Export HA Energy dashboard data

Is there a way to export the hourly data out of de HA energy dashboard towards CSV for example.

I would like to export that data to run calculations with regard to dynamic pricing.

6 Likes

That would be very nice :slight_smile:

2 Likes

It’s been a couple years since this post. Any chance this has actually happened?

1 Like

Bump.

Has a way been developed to extract the historical hourly interval data used by the energy dashboard?

Would be really nice i am now sitting and stringing into the SQL database but have not figured it out yet :stuck_out_tongue:

There is a custom card that can do CSV exports for long-term statistics:

2 Likes

Thanks but I don’t understand how that helps to export data from the Energy Dashboard.

The energy dashboard just displays long-term statistics (state_class: total or total_increasing) in nice graphs.

Thanks again but I’m not making the link.

How do I export historical interval data from the energy dashboard?

Let’s say I want to export my gas usage.
In Settings > Dashboards > Energy > Gas Consumption, I can see that the entity used to generate the gas usage graph is sensor.kotlik_gas_flow_integral.

After installing the custom card, add it to a dashboard:

type: custom:history-explorer-card
cardName: historycard-10713360
statistics:
  enabled: true
graphs:
  - type: line
    entities:
      - entity: sensor.kotlik_gas_flow_integral

Use the interactive zoom controls to select the time period that you want to export.
Press “Export statistics as CSV” in the menu (V next to the entity search box).
You should get a CSV file like this:

Time stamp,State,Mean,Min,Max
sensor.kotlik_gas_flow_integral
2022-10-11 12:00:00,1366.448,,,
2022-10-11 13:00:00,1366.448,,,
2022-10-11 14:00:00,1366.448,,,
2022-10-11 15:00:00,1366.448,,,
2022-10-11 16:00:00,1366.828,,,
2022-10-11 17:00:00,1366.828,,,
...
2 Likes

Got it. Thanks!

I was able to get this data from the console as well via the websocket API. For example, I used websocat and python. A quick snippet for those comfortable with cli:

websocat ws://my.ip.ad.dr:8123/api/websocket

Then paste in:

{"type": "auth","access_token": "super long HA long lived access here"}

After that auth’s, you can paste in:

{"type":"recorder/statistics_during_period","start_time":"2024-01-21T08:00:00.000Z","end_time":"2024-01-22T07:59:59.999Z","statistic_ids":["sensor.grid_import_energy","sensor.grid_import_energy_cost","sensor.grid_export_energy","sensor.grid_export_energy_compensation"],"period":"hour","units":{"energy":"kWh"},"types":["change"],"id":1}

Change the sensor names, dates, etc to suit your needs obviously.

For python I used Examples — websocket-client 1.7.0 documentation and just expanded on the first example with those same payloads above. If you go this route, know you need to ws.recv() right after connecting (before sending the auth payload).

As of 2024.2, you can download a csv file in the history dashboard. It does contain long-term statistics, but there seems to be only one of the values (i.e. mean/state, but no min nor max).