4.10c - The integrating factor method
Some first order equations become integrable when you multiply by the right function. You will build that function from the product rule, recognise when the method applies, and use initial conditions without losing the interval on which your solution is valid.
Building an integrating factor
How could you integrate when neither nor is known? Multiplying by makes the left-hand side a familiar derivative:
The obstruction has become a product-rule calculation in reverse.
You already know
The product rule gives . A general solution needs an arbitrary constant; an initial condition selects its value.
The method works for a linear first order equation: and occur only to the first power, and their coefficients depend on the independent variable alone. Thus is linear, whereas is not. The equation is linear too: a zero right-hand side causes no difficulty.
Integrating factor
An integrating factor for is a non-zero function which makes the multiplied left-hand side the derivative of .
To find it, compare the product rule with the expression you want:
Choose an interval on which and are continuous. The exponential factor is positive there. Its integral does not need an added constant: that would merely multiply the factor by a non-zero constant, and multiplying the entire equation by an extra constant changes none of its solutions.
Integrating factor method
For , with continuous and on the interval considered, take
Then
Any non-zero constant multiple of also works.
Not in the formula booklet: learn it.
The surprising part is that the integrating factor depends only on , not on . Once the coefficient of is fixed, the same factor handles constant, polynomial, exponential or trigonometric right-hand sides, provided you can do the resulting integral.
Key idea
The integrating factor is chosen to complete a product derivative, not to cancel the term.
Check whether the product rule really matches every term before integrating.
Constant coefficient linear equations
Write the equation with coefficient one on , identify , and multiply every term by . Only then replace the left-hand side by its product derivative.
Worked example 1: a constant forcing term
Question. Find the general solution of .
Plan. Use , multiply by , integrate and divide by the factor.
Why this step? Integration finds , so dividing by the factor is essential to obtain .
Answer. , for real .
Check. .
Change only the right-hand side from to . The integrating factor stays the same, but the integral now needs integration by parts.
Worked example 2: a linear forcing term
Question. Find the general solution of .
Plan. Use the same factor and integrate by parts.
Why this step? Choose the polynomial to differentiate in integration by parts, because its degree drops.
Answer. , for real .
Check. , so .
The factor has done its job before integration begins; do not change it to suit the right-hand side.
Watch out
Wrong: . This omits the extra product-rule term. Right: combine into first.
A new coefficient on the forcing term changes the integral, not the factor.
Normalising variable coefficients
The coefficient used in the integrating factor is the coefficient of after dividing by the coefficient of . For , it is , not .
Continue this lesson.
You’re previewing NovaLearn without an account. Create a free account to read the rest.
Create free account