Welcome to the World of Differential Equations!

In this chapter, we are going to learn how to solve equations that involve derivatives. While a normal equation like \(2x = 10\) helps us find a specific value, a Differential Equation helps us find a function. This is incredibly important because most things in the real world—from how a disease spreads to how a bridge vibrates—are described by how they change, which is exactly what a derivative is!

Don't worry if this seems a bit abstract at first. We will break it down into simple, step-by-step methods that will work every time.


1. First-Order Differential Equations

A First-Order equation only contains the first derivative, \(\frac{dy}{dx}\). In this course, we specifically focus on the Linear form:

\(\frac{dy}{dx} + P(x)y = Q(x)\)

The "Integrating Factor" Method

If an equation looks like the one above, we use a special tool called an Integrating Factor, denoted as \(I(x)\). It’s like a "magic multiplier" that turns the left side of the equation into a single product that is easy to integrate.

Step-by-Step Process:

  1. Ensure the equation is in the standard form (the coefficient of \(\frac{dy}{dx}\) must be 1).
  2. Identify \(P(x)\).
  3. Calculate the Integrating Factor: \(I(x) = e^{\int P(x) dx}\).
  4. Multiply the entire equation by \(I(x)\).
  5. The left side now simplifies to \(\frac{d}{dx}(I(x)y)\).
  6. Integrate both sides with respect to \(x\).

Quick Review: Remember that \(e^{\ln(f(x))} = f(x)\). This happens a lot when calculating the integrating factor!

Common Mistake: Forgetting to multiply the right-hand side (\(Q(x)\)) by the integrating factor. It’s a common "exam-pressure" error!

General vs. Particular Solutions

  • General Solution: The answer includes a constant \(+ C\). It represents a "family" of curves.
  • Particular Solution: If you are given coordinates (like \(x=0, y=1\)), you can solve for \(C\) to get one specific curve.

Key Takeaway: Use the Integrating Factor for equations where \(y\) and its derivative are added together.


2. Second-Order Homogeneous Equations

Now we step it up! These equations involve the second derivative, \(\frac{d^2y}{dx^2}\) (or \(y''\)). We start with Homogeneous equations, which equal zero:

\(ay'' + by' + cy = 0\)

The Auxiliary Equation

To solve these, we pretend the equation is a simple quadratic. Replace \(y''\) with \(m^2\), \(y'\) with \(m\), and \(y\) with \(1\). This gives us the Auxiliary Equation: \(am^2 + bm + c = 0\).

The type of solution depends on the Discriminant (\(b^2 - 4ac\)) of this quadratic:

  1. Two Distinct Real Roots (\(m_1, m_2\)):
    Solution: \(y = Ae^{m_1x} + Be^{m_2x}\)
  2. One Repeated Real Root (\(m\)):
    Solution: \(y = (A + Bx)e^{mx}\)
  3. Complex Roots (\(p \pm iq\)):
    Solution: \(y = e^{px}(A \cos(qx) + B \sin(qx))\)

Analogy: Think of the auxiliary equation as a "DNA test" for the differential equation. Once you find the roots, the "identity" (form) of the solution is revealed.

Key Takeaway: The roots of the auxiliary equation determine whether your solution grows exponentially, decays, or oscillates like a wave.


3. Second-Order Non-Homogeneous Equations

What if the equation doesn't equal zero? \(ay'' + by' + cy = f(x)\)

To solve this, we combine two parts:

General Solution = Complementary Function (CF) + Particular Integral (PI)

  • Complementary Function (CF): Solve the equation as if it equalled zero (using the auxiliary equation method above).
  • Particular Integral (PI): Pick a "trial" function that looks like \(f(x)\) and plug it in to find the coefficients.

Trial Forms for the PI:

If \(f(x)\) is...

  • A polynomial (e.g., \(x^2 + 3\)): Use \(y = \lambda x^2 + \mu x + \nu\)
  • An exponential (e.g., \(e^{5x}\)): Use \(y = \lambda e^{5x}\)
  • Trig (e.g., \(\sin(2x)\) or \(\cos(2x)\)): Use \(y = \lambda \cos(2x) + \mu \sin(2x)\)

Memory Aid: If your trial PI is already part of your CF, multiply the trial PI by \(x\) to make it unique!

Key Takeaway: The CF deals with the "natural behavior" of the system, and the PI deals with the "external force" being applied.


4. Modelling: Oscillation and Damping

This is where the math meets the real world! We use second-order equations to model things that vibrate, like a car's suspension or a guitar string.

Simple Harmonic Motion (SHM)

If there is no resistance, we get \(\frac{d^2x}{dt^2} = -\omega^2 x\). The solution is always a wave: \(x = A \cos(\omega t) + B \sin(\omega t)\).

Damping

In real life, friction or air resistance slows things down. This is Damping. We add a term proportional to the velocity (\(\frac{dx}{dt}\)):

\(m\ddot{x} + k\dot{x} + \omega^2 x = 0\)

  • Over-damped: Huge resistance. The system slowly returns to equilibrium without oscillating. (Auxiliary roots are real and distinct).
  • Critically damped: The fastest way to return to equilibrium without "overshooting." (Auxiliary root is repeated).
  • Under-damped: The system oscillates but the waves get smaller and smaller. (Auxiliary roots are complex).

Did you know? Engineers use "Critical Damping" for high-end door closers so the door shuts quickly but doesn't slam!

Key Takeaway: Damping "sucks" energy out of the system, turning oscillations into decay.


5. Coupled First-Order Equations

Sometimes, two things depend on each other. For example, in a Predator-Prey model, the number of foxes depends on the number of rabbits, and vice-versa.

System looks like:
\(\frac{dx}{dt} = ax + by + f(t)\)
\(\frac{dy}{dt} = cx + dy + g(t)\)

How to solve:

  1. Differentiate the first equation to get \(\frac{d^2x}{dt^2}\).
  2. Substitute the \(\frac{dy}{dt}\) term using the second equation.
  3. Replace any remaining \(y\) terms by rearranging the first equation.
  4. You are left with one second-order equation in terms of \(x\) only. Solve it normally!

Key Takeaway: To solve two linked variables, combine them into one second-order equation.


Final Checklist for Success

  • Identify the order: Is it first-order (Integrating Factor) or second-order (Auxiliary Equation)?
  • Check the RHS: If it's zero, it's homogeneous. If not, you need a CF and a PI.
  • Watch your variables: Don't mix up \(x, y,\) and \(t\). In modelling, time (\(t\)) is usually the independent variable.
  • Sketching: Be prepared to sketch solution curves. For complex roots with a negative real part, draw an oscillating wave that gets smaller over time (exponential decay envelope).

Good luck! Differential equations are a powerful tool. Master these methods, and you can model almost anything in the physical universe.