This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
@Steven Chase is it feasible to do the above problem using pen and paper only?
Log in to reply
It looks like it is. See Hosam's solution
Log in to reply
@Steven Chase Did you understand his whole solution completely?? I am not able to understand his solution.
Log in to reply
@Talulah Riley – No, that's not really my style either
Log in to reply
@Steven Chase – @Steven Chase So do you want to understand???
Log in to reply
@Talulah Riley – Not really. As long as I have a way of solving, I don't need to know every possible way
Log in to reply
@Steven Chase
–
@Steven Chase
Suppose just imagine if python programming is banned in USA, then what will you do?
You should have some alternative approach. Isn't it??
Log in to reply
@Talulah Riley – I'll switch to C++
Log in to reply
@Steven Chase – @Steven Chase if C++ and every type of this computational approach is banned then what??
Log in to reply
@Talulah Riley – Then the situation is so messed up that nobody will bother solving these types of problems in their spare time
Log in to reply
@Steven Chase
–
@Steven Chase
Then according to that if you don't mind this problem should posted in computer science section, because this problem only needs Computational approach.
The calculus section should full of only those problem that are good and trick and not should contain very very difficult maths which needs to be solved with computer only.
Isn't it??
Log in to reply
@Talulah Riley – People can solve however they want. I don't care what method is used. And this is absolutely not a computer science problem.
Log in to reply
@Steven Chase – @Steven Chase if this is absolutely not a computer science then why you are using computer ??
Log in to reply
@Talulah Riley – Because computer science has nothing to do with using computers
Log in to reply
@Steven Chase – @Steven Chase Then why “computer science”word contains “computer” ??
Log in to reply
@Talulah Riley – Computer science is really the study of algorithms and such. The properties of the algorithms are independent of the machines running them. Computer science problems typically deal with algorithms, data structures, sorting techniques, etc. Numerical calculus is not typically considered a computer science topic.
Log in to reply
@Steven Chase – @Steven Chase Then while doing Numerical integration why you use computer??
Log in to reply
@Talulah Riley – There is no other practical way to do numerical calculus.
Log in to reply
@Steven Chase – @Steven Chase Now finally I bring you to the end of the loop, that is why I am saying this should be posted in computer science section.
Log in to reply
@Talulah Riley – That is not even remotely the case
Log in to reply
@Steven Chase
–
@Steven Chase
what case?
I didn't understand
Log in to reply
@Talulah Riley – This is either a geometry problem, or a calculus problem depending on how you choose to solve it
Log in to reply
@Steven Chase
–
@Steven Chase
ok I agree
Then you should use geometry or calculus, but by neglecting both the things you are using computer why?
@Talulah Riley – I admire his solution very much; he did it without using any calculus. Basically, he found the rotation and translation of the equation, and found the original ellipse. It's all super algebraic and feels a bit arcane lol
That previous comment chain is so long that it is barely readable. I have not neglected calculus. If you are summing infinitesimal quantities to derive a result, you are doing integral calculus. Whether you do the summing by hand or using a computer, is just an implementation detail.
Log in to reply
@Steven Chase
Then why my mathematics teacher doesn't allow me to do integration using python.
When I showed him a python code of a good integration instead of complementing , he scolded me, he said me that you have to do with pen and paper only .
Whether you do the summing by hand or using a computer, is just an implementation detail.
Log in to reply
Apparently, he wants you to become proficient at doing integrals by hand, using anti-differentiation. That does not mean, however, that numerical calculus is not calculus.
In general, there are two main ways of evaluating an integral, both of which are calculus.
1) Compute an anti-derivative. This is typically what hand analysis consists of
2) Compute a Riemann sum. This is typically what numerical calculus does
Log in to reply
@Steven Chase
–
@Steven Chase
But why he wants me to become proficient, if we have another simple way.
All should have to use computers simply.
Log in to reply
@Talulah Riley – It's nice to know about anti-differentiation. It's a useful skill. But I agree with you that there is too much emphasis on it in calculus instruction. Numerical techniques should be taught at least as much as anti-differentiation, in my opinion.
Log in to reply
@Steven Chase
–
@Steven Chase
Till now I have not used that
useful skill
.
And if we go in higher levels like making rockets, a very big projects , the integration used in that purpose can't be done with hand.
