Application of the Genetic Algorithm for Optimizing the Room Temperature Regime through Balancing Valves of Risers

Abstract

The paper proves the relevance of the research of optimization algorithms for both the field of science on artificial intelligence and for solving various applied problems. The necessity of solving optimization problems of thermohydraulic buildings regimes is explained. The air temperature in the heated rooms depends on many factors, in particular on the parameters of the enclosing structures, the parameters of the coolant, the environment, and also the resistance of the balancing valves of the risers. The necessity of regulation of the hydraulic resistance of the nodes of a multi-circuit building heating system with the help of adjusting valves for eliminating the temperature imbalance of the building premises is substantiated. The paper presents a mathematical formulation of the optimization problem, as well as two methods for solving the problem posed: the coordinate search method and the genetic algorithm. A block diagram of the genetic algorithm, the probability formula for crossing, and the selection rules are given. The experimental results for these optimization methods are presented, including the graphs of the change of the objective function by iteration, the comparative table, the explanation of the results. Finally, the conclusion is drawn that the genetic algorithm shows better optimization results than the method of coordinate-wise search, but has a large computational cost. A hypothesis is proposed for further investigation.