Search

As an experimental technique for fluid mechanics, particle image velocimetry (PIV) can extract global and quantitative velocity field from images. With the development of artificial intelligence, designing PIV method based on deep learning is quite promising and worth studying. First, the authors in this paper introduce the optical flow neural network which is proposed in the computer vision community. Second, a dataset including particle images and the ground-truth fluid motions is generated to train the parameters of the networks. This leads to a deep neural network for particle image velocimetry which can provide dense motion estimation (one vector for one pixel) efficiently. The featuring of particle image extracted by the neural network is also investigated in this paper. It is found that feature matching improves the accuracy of estimation. The proposed network model is firstly evaluated by a synthetic image sequence of turbulent flow. A jet flow experiment is also carried out in this paper to validate the practicability. The experimental results indicate that compared with the traditional cross-correlation and optical flow methods, the proposed deep neural network has advantages in accuracy, spatial resolution as well as efficiency. [ABSTRACT FROM AUTHOR]

Mathematical modeling of aerodynamic data plays an important role in the performance evaluation of a flight vehicle. There are two kinds of aerodynamic characteristics modeling methods, i.e., the rational modeling method based on physical mechanism and the "black-box" modeling methods. In this paper, three types of "black-box" modeling methods including the classification and regression tree method(CART), the shallow learning method, and the deep learning method are investigated. The CART method and three shallow learning methods including Kriging method, radial basis function(RBF) neural network method, and support vector machines(SVM) method are applied to the aerodynamic data modeling of rocket, the unsteady aerodynamic characteristics modeling for delta wing with large angle of attack, the aerodynamic heat data fusion of wind tunnel test and CFD. The advantage and shortage of these modeling methods are compared. A deep learning neural network model of airfoil's aerodynamic coefficients is built. It considers the influence of airfoil picture and flow parameters such as Mach number, angle of attack on the prediction results at the same time through a "synthetic picture". It could significantly expand the applied range of the deep learning modeling method in aerodynamic research field. [ABSTRACT FROM AUTHOR]