Development of an Ensemble Model for Power Demand Prediction in High-Rise Hotels with Environmental Variables

Accurate load forecasting is essential for efficient energy management in high-rise hotels, especially during periods of unexpected social change, such as the COVID-19 pandemic, which complicates electricity demand predictions. This study proposes an ensemble deep learning approach that combines bidirectional long short-term memory and gated recurrent units to improve the accuracy of electricity demand forecasting. By incorporating hotel revenue management factors, such as average daily rate and revenue per available room, along with pandemic-related variables, including daily confirmed COVID-19 cases and government policies, the model accounts for both environmental and socio-economic factors. Using real-world data, the model demonstrated superior accuracy compared to traditional methods, providing a more effective framework for peak load forecasting in high-rise hotels during times of significant social disruption.