It comes as no surprise that the UK is on track to miss its revised smart meter schedule, and in the same week that this setback has been announced, Navigant has predicted that smart meter penetration is going to reach 59% in 2028, up from some 41.2% today. This begs the question, one that could be quite uncomfortable for some in this sector – do we actually need all these smart meters?
This is a problem that crops up quite frequently in IoT use cases. There is a very sharp case of diminishing returns that appears once you’ve passed a certain threshold of IoT stuff. At some point, throwing more sensors at the problem doesn’t improve the outcome, and we are not sure if the industry has done the sums for smart meters – especially in their current incarnation.
Now, for customer billing, the current generation of smart meters are certainly valuable. The in-home displays they are typically paired with also help provide better customer awareness, which can led to proactive take up of Behavioral Demand Response (BHD) technologies and products, such as encouraging the installation of more efficient appliances, and taking advantage of local government schemes that might provide free or financed upgrades to a home.
But all this effort to save customers a monthly or quarterly hassle, reading the meters and sending the numbers off to the utility, does not square with the expense. No, the utilities benefit from getting a much more granular view of how their customers are actually using the product they supply. Instead of knowing a home’s consumption on a monthly or quarterly basis, the utility can have up-to-the-minute views of each customer’s usage, with most utilities settling on an interval between 15 and 60 minutes.
Now, that has no immediately obvious benefit to the customer, but this data does let the utility purchase power from its wholesaler in a more efficient manner. In the short term, this lets utilities take better advantage of renewable resources of energy too, and if the utility can better envision how it will need to supply its customers.
The readings from the homes can be pulled back from the edge, transported on your network of choice, to a centralized business application, which can crunch all that data using fancy analytics packages, before spitting out some lovely graphs on that single-pane-of-glass that the executives love so much.
But at what point does this application have enough data? Surely more than 10% of homes? Perhaps 25% of homes would be enough to extrapolate the entire picture from? Surely, over 50% is already fast approaching overkill, and if you ever made it to 70%, how would you ever justify trying to get to 100% penetration? At what point in that process do you find yourself with enough data to run the business?
It is worth remembering that the current generation of smart meters do very little. They are certainly not worth of their ‘smart’ moniker, and the only new features that some have added are the ability to perform remote disconnects if a customer has fallen behind on their payments – the legality of which varies depending on country or even county.
No, the next generation of smart meters are the ones that might actually earn that title, acting as the conduit for demand response programs that can reach into the home and dynamically alter the load in a way to best take advantage of renewable energy. However, we have seen that early DR programs don’t need smart meters, relying mostly on the smart thermostats. To this end, if smart home devices are the conduit, then what exactly does a new generation of bring to the table?
Tighter integrations between the meter and the in-home devices would be nice, but that would require something that is anathema to the smart home – a single overarching standard that would dictate how a meter would be talked to by the myriad devices.
But as is the case in all things IoT, that sort of problem could be reconciled in the back-end, somewhere in the cloud, as an API exchange between two different environments. A cable or its wireless counterpart, between the smart home equipment and the meter, is not necessary in this dynamic. Virtual connections between the two would be done remotely, and so, a shiny new meter isn’t really adding anything to this architecture. Indeed, you could achieve the same sort of outcome in a home with a conventional analog meter.
So, if DR-type applications don’t need a smart meter in the home, and it looks like the additional work is just going to be a complete headache without obvious RoI, then are utilities going to be particularly motivated to get beyond 50% penetration? What is the net benefit of moving from 50% to 51% penetration? Is that net benefit the same as moving from 70% to 71%?