As promised, I rewrite my research plan here for your suggestion. So far, I have got GPs prediction running on the UK Carbon Intensity. The result of the UK carbon intensity can be acceptable. However, the prediction of the specific user consumption by GPs based on historical data is hard, probably impossible to formulate a reasonable covariance function. Hence, we need to use the annotated events from FigureEnergy system, (or even Non-Intrusive Load Monitoring (NILM) technique) to predict activities ahead, and improve the user demand prediction.
I assume those prediction, which mentioned above, could be done. After that, we focus on providing feedback such that users can be raised awareness of carbon intensity based on their everyday activities. Up to this point, we can have two options:
1 - From historical data, we can do some analysis and show information of devices usage and carbon intensity. From this information, we hope users can have more attention on their energy usage, so they can change their behaviour in a positive way.
2 - From the prediction, we can run optimisation to minimise the carbon intensity. Then, we can advise some action to users to reduce the carbon intensity in term of the use of their devices. Probably we can suggest users to defer some events from the high peak of the grid carbon intensity to other low peak.
Furthermore, we want to think of the feedback interface where users can colaborate with agents to plan their activities ahead.
Henry, this sounds good, but it is very high level.
ReplyDeleteDid you look into what features you could use for the prediction of events? (e.g. frequency, time of day..)
> Furthermore, we want to think of the feedback interface
> where users can colaborate with agents to plan their activities ahead.
Yes, indeeed, what are your ideas around this?
>> Did you look into what features you could use for the prediction of events? (e.g. frequency, time of day..)
ReplyDeleteYes, I agree. That's the idea, and I really want to make it clearer a bit. I think the frequency and time of day would definitely help to predict events, however I have not found out how to do it yet.
Regarding to the feedback interface, where users can colaborate with agents. The feedback should have all relavant material of a good feedback in literature research (I will describe it in the next post). Then I think I like an idea of making a collaboration between users and agents in calendar (for example, google calendar). Should I research more about this idea?
Hello.
ReplyDelete> I have not found out how to do it yet
Try to search on google for "event prediction gaussian" and also on google scholar.
As I mentioned via email, you may also find something helpful if you look into "predicting faults in machinery" and predicting "network traffic" (e.g. predictions of server requests)
Once you have more information about that, you should start to think about how to model this situation, e.g. what are the parameters that you can train from the data.. but that should be done in light of prior work in this area.
What I mean is: ideally you should try to find existing work in a different domain and translate those models to our situation.
Does it make sense?
> Then I think I like an idea of making a collaboration
> between users and agents in calendar (for example, google calendar). > Should I research more about this idea?
I like this idea, because there is already quite some research on that (agents to keep calendars and agents to book rooms) -- please look into that and report...
There is also a good existing infrastructure (e.g. google calendar, ICS format, etc..)
Gopal, what do you think?