To study the compaction mechanism of asphaltic concrete core material from a mesoscopic point of view, a discrete element model including the coarse aggregate, the asphalt mortar and voids were constructed. Then, an inversion method of meso-structural parameters for the PFC(particle flow code)modelling of asphaltic concrete was proposed based on support vector machine model(SVM)and adaptive difference algorithm(ADE). Finally, the model and method were validated by a series of static pressure tests with different temperatures and pressures. The results show that the regression model based on SVM can well build the relationship between the input(parameters to be inverted)and the output(simulated settlement). Additionally, the PFC-simulated compaction settlement with the inverted parameters was very close to that from the physical test. The inversed meso-structural parameters can provide a condition for further study on the meso-mechanism and compaction quality of asphaltic concrete core wall.