Machine Learning => Energy Efficiency: Optimum Energy & Data Guild Interview


Data Science Weekly Interview with Kari Hensien - Sr Director Product Development at Optimum Energy and Cameron Turner - Data Scientist at The Data Guild

We recently caught up with Kari Hensien - Sr Director Product Development at Optimum Energy and Cameron Turner - Data Scientist at The Data Guild. We were keen to learn more about their recent collaborations, bringing data science and machine learning to the world of energy efficiency...

Hi Kari and Cameron, firstly thank you for the interview. Let's start with your respective backgrounds and your first experiences with data...

Q - What is your 30 second bio?
A - Cameron - Architecture student becomes software engineer becomes data analyst becomes data scientist - still learning about all of the above.
[Editor note - Cameron co-founded ClickStream Technologies in 2003, which was acquired by Microsoft in 2009. Cameron holds a BA in Architecture from Dartmouth College and an MBA from Oxford University.]
A - Kari - Long-time product planner who made the jump from large software company to startup - currently building products that leverage data-science and machine-learning disciplines.
[Editor note - Kari leads product management strategy at Optimum Energy. Prior to Optimum, Kari spent 15 years at Microsoft, most recently as Senior Product Planner in the Windows Product Group, where she directed product planning for the Windows hardware application development platform.]

Q - How did you get interested in working with data?
A - Cameron - I believe that statistics and creative data science can create answers to some of the world's toughest problems. Sometimes solutions can be finessed by correlation and analysis, rather than brute force approaches that attempt to answer a question directly. Data (at times) can make sense of the world and unlock the universe's secrets.
A - Kari - I have a natural instinct to question and dig deeper to gather the data needed to make an informed decision. It's incredible to now be looking at how a system is changing over time based on what it is learning.

Q - What was the first data set you remember working with? What did you do with it?
A - Cameron - In high school my friends and I created a tiny FM transmitter and hid it in the teachers' lounge. We were hoping to develop a database of article ideas for the student newspaper. Incidentally, we didn't get that far before our bug was found in the lunch table napkin holder. They weren't too pleased with us when we fessed up (to get our bug back).
A - Kari - Honestly, I was a girl scout and the cookie season was upon us. I found myself trying to figure out how many houses I would need to stop at in order to sell enough boxes to get the Rubik's Cube.

Q - Was there a specific "aha" moment when you realized the power of data?
A - Cameron - While working at Microsoft in the 90s we developed an opt-in program for enthusiasts to share their software usage with us through nightly uploads. No one knew if people would agree to do this, but after the first night there were hundreds of new uploads to parse and analyze. I got goose bumps and remember thinking: "This is going to completely change how software is made." Of course now, the data is real time and the analysis can be done through learning algorithms. At the Data Guild we see opportunity everywhere for machine learning to massively disrupt industries under human control. We're humbled daily to be a part of this transformation.
A - Kari - At Microsoft we were trying to prioritize feature requests. There was no shortage of feedback from customers telling us what they do. Cam's work allowed us, for the first time, to put together a picture that compares what people say they do with what they actually do. We ended up prioritizing very differently as a result of data of actual use.


Very interesting background and insights - thanks for sharing! Let's change gears and talk more about Optimum Energy…

Q - What specific problem is Optimum Energy trying to solve? How would you describe it to someone who is not familiar with it?
A - Kari - Optimum Energy is focused on energy optimization in enterprise facilities with a solution that provides automated, continuous commissioning through dynamic adaptation of complex HVAC systems. Essentially, Optimum uses technology that manages HVAC systems directly and reduces the amount of energy that they consume.

Q - Which Optimum Energy technologies/solutions have been most successful?
A - Kari - Optimum Energy is best known for its OptimumLOOP technology, which provides continuous, system-level energy optimization of centrifugal chilled water plants. The technology continuously and dynamically adapts to fluctuating load, weather and occupancy conditions to yield the lowest possible energy draw while maintaining occupant comfort.

Q -Tell us a little more about the partnership with The Data Guild - how did it come about? And what projects will you be working on together?
A - Kari - As a startup, having the funds and ability to invest in the R&D required to implement machine learning is challenging. When we recognized that this was an investment we needed to make, we connected with Cam and the Data Guild team to help us with the expertise needed to begin our efforts building the discipline. We are focused on equipment-level and building-system level projects that enable autonomous optimization of an HVAC System.

