I"m in my first semester of calculus, so the difficulties I"m dealing with are about as tough as those on KhanAcademy calculus playlist. I"m at this time doing integration, a somewhat challenging part that the course. Doing derivatives is mechanics; recognize the integral is an art.

The two key techniques verified are

u-substitutionintegration by parts

My concern is: room there any kind of rules of thumb (preferably with a logical factor behind it) of when to use which?

Always carry out a $u$-sub if you can; if you cannot, consider integration by parts.

You are watching: Integral of 1/a^2+x^2

A $u$-sub have the right to be excellent whenever you have something comprise a role (we"ll contact this $g$), and also that miscellaneous is multiplied by the derivative the $g$. The is, if you have $\int f(g(x))g"(x)dx$, usage a u-sub.

Integration by components is at any time you have two functions multiplied together--one that you have the right to integrate, one that you deserve to differentiate.

My strategy is to try to "play that out" in my mind and try to view which one will work-related better. The best way to get much better at this sorts of integrals is to practice large sets of every type. Then, you start to think "Oh--this looks prefer a u-sub!" or, "maybe by-parts is much better for this." Practice is really the best means to get better at recognizing every type.

re-publishing
mention
follow
answered Oct 24 "13 in ~ 21:51

apnortonapnorton
$\endgroup$
1
include a comment |
7
$\begingroup$
U-substitution is for attributes that deserve to be composed as the product the another function and that derivative.$$\int u du$$

Integration by components is for features that deserve to be written as the product of another role and a 3rd function"s derivative.

$$\int u dv$$

A great rule of ignorance to follow would certainly be to shot u-substitution first, and also then if you cannot reformulate your duty into the exactly form, try integration by parts.

share
point out
follow
answered Oct 24 "13 at 21:52

$\endgroup$
1
$\begingroup$
If the integral is simple, you can make a basic tendency behavior: if you have composition the functions, u-substitution might be a good idea; if you have products of functions that you know just how to integrate, you can shot integration through parts.

But most challenging integrals have actually no instant ideas. Perhaps you have to use them both. Commonly I begin with substitution an approach so ns can get a fine know function and then usage integration by parts.

Sometimes nobody of these approaches will help you. More than likely you"ll require some algebra, simplification or wizardry through the integral before start make the efforts to combine it.

Some definite integrals have actually no way to solve other than integration by parts, by detect the same integral top top both sides of the equation. These cases are yes, really die-hard troubles if friend don"t go the way.

re-publishing
point out
follow
answered Oct 24 "13 at 21:52

Victor ChavesVictor Chaves
$\endgroup$
include a comment |
1
$\begingroup$
While various other answer provide the basic guideline it"s often not evident to see once to usage which.

First look in ~ a table that differentials to see if you deserve to find any type of pattern you deserve to work towards algebraically to uncover the anti-derivative. For example if your integrand has a square source in the denominator and a consistent numerator her anti-derivative might be among the inverse hyperbolic functions.

Consider integration by parts in these cases:

To eliminate logarithm, station trigonometric and inverse hyperbolic functions:

$$\int x^2 \neurosoup.orgrmln(x) dx = \fracx^33 \neurosoup.orgrmln(x) - \int \fracx^33x dx = \fracx^33 \neurosoup.orgrmln(x) - \fracx^39 + C$$Of food the other aspect should be easy to integrate.

To revolve exponentials, trigonometric and hyperbolic functions. The exponential function"s antiderivative is itself, sines and cosines switch, hyperbolic sines and cosines switch together well. This have the right to be offered to repeatedly identify the attached element while maintaining the exponential intact, hope something less complicated comes the end of it:$$\int f(x) e^x dx = f(x)e^x - \int f"(x) e^x dx$$This is especially helpful if your $f(x)$ is a polynomial.

Consider using substitution:

To eliminate linear subexpressions. You can constantly do this: $\int f(ax+b) dx = \frac1a\int f(u) du$, with $u = ax + b$.

If you view powers different by just 1, check if you can factor out the reduced power: $\int x^n-1 f(a x^n + b) dx = \frac1na \int f(u) du$ through $u = ax^n + b$. Climate you get rid of the power altogether.

See more: Coronavirus: ¿Cómo Puedo Hacer Ejercicio Si Tengo Covid, Cómo Puedes Volver A Hacer Ejercicio Después De

If you have actually $x^2$ only and square roots however no $x$, the $\sqrtax^2 + b$ is a candidate because that trigonmetric and also hyperbolic substution through $x = \sqrt\fracba \neurosoup.orgrmsinh\, \theta$ and also $dx = \sqrt\fracba \neurosoup.orgrmcosh\, \theta$. For this reason it i do not care $\sqrtb \neurosoup.orgrmsinh^2\, \theta + b = \sqrtb\sqrt\neurosoup.orgrmsinh^2\, \theta + 1 = \sqrtb\sqrt\neurosoup.orgrmcosh^2\, \theta = \sqrtb|\neurosoup.orgrmcosh\, \theta| = \sqrtb\neurosoup.orgrmcosh\, \theta$ .