Real-time energy optimization feasibility for nearly zero energy buildings (NZEB) concept

Airports have a critical importance in the international air transportation system, as they provide the connection of the region to other points and most importantly, because of the contribution they make to the development of the region in which they are located. With the growth of airport infrastructures, airport-related business, commercial, residential and spatial development is taking place around the airport, which is connected by land transportation infrastructure. Airports no longer consist solely of aviation-related functions (e.g. passenger, cargo and aircraft handling facilities); It has developed to include shopping and hotel complexes, conference facilities, industrial zones, logistics centres and public transportation centres. With this complex and mega structure, in developing the infrastructure of airports and ensuring the sustainability of airports, studies on reducing harmful emissions, clean and green energy concepts, reducing carbon footprint, meeting the requirements of green building certifications such as NZEB, LEED, BREEAM have become among the corporate goals and have become critical. In addition to these concepts, the Airport Carbon Accreditation (ACA) program, specific only to airport structures, can also be preferred for airports. ACA aims to reduce the carbon footprint of airport complexes by minimizing their negative impact and includes 7 levels. All these concepts and areas of influence aim to make a positive contribution to global climate change. The aim of this thesis is to reduce the current carbon footprint by turning to cleaner and greener energy in an airport operation with a significant volume of real-time energy consumption and production, and to obtain LEED and BREEAM certification so that the consumed energy can be met from clean energy sources produced or planned to be produced within the scope of the project. It aims to carry out an optimization study based on the requirements. Consumption analyses, architectural structure, window and glass areas, energy use and CO2 amount, heating, cooling, domestic water installation, air conditioning and ventilation and wastewater treatment systems, domestic water systems, electricity and energy centre systems will be examined, and designs will be made after the current situation analysis, and suggestions will be presented. A successful hybrid machine learning model has been developed to predict future energy consumption, and its impact has been analysed, due to the important role energy consumption data plays in determining energy efficiency, certification criteria, and the sustainability of proposed efficiency projects. In the hybrid machine learning model, a basic model is first selected and trained. The predictions of the basic model are then added as a new input feature to the model, forming the meta-model input data. The predictions made by the meta-model are combined with the predictions of the basic model. The meta-model's predictions are used to improve the accuracy of the next model's predictions. The reason for using such a hybrid model is to provide more accurate predictions.