This study combines Machine Learning (ML) techniques and Computational Fluid Dynamics (CFD) simulations to assess the Indoor Air Quality (IAQ) in office settings. The primary objective is to evaluate the IAQ in an open-plan office and accurately predict and forecast Carbon Dioxide (CO2) concentrations by leveraging CO2 time-series data. Given the critical importance of maintaining acceptable IAQ and its direct correlation with CO2 control, which traditionally relies on reactive measures, this research examines the potential of Random Forests (RFs) in enabling a proactive control strategy to enhance IAQ. To achieve these objectives, IoT sensors were deployed to train a Random Forest (RF) model, evaluating its performance in forecasting CO2 concentrations with a ten-minute lead time. Various timesteps and tree configurations were assessed, revealing that the RF model achieved the highest accuracy with a window size of six timesteps and eight hundred trees, resulting in a Mean Absolute Error (MAE) of 2.55. The findings of this study hold promise for advancing IAQ management strategies by leveraging ML techniques and CFD simulations to facilitate accurate CO2 concentration forecasts and enable informed decision-making regarding proactive ventilation strategies.
| File Size : | 1
file
, 2.3 MB |
| Note : | This product is unavailable in Russia, Belarus |
| Number of Pages : | 9 |
| Product Code(s) : | D-CH-24-C090 |
| Published : | 2024 |
