Can the derivative of the given integral be simplified to -A?

Click For Summary

Discussion Overview

The discussion revolves around the simplification of a derivative of a given integral involving trigonometric functions and logarithms. Participants explore the validity of a proposed trigonometric identity and the steps involved in differentiating the integral with respect to a variable \(A\). The scope includes mathematical reasoning and technical explanations related to calculus and trigonometric identities.

Discussion Character

  • Technical explanation
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • Some participants assert that the proposed trigonometric identity is invalid based on specific substitutions, leading to contradictory results.
  • Several participants discuss the application of Leibniz's rule for differentiation under the integral sign and present their derived expressions for the derivative of the integral.
  • One participant proposes a method to simplify the integral further using substitutions and half-angle formulas, leading to a more complex expression involving \(-A\cot(A/2)\).
  • Another participant claims to have reached a simplified form of the derivative as \(-A\) using the formula for the difference of arctangents, but expresses uncertainty about the correctness of their steps.
  • There are requests for clarity on attachments and equations, indicating a collaborative effort to understand and verify each other's work.

Areas of Agreement / Disagreement

Participants do not reach a consensus on the validity of the proposed identity or the correctness of the derived expressions. Multiple competing views and methods are presented, and the discussion remains unresolved regarding the final simplification of the derivative.

Contextual Notes

Some participants note that the expressions derived depend on specific substitutions and assumptions about the variable \(A\), and there are indications of potential algebraic complexities that may affect the final results.

Suvadip
Messages
68
Reaction score
0
Can it be proved?
$$\left(\frac{-2\sin A}{1-\cos A}\right)\cos\left(\frac{A}{2}\right)\tan^{-1}\left[\cos \left(\frac{A}{2}\right)\right]=\frac{\pi^2-4A^2}{8}$$
 
Last edited by a moderator:
Physics news on Phys.org
suvadip said:
Can it be proved?
$$\left(\frac{-2\sin A}{1-\cos A}\right)\cos\left(\frac{A}{2}\right)\tan^{-1}\left[\cos \left(\frac{A}{2}\right)\right]=\frac{\pi^2-4A^2}{8}$$

Hi suvadip, :)

This trigonometric identity is not valid. For example substituting \(A=\dfrac{\pi}{2}\) we get zero in the right hand side whereas \(-\dfrac{1}{\sqrt{2}}\tan^{-1}\dfrac{1}{\sqrt{2}}\neq 0\) on the right hand side.
 
Sudharaka said:
Hi suvadip, :)

This trigonometric identity is not valid. For example substituting \(A=\dfrac{\pi}{2}\) we get zero in the right hand side whereas \(-\dfrac{1}{\sqrt{2}}\tan^{-1}\dfrac{1}{\sqrt{2}}\neq 0\) on the right hand side.

Please look at the attached sheet. How to reach the final answer
 

Attachments

  • ArcSoft_Image2.jpg
    ArcSoft_Image2.jpg
    21.9 KB · Views: 138
suvadip said:
Please look at the attached sheet. How to reach the final answer

Your attachment is not very clear and I cannot read it. If you can post a more clear attachment or write down the equations using LaTeX I would perhaps be able to help. :)
 
Sudharaka said:
Your attachment is not very clear and I cannot read it. If you can post a more clear attachment or write down the equations using LaTeX I would perhaps be able to help. :)
I have to prove
$$\int_0^{\pi/2}\frac{log(1+cosA cosx)}{cosx}dx=\frac{\pi^2-4A^2}{8}$$

I have arrived at

$$\frac{d}{dA}(I)=-sinA \int_0^1\frac{2}{1+cosA +(1-cos A)z^2}dz$$ where I is the given integral. Am I correct so far and how to proceed at the answer from there? I have used the Leibnitz's rule for differentiation under the sign of integration.
 
suvadip said:
I have to prove
$$\int_0^{\pi/2}\frac{log(1+cosA cosx)}{cosx}dx=\frac{\pi^2-4A^2}{8}$$

I have arrived at

$$\frac{d}{dA}(I)=-sinA \int_0^1\frac{2}{1+cosA +(1-cos A)z^2}dz$$

How do you get this? After differentiating wrt A, I get,
$$\frac{dI}{dA}=\int_0^{\pi/2}\frac{-\sin A}{1+\cos A \cos x}dx$$
Can you integrate $\frac{dx}{1+a\cos x}$? Hint: Use $\cos x=\frac{1-\tan^2(x/2)}{1+\tan^2(x/2)}$.

