Frobenius Method

The Frobenius method is an approach to identify an infinite series solution to a second-order ordinary differential equation. Generally, the Frobenius method determines two independent solutions provided that an integer does not divide the indicial equation’s roots.

Consider the second-order ordinary differential equation given below:

Frobenius method 1

Here, a(x), b(x), and c(x) are “suitable functions”.

Let us assume that these are the suitable rational functions that means polynomials divided by polynomials, such as in Bessel’s equation of order 1/2.

(d2y/dx2) + (1/x) (dy/dx) + [1 – (1/4x2)]y = 0

For this equation, we can illustrate the Frobenius method. The primary approach of Frobenius aims at solutions in the form of power series around some given point x0 multiplied by (x − x0) to a certain power. This can be expressed as:

Frobenius method 2

Here, ak and r are the constants that can be identified through the Frobenius method.

Learn:Ordinary Differential Equations

Let’s start with the Frobenius method to solve the second-order ordinary differential equation.

Step 1: Choose a suitable value for x0. This can be done in two ways:

(i) If conditions are given for y(x) at some point, we can use that for x0.

(ii) If no conditions are given for y(x), we must choose x0 as per our convenience. Generally, we prefer to choose x0 = 0.

Step 2: If the given differential equation is of the form a(x) (d2y/dx2) + b(x) (dy/dx) + c(x) y = 0, then convert this, as mentioned above.

i.e., (d2y/dx2) + (1/x) (dy/dx) + [1 – (1/4x2)]y = 0 ………(1)

Then multiply the equation by 4x2 so that we can avoid fractions to make the simplification easy.

4x2(d2y/dx2) + 4x (dy/dx) + (4x2 – 1)y = 0

(or)

4x2 y′′ + 4xy′ + (4x2 – 1)y = 0……..(2)

Step 3: Let us assume that the solution is of the form

\(\begin{array}{l}y=y(x)=(x-x_0)^r \sum_{k=0}^{\infty}a_k(x-x_0)^k\end{array} \)
, where ak is an arbitrary constant such that ak ≠ 0.

Step 3: Now, bring the factor (x – x0)r inside the summation. That means,

\(\begin{array}{l}y=y(x)= \sum_{k=0}^{\infty}a_k(x-x_0)^{k+r}\end{array} \)

Step 4: From the assumed series of y, we need to calculate the respective modified power series for y′ and y′′ by differentiating “term-by-term”.

For this substitute x0 = 0. Thus, we get;

\(\begin{array}{l}y=y(x)= \sum_{k=0}^{\infty}a_k\ x^{k+r}\end{array} \)

Then, by differentiating y, we get;

Frobenius method 3

Step 5: Substitute the expressions for y, y′ and y′′ in equation (2). On simplification of this we get;

Frobenius method 4

Step 6: For each series obtained in the above equation, we need to change the index so that each series will be of the form:

\(\begin{array}{l}\sum_{n=something}^{\infty}[Term\ not\ containing\ x] (x-x_0)^n\end{array} \)

Step 7: Convert the sum of series in the obtained equation into one big series. A few terms will likely have to be written separately, so we need to simplify to a possible extent.

Step 8: Now, the first term of the obtained series may be of the form;

a0 [formula of r] (x – x0)something

Also, remember that each term of the series is 0.

From this, we can write the formula of r = 0

This yields a quadratic equation in r.

Solving this equation, we get two roots, r1 and r2.

Step 9: By substituting r1 in the last series equation, we get;

Frobenius method 5

From this, we can get;

n th formula of ak’s = 0 for n0 ≤ n

Then, solve this for highest index = formula of n and lower indexed ak’s

To simplify the terms at least a little, perform the change of indices so that the recursion formula will be derived and can be rewritten as;

ak = formula of k and lower-indexed coefficients

Step 10: Using the recursion formula or any corresponding formulas to the lower-order terms, we need to find all the ak ’s in terms of a0 and, maybe, one other am.

Step 11: Using r = r1 and the formulas, we have derived the coefficients. Now, write out the resultant series for y. Simplify it and factor out the arbitrary constant(s) if possible.

Step 12: Repeat steps 9 to 11 for substituting r = r2.

Finally, the last step may yield y as an arbitrary linear combination of two distinct series. Therefore, this is referred to as the general solution to the given differential equation.

