∫ ( -e^(1/t) / t² ) dt
This is actually really quite easy. Keeping the reverse chain rule in mind, if you have the following, integrals end up very easy:
∫ [ f'(g(t)) · g'(t) ] dt = f(g(t)) + C
If you go ahead and define the following functions, you have exactly that form:
f(t) = e^(t)
g(t) = 1/t
Therefore, f'(t) = e^t, and g'(t) = -1/t², so we have exactly the form above. So, your answer is simply:
∫ ( -e^(1/t) / t² ) dt = e^(1/t) + C.
2007-08-07 11:31:17
·
answer #1
·
answered by C-Wryte 4
·
0⤊
0⤋
one technique is to look at the terms you're trying to integrate, and make a substitution
u=f(t) such that du = f'(t) dt where f'(t) cancels off terms in the original integral.
In your case, if you use u = 1/t,
du/dt will have a 1/t^2 term which will take out your t^2 in the denominator in the original integral. Try this, and after adjusting for the constant, see if you can solve it.
2007-08-07 18:33:34
·
answer #2
·
answered by astatine 5
·
0⤊
0⤋
Use substitution of variables:
u= 1/t ==> du = -1/(t^2)
then int((-e^(1/t))/(t^2)dt = int(e^udu) = e^u + C
Replacing back the variable: = e^(1/t) + C
2007-08-07 18:35:06
·
answer #3
·
answered by vahucel 6
·
0⤊
0⤋
Doesn't look all that hard. Let 1/t= u and
du= -1/t^2 * dt. Doesn't that sort of look like a
e^u * du form?
2007-08-07 18:34:21
·
answer #4
·
answered by cattbarf 7
·
0⤊
0⤋
integ - e^(1/t) / t^2
integ - e^(1/t) (-t^(t^(-2))
= e ^ (1/t + 1) + c
2007-08-07 18:34:00
·
answer #5
·
answered by CPUcate 6
·
0⤊
0⤋
way over my head buddy
2007-08-07 18:34:37
·
answer #6
·
answered by Thatdudenextdoor 3
·
0⤊
0⤋