An IOT based smart HVAC control using heating load predictions

Global demand for Heating, Ventilation and Air Conditioning (HVAC) systems is expected to increase by 5.7% annually. HVAC systems will be more popular in commercial buildings. But according to research, HVAC accounts for nearly 40% of electricity used in commercial buildings. Traditional control strategies no longer meet the environmental requirements. Therefore, we need a new method with great efficiency which can significantly save energy, reduce unnecessary costs, and increase return on investment. Model Predictive Control (MPC) is a model-based advanced control method. It can integrate several factors, such as temperature, occupancy and humidity to calculate the most proper model parameters for the next control step. This dissertation concerns the research and application of model predictive control theory, and applying the theory in the HVAC system based on Raspberry Pi3. The main innovations and contributions in this project are listed as follows: (1) Sensor installation and programming: The sensor plays a very large role in this system, allowing him to monitor the environmental changes in each room in real time. Sensitivity and stability are very important. (2) The application and expansion of the Raspberry Pi board: Raspberry Pi is the core hardware of the whole system. Firstly, it is used as the control center. It needs to separately control the sensor module to detect the relevant data of the environment, and then carry out practical processing analysis on the collected data. (3) MPC algorithm programming in matlab and python: In matlab, we use existing data to simulate and test whether the algorithm works. After the simulation, we need to run the algorithm in linux environment with python language.Master of Science (Computer Control and Automation