Course contentsShow
← Mathematics
Lesson 2360 of 3,12148. Ordinary Differential EquationsFree lesson

The Variation of Parameters Formula for Second-Order ODEs

Deriving the general formula for particular solutions using two independent homogeneous solutions.

The Variation of Parameters Formula for Second-Order ODEs

What You'll Learn

You'll learn how to derive a general formula for finding particular solutions to non-homogeneous second-order linear ODEs using two independent solutions from the homogeneous equation.

The Core Idea

Recall that for a non-homogeneous ODE like y'' + p(t)y' + q(t)y = g(t), the general solution is y = y_h + y_p where y_h comes from the homogeneous equation.

The Method of Undetermined Coefficients works beautifully—but only when g(t) has special forms and coefficients are constant. What if g(t) is messy, like ln(t) or tan(t)? What if p(t) and q(t) aren't constant?

Variation of Parameters solves this by building on what we already know: the homogeneous solutions.

Suppose you've found two linearly independent solutions y₁(t) and y₂(t) to the homogeneous equation (confirmed by checking their Wronskian is nonzero). Instead of assuming the particular solution has a fixed form, we "vary the constants" in the homogeneous solution by replacing them with unknown functions:

y_p(t) = u₁(t)y₁(t) + u₂(t)y₂(t)

Through substitution into the original ODE and strategic simplification (imposing the condition u₁'y₁ + u₂'y₂ = 0 to avoid second derivatives of u), we derive:

u₁' = -y₂g / W
u₂' = y₁g / W

where W = W(y₁, y₂) is the Wronskian. Integrate these to find u₁ and u₂, then construct y_p.

This formula works for any continuous g(t) and doesn't require constant coefficients—it's the universal method!

Key Takeaway: The Variation of Parameters formula derives particular solutions by treating the homogeneous solution's "constants" as functions, resulting in integrals involving the forcing term g(t), the homogeneous solutions y₁ and y₂, and their Wronskian.