed data sharing problems (confidentiality), issues at the campus management level, and other problems

The Academic Symposium at PI World 2018 is the largest and most diverse Academic Community event that OSIsoft has hosted to date. In all, we had around 220 attendees, of which ~25% were from academia and ~75% represented our industrial customers. We hosted an AM Industry Innovation session and a PM Workforce Development session. The day included several podium presentations, panel discussions, a workshop, and a networking dinner.


Presentation recordings are now available on the OSIsoft website. Just click the preview images below to stream the videos.


To all who attended, thank you for spending the day with us! Your engagement and participation made this event a success.


AM Session: Industry Innovation


1. Welcome and Overview of Day

John Matranga, OSIsoft


2. Utopus Insights Journey: An innovative company launched from Vermont Electric Power Company

Kerrick Johnson, Utopus Insights


3. Partnering with Academia to Develop Advanced Analytics Platform To Accelerate US Manufacturing Competitiveness

Scott Miller, Texas A&M University


4. The Journey from a Classroom Idea to Commercialization of an Energy Management Company

Mike Broeker, Building Ideas Group; Mike Mihuc, OSIsoft


5. Lightning Round Talks: OSIsoft Call for Collaboration – 1st Year Innovation Showcase

Ashkan Negahban, Penn State; Denis Grancanin, Virginia Tech; Frank Lee, UC Davis


6. Panel Discussion: Student Engagement with Real-World Data

George Paterson, University of Iowa; Pratt Rogers, University of Utah; Steve Moore, Del Mar College; Leo Vitor, Radix; Erica Trump, OSIsoft



PM Session: Industry Innovation


7. Welcome from OSIsoft VP of HR

Dawn Sprague, OSIsoft



8. Keynote from Academia: Revamping Student Education with Real-World Data

Brian Davison, Lehigh University



9. Multidisciplinary Collaboration for Student Education - UC Davis PI Systems for Learning & Research

Kiernan Salmon, UC Davis; Jill Brigham, UC Davis



10. OSIsoft Cloud Offering: Transforming Student Education with the Academic Community Service

Zhiyuan Cheng, Industrial Learning Solutions; Erik Ydstie, Carnegie Mellon University; Erica Trump, OSIsoft



11. Student Education – Keeping Pace with Industry Trends: TransCanada Sees a Need for New Skills

Emily Rawlings, TransCanada; Keary Rogers, TransCanada


Erica Trump

PI World Hands-On Labs

Posted by Erica Trump Mar 5, 2018

Attention, academic attendees at PI World 2018!


We’re excited to have so many of you join us in San Francisco for the 2018 OSIsoft Academic Symposium. This event will kick off the conference on Monday, April 23rd. PI World 2018 will be full of opportunities to learn how the PI System is used to solve business challenges, to interact with industry representatives, and to learn about OSIsoft technologies.


Our PI World Hands-On Learning Labs provide one such opportunity. These hands-on labs occur on Thursday, April 26th and Friday, April 27th. Note that pre-registration is required. You may register for the labs when you register for PI World. If you have already registered, please email to update your registration.


These labs will be of particular interest to the Academic Users Group:


A Digital Plant Template for Operational Insights - An Enterprise Strategy

This lab uses a follow-the-money strategy to provide an example that integrates the AF/Asset Analytics/Event Frame data for running a plant. In this lab, you will learn how to bridge the gap between production planning and execution and to simplify implementation for production monitoring. We will run simple analyses that will help increase production yields and reduce operating costs with an integrated approach for Overall Production Management. Power User Lab.






Advanced Analytics for PI data for Data Scientists

PI System is the standard time series data infrastructure across several industries. Decades of operational data collection has turned each PI System into a treasure trove for data scientists to extract knowledge. Come to this lab and learn, as a data scientist, how you can access, shape, and use PI data to generate insight and knowledge. A basic understanding of data science basics is required for this lab. No knowledge of PI System is necessary.


