search

jorgebooks / measureit

Automatically exported from code.google.com/p/measureit
0 stars 0 forks source link

V115. PVOutput not showing smooth curves when intermixing generated and consumed power. #36

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago 
What steps will reproduce the problem?
1. Sending consumed and generated power on different sensors to PVOUTPUT
2.
3.

What is the expected output? What do you see instead?
A jaggy graph. If you join up all the peaks you would be an approximation of 
the correct graph.

What version of the product are you using? On what operating system?
V115

Please provide any additional information below.
Copy and pasted from the user measureit google user group.

When Measureit V115 sends data to PVOUTPUT it uses an individual sensor based 
approach.
This means that whatever sensor "hits the jackpot", and is the first one to 
send some data after the 5 minute summing interval (This is a PVOUTPUT 
requirement), will have its power data shown with any other sensors summed 
power data set incorrectly to 0.
This causes the graph to be jaggy. I've attached a picture. If you join up the 
peaks you will get the correct looking graph :-)

Note the intermixing.

At 10:15AM the consumed power sensor "hit the jackpot" and set Power to 0 and 
Power Used to 1690W. The red line goes up at the end the green one falsely goes 
to 0.

At 10:10AM, 10:05AM  and 10.00AM the generated power sensor "hit the jackpot" 
three times and set the Power to 2171W, 2049W & 2049W respectively and the 
Power used to 0.

Red lines goes to 0 and green line forms a little hump.

If you are sending just consumed power on one sensor then this bug will not 
show up!

Oh dear at 8:20AM my 3 phase air conditioner turned on and I'm paying Energy 
Australia a whole pile of money to heat my house :-( Big fat red line!

To correct this the python grabber needs to be slightly refactored to base the 
data sent to PVOUTPUT on the PVOutput System Id of each sensor and not on the 
individual sensors summed value.
There are a few ways to actually do this and I would be happy to exchange an 
email or two with Thomas if he wants.
Old English Proverb (probably German as well) "Don't teach your grandmother to 
suck eggs." 
Which means that this is Thomas's system and I don't want to presume to advise 
him how to code the solution. 

Just happy to help if needed. I am keen to get a low cost monitoring system up 
and running.

Original issue reported on code.google.com by sireland...@gmail.com on 15 Aug 2013 at 3:04

Attachments:

GoogleCodeExporter commented 9 years ago 
Another contributor (another Steve) on the measureit google group has commented 
that this may be fixed by supplying a -1 in the value that is not being 
reported so that PVOUTPUT ignores it. The issue with this solution is that 
depending who "wins the jackpot" then consumed or generated data will not be 
updated every 5 minutes.

Regards

Original comment by sireland...@gmail.com on 16 Aug 2013 at 1:50

GoogleCodeExporter commented 9 years ago 
Ok. I need some help to understood what happened there.

You have your PVOutput API key in the system settings?

At the sensor settings you set the "PVOutput System ID" on each sensor / clamp?

Do you use one "PVOutput System ID" on multiple sensors?

Original comment by lalelu...@gmail.com on 19 Aug 2013 at 8:56

GoogleCodeExporter commented 9 years ago 
I have the PVOutput key in the system settings as V115 has a bug if you put the 
key in the individual sensor. 

I have 3 transmitters

Transmitter 0 is consumed power on 3 clamps (3 phase) with PV Output System ID 
22113
Transmitter 1 is generated power on 3 clamps (3 phase) with PV Output System ID 
22113
Transmitter 2 is measuring off peak hot water  (1 Phase) with PV Output System 
ID 22183 <- THIS NUMBER IS DIFFERENT

I have also defined all of the clamps but have not enabled them for PV Output.

If you look at the original image I attached (pvopV115.png) you can see on the 
data grid that at 10:15AM the grabber sent some data that set the Power Used 
(last column) to 1,690W, but the Power (5th column) was set to 0. This column 
should be set to the summed generated power for the period 10:10 to 10:15.

Hope this helps.

Original comment by sireland...@gmail.com on 22 Aug 2013 at 5:30

Attachments:

GoogleCodeExporter commented 9 years ago 
[deleted comment]
GoogleCodeExporter commented 9 years ago 
Ok. I did not think about that it should be possible to send usage and 
generation data to a single PVOtput ID.

Currently you can only send either one of them. The other value is always 
setting to 0.

I had to change this.

The bug with the api key is solved. Just use the newest grabber:
https://measureit.googlecode.com/svn/trunk/measureit_system_files/python/data-in
put.py

Original comment by lalelu...@gmail.com on 24 Aug 2013 at 9:20

GoogleCodeExporter commented 9 years ago 
I solved the problem but I can not commit the current grabber because of there 
are to much changes in my version with the online version and some missing 
tables in version 115 that comes with the next version.

To solve it edit the data-input.py and change the following code in the 
function sensor_data_pvoutput_status_generate. Must be in +/- line 385

from:

type = 'v2=0&v4' if sensor_settings[sensor]['type'] == 0 else 'v4=0&v2'

to:

type = 'v4' if sensor_settings[sensor]['type'] == 0 else 'v2'

You had to restart the grabber after changing.

Original comment by lalelu...@gmail.com on 24 Aug 2013 at 11:13

GoogleCodeExporter commented 9 years ago 
[deleted comment]
GoogleCodeExporter commented 9 years ago 
[deleted comment]
GoogleCodeExporter commented 9 years ago 
It would be nice if we could map other sensors to PVOutput V7 - V12 too. 
Perhaps something for a future version.

Original comment by wot...@gmail.com on 24 Aug 2013 at 11:33

GoogleCodeExporter commented 9 years ago 
Added the code as suggested into version 115 and the solution works well.

Thanks

Steve I

Original comment by sireland...@gmail.com on 27 Aug 2013 at 11:48

Attachments:

GoogleCodeExporter commented 9 years ago 
Dears,
I have started to have the same problem (or it seems so) as above.
It worked for more than one month on my Raspberry without any problem and 
suddenly ...
I have already added, since the beginning, the changes on data-input.py.
Any suggestion?

Original comment by sabatini...@gmail.com on 7 Feb 2014 at 8:05

Attachments:

GoogleCodeExporter commented 9 years ago 
Solved in version 116 that is available to download

Original comment by lalelu...@gmail.com on 19 Apr 2014 at 9:08