A Vortex Identification Method Based on Extreme Learning Machine

Figure 2

The architecture of the method proposed in this paper, flow field data vortex feature identification process. The preprocessing part preprocesses the data and provides the required input for the network part. The network model part trains a network model for vortex feature recognition. The output of the network model part provides predicted small patches of data for the postprocessing part, and the postprocessing part performs flow field reconstruction on the small patches of data.