SENEC Monitor

This is a monitor accessing the unofficial senec api. It retrieves the data for your photovoltaic system
and saves it to a postgres database.

Info

Previous tracking of this project happend over at prev repo

Utility

• Websocket connection can be used by embedded devices to get real time data Check out my implementation with ESP32
• Creates a record of yield data that can be used for analysis and prediction ( also saves weather data )
• Combines current photovoltaic system data with weather data to allow for accurate prediction / forecast
• Could also be developed into a more sophisticated web interface in combination with weather data, graphs etc

Under the hood

• Retrieves yield data for every full day ( unofficial senec api )
• Gets current data every 10s ( your local api )
• Obtains weather data for you location on a hourly basis
• Retrieves Daily weather forecast every 24 hours and saves it to be combined with total yield values for later prediction purposes
• Provides an Api ( for specific data ) and a websocket connection for frequent updates ( weather, current power levels, total yield )

Endpoints

All the data from the Senec-Api

/full 

Data for a specific Timestamp

/data?ts=1234

Latest data from the local-Api updated every 10 Seconds

/localLatest 

Websocket Connection that gets every update of the local-Api & weather in real time

/subscribe

Instructions

Copy compose.yaml to your directory and run:

docker compose pull

Also copy the /docker-db/initFiles/init.sql to the directory containing the compose file. Please preserve the directory structure so that the init.sql file lies inside the /docker-db/initFiles/ directory to make sure the db
is intialized properly. This is important as it initializes the database with the required tables

Createa a ```.docker.env``` with the following values
```shell
# your my-senec login password
SENEC_PASS=

# your my-senec login username
SENEC_USER= 

# db host default should be  "docker.host.internal"
DB_HOST=docker.host.internal

# db port default should be 6000
DB_PORT=6000

# db password default is yourpass
DB_PASS=yourpassj

# db user default is myuser
DB_USER=myuser

# just search the coordinates of your location and put them here
LONGITUDE=
LATITUDE=

# the ip of your local senec dashboard
SENEC_IP=

Then run

docker compose up

Please note that currently this docker compose file utilizes watchtower to check for container updates.
If you do not plan on continuously working on this project yourself you can just remove everything associated with watchtower from the compose file.

Now you should be able to visit the service on:

localhost:4000/

Instructions (dev)

Create database and table with docker (to have a separate database)

• DB Setup with docker compose

Clone the repo

Add a .env file to your directory:

SENEC_PASS=
SENEC_USER=
DB_HOST=
DB_PORT=
DB_PASS=
DB_USER=
DB_NAME= 
LONGITUDE=
LATITUDE=
SENEC_IP=
# this is optional  
PREDICTION_URL=

For the prediction server I have developed an inference server for my own AI Model (decent but definitely improvable) that predicts the yield for the next 24 hours. Check it out here

Generate proto files}

protoc --go_out=. --go_opt=paths=source_relative ./proto/types.proto

Run via:

go run .

Or build to exe:

go build .

Visit on:

localhost:4000/

Building for docker arm64 requires this command if you want to run it as a container

docker buildx build --platform linux/arm64 -f Dockerfile -t your_username/test-repo:senec-monitor-latest --push  .

Todos

• add ci/cd pipeline for automatic container builds