One for Many: Transfer Learning for Building HVAC Control

Abstract

The design of building heating, ventilation, and air conditioning (HVAC) system is critically important, as it accounts for around half of building energy consumption and directly affects occupant comfort, productivity, and health. Traditional HVAC control methods are typically based on creating explicit physical models for building thermal dynamics, which often require significant effort to develop and are difficult to achieve sufficient accuracy and efficiency for runtime building control and scalability for field implementations. Recently, deep reinforcement learning (DRL) has emerged as a promising data-driven method that provides good control performance without analyzing physical models at runtime. However, a major challenge to DRL (and many other data-driven learning methods) is the long training time it takes to reach the desired performance. In this work, we present a novel transfer learning based approach to overcome this challenge. Our approach can effectively transfer a DRL-based HVAC controller trained for the source building to a controller for the target building with minimal effort and improved performance, by decomposing the design of neural network controller into a transferable front-end network that captures building-agnostic behavior and a back-end network that can be efficiently trained for each specific building. We conducted experiments on a variety of transfer scenarios between buildings with different sizes, numbers of thermal zones, materials and layouts, air conditioner types, and ambient weather conditions. The experimental results demonstrated the effectiveness of our approach in significantly reducing the training time, energy cost, and temperature violations. CCS CONCEPTS• Computing methodologies → Reinforcement learning; • Computer systems organization → Embedded and cyberphysical systems.

Reinforcement learning building control approach harnessing imitation learning

“…A short-term memory buffer that empties after collecting the most recent batches in lower numbers had the best performance when compared to long-term memory, a memory that never empties. Xu et al [26] used a novel TL method that is scalable to multi-zone buildings with different layouts and building materials by taking advantage of two neural networks -a front-end network generalized to all buildings and a building-specific back-end network.…”

Variational Nested Dropout

“…In the meantime, they are experiencing rapid growth in model size and computation cost, which makes it difficult to deploy on diverse hardware platforms. Recent works study how to develop a network with flexible size during test time [20,50,49,4,6,47], to reduce the cost in designing [44], training [21], compressing [13] and deploying [36] a DNN on various platforms. As these net-Department of Computer Science, City University of Hong Kong.…”