Fit for Purpose - Layers of Analytics using the PI System – AF, Matlab, Machine Learning

This hands-on lab covers scenarios to illustrate the different levels of analytics that are fit-for-purpose when using the PI System. For example, what calculations and analyses do you do in AF? When do you use Matlab and similar libraries for advanced calculations that hook into AF? When do you call on “data science and machine learning ?” Use cases will include those focused on an equipment, such as a pump, motor or compressor, and those focused on a process. We will also cover examples of advanced analytics that are part of the data collection from the edge devices and contrast these with predictive analytics – both the model development and model deployment (scoring new incoming real-time data) and include cloud vs. fog vs. edge scenarios.


Introduction to PI System Developer Technologies

This lab is designed for software developers new to the PI System who desire programmatic access to it. You will be introduced to various Developer Technologies that you can leverage in your custom application to interact with the PI System. You will have some hands on experience with each technology. At the end of the lab, you will walk away with knowledge of their basic capabilities and also know where to go if you want to dive deeper into them.

I recently attended the annual National Instruments customer conference, commonly known as NI Week.  NI is a leading edge provider of control and instrumentation systems in the areas of test and measurement, wireless (5G), smart grid, satellites, and other areas.  During the NI Week conference, OSIsoft had a speaking slot explaining the benefits of linking product manufacturing test information with product performance once it is deployed into production.  This is
done with the recently released LabVIEW - PI Connectivity toolkit, which allows data to be placed into the PI System from LabVIEW and into LabVIEW from PI.


There were two (2) companies that presented their experience in this area, National Oilwell Varco (NOV), and RoviSys Corporation.  Unfortunately, the presentations were not recorded, but here are the key takeaways:

NOV builds a blowout preventer for offshore oil rigs.  During the manufacturing and test quality process, they capture test data through LabVIEW and subsequently send batches of the test data to the PI System, so that once the blowout
preventer is put into production, NOV personnel can compare its performance in the field to how it behaved during the test phase.  I suppose it could also identify potential problems ahead and notify when it appears to either malfunction or when pressure is abnormally high.


RoviSys gave a similar presentation in which it provided the LabVIEW and PI System integration expertise for LG Corporation, who is manufacturing a 1MW fuel cell to be used for datacenter power where these centers are located in remote locations, such as deserts.  RoviSys was similarly able to meld the LabVIEW manufacturing test phase data with the production data in a common repository, the PI System.


Perhaps there may be academic use cases using this approach.  Please comment.  I'd be interested.

Thank you.


APAC Internship Project

Posted by mtippett Apr 27, 2017

Over the Australian summer holidays the three interns, Miwa Teranishi, Ryan Amaudruz, and Thomas Hahn (all from the University of New South Wales, Australia), worked on projects using PI and machine learning tools to investigate the Japanese and Australian electricity markets in OSIsoft's Tokyo, Sydney, and Perth offices.  Even in the short period of time they had to work with us, and a lack of any prior expertise with PI, they were able to develop some interesting models and gain some very interesting insights into these markets.


Some of the highlights of the results are:


Using PI and Falconry to predict the daily minimum electricity price in Japan


Using PI, Microsoft Azure, and only publicly available data to predict net aggregate solar generation in Western Australia, covering an area of 2.6 million square km.



An overview of their projects, along with some videos describing what they did, and how they did it are available in the blog post below:


Three Interns Describe Three Successful Machine Learning Projects


Additionally, a version of Miwa's video with Japanese subtitles is available on our OSIsoft Japan youtube channel at the link below:


OSIsoft Japan Intern Project 2016 – Power market forecasting with Falkonry - YouTube


This is a great example of how the PI system can easily be integrated with other platforms to provide a powerful results in a variety of fields.

For those of you that did not have a chance to attend the OSIsoft Academic Symposium on March 20th 2017, you can now view the sessions from the web site.  Some of the more noteworthy presentations were:


  • Panel Discussion: PI System for Data Science Research & Curricula.  Several panelists, including Don Paul of USC (ex-Chevron CTO), discussed how industry and academic are using the PI System for collaborative big data analytics and machine learning projects. Also, Lehigh University discussed their Data X program, where real-time data is being used as a component of big data analytics course at Lehigh University.  The link is here:


  • Pratt Rogers discussed how the University of Utah is including real-time data and the PI System into the metals and mining curriculum.  This enables the next generation of engineers to understand how industry leading companies handle their operations and maintenance data. The link is here: Big-Data-and-The-PI-System-–-Teaching-the-Next-Generation-of-Engineers



