For Raytheon, the technology company that develops defense, civil government, and cybersecurity solutions, tracking building energy use isn’t just about cutting back at peak times of the day. Instead, it’s about using tech to track exactly what’s happening and then optimize operations, explains Paul Kramkowski, senior manager of design and construction at Raytheon.
“The internet of things and the big data world is here now, and we need to be using all the information we can,” he says.
Kramkowski’s Raytheon colleague Lang Lawrence will be discussing peak shaving technologies at the 2017 Environmental Leader Conference in June. Recently we caught up with Kramkowski to find out what it really means to add intelligence to building automation systems.
How does peak shaving work and why is it important to understand?
Peak shaving is basically lowering your total energy uses at peak times of the day. A lot of energy rates are calculated based on how much energy you use at that time so for large employers and energy users, it’s a big value to be able to cut down your peaks.
A technology solution can be implemented using metering and building automation systems. For instance, if you do thermal energy storage or cutbacks on certain things at different times of the day to minimize your load, it can really affect energy costs. For us, it’s about tracking what we’re doing and when we’re doing it using a technology solution to do different things in the way we operate.
What is Raytheon’s approach to peak shaving?
Peak shaving is one element of our sustainability solution. There’s more data available, and that’s where the intelligence comes in. One of our major systems is called fault detection and diagnostics. Most building automation systems do an alarm — the temperature is too hot, too cold, the humidity is out of range. The software we implemented does algorithmic analysis using as many available data points as possible to determine [if we’re] operating efficiently or that valves are leaking, dampers are stuck open, things you wouldn’t otherwise know unless you took a look at them.
What kind of difference has that made?
Our analysis tool actually gives us the cost to be experiencing that fault. So some faults can be as costly as $100,000 a year in energy costs, and some may be as simple as a few hundred dollars a year. By having that information, you can prioritize and go fix those items — like chilled water valves stuck open or closed, simultaneously heating and cooling spaces. It helps us to know where to go to drive down our energy costs, especially peak time energy costs.
How long have been doing this?
We’ve been doing a building automation system solution for many years, but in the past year we implemented our additional layer of energy management and fault detection and diagnostics. We started adding meters around our site in 2015, and started implementing software to do the analytics and analyses throughout 2016. Late last year we started operating and taking the information generated — our peaks, our base usage — to go after opportunities. Peak shaving is one small portion. It’s more about using intelligence to drive energy savings across the board all hours of the day.
What does intelligence mean in this context?
The newer heating and cooling equipment generate data in some way, whether it’s temperature, humidity or just positional data. And that information can be used to determine that things are operating properly, improperly, or just efficiently. The buzzwords now are ‘the internet of things’ and ‘the system of systems’. Building automation systems are one of the earliest internet of things-type systems. We’re utilizing that platform to figure out where we can improve.
When you were working on adding meters, what was the scope of that?
I have responsibility for multiple facilities, but we started our project in Tucson, Arizona, on a large manufacturing facility that has about 3.5 million square feet of space and about 100 buildings. That included adding meters at all of the major energy-using buildings. We deployed about 50 meters to analyze at each building and even get data at certain high-energy equipment points.
The scope of that project was large but we worked our way from a facility level to a building level down to an equipment level. When you buy a chiller or an air handler, most of the time the equipment comes with embedded smarts. Old facilities like ours, sometimes you’ve got to add sensors or equipment. We tried to utilize everything we had in place already to start this process, and then we will eventually add more data-generating devices to it to make it smarter.
What are some of the hurdles involved for companies?
Some of it is cost. There is an initial investment, but you can do it almost by attrition. Start with a plan and gradually work your way there. It may take many years, but it’s a path. The biggest thing I learned in our implementation was how to start down that path. If you can’t spend the money right now but you have traditional maintenance activities or a replacement of equipment, make sure that decision is going to fit where you can eventually get on board with a big data intelligence system.
Once you start to look at the return on investment related to what you can and should do, it starts to make a lot of sense. It takes resources and money, but it sells itself.
In terms of intelligence, are there helpful tools available?
The sensors, the pieces of equipment are always in constant improvement mode. Those will take care of themselves. Probably the newest element is the software layer that different companies have started to put on the market. Software can take all these data points and use them in certain algorithms to identify where you’ve got issues or opportunities.
What about battery storage and on-site solar storage options?
Solar is a peak shaving method that can reduce your energy loads significantly or at least augment your energy use. As they condense and get more efficiency out of batteries, that’s going to lead to a more 24-7, self-sufficient type of operation. It will make that energy last longer throughout non-peak hours.
Does Raytheon have projects involving battery storage?
We’ve done some development and have found a little success. We implemented battery storage units that support our solar charging stations, where it charges off the grid and solar during day and then at night the battery packs kick in if we need them. It’s still not a fully viable solution for a large facility, but there is development happening. We’re waiting on that technology jump.
What do you think the future looks like?
The internet of things and the big data world is here now, and we need to be using all the information we can. The future is using the information that our equipment, our people, and our world generates — and analyze what we’re doing, how we’re doing it, and how we can improve. That’s what the future is: utilizing big data to make us better.
Raytheon energy manager Lang Lawrence will be speaking at the Environmental Leader Conference in Denver June 5-7, 2017. His track, Artificial Intelligence and Peak Shaving, starts at 10:10 am on June 6.