Paper on Daily Power Demand Prediction for Buildings at Large Scale has been published in Building Simulation.
Large scale power demand prediction for buildings plays a great role in stable operation and management for the grid. To predict large scale power demand in an accurate and fast way, our team has developed a new method called E-GAN, which combines a physics-based model (EnergyPlus) and a data-driven model (GAN), to predict the daily power demand for buildings at a large scale.
This work has been published under the title "Daily power demand prediction for buildings at a large scale using a hybrid of physics-based model and generative adversarial network " in the journal Building Simulation. The full paper is availbale here.
The first author of this paper, Chenlu Tian, was a visiting Ph.D. scholar in the SBS lab, where her research focused on building data analysis using machine learning methods.
Congratulations to Chenlu on publishing this paper!