This piece is a quick look at what I’ve been thinking about this week. Let me know what you think!
A client asked me to explain the competitive landscape for energy efficiency and advanced supervisory control platforms.
I decided to frame the conversation from the perspective of the building owner. If I’m a building owner and I want to reduce the energy consumption of my portfolio, what options do I have?
Among those options, which are simpler, cheaper, or more foundational? Which of the options enable the others?
Enter: the hierarchy of needs.
We first introduced the hierarchy a few weeks back when we cut through the hype of advanced supervisory control, the pinnacle of the hierarchy. Today, let’s walk through it from the ground up.
Level 1: Energy management analytics
Level one is where every building starts down the energy management path. Or at least where they should start. It’s very, very low cost relative to everything else in the hierarchy to perform analytics on utility bill data (left) and interval meter data (right).
With a small investment and SaaS fee, you can benchmark buildings against each other, identify opportunities for savings, create reports (GHG, etc) and cool dashboards, set baselines, and calculate savings (M&V). With firms like Urjanet in the mix, you can even have all of that automated. Far too few building owners are taking advantage of these foundational opportunities (see this rant for the numbers).
This level is foundational for a few reasons. As I said already, it’s cheap, which means it’s often easiest to sell. It can be done for every building with a utility bill, whereas upper-level solutions require a building automation system. Third, parts of it enable everything above it in the hierarchy:
- Utility costs and savings analytics can help justify further investment up above.¹
- Interval meter analytics can help with the analysis of the tariff, which makes energy savings calculations more accurate and is required for some parts of advanced supervisory control that control equipment based on time of day utility rates.
The problem with level 1, and why we must progress further: if you only have meter data, you don’t know what’s causing high usage and you can’t control it even if you did.
Level 2: The enablers
Level two is where you start to dig into system-level performance and data. Building automation systems control stuff and save energy. That’s vital and allows you to tune the system’s setpoints, sequences, schedules to save energy.
However, BAS vendors (and most firms that do energy retrofits) are often poor stewards of the data created by automation. The data must be made useful by some sort of centralization/visualization/normalization solution. This solution can take many forms, the details of which are beyond the scope of this essay, but here’s a list:
- Analytics/FDD/ASC providers freeing the data for their solution (e.g. KGS Buildings)
- Integration + supervisory control platforms (e.g. Niagara)
- Open data layer or data lake platforms (e.g. Onboard Data)
- Digital twin platforms (e.g. CityZenith)
- Cloud providers (e.g. Microsoft Azure Digital Twin) creeping into the smart building space
I’ll also mention that “Blank sheet of paper” BAS vendors (e.g. 75F and Passive Logic) are building the modern BAS in a way that doesn’t require the data to be freed. It’s already free from the start.
This level is called “the enablers” because it allows everything above it in the hierarchy. And if done creatively, RCx and energy retrofits can actually be a funding source for everything up above. However, if this level is done poorly, it can also inhibit everything up above.
The problem with level two, from an energy management standpoint, is that energy savings created at this level are likely to degrade or drift away. That’s why we must proceed…
Level 3: System-level analytics
Level three is where the analytics come in. Fault detection and diagnostics (FDD) find issues and anomalies. Those faults can be used to (automatically) create work orders and change O&M processes from reactive to proactive. Not only is the energy drift prevented, but new opportunities for savings are continuously discovered as the monitoring continues over time.
The analytics uncover two types of opportunities:
- Physical issues—leaking valves, loose actuators, vibrating pumps.
- Software issues—setpoints, schedules, and sequences (the Three S’s) in the building automation system need to be optimized.
This distinction is important as we move to level four. But first, there are two problems with level three²:
- The workflow requires human action. It’s an open loop. Closing the loop requires regular use by a well-resourced team to fully convert data to information to insights to action to verified results. That sentence is NOT easy and actual execution on it is rare.
- It’s limited by the performance, capabilities, and business restrictions of the underlying BAS. If the analytics discover an opportunity to optimize a control sequence, I need to call the BAS service contractor and have them write a new one. Easier and cheaper said than done.³
And we proceed…
Level 4: Advanced supervisory control
ASC can partially close the loop by automatically optimizing the three S’s. It’s inhibited by physical faults, among other issues, and therefore is dependent on solid levels two and three.
As you can see, the hierarchy represents not only the available energy conservation options, but also a roadmap for the owner of an existing building.
As a final note, you might notice how the hierarchy is constructed with distinct building blocks. Unfortunately, this is another way the silo problem rears its ugly head. Far too many companies in our industry sell their building block as a stand-alone solution, when in fact it’s all interconnected. Reality looks more like this:
What do you think?
¹ And ideally, the owner wouldn’t rely on potentially-biased vendors in the hierarchy to do their M&V. M&V should be independent or at least open and independently-verified.
² Here I must pause and say that if every building owner were doing level 3, our industry wouldn’t be wasting much energy. I could retire and take piano lessons or something.
³ It’s also limited by the sophistication of the open data layer, but that’s a different story.