Congratulations. Your results are vastly better than ours…Sense engineering tells me that my home (in particular my deep well constant pressure pump) isn’t “Sense friendly” and after 19 months it’s detected very few of our 117+ electrical devices…and those not very reliably. So, when I get a chance, I’m going to pull it out of the various panels (circuits in one, solar in another) and give up.
How are you supposed to hear a conversation across the room with someone yelling in your ear?
That is what you’re expecting of Sense. The sheer amount of electrical noise your house has just from so many devices on it is the constant yelling. It only makes ‘Sense’ that you’re house is challenging is not impossible.
I too recently installed geothermal (open loop with variable speed well pump like you), and depending on how much yours runs I can easily see how devices are ‘missed’ or never learned.
Here is the sequence JUST for the geothermal part of your electrical load:
- Heating call pump turns on
- Blower moter ramps up (if its a fan coil)
- Buffer tank setpoint hit, calls for heat pump to turn on
- Before heat pump turns on, open loop solenoid opens and allows your well water to flow
- Buffer tank circulation pump turns on
- Desuperheater pump turns on (if equipped)
- Heat pump compressor turns on (huge power event)
- Variable speed well pump may or may not have turned on yet depending on pressure tank size and where the pressure was at before the heat call
Number 8 is a big deal because it looks different almost EVERY TIME the system cycles because it can happen anywhere through the stage at EXACTLY or very close to other devices in sequence.
While I would love for Sense to identify all these things perfectly, its not realistic. If you want precision you’ll have to look into something purpose built like WELServer.
In the meantime I have my variable speed well pump hooked to a HS110…
Correct, but that’s never mentioned in the Sense marketing….have also never seen it in the FAQ.
I installed a WELserver a decade ago when we built our home, and it’s worked perfectly. It monitors the biggest consumers (Geothermal, hot water heater, radiant system, A/C, Solar, etc), but the reason I bought Sense was to get a handle on all the other devices (washer, dryer, stove, ovens, refridge, computers, printers, entertainment center, etc) and it’s done none of that…sigh. So, I’m giving up.
No, your experience is vastly worse than most others. There’s a difference.
You keep saying that.
Lemme buy your sense monitor, @andy
I bought a Wemo just to make a fair comparison. The default power on feature is a big deal. The Wemo community has been asking for this for several YEARS.
Because of this, I can’t put the Wemo on any critical devices.
TP-Link was also real easy to install and Sense found it faster.
Do you get useful data to help with your machine learning for finding devices that are not plugged into smart plugs? By using these devices, are we helping the process or just prolonging the eventual detection?
This is touched on in the blog mentioned above and in release notes: https://blog.sense.com/smart-plug-integration
The short answer is yes, we absolutely get useful data and that was one of the main impetuses behind this integration. Finally, we can get real ground truth data from devices that are otherwise pretty tough to nail down. As the blog states, this does not net immediate benefits for ML detection, i.e., you can’t “train” Sense with smart plugs, but it gives us loads of great data to build better models, which will improve detection for everyone down the line.
Machine Learning is typically done by training the ML/AI engine though, so I continue to question why Sense keeps insisting you can’t “train” Sense with more data? All ML things I’ve done/seen are typically done by training the models to recognize things by providing it lots and lots and lots of data samples. For example training an AI engine to recognize a cat in a photo you build your models and then feed it 1 million pictures of cats and your recognition should be somewhere in the > 99% recognition range.
Granted I understand there’s not 1 million Sense users (yet) and certainly not 1 million Sense users that would have the same device (make/model/etc) on one of these Smart Plugs to create that kind of an accuracy, but you definitely should be able to use the data from the Smart Plugs as “training” data for the ML engine to “recognize” that this particular set of data belongs to “device type x” of “brand y” and model “z” (assuming people fill this out in the fields available) to use that to train the ML engine. Every data point is another data point that will make the ML engine smarter/better, so feeding this data to the ML engine seems like the correct way to “help” the engine learn more/faster/better/etc.
Just my $0.02
My response could have been better worded. Usually when users write in about “training,” they’re talking about supervised learning done by them, i.e., going around the house and turning devices on/off to force detection. This is not possible with smart plugs at the moment. We will absolutely be using that data to build better models on our end, however. In this way, sure, we’re training the models. I just try and avoid that language given the connotations it has for many users.
