Abstract: In fluid simulation, the parameters of the constitutive model are difficult to predict accurately, which leads to the inconsistency between the simulation results and the visual effects of real videos. To solve this problem, a real video-oriented non-Newtonian fluid simulation reconstruction method is proposed. The model training phase takes non-Newtonian fluid simulation videos as input and learns the best low-dimensional latent space representation of a single frame of fluid simulation images. Inter-frame prediction is then performed in this latent domain, employing a convolutional long-short-term memory network to predict latent vector representations of future frames. Finally, the reconstruction parameters of the constitutive model are predicted based on the frame-by-frame latent representation encoding and inter-frame temporal features. In the model verification stage, the real video of the non-Newtonian fluid is used as the input to predict the parameters of the fluid constitutive model, and realize the simulation and reconstruction of the non-Newtonian fluid based on the Cross model.The experimental results show that the video-oriented simulation reconstruction method can obtain the fluid flow phenomenon more consistent with the real video than the reconstruction method based on the rheometer measurement. The proposed method has higher pixel accuracy and pixel accuracy at different times, and has a visual effect that is more in line with the actual flow.

Key words: video spatiotemporal features, data-driven model, simulation reconstruction, non-Newtonian fluid