Field experiment testing of a low-cost model predictive controller (MPC) for building heating systems and analysis of phase change material (PCM) integration

• Published in: Applied energy Vol. 360; p. 122750
• Main Authors: Wei, Zhichen, Calautit, John Kaiser
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.04.2024
• Subjects: Building energy storage system; case studies; cost effectiveness; Demand response; electrical equipment; electricity; electricity costs; energy efficiency; Experimental model predictive control (MPC); field experimentation; greenhouse gases; Internet; Internet of things (IoT); Phase change material (PCM); phase transition; raspberries; Raspberry Pi; temperature; wireless technology; Phase change material (PCM); Building energy storage system; Demand response; Raspberry Pi; Experimental model predictive control (MPC); Internet of things (IoT)
• DOI: 10.1016/j.apenergy.2024.122750
• ISSN: 0306-2619

Model Predictive Control (MPC) emerges as a promising solution to address the substantial greenhouse gas emissions from the building sector. By employing advanced control strategies, such as MPC, for peak energy shifting, there is a significant potential to enhance energy efficiency and reduce emissions through effective demand response within a smart grid. Although MPC has been the subject of extensive research, the practical implementation of cost-effective and easily deployable solutions in buildings remains limited. This study proposes a cost-effective MPC approach, employing the Internet of Things (IoT) and dynamic pricing. The control strategy, developed in MATLAB, is locally deployed on Raspberry Pi hardware via WiFi. The proposed MPC was tested in a controlled environment at the University of Nottingham, UK, where it regulated a radiator heating device to maintain indoor comfort in response to dynamic hourly electricity prices, using real-time indoor temperature feedback. The results confirmed the proposed MPC's accuracy in predicting indoor temperature responses and controlling indoor temperature within setpoints over a typical winter week. Performance analysis further revealed that the proposed MPC strategy resulted in a 20% electricity cost reduction compared to a conventional control strategy. Additionally, alongside the proposed MPC strategy, a Phase Change Material (PCM) wallboard system was integrated into a co-simulation platform. The developed PCM wallboard model underwent a verification and validation process, utilizing both numerical simulations and experimental data. The results demonstrate that the proposed MPC-controlled PCM wallboard system saved 35% on electricity cost compared with the original case study room. This study provides valuable insights into the development of intelligent localized demand response control for the built environment, offering a range of choices for IoT equipment. [Display omitted] •Proposed low-cost, IoT-based model predictive control (MPC) strategy for buildings.•MPC strategy implemented in real-world settings using Raspberry Pi hardware.•Integrated PCM wallboard system, achieving 35% electricity cost savings.•Co-simulation platform for interactive modelling of MPC and PCM-integrated buildings.•Peak energy shifting using MPC in PCM-integrated buildings.