Estimating Air Change Rate in Mechanically Ventilated Classrooms Using a Single CO₂ Sensor and Automated Data...

Area of Science:

• Environmental Science
• Building Science
• Machine Learning

Background:

• Indoor air quality (IAQ) is crucial in schools, driving the need for effective monitoring.
• Carbon dioxide (CO2) monitoring is increasingly used to assess ventilation and IAQ in classrooms.
• CO2 data can estimate outdoor air change rate (ACH), impacting health, performance, and energy use.

Purpose of the Study:

• To develop and apply a novel machine learning method for segmenting CO2 time series data.
• To estimate classroom ACH using CO2 mass balance principles.
• To compare ACH estimates from the novel method with traditional ventilation rate data.

Main Methods:

• Applied a machine learning algorithm to segment CO2 time series into build-up, equilibrium, and decay periods.
• Utilized CO2 mass balance equations to calculate ACH for each segmented period.
• Collected data from 40 classrooms across two K-6 schools with mechanical ventilation.

Main Results:

• The study generated multiple daily ACH estimates per classroom.
• The equilibrium method provided ACH estimates closest to those from the building automation system.
• Decay and build-up methods showed a slight underestimation of ACH compared to mechanical ventilation rates.

Conclusions:

• Machine learning segmentation of CO2 data offers a promising approach for IAQ assessment.
• Accurate ACH estimation using CO2 data faces challenges due to real-world variables like occupancy and air mixing.
• Further research is needed to refine CO2-based ACH estimation for reliable IAQ monitoring.