Bernoulli Differential Equations
Solve Bernoulli first-order differential equations by reducing them to a linear ODE via a substitution, with a worked example.
Continue where you stopped
The core idea
On this page
A Bernoulli equation is a first-order ODE of the form
where n ≠ 0,1.
This is one of the main “nonlinear, but still solvable” first-order types. In physics it can appear whenever the rate depends on a power of the state variable (a common modelling move for nonlinear loss, nonlinear absorption, saturating effects, etc.).
- Main hub: First Order Differential Equations
- Linear method used inside Bernoulli: First-order linear ODEs
- Method checklist: Steps for solving first-order ODEs
- Physics pages that use first-order ODE tools: UY1: RC Circuits, UY1: Resistive Forces
- Math hub: Mathematics for Undergraduate Physics
After a substitution, a Bernoulli equation becomes a standard first-order linear ODE.
Method (reduce to linear)
Start with
- Divide by yⁿ.
- Substitute u = y¹⁻ⁿ.
- Solve the resulting linear ODE in u.
- Substitute back to get y.
-
Divide by yⁿ:
y⁻ⁿy' + P(x) y¹⁻ⁿ = Q(x). -
Substitute
u = y¹⁻ⁿ.Then
u' = (1-n)y⁻ⁿy'. -
Multiply the divided equation by (1-n) to get a linear ODE in u:
u' + (1-n)P(x) u = (1-n)Q(x). -
Solve for u(x) using the integrating factor method, then substitute back y = u^(1/(1-n)).
If n = 0 or n = 1, the equation is already linear (not “Bernoulli” in the usual sense).
- The factor (1-n) matters: u' = (1-n)y⁻ⁿy'; dropping (1-n) will break the algebra.
- Don’t divide by yⁿ blindly: if y = 0 is a possible solution, handle it separately before dividing.
- Check the final answer: differentiate your y(x) and verify it satisfies the original ODE.
Worked example
Solve
This is Bernoulli with n = 2, P(x) = 1, Q(x) = x.
Divide by y²:
Let u = y¹⁻² = y⁻¹ = 1/y. Since u' = -y⁻²y', the equation becomes:
Solve the linear ODE u'-u = -x. The integrating factor is μ = e^(∫ (-1) dx) = e^(-x), so:
Integrate:
So
Worked example (physics-flavoured): linear loss + quadratic loss
In some physics contexts you meet “linear + quadratic loss” models. A common mathematical form is:
where I(x) is an intensity/flux-like quantity and α,β are constants. This is Bernoulli with n = 2.
Divide by I²:
Let u = I⁻¹ so u' = -I⁻²I'. Then:
Solve the linear ODE:
So:
If I(0) = I₀, then 1/I₀ = C-β/α, so C = 1/I₀ + β/α and
Practice (with hints + answers)
1) Identify n and solve: y' + 2y = 2y^2
Hint: This is Bernoulli with n = 2. Use u = y¹⁻ⁿ = 1/y.
Answer: Divide by y²:
Let u = 1/y so u' = -y⁻²y', giving
Solve: u = 1 + Ce^2x. Therefore
2) Classify (no solving needed): y' + (1/x) y = x^2 y^3
Hint: Compare with y' + P(x)y = Q(x)yⁿ.
Answer: Bernoulli with n = 3, P(x) = 1/x, Q(x) = x².
3) Show the first move: y' - y = e^x y^4
Hint: Divide by y⁴ and set u = y¹⁻⁴ = y⁻³.
Answer:
then with u = y⁻³ you get a linear ODE in u.