Abstract
A neural network technique that extracts underlying flow features from the original flow field data is newly proposed. The technique here is based on the convolutional and sparse autoencoder learning algorithms and is called sparse convolutional autoencoder. Unlike the typical convolutional neural network (CNN) that changes the size of the data itself in the intermediate layers, flow field data size is not changed in the learning process of this method and only the numbers of channels are changed in each layer. Different but the same size of the data as the input are obtained by convolution with multiple spatially overlapping flow field data under the assumption of sparsity. When data restoration is realized in this autoencoder system, the channel numbers of data in the intermediate layers turn out to contain different flow characteristics of the original flow field. The proposed method is applied to the low Reynolds number flows over a circular cylinder. The high-fidelity unsteady flow data obtained by solving two-dimensional compressible Navier–Stokes equations with a high-resolution numerical scheme are used as a test case. In the proposed method, sparsity introduced in the middle-hidden layer is essential for the successful separation of the original data. The results presented in the example seem to correspond to positive and negative magnitudes of the original data, but future studies will reveal other features of the method. The present method shows flow features different from those of proper orthogonal decomposition in each mode, which is probably due to nonlinear decomposition in the CNN process.
Highlights
The difficulty of finding flow structures and understanding flow characteristics has been recognized for many years
Unlike the typical convolutional neural network (CNN) that changes the size of the data itself in the intermediate layers, flow field data size is not changed in the learning process of this method and only the numbers of channels are changed in each layer
We presented a new data analysis method based on sparsity, convolution, and an autoencoder to separate the fields of fluid dynamics data into several characteristic fields
Summary
The difficulty of finding flow structures and understanding flow characteristics has been recognized for many years. Proper orthogonal decomposition (POD) is one of the dimensionality reduction methods of the large amount of data used to extract features of them. It was used in many research fields and introduced to the fluid dynamics community by Lumley.. Dynamic mode decomposition (DMD), which is a dimensionality reduction method of the data, can extract the underlying dynamics and coherent structures from large-scale data in fluid flow in terms of spatial distribution. The present method seems to show flow features in each decomposed mode different from those of POD probably due to nonlinear decomposition in the CNN process
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have