# Is my work correct? (1 Viewer)

### Users Who Are Viewing This Thread (Users: 0, Guests: 1)

#### bancux

1. The problem statement, all variables and given/known data
I need to work on a differential equation.
$$\frac{d^2T}{dx^2} - (m+n\ sin(kx))\ T = 0$$

2. Relevant equations
Is my work correct?

3. The attempt at a solution
$$\frac{d^2T}{dx^2} - (m+n\ sin(kx))\ T = 0$$
$$\frac{d}{dx}\left(\frac{dT}{dx} \right) = (m+n\ sin(kx))\ T$$

$$\int \frac{d}{T}\left(\frac{dT}{dx} \right) = \int (m+n\ sin(kx)) \ dx$$
$$\frac{1}{T}\left(\frac{dT}{dx} \right) = mx-\frac{n}{k}\ cos(kx)+C_1$$

$$\int \frac{dT}{T} = \int (mx-\frac{n}{k}\ cos(kx)+C_1)\ dx$$

$$ln T = \frac{m}{2}x^2-\frac{n}{k^2}sin(kx)+C_1x+C_2$$

$$T = e^{\frac{m}{2}x^2-\frac{n}{k^2}sin(kx)+C_1x}\ e^{C_2}$$

$$T = e^{\frac{m}{2}x^2-\frac{n}{k^2}sin(kx)+C_1x}\ C_2$$

#### betel

No, your third line is not correct. Separation of variables does not work for second order DE.

#### bancux

No, your third line is not correct. Separation of variables does not work for second order DE.
So what is the correct solution?

#### betel

Do you want me to solve your work or give you hints?

There is no obvious way to solve this equation. You could try a Fourier series expansion for T.

#### bancux

Well actually my question is because I am trying to solve the homogeneous part of a 2nd order inhomogeneous differential equation with variable coefficient.

Could you give some hints on the Fourier expansion, or any good reference to that?
My main goal is to solve the differential equation which I mentioned earlier.

Thanks

#### betel

So this DE is not some part of a homework assignment?
Because I did not find a nice analytic solution using mathematica.

The method to solve the DE depends also on what you want to do with the solution afterwards. Fouriersieries might give you a solution but you it could not be very useful.

So what is the background you are trying to solve this equation? Are m,n,k fixed or do they have to be chosen. For some values solutions might exist, for some not.

I'll try to find and explicit way to solve the equation in the meantime.

#### bancux

So this DE is not some part of a homework assignment?
Because I did not find a nice analytic solution using mathematica.

The method to solve the DE depends also on what you want to do with the solution afterwards. Fouriersieries might give you a solution but you it could not be very useful.

So what is the background you are trying to solve this equation? Are m,n,k fixed or do they have to be chosen. For some values solutions might exist, for some not.

I'll try to find and explicit way to solve the equation in the meantime.
It is not a homework though I do need some result on this.
I have attached a pdf file so that you can have a better view on my problem.

Thank you.

#### Attachments

• 18.8 KB Views: 67

### The Physics Forums Way

We Value Quality
• Topics based on mainstream science
• Proper English grammar and spelling
We Value Civility
• Positive and compassionate attitudes
• Patience while debating
We Value Productivity
• Disciplined to remain on-topic
• Recognition of own weaknesses
• Solo and co-op problem solving