Preparing Weather Data for Real-Time Building Energy Simulation
This work addresses the problem of inaccurate or incomplete weather data for building energy simulations, which can lead to flawed simulation results or unexpected terminations, particularly benefiting building performance analysts.
This study developed a framework for quality control of weather data, including anomaly detection and infilling missing values, which are critical inputs for building energy simulations. The framework, utilizing Neural Networks, demonstrated increased accuracy in imputing missing temperature and relative humidity data compared to other supervised learning methods.
This study introduces a framework for quality control of measured weather data, including anomaly detection, and infilling missing values. Weather data is a fundamental input to building performance simulations, in which anomalous values defect the results while missing data lead to an unexpected termination of the simulation process. Traditionally, infilling missing values in weather data is performed through periodic or linear interpolations. However, when missing values exceed many consecutive hours, the accuracy of traditional methods is subject to debate. This study demonstrates how Neural Networks can increase the accuracy of data imputation when compared to other supervised learning methods. The framework is validated by predicting missing temperature and relative humidity data for an observation site, through a network of nearby weather stations in Milan, Italy. Results show that the proposed method can facilitate real-time building simulations with accurate and rapid quality control.