pierre-haessig / solarhome-control-bench

open testbench for control and optimization methods for the energy management of a simple solar home

Solar home control bench

This repository contains an open testbench for control and optimization methods for the energy management of a simple solar home.

Pierre Haessig, IETR (AUTomatic Control team), CentraleSupélec

Solar home model

Control methods

This repository contains several examples for the energy management of the solar home, with a different method or in a different language (Python, Julia, Matlab).

Each method lives in dedicated subdirectory of the methods folder. It includes:

• Rule-based control (Julia, Matlab and Python)
• Model Predictive Control (MPC)
• Stochastic Dynamic Programming
• …

Comparison of control methods

In the comparison folder.

Solar and load data

Solar production (from PV panels) and home consumption data is taken from the Solar home electricity dataset by Ausgrid (distribution grid operator in the region near Sydney).

A dataset extract used for this testbench is placed in the data subfolder. A description of this data extract is provided in data/README.md. In particular, the 30 days starting at 2011-11-29 should be used for final testing:

In addition the dedicated ausgrid-solar-data repository contains much Python code to analyze the entire Ausgrid dataset. However, it should not be needed for this benchmark.