Abstract: Abstract: Compared with other drying methods, vacuum freeze-drying (VFD) has better rehydration property, which can keep the original quality and color to the maximum extent, and has broader application prospects in such fields as food, medicine and biological products. In order to realize the on-line monitoring of moisture ratio, it is important to research the rules of moisture diffusion and the parameter optimization of VFD process within fruits and vegetables. In this paper, the drying ratio in VFD process was expressed with the change of porosity for fruits and vegetables based on Lattice Boltzmann Method (LBM). During the VFD process, ice crystals' sublimation begins at the material surface, and then goes into the inside gradually; the drying layer gets thicker and thicker, and the frozen layer thinner and thinner. Accordingly, the porosity in different parts of the material is different: the porosity at the edge expands quickly with the earliest sublimation and the value is the biggest at last, while the porosity at the center stays stable, and its value is the smallest. Based on the analysis of VFD process of fruits and vegetables, the paper constructed a square porous media model. The porosity of the model was variable, and the model presumed that the porosity increased linearly from the center to the edges during the drying procedure. By using the LBM combined with the conservation laws of mass, momentum and energy, the VFD velocity distribution in the model was simulated. The results showed that the bigger the change of porosity, the higher the VFD velocity, which suggested that the change of porosity in porous media can express the drying rate in VFD process. Then the experiment took apples as samples to validate the results in the simulation. Cleaned and peeled, apples were cut into 10 mm × 17 mm × 17 mm rectangular samples by a sharp self-made knife and frozen in a refrigerator at -40℃ for 10-12 h. In a VFD machine, through the built-in computer in the machine and on-line moisture measurement system connected to the machine, the real-time data of drying apples were collected and the curve of drying was drawn. In different drying stages of apples (1, 3, 5 and 7 h), the distance between 4 different positions at water boundary and sample edge was measured with vernier calipers, the electron microscope scanning was used to observe apple samples surface, and the porosity of apples in different water boundary was got with the help of image processing software Photoshop 2015 and Matlab 7.0. Origin 8.5 software was adopted to analyze the correlation relationship between the location of water boundary and the porosity of apples with experimental data. The results found that the porosity of apples increased linearly from the center to the edges during the drying procedure. Origin 8.5 software was also adopted to analyze the correlation relationship between the moisture ratio and the porosity of apples with experimental data. The mathematical function between the drying rate and the change of porosity was deduced on the basis of the relationship between the moisture ratio and the porosity. The results found that the bigger the change of porosity, the higher the VFD velocity. All the results in the experiments were consistent with those in the simulation, which proved that the change of porosity of apples can express drying rate in VFD process and LBM can be used to analyze moisture migration diffusion and distribution in VFD process. The conclusions not only provide basis for parameter optimization of VFD process of fruits and vegetables, but also offer application infrastructure for on-line monitoring of moisture ratio.