simulated annealing algorithm

Improving thermal comfort in buildings is one of the important issues that is emphasized in today's buildings. Given the growth in energy consumption, saving resources in buildings is an indispensable necessity. Optimal use of free solar energy on the one hand provides a favourable environment for the occupants of the building and on the other hand reduces the building energy consumption in cold months of the year. Since the overall trend of residential construction in Iran has led to the development of apartment units, Applying climate solutions that can be used in such buildings is an important need that is getting less attention these days. The use of semi-open spaces such as balconies, terraces and porchs in apartment units is emphasized by the National Building Regulations of Iran and is widely used in contemporary buildings.  When cleverly designed, these spaces can be converted to a sunspace to be used as a passive system for the absorption and utilization of solar energy. The purpose of this study is to use artificial intelligence and algorithmic optimization to find the optimal dimensions, from and properties of semi-open spaces in warm and dry climates in Iran, to transform them into a sunspace to be used in cold months of the year. The research method in this study is simulation and logical reasoning. To reach this goal, a residential space in Shiraz was simulated, and then the parameters of the semi-open space (dimensions, indoor window area, semi-open space type, glass type and ventilation) were investigated in order to convert it into a sunspace. Then the optimal conditions of each of these parameters for the city of Shiraz were determined. The results show that such an space can optimally reduce up to 54% of the building's heating energy consumption and the best results came from terraces using double glazing cover.

Location and allocation analysis is one of the most important network and useful analyzes in GIS. This analysis involves various models that, each model is used to solve various problems. Recently, a new model called VAOMP (Vector Assignment Ordered Median Problem) has been developed that, it can solve many different problems. Given that, the location and allocation problems are very difficult, it is almost impossible to solve them in exact methods. Therefore, the present research intends to solve the proper allocation problem of the population to existing fire stations, which are the emergency facilities, and their optimal allocation have a great importance. The VAOMP model and the Genetic and Simulated Annealing algorithms with three goals such as minimizing the arrival time of the fire trucks to the location of demands, minimizing distance and maximizing the coverage of each station in a radius of 5 minutes, solved the problem. The results of this research showed that the Genetic algorithm produces more qualitative solutions in shorter time, while 10 stations in the 21st and 22nd districts of Tehran are not sufficient to serve the total demands in the study area and 55240 people without services will remain and at least, 13 stations in the region should be created.