# Solve PDE w/ Comsol 5.3: Numerical Solution & Time Evolution

• A
• umby
In summary, the conversation discusses the use of Comsol 5.3 to numerically solve a partial differential equation with specific boundary conditions and initial conditions. The equation includes a parameter, St, and the person is interested in tracking the evolution of a specific term involving St over time. However, there is concern about the correctness of the problem due to a term that cannot be a function of the unknown variable. Further study is needed to determine the correctness of the problem.
umby
TL;DR Summary
Numerical solution of a partial differential equation containing the derivative of the unknown at a point
What is the best way to solve numerically the following equation using Comsol 5.3.

##\frac{\partial T}{\partial t}=\frac{\partial ^2T}{\partial x^2}+\text{St}\left[1+\left(\frac{\partial T}{\partial x}\right)_{x=0}\right]\frac{\partial T}{\partial x}##
##T(0,t)=1##
##T(\infty ,t)=0##
##T(x,0)=\exp \left(-\frac{x^2}{\pi }\right)-x \text{erfc}\left(\frac{x}{\sqrt{\pi }}\right)##

where ##\text{St}## is a parameter which can varies from 0.01 to 100.
I am particularly interested in following the evolution of ##\text{St}\left[1+\left(\frac{\partial T}{\partial x}\right)_{x=0}\right]## with time.

Be care. This is not a PDE at all. Even the correctness of such a problem needs for a separate study.

umby
wrobel said:
Be care. This is not a PDE at all. Even the correctness of such a problem needs for a separate study.
You mean because of the coefficient of the derivative of ##T## with respect to ##x##? Maybe differential relation is better? Can you help me please in determining the correctness of this problem?

Last edited:
I mean the term
$$\frac{\partial T}{\partial x}\Big|_{x=0}$$

umby
Exsactly what I was mentioning, thank you for point it out. Coefficients cannot be function of the unknown, only of the indipendent variables, in my case ##x## and ##t##.
Can you please tell me more about the matter of its correctness?

You can probably write a finite difference code easier than using comsol. Then you can insert the nonlinear part of the PDE with ease.

umby and berkeman

• Differential Equations
Replies
3
Views
623
• Differential Equations
Replies
4
Views
1K
• Differential Equations
Replies
3
Views
1K
• Differential Equations
Replies
3
Views
2K
• Differential Equations
Replies
5
Views
2K
• Differential Equations
Replies
5
Views
176
• Differential Equations
Replies
7
Views
2K
• Differential Equations
Replies
3
Views
1K
• Differential Equations
Replies
1
Views
2K
• Differential Equations
Replies
10
Views
2K