Resumen de A comparative review and recommendations for modeling thermostatically controlled loads in buildings
Gabriel Gómez Ruiz, María Reyes Sánchez Herrera, Jesús Clavijo Camacho, José Manuel Andújar Márquez 
Thermostatically controlled loads (TCLs) are essential for enhancing building sustainability by optimizing energy consumption through smart control strategies. Model predictive control (MPC) stands out as one of the most effective methods for managing TCLs. It aims to minimize energy use, reduce costs, and maintain thermal comfort, all while adapting to dynamic external factors such as electricity prices and grid demand. Accurate modeling of TCL behavior is essential for achieving these objectives. The objectives of this work are organized into two primary phases. The first phase involves a detailed review of the three main modeling approaches―white-box, data-driven, and gray-box―applied to commonly used TCLs in buildings, including air conditioners, refrigerators, electric water heaters, and heat pumps. In the second phase, the most suitable modeling approach―whether white-box, data-driven, or gray-box―is determined and recommended for the TCLs under consideration. This recommendation is based on a comprehensive assessment of key factors, such as system complexity, data availability, required accuracy, computational resource demands, and scalability. The outcomes of this systematic comparison are intended to guide stakeholders―including building energy efficiency companies, energy service companies (ESCOs), utilities, and grid operators―in selecting the most appropriate model for specific applications, guided by detailed criteria.
