Deep-Learning-Based Natural Ventilation Rate Prediction with Auxiliary Data in Mismeasurement Sensing Environments

Predicting the amount of natural ventilation by utilizing environmental data such as differential pressure, wind, temperature, and humidity with IoT sensing is an important issue for optimal HVAC control to maintain comfortable air quality. Recently, some research has been conducted using deep learning provide high accuracy in prediction. Therefore, reliability required achieve predictions successfully. However, it practically difficult predict accurate NVR a mismeasurement environment, since inaccurate are collected, example, due sensor malfunction. we need way deep-learning-based prediction environments. In this study, overcome degradation mismeasurement, use complementary auxiliary generated semi-supervised selected importance analysis. That is, model reliably trained generating selecting data, then predicted integration bagging-based ensemble approach. Based on experimental results, confirmed that proposed method improved rate 25% compared baseline context various address realistic utilize rapidly changing or slowly characteristics which can improve observation data.