A fully automated energy-optimisation suite powered by ChatGPT + Home Assistant
If you’re running a SolaX inverter system, switch between Octopus Energy Flux and Cosy tariffs during Summer and Winter months, and want Home Assistant to manage it intelligently — I’ve published my full working setup on GitHub 
SolaX Automation in Home Assistant
What It Does
This automation suite manages your SolaX hybrid inverter system to:
Minimise Grid Charging- Minimise PV Clipping
- Maximise Export Revenue
- Run a Free Energy Session Controller
- Run an Energy Saving Controller
- Include External Temperature Compensation
- Automatically switch between Octopus Cosy & Flux tariffs
Learning Layer — Continuous Optimisation
At the core of the system is a robust automated learning layer that continually improves accuracy between forecasted and actual performance, learning from:
PV Generation- Daily Demand
Evening Demand (especially useful for ASHPs and the Cosy Tariff)
Minimum Daily Temperature (anticipates increased heating usage)
This learning layer automatically adjusts the system’s bias over time, keeping forecasts and plans realistic while protecting battery charge levels and grid-import costs.
Tested & Proven Setup
Designed for both single and parallel inverter systems — tested live with:
- 2 × SolaX X1 Gen4 Hybrid Inverters
- 2 × 11.2 kWh battery banks
- Fully compatible with any SolaX Modbus-enabled inverter (just map your entities and create the helpers).
All code is stable, production-tested, and running daily in my home.
Installation steps are included.
Linked Chat — How to Build YAML with ChatGPT
I’ve also documented how I used ChatGPT to generate and manage all my YAML automations:
ChatGPT-HA-Docs on GitHub
Please Note: This isn’t a product — just sharing the result of months of development and testing to help others streamline their energy management.
Would love to hear how it works for you — feedback, tweaks, or integrations welcome! 