And it’s not necessarily that user-driven training isn’t possible, it’s just wildly impractical and would likely be an incredibly frustrating experience to deliver us the fine-grained data necessary for detection.
Ok makes sense. That said though, if for example I would have an HS110 on my washer (say it wasn’t detected yet, in my case it is), Sense will now attribute the power usage it gets from the HS110 to the “Washer” device that it sets up as part of the HS110 integration. I would assume it’s going to continue to “try” to still identify the power signature of my washer the “old fashioned” ML way so it could happen that at some point in the future, it now detects my washer using the ML model (whether or not at this point it uses the HS110 data or not is irrelevant). At that point since it now has “detected” my washer the ML way, I could (or at least should be able to) remove the HS110 from my washer to utilize it somewhere else without this affecting the “ML detection” of my washer… right?
Does that make sense?
That definitely makes sense, but it’s not quite how the integration works. So, we’ll actually make every attempt to not detect a smart plug device via ML after you’ve connected it to a smart plug. This is very much intentional, as we would not want wattage to get duplicated from multiple iterations of the same device or device components. That said, we still collect data for it in the typical fashion and if you remove the device from the smart plug, that data will be used to help with an ML-based detection.
Interesting. That really makes no sense as to why you wouldn’t want to continue to try to detect the device via ML as you can feed the ML engine model for device type X very specific data about the device plugged into the smart plug (gathered from the integration) to help that process potentially go quicker.
I can see there could be some difficulty with double counting usages for the relatively short period of time between the time the device (component) is detected the ML way and the time the user confirms that “Device X” that was detected actually is (part of) the device(s) that are plugged into Smart Plug X. Once that identification is confirmed all that data can be attributed correctly to the combined “merged” device(s) (smart plug one + ML one) again though.
Keep in mind that smart plugs aren’t just about quicker detection. They’re also about control. That double counting becomes a lot more serious when a user intends to use a smart plug as a permanent control mechanism for the device. Still, I should add that this integration is new. As we actually see it put to use by users, functionality could change down the line.
I think I’d like to see us be able to configure a smart plug as either controllable or not controllable. There are some plugs that I have for control - lamp, subwoofer, etc. I have other smart plugs purely for measurement. For the latter, I’d like to see some indicator that says that Sense ML has detected the underlying signature of the device and that I can remove it from the smart plug because I don’t need the smart plug electrical usage monitoring anymore.
I’ve been thinking of buying some extra smart plugs for my washer and (gas) dryer. This would purely be for electrical usage monitoring, not for smart control. I’m not sure why my washer and dryer haven’t been detected yet.
As of the last patch, you can turn on/off the control functionality for individual smart plugs. You’ll see that options in the Device Settings screen.
As for the possibility of removing smart plugs after Sense has accurately detected the device, that is the goal here. We’re pulling in lots of ground truth data in order to implement that functionality in the future.
That makes all sorts of sense to me. Effectively the smart plug would be acting like a rapid response 24x7x365 human observer, accurately telling Sense when something turns on or off and probably more promptly than a human user could possibly do it. So it doesn’t interfere with Sense’s underlying architecture and it eliminates the need for a customer to be watching the alerts all the time.
And, since Sense would tell the user wen it’s identified each smart plugged device, the user can then decide when to remove the smart plug and let Sense do its job.
A related question. I have LED lighting all over my home including stand-alone lamps and many lamps on a single switch. Would it be reasonable to assume that using Hue or TPlink identification on the stand-alone lamps would focus a bit more learning on things that have not been identified?
As both a Hue and HS110 smartplug user, I’ll give my perspective. The beauty of Hue and smartplugs is that they result in more known devices in your house, and less power ending up in either Always On or Other categories. The HS110 also helps with learning the device it is attached to, but not device on outlet strips, since the smartplug supplies Sense with ‘ground truth’ for that specific device. ‘Ground truth’ is a key ingredient in the machine learning training process.
I don’t think having things identified via smartplug or Hue helps focus Sense on ‘other things’ within your house. It’s not like the Sense algorithms can subtract off found devices from the waveform. Training is more about Sense learning how to focus in on very selective aspects of the incoming waveforms, not ‘elimination’. But you do benefit from others who use smartplugs and supply ground truth for devices you have in your house, that you may or may not have used a smartplug on.