This repository contains the raw data used in "A Multi-Agent Reinforcement Learning Approach to Price and Comfort Optimization in HVAC-Systems" and the control code for article "Data-Driven Offline Reinforcement Learning for HVAC-Systems". It is strongly recomented to read these articles before deploing any of this code. Additional, will reading this work also help to understand the raw code.
The material used for "A Multi-Agent Reinforcement Learning Approach to Price and Comfort Optimization in HVAC-Systems" is located in "Control_codes_and_env" and "Data" Each folder represents the data gathered from one zoned UFH system and a four zoned UFH system. Each folder contains multiple CSV files, and a plotting file - dependencies are Matplotlib and Pandas. Be aware when plotting different variables, this is not the most user-friendly script – quite a lot of spaghetti code. We apologies in advance
The Control code used for "Data-Driven Offline Reinforcement Learning for HVAC-Systems" is located in "Offline_RL". In this folder is what is refered to as secnario A and scenario B.