Abstract
All over the world, there are several unoccupied spaces without adequate constant control mechanisms to reduce and prevent mold and provide good internal conditions and indoor air quality. A widespread way to reduce building humidity is through heating and dehumidification, which are costly to maintain and have high energy consumption. In addition, there are few studies on adjustable hygro ventilation systems, which do not consider the influence of temperature fluctuations. This work describes the operation of a prototype, which fills existing research gaps by considering not only the control of relative humidity (RH) but also the temperature peaks in indoor air conditions, allowing the maintenance of good air quality. The prototype Smart Hygrothermal Ventilation system uses two pairs of sensors related to RH and temperature, one pair placed inside an unoccupied compartment of the building and the other pair in the external environment, in order to activate a fan and the respective speed. The proposed prototype was applied in a compartment located on the ground floor in an unoccupied old rural building in a village near Porto during the winter period. The results show that the system performed adequately for different configurations of its functionalities. Therefore, the system offers an efficient alternative to minimize mold and the fluctuation of internal RH and temperature. Furthermore, it could be a vital mechanism for the conservation of historic buildings.

Abstract
Artificial Intelligence has recently expanded across various applications. Machine Learning, a subset of Artificial Intelligence, is a powerful technique for identifying patterns in data to support decision making and managing the increasing volume of information. Simultaneously, Digital Twins have been applied in several fields. In this context, combining Digital Twins, Machine Learning, and Smart Buildings offers significant potential to improve energy efficiency and operational effectiveness in building management. This review aims to identify and analyze studies that explore the application of Machine Learning and Digital Twins for operation and energy management in Smart Buildings, providing an updated perspective on these rapidly evolving topics. The methodology follows the PRISMA guidelines for systematic reviews, using Scopus and Web of Science databases. This review identifies the main concepts, objectives, and trends emerging from the literature. Furthermore, the findings confirm the recent growth in research combining Machine Learning and Digital Twins for building management, revealing diverse approaches, tools, methods, and challenges. Finally, this paper highlights existing research gaps and outlines opportunities for future investigation.

Abstract
Recent studies have focused on forecasting power demand in buildings to enhance energy management. However, the literature still lacks comparative analyses of power demand forecasting algorithms. In addition, more case studies involving different building typologies are needed, as each building exhibits distinct behavior and load profiles. This paper aims to develop machine learning models to forecast the power demand of a large shopping center in the northern region of Portugal. The main objective is to compare the performance of several machine learning models. The results are promising, demonstrating adequate performance even during most holidays.