  • A presentation was jointly made by National Instruments (NI) and OSIsoft that described how both companies are working together to provide common solutions.  OSIsoft's Dan Lopez followed with a short demo showing how to write PI System data to NI's LabVIEW and how to write NI LabVIEW data into the PI System.  This solution is available to OSIsoft university customers at no cost.  For the academic community, it was also announced that OSIsoft is developing a hosted, cloud-based PI System that will be used in chemical engineering unit operations labs.  The PI System will read and store data from NI equipment, archive the data, and allow students and staff to visualize the real-time and historical data.  Students will also be able to download their experiment data for machine learning and analytics using tools such as R, Python, and Matlab.   Please contact for more specifics.



Four OSI Johnson City co-ops competed in the 2017 OSI Users
Conference Visualization Virtual Hackathon and took First Place
in the ‘OSI Internal’ division.


The team submitted “PROJECT-OPERATION-PI-TANGO-DOOM-3D A proof
of concept for determining if Google's new Augmented Reality project, Google
Tango, would be useful and applicable to consumers of PI and its systems.
students are Simon Boka (UT-Knoxville), John Burns (UT-Knoxville), Phillip
Little (Clemson) and Andrew Bathon (ETSU). They are working this semester

on the Interface and AF Development Teams at OSI in Johnson City.


UC2017 Virtual Hackathon OSI Internal winners.jpg

Left to Right: Andrew Bathon, Phillip Little, John Burns, And Simon Boka

Here is a link to the Submission and a link to other Hackathon Category Winners


The Virtual Hackathon was the first of its kind at this s UC and lasted for the

3 weeks leading up to the Conference. Entries were submitted online.

Winners were selected on (1) Creativity and Originality (2) Potential Business Impact

(3) Technical Implementation (4) Data Analysis and Insight and (5) UI/UX.


Way to go students!

Joe Gantz

The talks from the orientation session were:


1. PI System 101 - General Introduction to the PI System for Academic Users

2. OSIsoft Collaborations with Academia - Pathway to Success and Grant Partnering

3. Getting Started with PI at your University - First Semester Tools for Success

4. OSIsoft Learning Resources - Transform Your Classroom

5. A Classroom Example and Hands-on Demo


Video recordings will also be available soon.

We put together YouTube Playlists to supplement the morning orientation session at the 2017 Academic Symposium.


Here they are:


PI System 101 - General introduction to the PI System for Academic Users

Academic Symposium - PI 101 - YouTube

Getting Started with PI at Your University – First Semester Tools for Success

Academic Symposium - Getting Started with PI - YouTube


Remember to check back after the event for slide decks from these and other sessions!

We’re sharing a demo based on an IoT device that we call RoomBot. This is a simple device that we designed for educational demos. It’s easy to assemble and easy to integrate with PI.


We want you to replicate this setup for use in your classroom! We documented the RoomBot setup steps and are sharing it here. We also share the sketch that we used to program the Arduino and the associated Python code that writes the data to PI using PI Web API.


We used this device to collect room data at the Johnson City office. We spotted interesting trends right away!





Getting Started:


This project had a small start-up cost. Arduino devices are cheap and the learning curve is small.


I’ll admit to being a little intimidated at the start of this project. I was not at all familiar with Arduino devices or any kind of microcontroller. I also don’t know Python, although I am competent with scripting in other languages.


I was excited to find that working with Arduino devices is simple! I was writing sensor data to PI within two hours of unpacking the device.


To start, I downloaded the Arduino IDE software, as instructed by the packaging, and loaded an example Sketch called “Accelerometer”. The sketch programs the device to read data from the sensors (accelerometer and gyroscope sensors) that are built into all Arduino 101 devices.


I then found an example Python script that reads data from Arduino devices using a quick Google search. I modified this quickly and added some lines of code to write data to PI using REST calls through PI Web API. And that was it! A quick and dirty prototype, with sensor data writing to PI.


We then passed the project over to a team of two engineers in the OSIsoft Johnson City Office: Steve Edwards and Caleb Steiner. They experimented with different Arduino sensors before coming up with the final RoomBot setup. The RoomBot has temperature and light intensity sensors, as well as a knob that is attached to a potentiometer; adjusting this knob mimics a set point change. They built an Arduino sketch for their setup, perfected and documented the associated Python code, created a schematic of the Arduino wiring, and documented their work. We’re sharing all of their work with you here.


