Sanjit Vijay
Pronouns: He/Him
Research Mentor(s): Lars Junghans
Research Mentor School/College/Department: Architecture / AUP
Program:
Authors: Lars Junghans
Session: Session 6: 3:40 pm – 4:30 pm
Poster: 100
Abstract
22/26 Midwest aims to create a novel heating/cooling system for houses by using a window opening mechanism to maintain the temperature of the house between 22 and 26 degrees celsius. This project was conducted once in Europe, and we are trying to adapt the project to the midwest climate. Additionally, while the previous project was targeted towards office buildings, we aim to create a plug-and-play solution that can be used in residential houses. Because this research was previously performed, we are using the findings from the previous research to guide our research. Additionally, we have looked into many other research sources that discuss residential environment monitoring and control systems in houses. The main focal point of the project is the window actuator, which can open and close depending on the data collected, which determines if the environment is too warm or too cold. Additionally, we are using a predictive AI model that can predict the room characteristics for the next three days. The target audience for this project are residential home owners. To collect data, we were required to configure sensors that collect temperature (degrees celsius), CO2 concentration (parts per million), and humidity (percentage) data to feed into our AI model to determine when to open and close the windows. We collected the temperature and CO2 data with a EE895 sensor and we collected the humidity with a DHT11 Temperature and Humidity sensor. We utilized a raspberry pi to collect this data, and used a python script running on the raspberry pi to collect this data. Because we are running multiple raspberry pis, we are using MQTT to communicate between raspberry pis. Additionally, we are running fans in order to ventilate the house, which the code was written for in C. For the AI aspect, we utilized the sklearn library in python.
