In order to construct a sensor network to monitor the fluctuation of underground parameters that reflect the running state of ventilation network in real time and realize the intelligent perception of mine ventilation parameters,the wind speed parameters that can best reflect the changes of the ventilation network were taken as the monitoring objects,and the sensor optimization number was used as the uncon-strained objective function.The cut set matrix algorithm was used to solve the function and determine the number of branches installed by the sensor,and the distribution of the influence degree of the branch was used to optimize the location of the sensor,so as to realize an optimal construction of the sensor network,and to collect data by via the wind speed sensor so as to obtain the global operation state of the ventila-tion system.In order to verify the reliability of network optimization,Liyazhuang Coal Mine was taken as the engineering background for the example optimization.The results showed that the whole mine ven-tilation parameter monitoring could be realized only by arranging sensors in 32 branches.