Q - What excites you most about bringing machine learning and energy issues together?
A - Cameron - There is an immediate opportunity here to substantially reduce carbon emissions through machine learning. I love the fact that I can draw on data science best practices that are working in other verticals and apply them to improving energy efficiency.

Q - What are the biggest areas of opportunity or questions you want to tackle?
A - Cameron: Three things

  1. Product line expansion and enhancement: equipment-and system-level HVAC efficiency
  2. Ecosystem enablement: real-world equipment operating specifications
  3. Customer targeting and opportunity analysis: initial plant assessment

Q - What machine learning methods have you found or do you envision being most helpful? What are your favorite tools/applications to work with?
A - Cameron - We use correlation/covariance analysis along with regressions to do basic modeling and build out our view of the landscape. We use both supervised and unsupervised learning to build clustering and identify untold structure in plant performance. We use recursive partitioning to identify custom rules for local set points based on global algorithm development. In terms of favorites: R/R-Studio, Python, Java, SQL, Tableau, Hadoop, AWS

Q - What publications, websites, blogs, conferences and/or books are helpful to your work?
A - Cameron - We feel indebted to our network of affiliates (http://www.thedataguild.com/people) for ongoing support and review of ideas and approaches. Specifically, Paco Nathan and Dennis Lucarelli for their continuing support of our work. We love O'Reilly and look forward to their Strata conference here in the Silicon Valley (on now!), as well as the Data Visualization Summit where Cameron Turner will be speaking this winter.

Q - What project have you been working on this year, and why/how is it interesting to you?
A - Cameron - We're moving into pilot-test phase with a project that focuses on equipment recommendations in HVAC systems. This expands on the energy optimization that Optimum Energy currently provides and sends additional information to a facility about the most efficient combinations of equipment to run at a given time. Initial tests have been promising, and we're excited for the next test stage.

Q - What has been the most surprising insight or development you have found?
A - Cameron - We wanted to better understand what was happening within a chilled water system around a chiller surge. Engineers know a chiller is going to surge instinctively. They just need to be at the plant and they can see it, and feel it. We are trying to create vibrational and acoustic classifications around a surge to be able to better understand and predict them.


Very interesting - look forward to following future progress as these projects reach completion! Finally, it is time to talk a little more about each of your accomplishments and what you think the future of data science/machine learning and energy efficiency looks like…

Q - What in your career are you most proud of so far?
A - Cameron - Developing high-performing teams of great data scientists with diverse backgrounds and skills.
A - Kari - Seeing the products I've helped to develop in use and valuable to customers. Most recently, I planned, designed, built and shipped my first mobile app: OptiCx Trend.

Q - What does the future of machine learning and energy look like?
A - Cameron - Big question. It is inevitable that near-to-real-time cloud-based decision support systems will come to the energy sector. In fact, energy and related fields will be the first to embrace and extend the concepts of machine learning and true big data opportunity, due to: 1) the closed form of some aspects of the problem (for example, lower kWh consumption, higher savings), 2) the enormous upside of successful implementation, 3) the critical impact CO2 emissions have on the earth's future.

Q - What's next for Optimum Energy?
A - Kari - We are compiling a lot of valuable real-world operating data: we currently have 250 cumulative years of data, and we add approximately 8 years each month. Our long-term plans involve continuously improving energy efficiency for its customers by leveraging this data using machine-learning and predictive-maintenance algorithms. It's an honor to be playing a part in this transformation.


Kari and Cameron - Thank you so much for your time! Really enjoyed learning more about your backgrounds and what you are working on together. Optimum Energy can be found online at http://optimumenergyco.com and The Data Guild at http://thedataguild.com.

Readers, thanks for joining us!


P.S.If you enjoyed this interview and want to learn more about
  • what it takes to become a data scientist
  • what skills do I need
  • what type of work is currently being done in the field
then check out Data Scientists at Work - a collection of 16 interviews with some the world's most influential and innovative data scientists, who each address all the above and more! :)

You might also enjoy these interviews because you are awesome:

Back

Sign up to receive the Data Science Weekly Newsletter every Thursday

Easy to unsubscribe. No spam — we keep your email safe and do not share it.