hi,
I am new to open energy monitor community. I am exploring an idea of training a machine learning models to predict power consumptions for different heat pumps, using the anonymous data in https://heatpumpmonitor.org/.
Is there any way I can get a dump copy of the underlying mysql database?
Thanks,
Weiqi
Hello @jiweiqi10 welcome to the forums!
It’s not possible to give a dump of the database as it contains passwords & emails. I imagine you would really require the underlying data which is stored seperatly to the mysql database as feed data in the emoncms application. We will be documenting API’s at some point and clarifying terms of use.
Could you tell us more about your project? Is this a university project and would you be able to contribute your work back to the community in an open source way? and reporting / findings of the work?
Hello @TrystanLea
This is an older topic but I assume it makes sense to continue here.
I am doing my Bachelors’ thesis on non-intrusive fault detection and diagnosis of heat pumps. The primary goal is the detection or prediction of hard faults and machine states but soft faults like refrigerant leaks may also be detected.
Access to as much of the historical feed data as possible would be very helpful for training and evaluation of fault detection models. The API works great for larger time intervals, but for the short 10s intervals the requests take a lot of time and probably a lot of computing resources on the community server, which I do not want to overload.
Are there any feed dumps available somewhere? Could they be provided via a fresh dump or are you aware of any users that may already be collecting feed data?
Thank you and best regards
Patrick
Hello @PatrickFrancke
Thanks for getting in touch.
Yes the API is not really designed for this type of use and may result in slower experience for other users.
I wonder if you could test a proof of concept for this using the Electrification of Heat dataset first? https://datacatalogue.ukdataservice.ac.uk/studies/study/9050#details, this is a 2 minute dataset of all the main metrics that we also collect. I downloaded and created the following easy access visualization tool for the data: https://eoh.heatpumpmonitor.org which should give a good idea of what the data looks like. There are also more heat pumps there with installation issues and poor performance so probably more opportunity for fault detection.
It would be great to chat however about what you are working on, it would be great to integrate something like this in HeatpumpMonitor! perhaps we could have a call about it to discuss?
Hello @TrystanLea
Thank you for pointing out the Electrification of Heat dataset. I was unaware of its existence and I believe it could be very well suited as basis for further research. Especially as the datasets are available in pre-cleansed format. This will be a great starting point.
Thank you for the offer to chat and discuss integration. Right now I am very much at the beginning and still researching approaches on how to use the data for models. I will gladly follow up on your offer at a later stage, when I have some findings. The goal is a PoC standalone service that may be integrated in other systems or other tools that may have this kind of data recorded.
Great to hear! good luck with the project and please let us know how you get on.
You may find the basic system classification that I’ve performed of the EOH vaillant systems useful, available here Analysis of Electrification of Heat trial data — OpenEnergyMonitor 0.0.1 documentation Download spreadsheet: EOH_WC_analysis.ods.
Thanks! I shall have a look at it.
If the following should be a new topic, please just tell me:
I am looking through the EOH data and am confused by the meaning of some sensor positions:
The figure for the ASHP sensor placement suggests that the heat meter was installed on the refrigerant lines, while for the GSHP the heat meter is installed on the water line.
Do I interpret it correctly, that the Heat Pump Return Flow is supposed to be measuring the refrigerant line that is coming from the condenser outlet while the Flow Temperatures are measuring the water lines inside the house?
If that is the case, is the ASHP figure missing a heat meter on the water side?
In turn, assuming my interpretation of the Heat Pump Return Flow measurement is correct, how is the temperature measured with the heat meter for GSHP?
GSHP:
