首页>
中文期刊>
《矿业科学技术（英文版）》
>Optimization of air quantity regulation in mine ventilation networks using the improved differential evolution algorithm and critical path method
Optimization of air quantity regulation in mine ventilation networks using the improved differential evolution algorithm and critical path method
In mine ventilation networks, the reasonable airflow distribution is very important for the production safety and economy. Three basic problems of the natural, full-controlled and semi-controlled splitting were reviewed in the paper. Aiming at the high difficulty semi-controlled splitting problem, the general nonlinear multi-objectives optimization mathematical model with constraints was established based on the theory of mine ventilation networks. A new algorithm, which combined the improved differential evaluation and the critical path method（CPM） based on the multivariable separate solution strategy,was put forward to search for the global optimal solution more efficiently. In each step of evolution,the feasible solutions of air quantity distribution are firstly produced by the improved differential evolution algorithm, and then the optimal solutions of regulator pressure drop are obtained by the CPM.Through finite steps iterations, the optimal solution can be given. In this new algorithm, the population of feasible solutions were sorted and grouped for enhancing the global search ability and the individuals in general group were randomly initialized for keeping diversity. Meanwhile, the individual neighborhood in the fine group which may be closely to the optimal solutions were searched locally and slightly for achieving a balance between global searching and local searching, thus improving the convergence rate. The computer program was developed based on this method. Finally, the two ventilation networks with single-fan and multi-fans were solved. The results show that this algorithm has advantages of high effectiveness, fast convergence, good robustness and flexibility. This computer program could be used to solve large-scale generalized ventilation networks optimization problem in the future.
展开▼
