Optimization of Air distribution in Mine ventilation networks based on Genetic Algorithm (case study: Kalariz Coal Mine)

The optimal distribution of air in mining operations can be called the most important executive solution to reduce the operating and capital cost of the ventilation system that causes by choosing the correct location and characteristics of the fans for the actual and adequate distribution of air flow in The network. Due to the multiplicity of parameters affecting the performance of the active ventilation system and its various components, achieving the optimum network parameters is time-consuming. Applying Metaheuristic techniques can help solve problems quickly and find optimal values of ventilation network. The ventilation network simulation was performed in the two-dimensional implementation of the eastern Kalariz coal mine, identifying 43 nodes, 61 branches and 22 network loops. Then values of the intersection matrix and the fundamental matrix of the loop were identified. After simulation of the current design with the Ventsim software, the air flow rate in some of branches was less than the required level to establish the required minimum air flow rate. Therefore, the optimization of air distribution in the Kalariz coal mine was conducted in two separate conditions including the current conditions of the mine and supplement of minimum air flow intensity in all branches of the network. For network modeling, the semi-controlled flow type II was determined. Then the coding of the objective function in Matlab R2014 software was based on the minimization of energy consumption with regarding to flow rate and the pressure loses of the Kirchhoff's laws. Also the effect of the parameters of the genetic algorithm (mutation rate and crossover rate) along with population size on optimal response was investigated. The results of the implementation of the model indicate a decrease in energy consumption from 10054 to 10040 watts. In the second condition, the energy consumption increased to 13696 watts and the current intensity was 32 m3/second. This values due to the fact that there were more than 22 mining branches, which in the first case did not observe the minimum required air flow. The efficiency of the genetic algorithm was determined in the analysis of the optimal fan performance and the amount of pressure drop required for the regulator doors. Increasing the number of people to a certain degree increases the chances of achieving the optimal answer. As the coefficients of the crossover and mutation become larger, the accuracy of the calculation decreases and the time to reach the optimal response increases.