Let’s introduce baby integration as reverse-differentiation, which, comically, I have criticised elsewhere on my blog.
Given functions and
, write
We call the left-hand equation the indefinite integral of with respect to
.
(Wrong) Example 1. Recall that . Therefore,
Now strictly speaking, our integrals are wrong.
Lemma 1. Suppose and
are functions such that
Then, for any real number ,
Proof. Using the linearity of differentiation,
Therefore, we use the following definition for indefinite integrals.
Definition 1. Given functions and
, write
where is an arbitrary constant of integration.
Example 1. Recall that . Therefore, for
,
Now, Definition 1 is far from complete, but works well enough for most secondary-school assessments.
Theorem 1. Indefinite integration is linear: given functions and any real constant
,
Proof. Write
so that
By the linearity of differentiation,
For the second result, the linearity of differentiation yields
By Definition 1,
Example 2. For any , evaluate
. In particular, for any constant
, evaluate
Solution. From Example 1, if , then
Hence, writing , for
,
By the linearity of integration,
Therefore,
Since is still an arbitrary real constant,
The other results follow from linearity and the law of exponents:
Example 3. Evaluate .
Solution. Expanding yields
Since integration is linear,
Remark 1. Where possible, it is good practice to present your final answer as a sum of terms of the form , where
are constants with
being a positive integer and
is not a polynomial.
Example 3 should suggest that computing integrals exactly would require a lot more effort than computing derivatives exactly. That is correct. For instance, we don’t actually yet know how to calculate
since Example 2 only works for .
Eventually, we will see that
which should surprise us—how did the logarithm magically appear there? For another fun result,
What else could we do with integration?
Example 4. Let be real constants. Evaluate
. Hence, evaluate
Solution. Using the chain rule,
By Definition 1 and the linearity of integration,
where is an arbitrary real constant of integration.
Using the same idea in Example 4, we have the following special form of integration.
Theorem 2. If , then for real constants
,
Proof. Using the chain rule,
By Definition 1 and the linearity of integration,
where is an arbitrary real constant of integration.
Remark 1. Theorem 2 is a very special case of integrating by substitution, which is, in spirit, simply the reverse of the usual chain rule in differentiation.
Example 5. Differentiate . Hence, using Example 3, evaluate
Solution. Using the product rule and the chain rule,
Integrating on both sides and using linearity,
Doing algebra and using Example 3,
where is an arbitrary real constant of integration.
Remark 2. Example 5 is a guided example of integrating by parts, which is, in spirit, simply the reverse of the usual product rule in differentiation.
There is a lot more to say about integration, especially since we have not even discussed calculus involving the non-algebraic functions like the trigonometric functions ,
,
, as well as the exponential function
and its inverse the natural logarithm
.
Nevertheless, we will push these discussions to the final post on integration, and pursue the essentials of the topic first. These functions have massive uses in the STEM fields, but for now we want to cover the big-idea bases of calculus that do not necessarily require them.
As such, we will apply integration to computing areas, as well as discuss some basic physics.
—Joel Kindiak, 18 Jan 25, 1200H
Leave a comment