Last answered:

08 Nov 2021

Posted on:

03 Apr 2020

0

Solving Partial differential Equations using Neural Network

Dear All, Anyone who is using feed forward Neural Network for solving Partial differential equations in tensorflow. Then, please share some ideas, algorithm and codes. Thanks
5 answers ( 0 marked as helpful)
Instructor
Posted on:

05 Apr 2020

1
Hi Sumant, I can recommend taking a look at this (quite old) article: https://arxiv.org/pdf/physics/9705023.pdf
It should cover the theoretical background for the task. For a more practical perspective on using TensorFlow for solving PDEs, take a look at this tutorial:
https://data-flair.training/blogs/tensorflow-pde/ Hope these resources were helpful! Best,
Iskren
Posted on:

06 Apr 2020

0
Hello Iskren,  The tutorial link and research paper are given by you really helpful. I would ask further help required while understanding and coding point of view.   Thanks
Posted on:

24 Sept 2020

0
Hello Iskren,     I have solved the PDE using tensorflow successfully. But, for now I am stuck at solving the coupled equation using the tensorflow with the algorithm mentioned in the https://arxiv.org/pdf/physics/9705023.pdf   . Could you please take a look on my code and suggest the needful improvement in the code.   Thanks      
Instructor
Posted on:

25 Sept 2020

1
Hi Sumant,

Glad to hear you managed to solve the PDE!

I can try taking a look for you if you would send me your code so far (maybe also include the working PDE solver example as reference).

Cheers,
Iskren
Posted on:

08 Nov 2021

0

Hi Sumant,

I have also encountered problems in writing a neural network to solve coupled differential equations, have you solved your problem regarding this area? If so, can I refer to your code? Please forgive me if I offend!



Thanks

Submit an answer