A METHODOLOGY TO DESIGN A KNOWLEDGE-BASED TOOL FOR RESIDENTIAL BUILDINGS SIMULATION

The actual growing world energy demand is generating strong attention to the energy efficiency and to the environmental sustainability. The residential sector is one of the most energy-intensive reaching about 25% of global energy consumption. Furthermore, it is difficult to understand the real energy use in residential buildings suggesting the development of methodologies and tools to monitor and assess their energy performances. Such an analysis requires defining all the actors, their interaction rules and the intelligent management of a large amount of data. In addition to this, the energy performances of the home environment are closely related to the specific case under investigation, in fact it imposes the analysis of the particular application scenario and the target users to extract parameters able to describe the building behavior. They are related to the technological characteristics of the systems, to the external environment (e.g. external temperature, solar irradiation, etc.) and to the user needs and habits. It introduces some complexities, because many of these data are difficult to find and to predict. Another cause of complexity was introduced by the information and communication technologies (ICT) that creates new relationship between home devices fostering the spread of smart systems. In this scenario, simulation tools have been developed to grasp the real energy performance of a building, but they require a high level of detail for the input data which are often difficult to find. Otherwise, other tools are built for much simplified simulations and typically addressed to inexperienced users neglecting the real complexities of the system. For this reason, the paper presents a new design approach that aims to develop knowledge-based tools with architecture able to simulate the real behavior of the building considering all the actors and their interaction rules, but at the same time containing the features that allow them to be used in fast simulations or by inexperienced users. One of the focuses is the development of an intelligent user interface that, when requested, is able to automate and simplify data entry process. It is particularly useful when the user does not know the necessary input data for the simulation, e.g. energy consumption profiles, detailed features of the building, etc. All the steps of the proposed approach are described in the paper. In the first section is highlighted how the key parameters have been found and how they are organized as creating a knowledge base, while in the second section is shown the operation of the smart user interface and of the simulation tool. The results demonstrate that the proposed knowledge-based approach generates real benefits by simplifying and speeding up the data entry. Furthermore, the study shows how the smart user interface allows performing complex simulations also to novice users.