In order to address the issue of improper control of air volume in mine automatic ventilation systems,which has the potential to pose personal safety risks,a method based on the SA-BP algorithm is proposed as a means of optimizing such systems.Environmental sensors are employed to gather data regarding methane concentration,dust,temperature,and other pertinent variables,which are then subjected to analysis using the Simulated Annealing(SA)algorithm.The negative back propagation characteristics of the BP neural network are utilized to optimize the weights and thresholds of the BP neural network through the SA iteration.The BP neural network output error is employed as the adaptive function of the SA algorithm,facilitating the selection of the optimal target between ventilation volume and CH4,dust,etc.The optimal objective value between ventilation and CH4 and dust is selected.Experimental results demonstrate that this method can accurately predict the ventilation of the mine,exhibiting enhanced robustness and generalization ability.
维普
万方数据