8. Solution by Variation of Parameters

In many engineering applications, especially in civil engineering, differential equations model real-world systems such as structural deflection, fluid flow, and heat conduction. Often, the equations encountered are non-homogeneous linear differential equations, where the method of undetermined coefficients may not be applicable due to the form of the non-homogeneous term. In such cases, the method of variation of parameters becomes a powerful and general technique to obtain a particular solution.

In engineering, we often need to solve differential equations that describe physical systems. These equations can be complex, especially when they are non-homogeneous, meaning they include a forcing function (or external influence). The traditional method of undetermined coefficients works well for specific types of forcing functions, but when the function doesn't fit these types, we turn to variation of parameters. This method allows us to find a particular solution to a broader range of problems by varying parameters instead of undetermined coefficients.

Think of it like trying to tune a musical instrument. If you only know how to play one specific tune (the method of undetermined coefficients), you might struggle when asked to play different songs (different forcing functions). Variation of parameters is like learning to improvise—you can play along to any song, adjusting your approach as needed.

A general second-order linear non-homogeneous differential equation is given by: y′′+p(x)y′+q(x)y =g(x) Where: • y is the dependent variable (typically displacement, temperature, etc.) • x is the independent variable (time, distance, etc.) • p(x), q(x): Coefficient functions • g(x): Non-homogeneous term (external input or forcing function) The solution to this equation is given by: y(x)=y_h(x)+y_p(x) Where: • y_h(x): General solution to the homogeneous equation y′′+p(x)y′+q(x)y =0 • y_p(x): Particular solution to the non-homogeneous equation.

The presented equation is fundamental to understanding how we solve differential equations in engineering. The equation consists of the second derivative of y, its first derivative, and the function itself, alongside coefficients and an external influence g(x). The total solution y(x) is made up of two parts: the homogeneous solution, which addresses the system without external forces, and the particular solution, which accounts for those external forces. This decomposition helps us solve complex systems by breaking them down into manageable parts.

Imagine you're trying to determine how much a bridge sways (y) under normal conditions (homogeneous) and during a storm with strong winds (non-homogeneous). First, you would calculate how the bridge behaves without the wind (the homogeneous solution). Then, you would add in the effect of the wind (the particular solution) to get a complete picture of the bridge's sway.

Suppose we already have the solution to the corresponding homogeneous equation: y′′+p(x)y′+q(x)y =0 Let the two linearly independent solutions of the homogeneous part be y1(x) and y2(x). Then the general solution of the homogeneous equation is: y_h(x)=C1y1(x)+C2y2(x) To find a particular solution y_p(x) to the non-homogeneous equation, we assume: y_p(x)=u_1(x)y_1(x)+u_2(x)y_2(x) Here, u_1(x) and u_2(x) are functions to be determined.

The principle of variation of parameters hinges on leveraging the known solutions of the homogeneous equation to develop a particular solution. By assuming that the particular solution can be constructed as a linear combination of these solutions, weighted by unknown functions u_1 and u_2, we can derive the particular solution to the non-homogeneous equation. This approach allows us to account for the external influences that change the behavior of the system.

Consider a chef who knows how to make a basic bread recipe. If they want to add unique flavors—like herbs or spices—they can use the original recipe (the homogeneous solution) and replace standard ingredients with new ones (the weighted functions). This way, they create a unique variant of the bread that still maintains the foundational structure.

Differentiate y_p(x): y_p(x)=u_1(x)y_1(x)+u_2(x)y_2(x) To simplify the derivation, we impose a constraint: u_1(x)y_1(x)+u_2(x)y_2(x)=0 Then: y_p'(x)=u_1'(x)y_1(x)+u_1(x)y_1'(x)+u_2'(x)y_2(x)+u_2(x)y_2'(x) Differentiate again: y_p''(x)=u_1'(x)y_1'(x)+u_1(x)y_1''(x)+u_2'(x)y_2'(x)+u_2(x)y_2''(x) Now substitute into the original non-homogeneous equation: y''+p(x)y'+q(x)y=g(x). After substituting and simplifying using the homogeneous equation, we arrive at a system of equations for u_1 and u_2.

In order to derive the formula for the variation of parameters, we need to differentiate our assumed solution multiple times. The key is imposing a constraint that simplifies the calculations. By ensuring that the linear combination of u_1 and u_2 vanishes, we reduce complexity while solving for these functions. Once we perform substitutions and simplifications, we obtain a manageable system of equations that helps isolate u_1 and u_2, leading us to our particular solution.

Think of this as a detective working to solve a mystery. The detective collects clues (differentiates) from various sources (the initial assumption). To keep the investigation focused, they set certain rules about what types of information to pursue (the constraint). As these clues are analyzed (substituted into the main theory), they help the detective narrow down potential suspects (find u_1 and u_2) until the case is solved.

1. Solve the homogeneous equation to find two linearly independent solutions y_1(x), y_2(x).
2. Compute the Wronskian W(x)=y_1y_2'-y_1'y_2.
3. Compute u_1(x) and u_2(x) using: u_1(x)=−(y_2g)/(W), u_2(x)=(y_1g)/(W).
4. Integrate to find u_1(x),u_2(x).
5. Construct the particular solution: y_p(x)=u_1(x)y_1(x)+u_2(x)y_2(x).
6. Write the general solution as: y(x)=y_h(x)+y_p(x).

The step-by-step procedure provides a systematic way to approach problems involving non-homogeneous differential equations. In the first step, we find the solutions to the associated homogeneous equation, which serves as a foundation. The Wronskian helps us measure the independence of our chosen solutions. Subsequent steps involve calculating the functions u_1 and u_2, integrating to find their explicit forms, and then using these to build the particular solution. Finally, we combine this with the homogeneous solution to arrive at the complete solution.

Think of following a recipe to bake a cake. You start by mixing the basic ingredients (solving the homogeneous equation). Next, you check that your ingredients are fresh and usable (compute the Wronskian). You then follow specific steps to add flavors and decorations (compute and integrate u_1 and u_2). Finally, you bake the cake (construct the particular solution) and serve it (write the general solution) for everyone to enjoy!

--