At present

in particle flow theory

mesomechanical parameters can only be obtained by varying them until the macro mechanical parameters of the numerical sample match that of the laboratory rock-soil mass sample. The adjustment process is inefficient with some blindness; so a new method should be introduced to establish the relationship between macro mechanical parameters and mesomechanical parameters. Based on PFC3D program

a nonlinear network model linked macro mechanical parameters and mesomechanical parameters is founded by adopting back propagation (BP) neural network; so mesomechanical parameters can be inversed rapidly and accurately by inputting macro mechanical parameters. Some study results are gained as follows: (1) Precision of macro mechanical parameters calculated by inversed results are generally over 90%. (2) Inversion performance of BP neural network model is best when RES

which means the number of particles across the minimum scale of the model

is equals to 10 and the hidden layer has six neurons. Application results show that BP neural network model exhibits an excellent inversion ability of mesomechanical parameters of rock-soil mass and provides a new technical approach for application of particle flow theory.