How to solve this type of differential equations?

In summary, the conversation discusses a differential equation that describes the change in area opening (ε) over time (t) as a function of velocity (u), length (Z), and constants (k1, k2). The conversation also touches on the method of characteristics and the need for additional information or boundary conditions to solve the equation. There is a discrepancy in the formulation of the equation, which may be due to the physical system being described.
  • #1
svenki7
5
0
dε/dt=d(uε)/dZ+[(e^(-k1t) - e^(-k2t)]

where ε=% area opening, u= velocity, Z=length , k1, k2= constants, t= time

Please help me how to solve the ODE
 
Last edited:
Physics news on Phys.org
  • #2
Context would be helpful - and what you have tried already.
You have not said if u and Z are constant - are they?

"dε/dt=d(uε)/dZ-[(e^(-k1t) - e^(-k2t)]" translates into ##\LaTeX## as:

$$\frac{d\varepsilon}{dt}-\frac{d(u\varepsilon)}{dZ} = e^{-k_1t} - e^{-k_2t}$$
Appears to be a coupled differential equation - so there must be another one for ##\frac{d\varepsilon}{dZ}## or some other information to help you out.
 
  • #3
If u is a constant or a function only of t, then this problem can be solved easily using the method of characteristics. I assume those are partial derivatives with respect to t and Z. Just factor out the u from the partial with respect to z.
 
  • #4
Q=u*ε;
Q= flow rate , u= velocity, ε=area
Q=flow rate is constant;

boundary conditions are
Z=0, t=0, ε=1 and u=uo
Z=0, t>0, ε=1 and u=uo


where uo= initial velocity
 
  • #5
I guess you mean Q is constant wrt to t, but not wrt Z. Seems that u and epsilon are functions of both. So it would be natural to use Q in the equation instead of u.
Can you describe the physical system? It would help ensure we're all on the same page.
 
  • #6
svenki7 said:
Q=u*ε;
Q= flow rate , u= velocity, ε=area
Q=flow rate is constant;

boundary conditions are
Z=0, t=0, ε=1 and u=uo
Z=0, t>0, ε=1 and u=uo


where uo= initial velocity

In my judgement, there is something wrong with this formulation. If the problem were truly as stated, then the throughput rate Q would be constant with z and t, and the PDE would reduce to an ODE.

This looks like the equation for the void fraction variation in some type of fixed bed operation, where the porosity is changing as a result of say dissolution or chemical reaction at the interface. Also, in my judgement, almost certainly, the d(εu)/dz term on the right had side has the wrong sign. Please provide a detailed description of the physical problem being solved so that we can check the formulation. The first step in any math modeling of a physical system is to articulate the physical mechanisms involved, and to correctly translate these physical mechanisms into the language of mathematics.
 

FAQ: How to solve this type of differential equations?

What are the general steps to solve a differential equation?

The general steps to solve a differential equation are:
1. Identify the type of differential equation (e.g. first order, second order, etc.)
2. Rearrange the equation so that the dependent variable and its derivatives are on one side and the independent variable is on the other side
3. Determine the appropriate method to use (e.g. separation of variables, integrating factor, etc.)
4. Apply the method and solve for the dependent variable
5. Check the solution by substituting it back into the original equation and verifying that it satisfies the equation.

What is the difference between an ordinary differential equation and a partial differential equation?

An ordinary differential equation involves only one independent variable, while a partial differential equation involves multiple independent variables. This means that the solution to a partial differential equation will involve multiple functions, whereas the solution to an ordinary differential equation will involve a single function.

Can all differential equations be solved analytically?

No, not all differential equations can be solved analytically. Some equations may not have closed-form solutions, meaning that they cannot be expressed in terms of elementary functions. In these cases, numerical methods or approximation techniques may be used to find an approximate solution.

How can I check if my solution to a differential equation is correct?

You can check your solution by substituting it back into the original equation and verifying that it satisfies the equation. Additionally, you can also check your solution using initial or boundary conditions, if they are given. If your solution satisfies both the equation and the given conditions, then it is considered to be correct.

Are there any software programs that can solve differential equations?

Yes, there are many software programs that can solve differential equations, such as MATLAB, Mathematica, and Maple. These programs use numerical methods and algorithms to find solutions to differential equations. However, it is still important to understand the steps and methods involved in solving differential equations, as these programs may not always give accurate or meaningful solutions.

Similar threads

Back
Top