Estimation of meteorological data with intelligent systems used in the design of thermal systems in Turkey

It has been seen that investigations have been focus on energy saving due to the limited energy sources and increasing of demand of the energy with increasing population in the world. Especially due to the importance economical consumption of available energy source researchers mostly have studied in this field. Forward-looking estimation of the meteorological data in the design of thermal systems is of very great importance.In this study, a program was developed with intelligent systems for the prediction of meteorological data used in the design of thermal systems in Turkey representing to 50 cities throughout the country. In this thesis, the meteorological data obtained from Turkish State Meteorological Service were modeled by using artificial neural networks and adaptive fuzzy neural network-based modeling systems. In the modeling the Matlab package program was used and thus used data in the design for future thermal system were estimated with high precisions. The heating degree-day and cooling degree-day zones were suggested with the help of these estimated data. Using estimated data for Turkey, a prospective map of humidity, temperature, solar energy potential, heating degree-day and cooling degree-day were prepared with Surfer and ArcGIS package programs. A graphical user interface was designed with Matlab GUI package program to take advantage of comfortable and successful usage of designed model for end-user. An economic analysis was conducted for two different wall structures by taking into account the number of new heating degree-day from the model.Consequently, meteorological data such as humidity, temperature and radiation intensity values which are used in the design of heating, ventilation and air conditioning systems (HVAC) have been estimated with an accuracy of 99 percent by the help of intelligent systems.