This video is very informative, I have only a small nitpick in the code: using omega = 1 we divide by sin(omega * a) = sin(a) = sin(3pi) = 0! The matrix indeed shows huge values (not infinity because of floating point rounding errors)
very good video sir i did watch lot of video but did not satisfied but after this becouse the governing position means before programe u did explain very well easilly understand so that not face any problem in ans
I can't thank You enough, sir! Very understandable. I had problems understanding how to work in MatLab, but I followed your steps and it was getting easier!
Hello! I have some questions: 1. How do deal with the corners when setting BC? It seems that the order at which each BC is defined matters. 2. When calculating the analytical part and plotting it, does it automatically account for BC, or we should also separately define BC and evaluate analytical part only in the middle? Because some solutions only account inside the rectangle in the continuum limit, but I do not know what to do when we are discretising it.
Hello. about the Q.1, because number of grid is high there is no difference to chose the value of corner from bottom or left boundary (for lower left corner for example) About Q.2 when we solve the analytical problem in final answer we considered boundary conditions, otherwise the number of solution for a partial deferential equation could be infinity without boundary conditions. You can find the analytical solution of Laplace equation in cartesian coordinate in a separate video in the channel. Here is the link : th-cam.com/video/2aJ3fFwET68/w-d-xo.html
@@abolfazlmahmoodpoor_ Thank you for the answers! Although, I understand about corners when dx and dy are small. My question was rather how to deal with them when the grid size is not so fine, sorry I did not precise this before.
If you have a large grid you will have large error in numerical solution, simply you can check it in the code, just decrease number of grids. So you have to have small grid, and in small grid no matter from which boundary you chose the corner value!
Respected Sir, I am doing same code in my system but I am getting some errors like " Error using max Trailing string input must be 'omitnam' or 'includenan' Please solve my this problems. Thanks in advance.
Thank you so much sir for this great video. But can you please help me on how to implement the Neumann boundary conditions on one, two, three or four sides of the Laplace equation using FDM
I received your email. If following explanation is not enough let me know using the email, it is easy, just using the definition of first order derivative. We used it a lot in the videos, If you use it you will see that one node after boundary the value of function at that point is given by the condition that imposed by derivate (an example: df/dx = a --> f(1)- f(0)= a*dx --> f(1)=a*dx + f(0) so the value of function is known in either node number 0 and node number 1 and you should just start calculation from node number 2 )
sir can u solve a governing equation of transient heat conduction with heat generation and convection for 2D using finite difference method? with matlab code
This video is very informative, I have only a small nitpick in the code: using omega = 1 we divide by sin(omega * a) = sin(a) = sin(3pi) = 0!
The matrix indeed shows huge values (not infinity because of floating point rounding errors)
very good video sir i did watch lot of video but did not satisfied but after this becouse the governing position means before programe u did explain very well easilly understand so that not face any problem in ans
I can't thank You enough, sir! Very understandable. I had problems understanding how to work in MatLab, but I followed your steps and it was getting easier!
awesome content finally
Doc . good job please dont left please upload a series of videos with help of matlab of finite differece method
Excellent Sir:))
I am really loving your work sir:))
Thank you
This is relevant. Thanks a lot!
Thanks Abolfazl ,
your video works more than too much for me ,it was great, 🌹🌹🌹hope to see more from you.
Br,
Excellent man.
Great video sir! Thank you!
Think you very much for the very good video.
Thank you very much sir, great lecture !!!
This is very great, can you do circular domain
Oh Man.... Thank you so much ❤
Hello! I have some questions: 1. How do deal with the corners when setting BC? It seems that the order at which each BC is defined matters.
2. When calculating the analytical part and plotting it, does it automatically account for BC, or we should also separately define BC and evaluate analytical part only in the middle? Because some solutions only account inside the rectangle in the continuum limit, but I do not know what to do when we are discretising it.
Hello.
about the Q.1, because number of grid is high there is no difference to chose the value of corner from bottom or left boundary (for lower left corner for example)
About Q.2 when we solve the analytical problem in final answer we considered boundary conditions, otherwise the number of solution for a partial deferential equation could be infinity without boundary conditions. You can find the analytical solution of Laplace equation in cartesian coordinate in a separate video in the channel.
Here is the link :
th-cam.com/video/2aJ3fFwET68/w-d-xo.html
@@abolfazlmahmoodpoor_ Thank you for the answers! Although, I understand about corners when dx and dy are small. My question was rather how to deal with them when the grid size is not so fine, sorry I did not precise this before.
If you have a large grid you will have large error in numerical solution, simply you can check it in the code, just decrease number of grids. So you have to have small grid, and in small grid no matter from which boundary you chose the corner value!
@@abolfazlmahmoodpoor_ ok, I see. Thank you!
loved it :) thank you
thank you. It is very useful.
Thanks for this concise and effective 2D solution at t=0.
Do you have another video showing the wave behaviour when u(t) in time?
No, sorry
Respected Sir,
I am doing same code in my system but I am getting some errors like " Error using max
Trailing string input must be 'omitnam' or 'includenan'
Please solve my this problems.
Thanks in advance.
Dear Nitish, Sorry for such a late reply, If you didn't solve the problem yet you can send it to me, I will check. abolfazlmahmoodpoor@gmail.com
Excellent
Masha Allah nice explanation, Sir please if you could help me with my Semi-implicit Finite Difference code
This is beautiful!could you extend it to 3D wave and structure application and do a tutorial on it?thanks in advance
hello . sir i wite same to same in matlab14 but this code is not working
Thank you so much sir for this great video.
But can you please help me on how to implement the Neumann boundary conditions on one, two, three or four sides of the Laplace equation using FDM
I received your email. If following explanation is not enough let me know using the email, it is easy, just using the definition of first order derivative. We used it a lot in the videos, If you use it you will see that one node after boundary the value of function at that point is given by the condition that imposed by derivate (an example: df/dx = a --> f(1)- f(0)= a*dx --> f(1)=a*dx + f(0) so the value of function is known in either node number 0 and node number 1 and you should just start calculation from node number 2 )
@@abolfazlmahmoodpoor_ can you send mail to me about how to apply neumann boundary condition. my email is shivaraghav@200701@gmail.com
How to solve for unsteady state?
Thanks a lot , great video
Glad you liked it!
thank you❤❤❤
sir can u solve a governing equation of transient heat conduction with heat generation and convection for 2D using finite difference method? with matlab code
Did you find out how to do this?
ERROR IMPLEMENTING MAXU. PLEASE SIR WHAT SHOULD BE THE WAY OUT
i tried to write the same code but the error not equal to zero in the final
Thank you!!!!
th-cam.com/video/XR578io_3eg/w-d-xo.html to get the analytical and numerical solutions of a Laplace/Poisson equation with MATLAB
Thanks!!!
We want finite difference for Fokker-Planck Equation
In near future, I hope!
👍👍
Sir, thank you so much
glad it helped
@@abolfazlmahmoodpoor_ Sir, is there any way to import an excel spreadsheet into MATLAB and store it into any matrix ?
[num,txt,raw] = xlsread('myfile') ;
num gives only numbers.
txt gives only text data.
raw gives the complete file.
plz give the code sir
Please follow the video and try to repeat the same code, I think everything is clear in video.
@@abolfazlmahmoodpoor_ okay thanks sir
During writing the code if you have some problem feel free to ask
@@abolfazlmahmoodpoor_ sure sir , thanks for the help
#pimathlab