Internet-of-Things Based Hardware-in-the-Loop Framework for Model-Predictive-Control of Smart Building Ventilation

Area of Science:

• Building energy management
• Smart building control systems
• Predictive control algorithms

Background:

• Smart buildings require advanced control strategies for energy efficiency and occupant comfort.
• Existing control methods may not fully leverage real-time data and predictive capabilities.
• Hardware-In-the-Loop (HIL) simulations offer a robust platform for testing control algorithms in realistic conditions.

Purpose of the Study:

• To introduce and evaluate a Hardware-In-the-Loop (HIL) framework for implementing and assessing predictive control in smart buildings.
• To integrate Internet of Things (IoT), big data, machine learning, and Model Predictive Control (MPC) for HIL simulations.
• To optimize ventilation systems for maintaining indoor air quality and enhancing energy efficiency.

Main Methods:

• Development of an HIL framework combining IoT, big data, machine learning (LSTM), and MATLAB-based MPC.
• Deployment of the framework in a real-case scenario at the EEBLab test site.
• Implementation of the MPC controller on Raspberry Pi (RPi) hardware, utilizing forecasted occupant numbers to determine optimal ventilation flow rates.

Main Results:

• The proposed HIL framework successfully enabled the implementation and assessment of the MPC algorithm for a standalone ventilation system.
• Energy consumption was reduced by approximately 16% compared to an ON/OFF control strategy.
• Indoor air quality, specifically Carbon Dioxide (CO2) concentration, was maintained within the standard comfort range.

Conclusions:

• The developed HIL framework is effective for testing and validating predictive control strategies in smart buildings.
• The MPC approach, enhanced by real-time data and forecasting, significantly improves energy efficiency while maintaining indoor comfort.
• This integrated approach demonstrates a promising pathway for optimizing smart building operations.