Engineering Math Series 8: Series Solutions and Ordinary Points

Engineering Math Series 8: Series Solutions and Ordinary Points

So far we have seen problems that solve relatively neatly using characteristic equations or standard formulas. But in practice, not all problems work that way. The important idea that emerges here is assuming the solution as a power series.

Why Are Series Solutions Needed

Some differential equations have no solution that can be expressed neatly as elementary functions. But that does not mean we give up on finding a solution. Instead, we can assume the solution as

$y(x) = \sum_{n=0}^{\infty} a_n x^n$

an infinite series, and find the coefficients.

This approach is the starting point that later connects to special functions like Bessel functions and Legendre functions.

What Is an Ordinary Point

When a second-order linear equation is written as

$y'' + P(x)y' + Q(x)y = 0$

if $P(x)$ and $Q(x)$ are analytic near the point $x_0$ where you want to expand, that point can be considered an ordinary point. At the introductory level, understanding it as "a point near which the coefficient functions are sufficiently smooth and the coefficient of the highest derivative is not zero, so you can safely expand" is adequate.

Basic Procedure

The flow of the power series method is usually as follows.

1. Assume the solution as $\sum a_n x^n$
2. Differentiate to write $y'$ and $y''$ as series too
3. Substitute into the original equation
4. Collect terms of the same power
5. Obtain a recurrence relation by comparing coefficients

The core of the computation is "pushing the infinite series into the differential equation and comparing coefficients."

Worked Example

Consider the following equation.

$y'' + y = 0$

Assuming the solution as

$y = \sum_{n=0}^{\infty} a_n x^n$

then

$y'' = \sum_{n=0}^{\infty} (n+2)(n+1)a_{n+2}x^n$

Therefore

$\sum_{n=0}^{\infty} \left[(n+2)(n+1)a_{n+2} + a_n\right]x^n = 0$

and since each coefficient must be zero,

$a_{n+2} = -\frac{a_n}{(n+2)(n+1)}$

This recurrence relation means that the even and odd coefficients are determined separately, and expanding them actually produces the cosine and sine series.

What This Method Gives Us

This example is for an equation whose solution we already know, but it is good for seeing how series solutions work. The important thing is that "even if you cannot find the solution as a single function, you can build the coefficients one by one in order."

Engineering Applications

Starting Point for Special Functions

When solving problems in cylindrical or spherical coordinates, Bessel functions and Legendre functions appear. Many of these functions are defined or understood through series solutions.

Approximate Calculations

When you need to quickly approximate a solution near a certain point, power series are very useful.

Algorithm Implementation

From a developer's perspective, the recurrence relation is essentially an algorithm. Translating the coefficient generation rule into code becomes a piece of a numerical computation library.

Common Mistakes

Not organizing the index shifts

Series solutions quickly become confusing if you do not handle index shifts cleanly. The habit of accurately matching $n$, $n+1$, and $n+2$ is important.

Overthinking the ordinary point concept

At the introductory level, starting with "are the coefficients well-behaved near the point I want to expand around" is sufficient.

Writing the recurrence relation but missing its meaning

The recurrence relation is not just a computational device but information that reveals the coefficient structure and degrees of freedom of the solution.

One-Line Summary

Series solutions are a powerful method for recovering solutions as rules about coefficients in differential equations where closed-form solutions are not readily available.

Next Post Preview

In the next post, we will take a brief pause from the ODE section and move to the Laplace transform, a powerful tool that converts problems into algebraic ones.

References

• Erwin Kreyszig, Advanced Engineering Mathematics, 10th Edition
• George B. Arfken, Mathematical Methods for Physicists
• James Ward Brown, Ruel V. Churchill, Complex Variables and Applications