So at the end we have to use python/Matlab.
So what is the use of that skill?
Log in to reply
@Talulah Riley – I would argue that anti-differentiation is still important to know, at least at a basic level. But as you say, in many cases, numerical integration is required. I think the key is just to use the right tool for the job. And there are certain cases in which numerical methods have problems, and hand analysis does better. And if I run into a situation where I need to solve a problem and I can't use numerical methods, I will try some other way.
Log in to reply
@Steven Chase
–
@Steven Chase
in which case you are not able to use numerical methods,
If you are not able to do, it doesn't mean that, it can't be done with numerical techniques.
Maybe you don't know that type of numerical technique.
Please don't mind.
Log in to reply
@Talulah Riley – For example, if the integrand takes on infinite values, or if the integration range is infinitely big (from zero to infinity, for example), numerical integration can potentially have trouble.
Log in to reply
@Steven Chase – @Steven Chase if the value of y is increasing if x is increasing then it is obviously that answer will be infinite and it is not a problem of numerical integration. But the y should decrease as x increase, then by putting x as 1000and then 1500 and then 2000,we can definitely reach to the answer.
Log in to reply
@Talulah Riley – Yeah, I think the infinities in the integrand is a bigger problem. For example, this problem.
https://brilliant.org/problems/integrate-cotx-sin4x-from-0-to-pi/
At x = 0 , the integrand is undefined 0 0 , and yet the integral still presumably converges to some finite value. For that type of problem, hand analysis is better than numerical integration.
Log in to reply
@Steven Chase
–
@Steven Chase
I think you should integrate from x=0.01 to x=π.
When you put x=0 to x=π . The computer doesn't able to swallow that information because at x=0 the graph is approaching infinity.
Log in to reply
@Talulah Riley – And then you would have to try again starting at x = 0 . 0 0 1 , etc. and see if the answer is the same (approximately).
@Steven Chase
–
@Steven Chase
But you don't need to worry sir .
When I will go to university, I will make such programming software which can integrate functions from x=0.
Log in to reply
@Talulah Riley – I look forward to it :)
Log in to reply
@Steven Chase – @Steven Chase what will you look forward? :)
Log in to reply
@Talulah Riley – To your revolutionary new software program. Maybe I can get a free license
Log in to reply
@Steven Chase – @Steven Chase Why not, you will be the first user of that software. And you will get free license lifetime. Just chill sir.
Log in to reply
@Talulah Riley – Bro can I get a free license?
Log in to reply
@Krishna Karthik
–
@Krishna Karthik
no bro sorry.
Only
@Steven Chase
sir
@Talulah Riley – Actually, artificial intelligence will easily be able to solve any type of integral.
21st century will bring about great computerised advancement.
@Steven Chase – Numerical techniques are more useful, if anything, than analytical calculus. Most integrals/differential equations found in real life will be barely solvable using analytical approaches (Navier-Stokes equation is an example). I agree.
@Talulah Riley – Also a historical note. For the great majority of history, we haven't had powerful digital computers. Therefore, all of the standard curriculum has been focused on anti-differentiation by hand, since the direct computation of Riemann sums has been impossible until lately. So the old way of teaching still has a huge amount of momentum.
Log in to reply
@Steven Chase – @Steven Chase So it is job of education system to destroy that momentum, because that techniques take very time instead of that use python and enjoy .
I personally think your maths teacher is living in the 19th century. That's a very old fashioned way of thinking. Physics and maths in the 21st century can't be done using only hand. It's stupid.
Listen up: if you wanna be a good physicist, you will have to learn numerical techniques. Tell him that no one is using blackboard and chalk anymore.
By the way, in contrast with living in India vs living in Australia, I have found Indian teachers to be much more ancient and old-fashioned. Just because your teacher isn't keeping up with the rest of the world, don't get disheartened. The Indian system limits imagination.
One way to solve this numerically is to sweep x over a range. For a particular value of x , there is a quadratic equation to solve for y . Call the two solutions y p and y n , where y p ≥ y n . The infinitesimal area is d A = ( y p − y n ) d x . This method is applicable to any ellipse; even ones that do not contain the origin.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 |
|
@Steven Chase Thanks , love you sir.