Frobenius method 6

This can be further simplified as:

Frobenius method 7

Read more:

Frobenius Method Solved Example

Example:

Find the solution of 4xy′′ + 2y′ + y = 0 by Frobenius Method.

Solution:

Given differential equation is:

4xy′′ + 2y′ + y = 0….(1)

Let

\(\begin{array}{l}y=y(x)= \sum_{k=0}^{\infty}a_k\ x^{k+r}\end{array} \)
be the solution equation.

So,

Frobenius method 8

Substituting the expressions of y, y′ and y′′ in equation (1), we get;

\(\begin{array}{l}4x\left [ \sum_{k=0}^{\infty}a_k (k+r)(k+r-1)x^{k+r-2} \right ]+2\left [ \sum_{k=0}^{\infty}a_k (k+r)x^{k+r-1} \right ]+\sum_{k=0}^{\infty}a_kx^{k+r}=0\end{array} \)

This can be written as:

\(\begin{array}{l}\left [ \sum_{k=0}^{\infty}a_k 4(k+r)(k+r-1)x^{k+r-1} \right ]+\left [ \sum_{k=0}^{\infty}a_k 2(k+r)x^{k+r-1} \right ]+\sum_{k=0}^{\infty}a_kx^{k+r}=0\end{array} \)
…..(2)

Dividing the above equation by xr-1, we get;

\(\begin{array}{l}\left [ \sum_{k=0}^{\infty}a_k 4(k+r)(k+r-1)x^{k} \right ]+\left [ \sum_{k=0}^{\infty}a_k 2(k+r)x^{k} \right ]+\sum_{k=0}^{\infty}a_kx^{k+1}=0\end{array} \)

Changing the indices of the bases, we get;

\(\begin{array}{l}https://latex.codecogs.com/svg.image?\left [ \sum_{n=0}^{\infty}a_n 4(n+r)(n+r-1)x^{n} \right ]+\left [ \sum_{n=0}^{\infty}a_n 2(n+r)x^{n} \right ]+\sum_{n=1}^{\infty}a_{n-1}x^{n}=0\end{array} \)

Now, we need to expand the summation to make the indices equal. This can be done as follows.

\(\begin{array}{l}a_0 4(0+r)(0+r-1)x^0+\left [ \sum_{n=1}^{\infty}a_n 4(n+r)(n+r-1)x^{n} \right ]+a_0 2(0+r)x^0\left [ \sum_{n=1}^{\infty}a_n 2(n+r)x^{n} \right ]+\sum_{n=1}^{\infty}a_{n-1}x^{n}=0\end{array} \)

Now consider the term which is of the form a0[term of r] = 0 .

i.e., a0[4r(r – 1) + 2r] = 0

4r2 – 4r + 2r = 0

4r2 – 2r = 0

r2 – (½)r = 0

r[r – (½)] = 0

Thus, r = ½, r = 0.

Now, by changing the indices of equation (2), we get;

Frobenius method 9

For k ≥ 0, we get;

4(k + r + 1)(k + r)ak+1 + 2(k + r + 1)ak+1 + ak = 0

From this, we can write ak+1 as:

ak+1 = -ak/ [(2k + 2r + 2)(2k + 2r + 1)] ; k = 0, 1, 2, 3, etc., ………..(3)

Substitute r = r1 = ½ in equ (3).

ak+1 = -ak/ [(2k + 3)(2k + 2)]

Substituting k = 0, 1, 2, 3, and so on, in the above equation, we get;

a1 = -a0/3.2 = -a0/3!

a2 = -a1/5.4 = a0/5!

a3 = -a2/7.6 = -a0/7!

As we know, a0 is an arbitrary constant. So, let a0 = 1.

Therefore, ak (r1) = (-1)k/ (2k + 1)!; k = 0, 1, 2, 3,….

Hence,

Frobenius method 10

Similarly, by substituting r = r2 = 0 in equ (3), we get;

Frobenius method 11

Thus, we obtained the solutions of the given differential equation.

Practice Problems

  1. Solve xy′′ + 2y′ + xy = 0 by Frobenius Method.
  2. Solve: 5x2 y′′ + x(1 + x)y′ − y = 0
  3. Consider the differential equation xy” + y’ + 2xy = 0. Obtain the solution by Frobenius method.

Comments

Leave a Comment

Your Mobile number and Email id will not be published.

*

*