The cavitation water jet has the characteristics of high flow velocity and strong turbulence,and its cavitation flow structure is very complex.Therefore,it is necessary to accurately and quickly extract key information from the massive data of unsteady flow field and analyze the flow characteristics of cavitation water jets.The turbulent viscosity of RNG model was modified based on FBDCM model,and the cavitation model was modified by considering the influence of shear stress on cavitation bubble formation,which was compared with the experimental results.Then,the mode decomposition of cavitation bubble volume fraction field of the cavitation flow field was carried out by the dynamic mode decomposition (DMD) method.The results show that the first-order mode is the average flow field,which is the main structure of the flow field; the second-order mode is related to the dynamic behavior of the periodic shedding and collapse of the cavitation cloud; the third-order and fourth-order modes show the high-frequency behavior of the flow field.With the increase of inlet pressure,the evolution of cavitation bubble shedding and collapse becomes more and more intense,the size of cavitation bubble and the length of cavitation hole increase gradually,and the frequency of the second-order mode decreases gradually,which makes the flow period of cavitation jet longer.Finally,the DMD reconstructed flow field was compared with the real flow field.It is found that as the inlet pressure increases,the error of the DMD reconstructed flow field increases,but the maximum error of the reconstructed flow field is still less than 1%,it shows that the DMD method can accurately extract the flow characteristics of the cavitation flow field.