-
Notifications
You must be signed in to change notification settings - Fork 21
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Integrate with an ISO to receive grid signals #79
Comments
I will do this after #78 |
Now that we have cleared the decks with #88, getting started on this. |
So we can see why `SetChargingProfile` is not implemented EVerest/everest-demo#79 (comment)
This is super hard to debug because I can't build the images locally. However, launching an existing container (with modules downloaded) and building in it works. Changes to enable this hacky build: I can then run Using this to enable logging, I see this comparison.
an example of a successful run initiated by a script is
an example of a successful run initiated by the CSMS is
It seems clear that we are trying to forward the message to the charge station. It also seems I am not sure if it is failing in the send or the response, but the |
Ok so the the CallRoutes map does not list the `SetChargeProfile` entry.
But I checked the code, and it is not defined https://github.com/thoughtworks/maeve-csms/pull/44/files#diff-2e87aed153ea3c66a759c947e51d51ec1b0275802fc12e8f987c0d570d2c03c7 Let's quickly check the @the-bay-kay are you sure this worked for you? Do you still have the image/container that you tested against? It would be good to check the code and copy it over as additional patches... |
I tried to run using the instructions here #74 (comment) |
My personal laptop is super slow (maybe even slower than the work laptop that @Abby-Wheelis has. So it is still building the maeve images. But I checked the call map and can verify that there is no additional code. Writing it now... |
Ok, so this appears to be an issue with EVerest (phew!, I am not crazy). The error that we were getting was while handling the response from the We should still define it, but its absence should not affect the message sent to the station. If we clear out the gateway logs and then send the message again, we see that it does in fact send a websock message and receive a response, which it tries to send to us. Let's add some additional logs in the gateway to validate this, and then see what is happening on the station...
|
Ok, so it looks like it was actually working?! We just didn't notice it because apparently that is not being logged on the charger?! Let's take a look at the logging on the charger, and also see if we can connect to its MQTT bus so we can see that the value has been received.
|
Which is currently a NOP, but we want to register it in the map so that it doesn't error out. This fixes EVerest/everest-demo#79 (comment)
The additional logging to show when we received the profile is in Patch has been applied properly; let's bump it up to ERROR to see if it shows up |
Verified that after bumping it up to ERROR, we do see it. But we see other
|
When rebuilding after configuring libocpp, it looks like it was still stuck at ERROR. Need to investigate this, but moving on finish the other changes first. |
I just recreated all the containers, so checking this again...
Just need to install after compiling! |
Basic MIDAS integration done - the scripts at https://github.com/morganmshep/MIDAS-Python-Repository were crude but very helpful and the system as a whole was functioning well. It was also fairly easy to port to javascript/nodered. HOWEVER, the sample rate schedule does not have any curtailment, only tariffs
I tried a different rate schedule (retrieved via
Will see if we can get info from CARB, otherwise, in the interim, will just try to munge the TOU values into pmax. |
Couple of notes: the charging schedules seem to show up at `everest/ocpp/ocpp_generic/var`, but it seems like the full list and not the composite
If we change the max_current slider in the UI, it publishes to `everest/iso15118_charger/charger/cmd`
Hm... but from the UI, it looks like we publish to searching for "sched", I also find potentially relevant entries in `everest/evse_manager_1/evse/var` and `everest/evse_manager_1/evse/cmd`
`everest/grid_connection_point/energy_grid/var` and `everest/grid_connection_point/energy_grid/cmd` and `grid_connection_point` with similar paths
and of course, ocpp publishes the basic schedules at `everest/ocpp/ocpp_generic/var`
Need to look through the codebase to see where these are published, what they mean and whether any of them represent the composite schedules. |
Looks like every module has an mqtt interface to pub/sub
or
or
|
Starting from scratch: Looking at the published API values, I only see status I am not seeing connection schedules from I also note (as I have before), that the code to publish the composite schedules goes into
But I can't see it in the code. The I manually set the schedule through the CSMS, and I still don't see anything in the API topic. `everest/ocpp/ocpp_generic` does show some schedules but they are wrong.
Let's try to edit to make it composite. After trying to set the current to 10,
we end up getting the following, which is clearly wrong.
Let's print out some composite schedules and figure out what is going on... |
Composite schedule calculations appears to be completely messed up, unless I am misunderstanding how it is supposed to work. It works for one single clamp but appears to fail completely once we have a second, overlapping clamp. I don't have time to investigate this now (I need to finish the grid integration and then take a look at the power dial and maybe even SP2/SP3. So filing a new, follow up issue to track that and coordinate with the community (#90) Turns out this was mainly a false alarm (#90 (comment)) so let's wrap this up and then see what's going on with the power meter. |
- Login to MIDAS with username and password - Get token - Use token to retrive all rates - Display the retrieved rates in a table Related: EVerest#79 (comment) Testing done: - Ran flow and confirmed that the table showed data Signed-off-by: Shankari <[email protected]>
Since CharIN is in California, we will want to integrate with the California MIDAS API
https://www.energy.ca.gov/publications/2021/market-informed-demand-automation-server-midas-documentation-version-12
Design considerations:
Given the time frame, I would suggest that we hack together something with username + password, and curl to pull data and send it to the CSMS.
Can refine further once that is done
The text was updated successfully, but these errors were encountered: