Proposal for Home Assistant Integration to Support Grid Demand Management Programs

Feature Requests
1

Dear Home Assistant Team,

I hope this message finds you well. I’m writing this annonymously for now as this is a non-official inquiry but I work for a Utility Company. We are currently exploring innovative ways to leverage smart home technology to benefit our customers and improve grid reliability. We believe that integrating grid demand management capabilities into Home Assistant could provide significant value to both homeowners and utilities, and we are eager to partner with you on this initiative.

Background

As a utility, we face the challenge of managing electricity demand, especially during peak periods. High demand can strain the grid, leading to increased costs, potential outages, and greater environmental impact. Conversely, reducing demand during these periods can lead to significant benefits, including cost savings, enhanced grid stability, and reduced emissions.

Proposal

We propose creating an interface within Home Assistant that allows utilities to interact with smart home devices for the purpose of demand management. This integration would enable utilities to send commands to enrolled devices, such as thermostats, water heaters, and EV chargers, to temporarily reduce their energy consumption during peak demand periods. Here’s how the integration would work:

  1. Device Visibility and Selection: Users can choose which smart devices they wish to make visible to the utility. This ensures privacy and control over their home environment.
  2. Demand Response Commands: During peak periods, the utility can send commands to these devices to turn them off or adjust their settings (e.g., lowering a thermostat by a few degrees or pausing an EV charger).
  3. User Control and Override: Users retain full control over their devices. They can override any utility commands at any time, ensuring they are never inconvenienced or left without essential services.

Benefits to Users

  1. Energy Savings: By allowing the utility to adjust device settings during peak periods, users can benefit from lower energy bills due to reduced consumption.
  2. Financial Rewards: Users can earn bonuses from the utility for participating in demand response events, providing a direct financial incentive.

Benefits to Home Assistant

  1. Enhanced Value Proposition: Integrating grid demand management features positions Home Assistant as a leader in smart home technology, providing users with innovative ways to save money and support sustainable energy practices.
  2. Increased Adoption: Utilities can promote Home Assistant as a recommended platform for demand management programs, driving adoption among energy-conscious consumers.
  3. Community Support: This initiative aligns with Home Assistant’s mission of empowering users to take control of their home environment, while also contributing to broader societal benefits.

Addressing Potential Concerns

  1. Privacy and Security: We understand the importance of privacy and data security. Any integration would adhere to strict data protection standards, ensuring that only necessary information is shared and that all communications are secure.
  2. User Experience: The interface will be designed to be user-friendly and non-intrusive. Users will have clear visibility and control over which devices are included and can easily override commands if needed.
  3. Reliability: We will work closely with the Home Assistant community to ensure the integration is robust, reliable, and well-supported. This includes thorough testing and ongoing maintenance.

Conclusion

We believe this integration presents a unique opportunity to enhance Home Assistant’s capabilities while providing significant benefits to users and utilities alike. By working together, we can create a smarter, more efficient, and sustainable energy future.

We would be delighted to discuss this proposal in more detail and explore how we can collaborate on this exciting initiative. Thank you for considering our proposal.

2

How would you determine the priority of devices to turn off or reduce the power consumption of? Would you target power usage or device type, for example?

3

This is a though experiment right now but there is a big drive over here of reducing our grid demand during peak hours.

Our ideal case would be to target high use devices in a way that would impact customers as little as possible. The sweet spot for us would be to reduce customer energy usage without the customer even being aware of it (unless they check).

Some of the ways we are doing this today is turning down thermostats a few degrees during peak demand. We’re also turning off electric car chargers for periods of a time as well.

We’re also alerting customers during peak demand periods and letting them choose the devices. A way this might work in an automated way is a customer may choose to allow us to “disable” their washing machine and dryer during peak usage periods. Obviously this could be overriden but this would keep someone from “accidently” using the device during this time.

4

Nope, not giving control to anyone. It would require a level of trust that would be hard to establish.

5

Its an interesting proposal and one that I think is somewhat forward thinking.
My personal view is that very specific logic would need to be implemented, arguably something that could be voted on by customers.

I like the idea of short interruptions based things outside of historic trends, but I think maybe different layers of interruption could be implemented too? Limit to small changes, limit to small and medium, and so on. :thinking:

6

I understand your hesitation. Ideally this would be an integration that a user could both choose to enable and configure which devices they choose to expose.

7

From our point of view, if the customer is negatively impacted this doesn’t accomplish our goal. We would want to be as “invisible” as possible because, if the user is impacted, they are likely to simply override the change which defeats the whole purpose.

The user would choose to participate or not, choose which devices they wish to enable or not, and perhaps even be able to put “guard rails” in place to ensure their comfort (for instance, temperature thresholds if adjusting a thermostat).

8

As an end user, I’d perhaps be interested in the utility sending a demand response notification that’s accessible to my Home Assistant instance, perhaps with a duration and a requested or target reduction, but I certainly wouldn’t give direct control to, or expose any any devices.

Instead I might want HA to either prompt me to take action, or have an automation that I’m in control of to take some reduction steps. In lieu of reducing grid usage I might have my house draw from batteries instead, or pass through more solar to the grid.

