The Research on Demand-Based Regulation of Mine Airflow Based on Niche Genetic Algorithm

As the core of coal mine safety production, the mine ventilation system plays a crucial role in ensuring the safety of the underground environment and reducing energy consumption. Demand-based regulation of volumetric airflow is essential in this context. To address the limitations of traditional methods in regulating optimal branch selection and multi-branch coordinated control, this paper proposes an intelligent volumetric airflow regulation method for mines based on the Niche Genetic Algorithm (NGA), focusing on solving the nonlinear optimization problem of multi-branch adjustment in complex ventilation networks. The study establishes a ventilation network optimization model with the objective of minimizing the total power consumption of the fans, incorporating the penalty function method and simulated annealing algorithm to handle constraint conditions. In terms of regulating branch selection, a method is proposed for determining the adjustable range by comprehensively considering sensitivity attenuation rate and volumetric airflow constraints. Additionally, the mutual disturbance effects during multi-branch regulation are quantified based on Taylor series expansion theory. By introducing an improved niche genetic algorithm and initializing the population using Latin hypercube sampling, a fitness sharing strategy is adopted to maintain population diversity, enabling the rapid acquisition of multiple sets of near-optimal solutions. Experimental results show that, for single-branch regulation, the error of fitting the resistance-airflow relationship using a power function is less than 0.6%. In multi-branch regulation, the airflow prediction accuracy of the second-order Taylor expansion is significantly better than that of the first-order. The optimization algorithm demonstrates effective convergence in both dual-branch and triple-branch regulation cases, providing multiple feasible regulation schemes with significant improvements in fan power optimization. The research results offer an effective solution for the dynamic optimization and energy-efficient operation of complex ventilation networks and can be further extended to application scenarios such as coordinated control of multiple fans and optimization of multiple demand airflow branches.