Explore the vast range of titles available.

An intelligent modelling method driven by flow field images for predicting steady and unsteady flow filed around aerofoils has been developed. Signed Distance Field (SDF) images achieve dimensionality enhancement of aerofoil geometric information, and ‘synthesised images’ achieve dimensionality enhancement of the angle of attack of the aerofoil and Mach number. An intelligent aerodynamic model for steady flow field of aerofoils is constructed based on the U-Net neural network architecture, and further incorporating a long short-term memory (LSTM) module to construct a U-Net-LSTM neural network architecture to extract the temporal features. Typical NACA aerofoils results show that, the prediction error for steady flow is less than 1.98%, while the prediction error for unsteady flow is less than 2.56%. Additionally, the model demonstrates good generalization capability, with a generalization error for steady flow less than 2.45% and a generalization error for unsteady flow less than 3.34%. This research provides a new method for intelligent aerodynamic modelling based on physical representations. Compared to existing methods, this method avoids the need for extracting aerofoil geometry information and eliminates the necessity of predicting the flow field point by point, making it more concise and efficient. Highlights 1. An aerodynamic model was constructed using U-Net to rapidly predict the steady flow field around airfoils. 2. A Long Short-Term Memory (LSTM) module was incorporated to capture temporal information, enabling the rapid prediction of the unsteady flow field around airfoils. To address the problem of ‘dimension loss’ in the modelling datasets, effective data dimensionality enhancement was achieved using SDF images and ‘synthesized images’.

Line integral convolution (LIC) is a useful visualization technique for a vector field. However, the output image produced by LIC has many problems in a marine vector field. We focus on the visual quality improvement when LIC is applied in the ocean steady and unsteady flow field in the following aspects. When a white noise is used as the input in a steady flow field, interpolation is used to turn the discrete white noise into continuous white noise to solve the problem of discontinuity. The “cross” high-pass filtering is used to enhance the textures of streamlines to be more concentrated and continuity strengthened for each streamline. When a sparse noise is used as the input in a steady flow field, we change the directions of background sparse noise according to the directions of vector field to make the streamlines clearer and brighter. In addition, we provide a random initial phase for every streamline to avoid the pulsation effect during animation. The velocities of vector field are encoded in the speed of the same length streamlines so that the running speed of streamlines can express flow rate. Meanwhile, to solve the problem of obvious boundaries when stitching image, we change the streamline tracking constraints. When a white noise is used as an input in an unsteady flow field, double value scattering is used to enhance the contrast of streamlines; moreover, the “cross” high-pass filtering is also adopt instead of two-dimensional high-pass filtering. Finally, we apply the above methods to a case of the surface wave field in typhoon condition. Our experimental results show that applying the methods can generate high-quality wave images and animations. Therefore, it is helpful to understand and study waves in typhoon condition to avoid the potential harm of the waves to people’s lives and property.