Non-Separable Differential Equations- Solving Techniques

What Are Non-Separable Differential Equations?

Most students learn separable equations first. You isolate y, separate dx and dy, integrate both sides, done. Clean. Simple. Almost satisfying.

Then you run into the real world.

Non-separable differential equations are the ones where you cannot write dy/dx as a function of y times a function of x. No amount of algebra will let you separate the variables. That's not a failure of technique—that's just how these equations work.

Get used to it. Most practical differential equations fall into this category.

Why You Can't Just Separate Everything

Here's the fundamental problem. Separable equations have this form:

dy/dx = g(x) · h(y)

Non-separable equations don't. Examples:

• dy/dx = x² + y²
• dy/dx = (x + y) / (x - y)
• dy/dx + 2y = eˣ

Notice how the x and y terms are tangled together. You can't untangle them by algebra alone. That's where other techniques come in.

The Main Solving Techniques

1. Linear First-Order Equations

This is probably the most common non-separable type you'll encounter. The standard form:

dy/dx + P(x)y = Q(x)

The trick is the integrating factor. You multiply both sides by μ(x) = e^(∫P(x)dx). This makes the left side collapse into a perfect derivative.

After multiplying, you get:

d/dx [μ(x) · y] = μ(x) · Q(x)

Then integrate both sides. That's it. The solution is:

y · μ(x) = ∫μ(x) · Q(x) dx + C

2. Exact Equations

An equation M(x,y)dx + N(x,y)dy = 0 is exact if:

∂M/∂y = ∂N/∂x

If that condition holds, there's a potential function ψ(x,y) where:

• ∂ψ/∂x = M(x,y)
• ∂ψ/∂y = N(x,y)

Find ψ by integrating M with respect to x (keeping y constant), then find the missing function of y by comparing. The solution is ψ(x,y) = C.

If the equation isn't exact, sometimes you can find an integrating factor μ(x) or μ(y) that makes it exact.

3. Homogeneous Equations

These look intimidating but follow a pattern. An equation is homogeneous if M(tx, ty) = tᵐM(x,y) and N(tx, ty) = tᵐN(x,y). In plain terms: every term has the same degree when you count x's and y's.

The substitution y = vx (or y/x = v) transforms the equation into something separable. Then you solve for v, substitute back, and you're done.

4. Bernoulli Equations

These look like:

dy/dx + P(x)y = Q(x)yⁿ

Where n ≠ 0 and n ≠ 1. The substitution z = y^(1-n) converts this into a linear equation. Solve for z, then substitute back to get y.

5. Substitution Methods

Sometimes you need to get creative. Common substitutions:

• u = x + y works when you see x + y in the equation
• u = y/x for equations involving ratios
• u = ax + by + c for linear combinations

The substitution depends on recognizing patterns. That comes with practice.

Comparison of Methods

Getting Started: A Practical Approach

Here's how to actually solve these things when you encounter one:

Step 1: Classify the equation

Does it fit linear form? Is it exact? Is it homogeneous? Check each possibility in order. Most textbooks teach a decision tree:

1. Can you separate variables? If yes, do that. If no, continue.
2. Is it linear (dy/dx + P(x)y = Q(x))? If yes, use integrating factor.
3. Is it exact? Check ∂M/∂y = ∂N/∂x.
4. Is it homogeneous? Check if M and N are both homogeneous of same degree.
5. Can you use a substitution to convert it to something you recognize?

Step 2: Apply the technique

Once you identify the type, apply the standard method. Keep your work organized. One mistake in algebra and you'll get a wrong answer that looks plausible.

Step 3: Check your work

Differentiate your solution. Plug it back into the original equation. If it doesn't satisfy the equation, you made a mistake somewhere.

Common Mistakes

• Forgetting the integrating factor: Students often try to integrate linear equations directly without multiplying by μ. That doesn't work.
• Wrong exactness test: ∂M/∂y must equal ∂N/∂x. Not the other way around.
• Substitution errors: When substituting y = vx, remember dy/dx = v + x(dv/dx). People forget the v term.
• Integration mistakes: The hard part is often just the integration itself, not the differential equation theory.

When Nothing Works

Some differential equations have no closed-form solution. That's not a reflection of your abilities—that's just reality. In those cases, you use:

• Numerical methods (Euler's method, Runge-Kutta)
• Series solutions
• Qualitative analysis
• Computer algebra systems

Knowing when to stop chasing an analytical solution is important. Some problems are meant to be solved numerically.

Bottom Line

Non-separable differential equations require you to recognize patterns and apply the right technique. Linear equations use integrating factors. Exact equations use potential functions. Homogeneous equations use y = vx. Bernoulli uses a power substitution.

Study the standard forms. Practice identification. The patterns become obvious after enough problems.