Abstract: The flow-focusing droplet microfluidics achieves continuous generation of monodisperse microdroplets by means of flow-focusing effects and interfacial destabilization phenomenon of discrete-phase liquid filament. The multiphase interfacial flow in this technique exhibits dependence on configuration parameters and shows rich microfluidic device developed in our previous study, numerical simulations are used to investigate the influences of key configuration parameters on droplet generation modes and droplet dimensions. After reasonable simplifications, the study establishes an axisymmetric model of the actual device and combines the adaptive mesh refinement technique to improve the efficiency of the numerical simulation. The accuracy of the numerical simulation is verified through the comparison of several experimental operating conditions. It is found that within the selected fluid combination, geometry, and flow parameters, the droplet generation process exists in four modes: dripping, streaming, jetting, and unstable. Under the fixed discrete phase and continuous phase flow rate combinations, the variation of the distance between the upstream and downstream capillary ends changes the droplet length in the dripping and streaming modes, while it has little effect on the droplet size in the jetting mode. Under the fixed geometry parameters, when the flow rates vary, the change of droplet length is nearly continuous at the transition between dripping and streaming modes, but produces a sudden drop at the onset of the jetting mode. The internal diameter of the downstream capillary has a significant effect on the phase diagram, the dripping mode dominates for the large diameter internal diameter and the jet length changes more significantly in the jetting mode, while the jetting mode dominates for the small internal diameter and unstable modes are found at large continuous phase flows. The results of this paper show that the key configuration parameters have important effects on the flow-focusing microfluidic droplet generation, and the applicable alteration of these parameters can control the droplet size and improve the droplet monodispersity, which provides a basis for the design and optimization of flow-focusing microdroplet generation devices.