Building occupants find it difficult to achieve the optimal indoor environment quality (IEQ) through natural ventilation. To solve this problem, this study aimed to develop an automatic ventilation control algorithm considering the IEQ factors and occupant ventilation behavior. The algorithm was developed in four steps: (i) real-time collection of data on the IEQ factors and occupant ventilation behavior; (ii) development of the automatic ventilation control algorithm using logistic regression; (iii) determination of the automatic ventilation control algorithm using receiver operating characteristic curve analysis; and (iv) evaluation of the automatic ventilation control algorithm's performance according to the indoor environmental standards. Through this process, the logistic regression model with ridge regression (area under curve: 0.865), with the highest classification accuracy, was selected. Then Youden's index was used to define the decision criterion (i.e., optimal cutoff value) for the logistic regression model. As a result, the decision criterion for opening and closing the windows or doors was 0.533. When the developed algorithm was compared with the indoor environmental standards to analyze its performance, the compliance rate of the opening of the windows or doors based on the monitored data was 77.6%, but it increased to 99% based on the data classified by the developed algorithm. It is expected that if the automatic ventilation control algorithm is embedded in a building ventilation system, which is connected to various IEQ measurement sensors, it will offer a customized building ventilation system to the building occupants.
All Science Journal Classification (ASJC) codes
- Environmental Engineering
- Civil and Structural Engineering
- Geography, Planning and Development
- Building and Construction