It is interesting to note how easy it is to recognise some specific appliances.
- Fridge - 35-40 cycles throughout the day. The fridge's square power draw is clearly visible throughout the night, and at various points throughout the day. However, it is less obvious in the evening when many other appliances are operating.
- Kettle - 10.00, 13.00, 17.30, 19.00. The kettle is distinguishable by its power draw of 2.5kW and its short duration for 2-3 minutes.
- Washing machine - 13.00, 15.30. The washing machine has an alternating power draw between 600W and 0W for its 1 hour cycle duration. However, it also has a solid power draw of 2kW for about 20 minutes.
- Oven - 18.30-19.30, 21.30-22.30. The oven's temperature is regulated by a thermostat so its power draw predictably alternates between 0W and 2kW. However, there is a period while it initially heats up to temperature when it is on continuously. For the first use the temperature was set to 170 degrees, while for the second use it was set to 110 degrees. This is reflected by the duration between 'on' cycles, similar to Alex's fridge's behaviour.
It is also interesting to analyse the lighting circuit's power draw. Since each light only has two states, 'on' and 'off', techniques such as steady-state analysis or combinatorial optimisation would work well here.
Now we know what you've been up to! :)
ReplyDeleteOn a more serious note - do you have an answer to the question: which device would you need to monitor to minimise the uncertainty about the monitoring of all the 'other' devices in the system. Basically, if you had a clamp, where would you put it so that you can take it out from your data and recognise the others better?
I think the lighting circuit is probably the most useful.
ReplyDeleteLighting accounts for roughly a quarter of my electricity consumption, and includes about 8 different light switches. Sub-metering the circuit makes it quite easy to distinguish between each room's lights. Also, I believe after subtracting the lighting circuit the data appears significantly less noisy. However, remember you need to be a qualified electrician to open up the circuit breaker box in someone else's home (I assume it's okay in my own!)
However if the question was: given a smart plug, which appliances would you meter? It would be much harder! I'll have to have a think about that.
Hello Oli,
ReplyDeleteI am working on NILM for my PhD myself and I am very glad I found this website. I would like to get more information on some of the routines you have developed in order to disaggregation. So far, I have seen many approaches, but your ideas are very interesting.
Hi,
ReplyDeleteGreat to hear from you! I'm only 6 months in to my PhD, and have so far only experimented with a few techniques. You can read about what I've done so far by having a look at previous posts on this blog. My website is: https://sites.google.com/site/oliparson/research for an overview of my research or links to other relevant people and projects.
Do you have any information about your PhD topic, research group or any reports you'd be willing to share?