How can we help?


Parser for german DWD (Deutscher Wetterdienst)

Comments

29 comments

  • Avatar
    Nicholas

    Great work, code looks clean. I'm adding this to test queue and maybe we can include in the next builds with your permission to modify if needed.

  • Avatar
    Sebastian

    Hi Nicholas, 

    since your source repo is GPLv3, the repo fork is of course the same license.

    This means that you are of course allowed to modify if needed and include it into Rainmachine FW deliveries.

    To highlight this, I added the GPL header to the file in my repo.

  • Avatar
    Tomas

    Many thanks! Vielen Dank!

  • Avatar
    Tomas

    Hi Sebastian,

    please check my post regarding a bus with the DWD parser

    https://support.rainmachine.com/hc/en-us/community/posts/360012576333/comments/360001312674

    after adding the DWD parser my RM calculates a very high EvapoTranspiration and wants to extend the watering massively

  • Avatar
    Sebastian

    Hi Tomas, 

    thanks for the hint. I found the problem an I will add an updated version later.

    Greetings

     

    Sebastian

  • Avatar
    Sebastian

    And fix is here:

    https://github.com/SgtSeppel/rainmachine-developer-resources/blob/master/sdk-parsers/RMParserFramework/parsers/dwd-parser.py

    Solar radiation was not converted from MJ/m² to kJ/m². My bad ...

    With this, it looks a lot better now.

  • Avatar
    Tomas

    cant upload the new file.

    i replaced the code of my existing file and created a new one.

    but RM says : error uploading dwd-parser.py

  • Avatar
    Sebastian

    Its too late to be doing such things...

    Can you please move the line

    # coding=utf-8

    to the first line of the file, before the disclaimer?

    Otherwise I'll reupload tomorrow.

  • Avatar
    Tomas

    Thanks. Will try tomorrow. Just shut down the pc

  • Avatar
    Tomas

    OK, uploaded successfully. BUT: the Evapo Transpiration goes up again from 3.8 to avg 25.

    Is this realistic?

  • Avatar
    Nicholas

    No. I looked at the source and I have 2 comments:

    1. I would disable SOLAR RADIATION as it might come wrongly from the station/service. We had a similar issue with SOLAR RADIATION from Weather Underground, coming with either wrong values or different unit of measurement than the one documented.

    2. The min/max temperatures seems to be for yesterday (in CSV) but the parser inserts them for today. This has a small impact (but nothing of that 25 ET0 magnitude). I would add them with something like:

    yesterdayTimestamp = rmGetStartOfDay(timestamp - 12 * 60 * 60)
  • Avatar
    Sebastian

    Hi Nicholas,

    Thanks for the hints.
    Im currently installing my irrigation.
    As soon as I have closed all the trenches and installed the RM at the final install location, I will have more time to toy around with the RM and take another look at parser.

  • Avatar
    Nicholas

    I pushed some small changes to DWD parser into our git  and I also disabled SOLAR RADIATION. Still needs more testing before inclusion.

  • Avatar
    Tomas

    hi nicholas,

    I tested ur version and I still have evapotranspiration of 170-300

  • Avatar
    Nicholas

    Thanks ! Can you give me your station id ?

  • Avatar
    Tomas

    10736

  • Avatar
    Nicholas

    I updated the parser again on git. I missed a second Solar Radiation being fetched. Running on my machine seems fine. 

    To update parser do (in Web UI):

    1. DELETE parser from User Uploaded tab

    2. Press REFRESH ALL 

    3. Upload the parser again, configure it and SAVE

    4. Press REFRESH ALL

  • Avatar
    Tomas

    I have copied the code from here:

    https://github.com/sprinkler/rainmachine-developer-resources/blob/a6eacc43f12eeb46d60c49e2d700b9badb82cb23/sdk-parsers/RMParserFramework/parsers/dwd-parser.py 

     

    and put it into my dwd-parser.py file.

    but I still have 160-300 evapotrans

    did I take the wrong file?

    could you show me the link of the latest file?

  • Avatar
    Tomas

    hey all,

    After my last dwd test I cant bring the RM-Mixer to recalculate the values.

    The ET ist still 160-300 and RM wants to extend the watering time up to 200%.

    Can I reset the RM-Mixer?I turned off my 3 weather services or used only the dwd. nothing made the mixer recalculate ET

  • Avatar
    Tomas

    I have also disabled all weather correction and rebooted the RM twice... no change

  • Avatar
    Tomas

    ok, it works again

  • Avatar
    Tomas

    Its me again. I see some strange behavoir of my RM. After 2 days of significant precipitation my watering forecast looks like this:

    loan is watered every day and the bushes not at all...

  • Avatar
    Nicholas

    What's the field capacity of Zone 1 ? Can you compare the graphs ("show graphs" button) next to field capacity for Zone  1 and Zone 2 ?

  • Avatar
    Tomas

    here we go

  • Avatar
    Nicholas

    We should move the discussion outside this thread but it's pretty strange why zone 1 didn't register that large amount of rain. Guess the program wasn't created after the rain event ? Maybe some zone soil setting won't let the water accumulate that much (I see much lower field capacity on Zone 1). Let me know if it doesn't auto-correct itself at next run.

  • Avatar
    Sebastian

    Hello together, 

    the changes discussed above, double entry of solar radiation, wrong conversion of some units, are included in the current commit on my git.

    If I just checked again. Without the DWD parser the ET from the mixer is 6.60mm for June 31st. 

    Thats the same value that I get with the parser?

  • Avatar
    Tomas

    I cant check the dwd parser. Made a factory rest becaue of the strange zone2 calculation. now my  field capacity ist still 21 and 91 but the watering time is correct. zone 2 is beeing watered again every day. I will try the dwd parser when its running without any errors. otherwise I risk again to reset everything

  • Avatar
    Sebastian

    I just added the current version of my parser to github and created a pull request. 

    I have been running that now for several weeks without issues. ET is now totally within expected ranges.

Please sign in to leave a comment.