Optimizing measurement schemes to improve indoor airflow and temperature CFD–EnKF joint simulation
Weixin Qian,Hu Gao,Yanyu Lu,Sheng Lyu,Lei Zhuang,Site Hu,Lixiang Wang,Jing Liu
DOI: https://doi.org/10.1016/j.buildenv.2023.111070
IF: 7.093
2023-11-01
Building and Environment
Abstract:An ensemble Kalman filter (EnKF) effectively rectifies simulation inaccuracies arising from uncertain flow field boundary conditions. In this methodology, measurement data plays a pivotal and indispensable role. This study investigated the requirements for observation data in joint EnKF and computational fluid dynamics (CFD) simulation models to correct errors. It established guiding principles for measurement schemes and validated them using numerical experiments on the airflow and temperature fields in a small chamber. The optimized scheme significantly reduced simulation errors, enhancing assimilation accuracy. In addition, the influence of measurement noise was considered, revealing a positive correlation between assimilation and measurement errors, thus highlighting the significance of high-precision instruments. Additionally, this study confirmed that the optimal number of measurement points corresponds to the number of uncertain boundary condition variables. Fewer measurement points hinder accurate estimation, whereas increasing their number does not significantly improve the data assimilation accuracy, regardless of measurement noise. Overall, this study provides guidelines for implementing EnKF data assimilation to ensure reliable and precise results.
engineering, environmental,construction & building technology, civil