On the other hand, my utility only really lets me get usage data from Opower, which lags days behind. If I were to set this up at all, I’d also want the integration to be providing real-time metering (and current rates), so I get some value out of it (a clearer and quicker picture of my usage.) I guess any demand response requests could be exposed in the energy dashboard as well.

1 Like
9

I hear your pain on the data issue. We’re looking at next-gen smart meters now, which should be wifi enabled. This will let us share realtime numbers with all of our customers going forward. We’re also doing optional programs that lets customers buy a bridge unit in the meantime to get realtime data.

As for the control, the programs don’t really work if we have to depend on manual confirmations on the user side. That’s why we were thinking we could have the users select ahead of time what they were willing to let us send command signals to, and the guardrails they wanted to put in place. This would allow us to calculate dependably the impact on the grid. Users could always choose to overide after the fact if they are impacted.

This is the balance between choice (everything is opt in, at a granular level, while removing any barriers for reducing grid load. Happy to talk about potential other solutions but based on all of our past programs, the biggest barrier is always getting people to “throw the switch” at the right time.

10

Let’s take a step back from the idea of an end user making a manual intervention. If I were using a demand response program, yeah, I’d want the behavior to be automatic.
However- I’d want that automatic behavior to be determined (and controlled) locally, not letting some external entity reach directly into my home.

Outside of demand response, looking at the forums, there’s a lot of people who try to limit usage for many reasons- current rates in regions where dynamic energy costs are normal, maximizing usage of solar generation in places where there’s not net metering, or even plain limits on the home’s electrical infrastructure (panel limits, service limits).

I think building something more general (& local) would get a lot more buy-in and interest. A local service to manage load shedding, with the side benefit that it’s demand response ready.

A service like that would have a few features you’ve mentioned, but operate a bit differently.
As an end user, I’d use some HA feature to mark devices as eligible for load-shedding with the service. (Tags? device attributes? entities? doesn’t matter, so long as they’re settable via automation if I wish to dis-enroll a device automatically under certain conditions)
As an end user, I’d want to define the parameters of what an eligible device can do to shed load (max target temp swing from current, floor/ceiling on the window, can the device just be turned off- and if so, do other actions need to be taken to shut it down gracefully? Maybe the load shedding behavior needs to be script- or automation-like.)
As an end user, I’d want to be able to trigger load shedding myself (based on solar generation, upcoming price change from the utility, etc)
As an end user, I’d want to be able to prioritize the order in which loads are shed.

From the utility (or non-demand response customer wanting to shed load) side, we’d want to know the amount of load that’s sheddable in a given moment.
From that side, we’d want to be able to request a specific reduction in load (optionally/alternatively, a maximum load) from the load shedding service.
From that side, we’d also want to know the amount of load that was actually shed.

So… given that, it might be worth engaging with the HA architecture team on how load management/shedding might be built into the existing energy features (ie: energy dashboard.) Once that’s in place, offering an integration for the demand response service that can interact with the generic load management/shedding features in the core energy tools of HA would be the best option. The integration you propose would get pushes from the utility side with the requested load reduction/duration, and would emit details on the sheddable load to the utility.

11

In my region (NSW, Australia) the local grid network operators already have controlled loads, mostly electric water heaters but also pool pumps and slab heating.

But they have priced themselves out of the market, charging WAY more for the off-peak controlled load tariff than we get credit for exporting excess PV to the grid. So naturally people with smart homes ditch control by the utility and manage it themselves.

Frankly all a utility need do is get their pricing right and people with smart homes will respond accordingly.

Those who don’t have smart homes can’t respond to an automated signal anyway, so that would be something basic like a phone app alert for them to manually choose what, if anything, they do in response.

By all means provide a signal consumers can respond to but as to taking control, we have a saying here “tell 'em he’s dreaming”.

As an example of the problem:
All larger air conditioning units installed here over the last decade or two have required the ability to use DRED control but its use is optional. Guess how many homes with larger aircon have taken up the option of utility managed DRED control?

Pretty much ZERO.

Grid utilities are on the nose and have been for a long time. Too many instances of screwing over consumers. Gaining the trust of (enough) households to allow the utility to have control over their energy consumption to make it a worthwhile endeavour would I imagine be a difficult task.

12

The only way I can see this working is for the utility to provide, say, a url for a JSON file that specified current energy situation. Then we could have HA read this file periodically and choose whether to change any systems based on it. That leaves control entirely in the hands of the end user.

13

Surely this could start with a simple “Market Alert” signal being made available.
I am turning my air con & dishwasher on when I have excess solar, based on my net power from my solar system, even if I am not home.
I am finding that getting the house cool during the day means I may not even need to run it when I get home (when peak charges kick in)!
I would be happy to just automatically turn that off if there was a market alert.

14

I think the very first step would be to implement an OpenADR integration in HA. A lot of utilities are already 90% set-up for this sort of thing.

My energy provider, BC Hydro, has several pilot projects in this space, some of which involve open standards like OpenADR. They offer rebates already for specific devices using proprietary protocols.

Especially now that they’re doing OpenADR for businesses, an HA integration would make a ton of sense and probably wouldn’t require too much involvement/investment from BC Hydro to get started.

Since we as HA users value autonomy, the utility would need to implement functionality to apply discounts based on actual compliance as opposed to opt-in. That’s already being done with time-of-day billing, so it isn’t much of a stretch.