Flow Rate Control in Smart District Heating Systems Using Deep Reinforcement Learning

TL;DR

This paper presents a deep reinforcement learning approach for flow rate control in district heating systems, demonstrating significant energy and water savings through simulations and a practical cloud-based implementation.

Contribution

It introduces a novel deep reinforcement learning method for optimizing flow control in district heating, with a practical cloud-based system and real-world validation.

Findings

Approximately 1985 gigajoules of heat saved per hour

Around 42,276 tons of water conserved per hour

Effective control demonstrated in real-world case study

Abstract

At high latitudes, many cities adopt a centralized heating system to improve the energy generation efficiency and to reduce pollution. In multi-tier systems, so-called district heating, there are a few efficient approaches for the flow rate control during the heating process. In this paper, we describe the theoretical methods to solve this problem by deep reinforcement learning and propose a cloud-based heating control system for implementation. A real-world case study shows the effectiveness and practicability of the proposed system controlled by humans, and the simulated experiments for deep reinforcement learning show about 1985.01 gigajoules of heat quantity and 42276.45 tons of water are saved per hour compared with manual control.