Abstract: In view of the current situation that the workload of designing blasting parameters for underground excavation in mines is large and inefficient,and the construction layout drawings are mostly two-dimensional plans,it is proposed to use particle swarm optimization algorithm and neural network algorithm to establish an improved model for the design optimization of blasting parameters for underground excavation to accurately predict the optimal blasting parameters for the roadway to be excavated.Using the designed optimal blasting parameters as data support,3D visualization techniques are applied to automatically construct 3D fine models of blast holes,charge structures,etc.The study was carried out in a mine in Shandong Province to optimize the design of blasting parameters and construct a 3D visualization model.The results show that the error between the blasting parameters designed in this study and the blasting parameters of typical successful cases is small,MAE,RMSE and R2 are 0-154，0-096 and 0-914 respectively.The established 3D model of blasting borehole is of high quality and can directly describe the information of blasting boreholes such as spatial distribution,charging structure and so on.

Key words: tunnel driving,blast drilling,optimization design of blasting parameters,3D model generation,Python