Genetic algorithm-based energy optimization for on-demand ventilation systems
In view of the problem of branch airflow optimization in the operation and maintenance of position en-gineering ventilation systems(that is to minimize the total ventilation power consumption while satisfying the on-demand ventilation requirements of each branch),this study constructed a ventilation model based on graph data-bases and graph algorithms,and designed a step-by-step optimization method for ventilation energy consumption by using genetic algorithms.First,it described the energy consumption optimization problem of the ventilation network under on-demand ventilation conditions,and established a target function for optimization.Secondly,a graph database model of the ventilation system was constructed to address the nonlinear constrained optimization issue of the hybrid ventilation system,and the graph query language and Prim′s algorithm were used to analyze,the relationships within the model′s components and to generate independent circuits.A genetic algorithm solution process was then proposed for the residual tree string parameters,involving the design of encoding rules to solve for airflow and pressure regulation parameters,followed by the allocation of airflow and pressure regulation calcu-lations for the independent circuits.The results were then evaluated against the target function.Finally,an opti-mization analysis was conducted by using a specific engineering case,which showed that the optimized ventilation system had a total power consumption reduction of 6.62%,and proved the high efficiency of the model and the rationality of the algorithm.