Study on long-term variation characteristics of indoor CO2 concentrations based on a data-driven method

Abstract

People sometimes stay indoors for a long time, such as sleepiness at night in bedrooms, study indoors with doors and windows closed, and it is significant to study long-term variation characteristics of indoor carbon dioxide (CO2) concentrations for studying the effects of the CO2 concentrations and improving indoor air quality. However, the researches on the variation of the indoor CO2 concentrations and its effects on humans were mainly focused on short-term time (≤2h) caused by some problems, such as physiological needs of subjects (for example, bathroom breaks) and insufficient financial resources for long-term measurements. To overcome these problems, a data-driven CO2 model was proposed to analyze the long-term (≥8h) variation characteristics of the indoor CO2 concentrations, in which only short-term data need to be measured. The CO2 concentrations of this study was measured within 1 h under different ventilation conditions in a 30 m3 experiment cabin. The measured data of some conditions were used to calibrate the key parameters of the data-driven CO2 model, such as indoor CO2 release rate and natural ventilation rate, and other data were used to validate the accuracy of the data-driven CO2 model under mechanical and natural ventilation conditions. And then the long-term CO2 concentration variations and their stability values were obtained and analyzed under the different ventilation conditions. Results indicate that the simulation errors of the model in mechanical and natural ventilation conditions were less than 3%. The indoor CO2 concentrations will reach 1000 ppm and 5000 ppm in about 0.5 h and 5.9 h under natural ventilation conditions, respectively. And a ventilation rate of 30 m3/h per person can ensure the indoor CO2 concentrations complying with the criteria of less than 1000 ppm. Therefore, the research methods and results of this study can provide references and suggestions for the studies about CO2 concentration variations and influences, and then to improve the indoor air quality.