Improving the Efficiency of the Optimization Algorithm of the Temperature Regime of Multi-Apartment Building Indoor Areas

The article substantiates the relevance of optimization algorithms research as for solving various applied problems as for the science of artificial intelligence. The need to solve problems of optimizing the thermal and hydraulic modes of buildings as part of the project "Smart City" explains. The article presents a mathematical formulation of the problem of optimizing the temperature mode of rooms using adjustable devices, as well as two methods for solving the posed problem: the coordinates search method and the genetic algorithm. A description of the above algorithms, including the mathematical apparatus used, is given. The objective function is described as the standard deviation of the temperature of the heated rooms. A method for calculating the temperature of air in a heated indoor area is given by solving the heat balance equation, including a method for calculating incoming heat fluxes from a radiator and outgoing heat fluxes through enclosing structures. Calculation formulas are given. The idea of improving these methods through preliminary insulation measures, which consists in installing additional sections of radiators in “cold” rooms, is considered. An algorithm for automatically issuing recommendations on the installation of additional sections of radiators is described. This algorithm consists in reading information from the database and comparing the calculated heat input from radiators with their rated power. In the event of a power lack, recommendations are issued on the installation of additional radiator sections. The results of the computational experiment for the considered optimization methods are presented. In conclusion, it was shown that the genetic algorithm shows better optimization results than the coordinates search method, but it has a large computational cost. The hypothesis was confirmed that in order to increase the efficiency of solving the considered class of problems, it is necessary to combine the genetic algorithm and gradient methods.