Aiming at the problems of many meso parameters of PFC 3D， difficulty in analyzing the influence mechanism between macro and meso parameters， and insufficient interpretability of parameter calibration model， an improved BP algorithm is proposed employing the function with steepness factor， adaptive Nesterov momentum method and adaptive learning rate for lightweight analysis of meso parameters. The comparison with the traditional BP algorithm shows that this algorithm improves the stability and convergence speed of the traditional algorithm and can reasonably and accurately quantify the sensitivity between macro and meso parameters， screen out the key meso parameters and obtain a lightweight macro-meso parameter analysis model. Based on the lightweight model， the correlation and influence mechanism between macro and meso parameters are analyzed， and the influence law of single and multiple key meso parameters on macro parameters is studied. The research shows that the elastic modulus increases linearly with the increase of the equivalent modulus. The parallel bond stiffness ratio can inhibit the growth of elastic modulus， and the elastic modulus decreases greatly when the stiffness is relatively high. There is a linear relationship between Poisson's ratio and stiffness ratio， and the increase of friction coefficient causes the nonlinear decrease of Poisson's ratio. The strength parameters have an incentive effect on the peak strength. The combination of the parallel bond tensile and shear strength at a high level has the most significant impact on the increase in the peak strength. The internal friction angle mainly has an influence on the post-peak strength of the rock. The internal friction angle has a certain control effect on the particle sliding on both sides of the shear zone when the sample is loaded after the peak. The research results have reference values for PFC parameter calibration.