Log in to reply
You're welcome. One thing to note is that the simple polar way only works if the ellipse contains the origin
We will make use of the following three facts :
Any transformation of axes system can be represented as the combination of translation and rotation .
It is immaterial which operation is executed first, translation or rotation. The final result is the same .
Transformation of axes system in a direction is equivalent to the same transformation of a figure in the opposite direction
Shifting the origin of coordinates from ( 0 , 0 ) to ( 1 1 1 4 , − 1 1 4 3 )
and then rotating the axes system counterclockwise through an angle of tan − 1 ( 3 1 ) , the given equation of the curve is reduced to
1 2 1 2 5 2 X 2 + 1 1 2 5 2 Y 2 = 1
So, length of the major semi-axis is
1 2 1 2 5 2
and the length of the minor semi-axis is
1 1 2 5 2
Therefore the area enclosed by the curve is
π × 1 2 1 2 5 2 × 1 1 2 5 2
= 1 1 1 1 2 5 2 π ≈ 2 1 . 7 .
Details
First give the translation of the axes system according to the law
x = x ′ + h , y = y ′ + k such that the linear terms in the given equation vanish. This will lead to h = 1 1 1 4 , k = − 1 1 4 3 .
Then give the axes system a counterclockwise rotation through an angle α to obtain a new system X , Y according to the law :
x ′ = X cos α − Y sin α
y ′ = X sin α + Y cos α
such that the x ′ y ′ terms vanish. This will lead to α = tan − 1 ( 3 1 ) .
Finally simplify the equation to obtain the required form.
The same thing can be done using the translation and the rotation matrices, which is elaborated by Hosam Hajjir :
X = T R x = R T x , where X and x are the transformed and the original coordinate matrices, T and R are the translation and the rotation matrices respectively.
Log in to reply
@Foolish Learner Thanks I understood this clearly. But after that shifting, what did you do?
Log in to reply
It's told in the details. Read carefully.
Log in to reply
@A Former Brilliant Member – @Foolish Learner I am not able to understand.
@Foolish Learner after shifting the origin to 1 1 1 4 , 1 1 − 4 3 , what did you do??
Let r = [ x y ] T be a column vector representing the coordinates x and y , then the given equation can be written as
r T A r + b T r + c = 0
where,
A = [ 5 2 3 2 3 1 ]
b T = [ − 1 4 ]
c = − 3
The first step is to eliminate the linear term ( b T r ), and to do that, we define
r 0 = − 2 1 A − 1 b
Then, one can see that, similar to completing the square, we can write,
r T A r + b T r = ( r − r 0 ) T A ( r − r 0 ) − r 0 T A r 0
To see why this is so, note that
( r − r 0 ) T A ( r − r 0 ) − r 0 T A r 0 = r T A r − 2 r T A r 0 = r T A r + r T b = r T A r + b T r
So now the given equation can be written as,
( r − r 0 ) T A ( r − r 0 ) = − c + r 0 T A r 0 = d
After calculating r 0 , the right hand side of the above equation comes to d = 1 1 1 1 5 = 1 1 1 2 6
To put the above in the standard format, we divide both sides by d , and this gives,
( r − r 0 ) T ( d 1 A ) ( r − r 0 ) = 1
One can check that matrix A is positive definite (i.e. both its eigenvalues are positive), by noting that A 1 1 > 0 and ∣ A ∣ > 0 , and therefore, the above equation represents an ellipse. We need to find the product of the lengths of the semi-axes, and this is equal to the square root of the reciprocal of the determinant of d 1 A . In other words,
a b = ∣ d 1 A ∣ 1
Since A is 2 × 2 , then ∣ d 1 A ∣ = d 2 1 ∣ A ∣
Hence,
a b = ∣ A ∣ d = 5 − 4 9 ( 1 1 1 2 6 ) = 4 1 1 ( 1 1 1 2 6 ) = 1 1 1 1 2 5 2
And the area is A = π a b = 1 1 1 1 2 5 2 π ≈ 2 1 . 7
@Hosam Hajjir
Hello sir
What is the meaning of
T
in your solution?
Log in to reply
T means 'Transpose', so if r is a column vector , then r T is a row vector with the same entries.
Yes; A T refers to the transpose of a matrix. What this means, basically, is that the matrix is flipped over on its side. An alternative way of notating this is A ′ .