EDIT: I see that you already tried that. Try this, it makes the algebra a bit easier to handle.
$$\frac{dI}{dA}=\int_0^{\pi/2}\frac{-\sin A}{1+\cos A \cos x}dx=\int_0^{\pi/2}\frac{-\sin A}{1+\cos A \sin x}dx$$
Use $\sin x=\frac{2\tan(x/2)}{1+\tan^2(x/2)}$ to get
$$\frac{dI}{dA}=-\sin A\int_0^{\pi/2} \frac{\sec^2(x/2)}{1+\tan^2(x/2)+2\cos A\tan(x/2)}dx$$
Use the substitution $\tan(x/2)=t$,
$$\Rightarrow \frac{dI}{dA}=-2\sin A\int_0^1 \frac{dt}{t^2+2t\cos A+1}=-2\sin A\int_0^1 \frac{dt}{(t+\cos A)^2+\sin^2A}$$
I suppose you can solve after this. :)
 
Last edited:
suvadip said:
I have to prove
$$\int_0^{\pi/2}\frac{\log(1+\cos A \cos x)}{\cos x}dx=\frac{\pi^2-4A^2}{8}$$

I have arrived at

$$\frac{d}{dA}(I)=-\sin A \int_0^1\frac{2}{1+\cos A +(1-\cos A)z^2}dz$$ where I is the given integral. Am I correct so far and how to proceed at the answer from there? I have used the Leibnitz's rule for differentiation under the sign of integration.
You are correct so far (having differentiated under the integral sign and then made the substitution $z = \tan(x/2)$). The next step is to write this as $$\frac{d}{dA}(I)=\frac{-2\sin A}{1-\cos A} \int_0^1\frac{1}{z^2 +\frac{1+\cos A}{1-\cos A}}dz.$$ Now use the fact that $$\frac{1+\cos A}{1-\cos A} = \cot^2(A/2)$$ to get $$\frac{d}{dA}(I)= \frac{-2\sin A}{1-\cos A} \int_0^1\frac{1}{z^2 +\cot^2(A/2)}dz = \frac{-2\sin A}{1-\cos A} \Bigl[\tan(A/2)\arctan\bigl(z\tan(A/2)\bigr)\Bigr]_0^1 = \frac{-2\sin A}{1-\cos A}\frac A2\tan(A/2)$$ (provided that $|A| < \pi$, so that $\arctan\bigl(\tan(A/2)\bigr) = A/2$). Using half-angle formulas again, you can write this as $$\frac{\frac{-2}{1+\tan^2(A/2)}}{1- \frac{1-\tan^2(A/2)}{1+\tan^2(A/2)}}A\tan(A/2) = \frac{-A\tan(A/2)}{\tan^2(A/2)} = -A\cot(A/2).$$ If that last expression was just $A$ (without the $\cot(A/2)$) then you could integrate it to get $-A^2/2$ which, together with the initial value $0$ when $A = \pi/2$, would give the formula that you are looking for.

But, unless I have made some silly mistake, that extra $\cot(A/2)$ is there, and that gives a function $-A\cot(A/2)$ that does not have an elementary integral.
 
Hi Opalg! :)

I am not sure but I seem to be getting the final answer. If the integral I reached is evaluated further, I get
$$\frac{dI}{dA}=2\left(\tan^{-1}\left(\frac{\cos A}{\sin A}\right)-\tan^{-1}\left(\frac{1+\cos A}{\sin A}\right)\right)$$
Using the formula for $\tan^{-1}a-\tan^{-1}b$, I simplified it to
$$\frac{dI}{dA}=-A$$
Please check if I did anything wrong, thank you. :)
 
Last edited:
Pranav said:
Hi Opalg! :)

I am not sure but I seem to be getting the final answer. If the integral I reached is evaluated further, I get
$$\frac{dI}{dA}=2\left(\tan^{-1}\left(\frac{\cos A}{\sin A}\right)-\tan^{-1}\left(\frac{1+\cos A}{\sin A}\right)\right)$$
Using the formula for $\tan^{-1}a-\tan^{-1}b$, I simplified it to
$$\frac{dI}{dA}=-A$$
Please check if I did anything wrong, thank you. :)
I'm very willing to believe that you're right, but I'll leave it to the OP to sort out the details. (Wink)
 

Similar threads

  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 12 ·
Replies
12
Views
3K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 16 ·
Replies
16
Views
2K