Steve and Caleb also did some setup in PI for the demo and documented their steps. They setup the Arduino device in PI, using a template, and created a web-based visualization. We later brought a second Arduino online. Now we’re able toggle the display to look at data from either device.


Other Projects:


We've had a number of projects collecting data from Arduino devices and sending to PI that were posted to PI Square.  These include the following posts:

  1. Daniel Lopez used Powershell scripts to pull in data from various sensors, including solar cell output, ambient sound levels, equipment vibration, air temperature and humidity,     and motor current and position.
  2. Barry Shang hooked up his Arduino with a simple temperature sensor to monitor in his office, and collected the data with similar Python code, also writing data through the PI Web API.
  3. Butch Payne had a similar setup as Barry to collect temperature data from an Arduino, but instead passed the data through a Microsoft Azure Event Hub, before consuming the data with the PI UFL Connector


We’ve found many ideas for IoT projects with Arduinos or similar devices. Using the above methods, this data could also be collected in PI.


Some links with these project ideas include:


Some interesting ideas included:

  • Collecting additional environmental data (humidity, pressure, soil moisture)
  • Intelligent bug zappers
  • Gas detection (e.g. natural gas leaks, radon)
  • Smart meat smoking

Hi everyone,


Just a quick note to let everyone know OSIsoft will be exhibiting at National Instrument's NI Day UK on 29th November.


You can register for this event (free of charge) and also find the agenda here.


Stop by and see us at the QEII Conference Centre in London! We'll have a demo to show how you can leverage the combination of NI and OSIsoft in your environment.



OSIsoft recently hosted 2 events here in the Houston area that provided how customers are using the PI System for enterprise-wide operational insight during the current energy industry downturn:

-- Houston Regional Seminar (October 11)

-- Big Data Forum (October 12)

Although both events were open to customers and partners from all industry segments, both days seem to focus on how oil & gas customers are mining PI System data and combining the information with other big data to continually increase production productivity and reduce unscheduled equipment downtime.  Various companies presented how advanced analytics are enabling them to further squeeze the cost of producing a barrel of energy.  Notable presentations were given by Devon Energy, Apache, National Oilwell Varco, and others.  Customers and partners explained how the PI System Integrators to Esri ArcGIS, SAP HANA, Microsoft Azure, Spotfire, and other big data analytic systems significantly reduced their data extraction efforts.

The links to all video presentations from both days are found here:

Houston Regional Seminar:
Houston Big Data Forum:,advancedanalytics,bigdataforum

I hope you find the presentations as informative as I did.

If you're in the greater Houston area the week of October 10th, please register to attend our no-cost Houston Regional Seminar to be held at the JW Marriott, located in the Galleria area.  


Here is the link to sign up:


Also, there is a no-cost big data and advanced analytics workshop the next day at the same venue.  The link to register is here:


See you there.

As universities continue to reduce energy consumption on their campus facilities, buildings are getting "smarter", due to the deployment of IoT (Internet of Things) power consumption sensors and improved analytics using meter and building automation system (BAS) data.  That's good for everyone. In the past few years, the PI System has also been used to monitor energy usage at Major League Baseball stadiums and ice quality at NHL hockey arenas.  Here's just a few of the use cases:


Recent article on smarter stadiums:  How the smart stadium will transform the smart city

Short video on MLB's San Diego Padres vision for monitoring Petco Park:  OSIsoft: San Diego Padres & SDG&E: A Community System Story - YouTube

Short video about the NHL's Minnesota Wild and their journey to improve playing ice quality: The Minnesota Wild on Better Data, Better Communication, Better Ice - YouTube


Why not save some of that wasted power at your stadium?

Do you use MATLAB in your academic projects? If so, you should check out the recently released PI-to-MATLAB Utility! This simple, graphical tool allows you to import PI data into MATLAB. In almost no time, you'll be making use of MATLAB's computational abilities to enhance your time-series data.


Screenshot-PI System to Matlab1.png


For more information, reference the PI Developers Club White Paper: Using the PI-to-MATLAB Utility. The install package and the C# code for the utility are freely available on GitHub.