Demand Response of HVAC Systems Using Data-Driven Approaches and Modelling Procedure

Abstract

Demand Response (DR) from Heating, Ventilation, and Air-Conditioning (HVAC) systems is quantified by studying performance of energy buildings. Physical models, hybrid methods and data-driven approaches are used to predict the performance of building energy. Physical models require numerical equations that account for specific physical attributes and characteristics of building envelope materials. While the physical models are advantageous in describing heat transfer mechanisms, they are time-consuming, require expertise, are difficult to make proper assumptions and may not adapt to environmental or socio-economic variabilities. Hybrid models have similar drawbacks as physical models and require expertise and improper assumptions. But, data-driven approaches build models based on statistical data and overcome the shortcomings of model-based and hybrid approaches. Due to these advantages, data-driven approaches have gained popularity in recent years. In this context, this paper attempts to summarise and develop an overarching view of data-driven approach for building DR. Moreover, this review highlights the comparison of model-based and data-driven approaches for building DR and highlights the key benefits of the data-driven approach for building DR in power systems.