Towards Optimal District Heating Temperature Control in China with Deep Reinforcement Learning
This work aims to improve the energy efficiency and reduce the carbon footprint of district heating networks in China, benefiting both the environment and consumers through lower energy costs.
This paper addresses the challenge of optimizing district heating networks in China to reduce their carbon footprint. The authors developed a deep reinforcement learning approach, utilizing a recurrent neural network to predict indoor temperatures, which then trains DRL agents to control supply water temperature. Their method achieved higher thermal comfort and lower energy costs compared to an optimized baseline strategy in multi-apartment settings.
Achieving efficiency gains in Chinese district heating networks, thereby reducing their carbon footprint, requires new optimal control methods going beyond current industry tools. Focusing on the secondary network, we propose a data-driven deep reinforcement learning (DRL) approach to address this task. We build a recurrent neural network, trained on simulated data, to predict the indoor temperatures. This model is then used to train two DRL agents, with or without expert guidance, for the optimal control of the supply water temperature. Our tests in a multi-apartment setting show that both agents can ensure a higher thermal comfort and at the same time a smaller energy cost, compared to an optimized baseline strategy.