WORK IN PROGRESS. Private test project

Code (and READMEs) from AI, will be checked and maybe adjusted.

TODO: Check AI code for correctness and security.

---

GoSmartMeterGo

GoSmartMeterGo is a tool written in Go to collect, archive, and visualize electricity consumption and feed-in data on a Raspberry Pi.

---

How it works

The tool reads and combines two data sources:

1. SmartMeter (SML): Reads the meter index (import and export) of the electricity meter via an optical reading head using the serial SML protocol (Smart Message Language).
2. SMA HomeManager (Speedwire): Listens on the local network for SMA HomeManager multicast packets (UDP Speedwire protocol).

---

System Architecture

The project is split into a local collection daemon on the Raspberry Pi and an optional deployment on a Strato webspace:

graph TD
    subgraph pi["Raspberry Pi (local)"]
        A[Infrared Reader] -->|SML via Serial| B(Go Daemon)
        C[Network / PV System] -->|SMA Speedwire Multicast| B
        B -->|Live Stream SSE| D[Local Dashboard :8080]
        B -->|5-Minute Intervals| E[(Local SQLite DB)]
        B -->|Periodic Rsync / SSH| F[Strato Webspace]
    end

    subgraph cloud["Strato Cloud"]
        F -->|Backups| G[(sqlite.db Copies)]
        H[Browser / User] -->|HTTPS Basic Auth| I[Online Dashboard]
        I -->|api.php| G
    end

Loading

---

Features

• Real-time Monitoring & Local Dashboard: A local web server that displays live data in your browser using Server-Sent Events (SSE).
• History & Reports: Metrics are aggregated every 5 minutes and stored in a SQLite database to generate daily, weekly, monthly, and yearly statistics.
• Telegram Notifications:
  • Watchdog: Sends alerts via a Telegram bot if the SMA HomeManager stops broadcasting data.
  • Daily Report: Daily summary of the previous day's consumption and feed-in.
• Strato Backup & Online Viewer: The SQLite database is uploaded to a Strato webspace at configurable intervals (or manually) using rsync (via SSH). A PHP script on Strato reads the database, allowing password-protected dashboard access on the go.

---

Technical Details

• Programming Languages: Go (Golang) for the backend, HTML/CSS/JS for the dashboard.
• Database: SQLite for local storage.
• Deployment: Can be built as a Debian package (.deb) using a Makefile and installed as a systemd service on the Raspberry Pi.