Distributed model-based predictive control for peak load management using Arduino-based LoRa communication

Model-based predictive control (MPC) in buildings is an advanced method to reduce energy consumption, lower operational costs, and manage loads more effectively. However, the implementation of MPC in actual buildings necessitates real-time indoor data collection, communication with the heating, ventilating, and air-conditioning (HVAC) systems, and automatic control. In previous research, the communication methods between optimal control results and HVAC systems were addressed. In this study, a multi-zone distributed MPC (MZ-DMPC) is proposed to handle real-time indoor data collection and automatic control in multi-zone buildings based on an Arduino-based LoRa communication system and a distributed optimization algorithm.

First, building and HVAC system modeling is conducted. The target building is a four-story campus building, modeled to include a multi-zone experimental chamber (two zones) on the first floor and a single-zone research laboratory on the fourth floor. Grey-box modeling was performed using historical data, which was acquired through Arduino-based LoRa communication. This communication system is intended to be used for real-time data collection in the actual implementation. Subsequently, distributed MPC simulations are conducted, linking models to achieve energy savings, operating cost reduction, and peak load reduction.

Peak loads are reduced through the interconnection of loads occurring in each zone, and operating costs are lowered by applying a time-of-use (TOU) plan. Additionally, energy savings are achieved through the performance of the HVAC systems and the building’s thermal capacity. This is accomplished through operational schedule control, such as pre-cooling, and setpoint temperature adjustments.

The LoRa communication-based MZ-DMPC proposed in this study is a low-cost and easy-to-apply method for MPC implementation in smart cities. It is expected to significantly reduce peak loads in single buildings and building complexes, alleviate demand load issues, and contribute to smart grid research. Additionally, reducing energy consumption and operating costs is anticipated to significantly contribute to carbon neutrality and lower the operational cost burden of HVAC systems for consumers.