If the maximum value of
f ( x , y , z ) = 2 x 2 + 3 y − 4 z
subject to the constraints
{ 2 x − 4 y − 2 z = 5 x 2 + y 2 = 1
is equal to M , then evaluate ⌊ M ⌋ .
Note: It is advised to take the aid of a calculator for mathematical computations in this problem.
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.
Let's say we want to proceed from here
f ( θ ) = 2 cos 2 θ + 1 1 sin θ − 4 cos θ + 1 0
without plotting graph. How can we do that?
Log in to reply
I was thinking of solving for d θ d f = 0 . Use Newton's method (numerical) to find θ , then find f ( θ ) . It would be very long and it is still numerical.
I also proceed up to this but I got s degree 4 polynomial in x which is no solvable by normal method
So, do u knows any other ways ?
Log in to reply
Well this method of taking the polar form is very efficient, way better than what I has used (Lagrange multipliers) and by that method too I ended up in a biquadratic which I had to take aid of computer to solve. So, yes a graph plot is a good way.
Log in to reply
@Tapas Mazumdar – I doubt you can get a biquadratic expression. If what you said is true, then it's probably the most elegant solution here.
Can you share with us on how you get this result?
I used Lagrange Multipliers to get a system of nonlinear equations. Then used Multivariate Newton Raphson.
Log in to reply
@Steven Chase – Same, Langrange was giving me a quartic equation for the λ 2 term, that was better to solve.
@Steven Chase – I think you meant Hessian Matrix. It's impossible to prove the maximality via Lagrange Multipliers alone.
Log in to reply
@Pi Han Goh – I took it for granted that it would be a maximum, since the question said so :). The Lagrange Multipliers gives a nonlinear system. To solve the nonlinear system, you can use the matrix (multivariate) version of Newton Raphson. But yes, strictly speaking, even when you have a solution, you don't know whether it's a max, min, or inflection point.
The maximum value occurs when θ = 2 π k + 2 arctan ⎣ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎡ 1 1 4 + 2 1 1 2 1 6 4 + 3 3 2 / 3 2 3 − 1 4 4 + 6 4 6 5 9 − 3 3 ( 6 4 6 5 9 − 1 4 4 ) 2 × 1 1 2 / 3 + 2 1 1 2 1 1 2 8 − 3 3 2 / 3 2 3 − 1 4 4 + 6 4 6 5 9 + 3 3 ( 6 4 6 5 9 − 1 4 4 ) 2 × 1 1 2 / 3 + 1 3 3 1 1 2 1 6 4 + 3 3 2 / 3 2 3 − 1 4 4 + 6 4 6 5 9 − 3 3 ( 6 4 6 5 9 − 1 4 4 ) 2 × 1 1 2 / 3 1 0 2 4 ⎦ ⎥ ⎥ ⎥ ⎥ ⎥ ⎥ ⎤ for any integer k .
This value can be determined using half angle tangent substitution followed by the quintic formula .
Log in to reply
Woah! That's a really big expression! This is why I recommended using computational aid in this step.
Wow, you are real good.
Log in to reply
@Chew-Seong Cheong – Thanks for the compliment.... although this is just pure bashing without any real creativity.
Here's an alternative approach (without half angle substitution)
f ( θ ) can also be written as f ( θ ) = 1 2 − 2 sin 2 x + 1 1 sin x − 4 cos x .
Then f ′ ( θ ) = 1 1 cos x + 4 sin x − 4 sin x cos x . When at critical point, f ′ ( θ ) = 0 , so
4 sin x cos x − 4 sin x = 1 1 cos x 4 sin x ( 1 − cos x ) = 1 1 cos x 1 6 sin 2 x ( 1 − cos x ) 2 = 1 2 1 cos 2 x 1 6 ( 1 − cos 2 x ) ( 1 − cos x ) 2 = 1 2 1 cos 2 x
Then by the quintic formula, we can solve for cos x , which is an algebraic number. What's left is to prove that f ′ ′ ( θ ) < 0 and we're done.
To Tapaz: A good (albeit, super duper extremely tedious) follow up question to ask is "Find the exact closed form of the maximum value of f ( θ ) ".
Problem Loading...
Note Loading...
Set Loading...
Since x 2 + y 2 = 1 , we can substitute x = cos θ and y = sin θ , then from 2 x − 4 y − 2 z − 5 , we have 2 z = 2 cos θ − 4 sin θ − 5 and:
f ( x , y , z ) = 2 x 2 + 3 y − 4 z = 2 cos 2 θ + 3 sin θ − 2 ( 2 cos θ − 4 sin θ − 5 ) = 2 cos 2 θ + 1 1 sin θ − 4 cos θ + 1 0
By plotting we find that M ≈ 2 2 . 0 3 1 ⟹ ⌊ M ⌋ = 2 2 .