If you transposed this vector: v = < x , y , z > , you would basically make it a column vector like so:
v = ⎣ ⎡ x y z ⎦ ⎤
Same with a matrix. Think of transposing a matrix or a vector as flipping it onto it's left side, or turning it 90 degrees anti-clockwise.
Transpose, which for a unitary orthogonal matrix is also the inverse of the matrix and much easier to compute
From Wolfram Mathworld's Ellipse article :
a = 5 ; b = 2 3 ; c = 1 ; d = − 2 1 ; e = 2 ; f = − 3 ;
a x 2 + 2 b x y + c y 2 + 2 d x + 2 e y + f = 0 → 5 x 2 + 3 x y − x + y 2 + 4 y − 3 = 0
Δ = ∣ ∣ ∣ ∣ ∣ ∣ a b d b c e d e f ∣ ∣ ∣ ∣ ∣ ∣ → − 2 6 3
Ξ = ∣ ∣ ∣ ∣ a b b c ∣ ∣ ∣ ∣ → 4 1 1
I = a + c → 6
I Δ → − ( 2 1 / 4 )
Since Δ = 0 , Ξ > 0 , and I Δ < 0 , assuming the conic section is not degenerate (which if it were, then the problem would be nonsense), then it is an ellipse.
δ = b 2 − a c → − 4 1 1
The center of the ellipse is at ( δ ∣ ∣ ∣ c b e d ∣ ∣ ∣ → 1 1 1 4 , δ ∣ ∣ ∣ a d b e ∣ ∣ ∣ → − 1 1 4 3 )
Translating the ellipse so that its center is at the origin, which does not change its area. This is done by substituting new values for the variables and simplifying.
5 x 1 2 + 3 x 1 y 1 + y 1 2 − 1 1 1 2 6 = 0
Note that linear terms are now gone.
The next problem is rotating the ellipse onto the axes, which also does not change the area.
The rotation angle is 2 1 tan − 1 ( a − c 2 b ) → ( cos ( 2 1 tan − 1 ( 4 3 ) ) sin ( 2 1 tan − 1 ( 4 3 ) ) − sin ( 2 1 tan − 1 ( 4 3 ) ) cos ( 2 1 tan − 1 ( 4 3 ) ) ) , in which you can see the Pythagoras 3-4-5 triangle, which permits that rotation matrix to be simplified in turn to ( 1 0 3 1 0 1 − 1 0 1 1 0 3 ) .
After substituting the rotated coordinates and simplifying: 1 1 1 ( 1 1 ( 1 1 x 2 2 + y 2 2 ) = 2 5 2 )
By substituting zero for x2 and y2 in turn, the semi-radii are 1 1 6 7 and 6 1 1 7 , respectively.
Therefore, the area is π 6 1 1 7 6 1 1 7 → 1 1 1 1 2 5 2 π ≈ 2 1 . 7 0 0 1
Problem Loading...
Note Loading...
Set Loading...
For an ellipse or hyperbola of the form A x 2 + B x y + C y 2 + D x + E y + F = 0 , it can be transformed into the following standard form ( reference ):
− S / ( λ 1 2 λ 2 ) x ′ 2 + − S / ( λ 1 λ 2 2 ) y ′ 2 = 1
where S = det ⎣ ⎡ A B / 2 D / 2 B / 2 C E / 2 D / 2 E / 2 F ⎦ ⎤ ; and λ 1 and λ 2 are eigenvalues of [ A B / 2 B / 2 C ] or the roots of λ 2 − ( A + C ) λ + ( A C − ( B / 2 ) 2 ) = 0 . Putting in the values, we have
S = ∣ ∣ ∣ ∣ ∣ ∣ 5 3 / 2 − 1 / 2 3 / 2 1 − 2 − 1 / 2 2 − 3 ∣ ∣ ∣ ∣ ∣ ∣ = − 2 6 3
λ 2 − 6 λ + 4 1 1 4 λ 2 − 2 4 λ + 1 1 ( 2 λ − 1 ) ( 2 λ − 1 1 ) ⟹ λ 1 , λ 2 = 0 = 0 = 0 = 2 1 , 2 1 1
Since area of an ellipse is given by π a b , where a and b are the semi-axes, then π a b = − S ( λ 1 λ 2 ) 2 3 ≈ 2 1 . 7 .