STUDY ON VORTEX CONTROL MECHANISM OF SYNTHETIC JET BASED ON FEATURE EXTRACTION METHODS

A synthetic jet with an inclination angle of 60 ° is applied at the bottom of a crossflow, and its periodic vortex motion is numerically simulated by large eddy simulation (LES). The generation and evolution mechanism of the vortex interaction between the synthetic jet and the cross flow are studied under different driving parameters. The third-generation vortex identification method (Liutex vector) is used to track the vortex evolution in the flow field. The rotation intensity and vortex core size of the vortex structure, which are generated by the synthetic jet under different driving frequencies (St=0.25, 0.5, 0.75, 1) and driving amplitudes (A₀=1, 1.5, 2, 2.5), are quantitatively recorded. In addition, the proper orthogonal decomposition (POD) and dynamic mode decomposition (DMD) are used to extract the structure and frequency features of the flow field with specific frequency (St=0.25) and amplitude (A₀=2.5) of the synthetic jet, and the main characteristic modes affecting the flow field are analyzed. The results show that amplitude has positive contribution to the rotation intensity and vortex core size, while frequency has negative contribution to the both. The vortex motion mechanism of synthetic jet in crossflow can be summarized as follows: under the condition of lower frequency and lower amplitude, the clockwise vortex structure is favorable, while it is beneficial to the generation of counter-clockwise vortex under higher frequency and higher amplitude. Based on the spatial structure and spectrum characteristics of POD mode, the main characteristic structure of the flow field is shedding vortex with a shedding frequency of St=0.25, which is consistent with the jet driving frequency. The primary clockwise vortex structure is found to be the dominant structure for jet vortex control, while each mode is multi-frequency coupling. The flow field information is decoupled in time and space by DMD, and the dynamic characteristics in terms of shear induction and vortex merging are captured. Therefore, the combination of various feature extraction methods contributes more to the profound study of the mechanism of vortex control of synthetic jet.