This is our first apply ConvLSTM to CFD successfully! although the case is simple and under control of lots of factors. The ground factor is generated by Openfoam, and the custom model is predNet from
coxlab. We trained three models in this time.
1.Training: use Nth frame to predict (N+1)th frame
Prediction: use 1-10th frame to predict 2-11th frame, then combined 11th frame in the predicted output with 1-10th frame. With new input(2-10th are ground truth, 11th is predicted), we can keep predicting 3-12th frame. In this experience, we predict the frames until 20th where sliding window = 1 frame. ( only first few frame are good, since we use the predicting frames to do the prediction)
2.Training: Nth frame to predict (N+10)th frame
Prediction: use 1-10 frame to predict 11-20 frame, no sliding window. ( this is very good since all input are ground truth)
3.Training: use Nth frame to predict (N+1)th frame
Prediction: use one frame to predict next frame, like driving prediction (animation has a little problem. right side is prediction, left side is ground truth)
4 Responses on “First Running! Deep learning to CFD”
Sandipan Banerjee says:
Great work, I am trying to do something similar, and want to ask you, the following questions:
1. Are your inputs in images or are they as physical quantities like the velocity field, pressure, etc?
2. If the inputs are physical quantities, then for each grid point you have how many features, and would it be possible to provide a sample of your code specially the section where you include the data in the spatial and temporal format?
1. I input images, but you can input physical quantities as well, like multi dimensional vectors(in a multi dimensional gird).
2. you can find my code here: https://github.com/neoaksa/Application-of-DL-to-CFD.
I am very interesting how you use vectors as input, I want to do it but I hadn’t enough time. Keep in touch.
it’s very interesting. can you share your dataset?? i have a similar project with 400 diferent 2Dshape and their cfd analyses.
each shape has 100 frame from time 0 to 100. how can i train my model? you train your mode onle with 1 shape??