d2jsp
Log InRegister
d2jsp Forums > Off-Topic > General Chat > Homework Help > How To Solve These Equations Simutanelously?
Prev12345Next
Add Reply New Topic New Poll
Member
Posts: 16,662
Joined: Nov 24 2007
Gold: 15,245.00
Trader: Trusted
Sep 20 2015 09:45am
Good catch from Amaston. Except each of the 3 first equations is the equation of a plane in a 3d-space (and not a line).
The intersection of 3 planes is a point (except if 2 or more are parallel, which is not the case here).

Quote (FamilyGuyViewer @ Sep 20 2015 04:33pm)
(...)
did that, but it has to satisfy x^2 + y^2 + z^1 = 1
(...)


It can't, so you're stuck.

This post was edited by feanur on Sep 20 2015 09:45am
Member
Posts: 35,671
Joined: Aug 11 2008
Gold: 2,130.00
Sep 20 2015 09:46am
Quote (feanur @ Sep 20 2015 10:45am)
Good catch from Amaston. Except each of the 3 first equation is the equation of a plane in a 3d-space (and not a line).
The intersection of 3 planes is a point (except if 2 or more are parallel, which is not the case here).



It can't, so your stuck.


theres a solution that satisfies all 3. my professor alrdy posted the solutions and he is right as well

Quote (RzChaos @ Sep 19 2015 05:54pm)
I did it by solving for z in terms of x and y, and substituting that into the first two equations.

z = (1 - x^2 - y^2)

Solve for x and y, got x = -.09123 and y = -.87391

Sub those into the equation for z and got z = .47745

Member
Posts: 996
Joined: Oct 18 2012
Gold: Locked
Trader: Scammer
Sep 20 2015 09:47am
It can't satisfy all 4 equations because the inverse of A exists => unique solution => (0,0,0) is the *only* solution to the first 3 equations.

This post was edited by Amaston on Sep 20 2015 09:50am
Member
Posts: 996
Joined: Oct 18 2012
Gold: Locked
Trader: Scammer
Sep 20 2015 09:52am
Quote (feanur @ Sep 20 2015 10:45am)
Good catch from Amaston. Except each of the 3 first equations is the equation of a plane in a 3d-space (and not a line).
The intersection of 3 planes is a point (except if 2 or more are parallel, which is not the case here).


Haha, yup, replace everywhere I said line with plane :p
Member
Posts: 35,671
Joined: Aug 11 2008
Gold: 2,130.00
Sep 20 2015 09:53am
Quote (Amaston @ Sep 20 2015 10:47am)
It can't satisfy all 4 equations because the inverse of A exists => unique solution => (0,0,0) is the *only* solution to the first 3 equations.


ok and it doesnt satisfy the 4th equation. test out the numbers posted in the 1st page, and they check out
Member
Posts: 16,662
Joined: Nov 24 2007
Gold: 15,245.00
Trader: Trusted
Sep 20 2015 09:59am
Quote (FamilyGuyViewer @ Sep 20 2015 04:46pm)
theres a solution that satisfies all 3. my professor alrdy posted the solutions and he is right as well


RzChaos is not right at all.
He gave you an approximate "solution", that is a triplet (x; y; z) for whom the first 3 calculations give approximately 0 and the last gives approximately 1.

I already gave you another approximate "solution", that is the triplet (0; 0; 0), for which the first 3 calculations give exactly 0, and the last gives approximately 1 :
0² + 0² + 0² = 0
and 0 is approximately 1.

Notice that the huge difference of scale among the coefficients (300.21 , 114.21 , ... and 1) makes this thing possible.

It's like saying : the point at the intersection of those 3 planes almost lies in the sphere (it's really close to it !).
Member
Posts: 996
Joined: Oct 18 2012
Gold: Locked
Trader: Scammer
Sep 20 2015 09:59am
It's *almost* a solution (they don't check out by errors on the order of 10^-2).

There are two *almost* solutions if we look graphically:



This post was edited by Amaston on Sep 20 2015 10:09am
Member
Posts: 996
Joined: Oct 18 2012
Gold: Locked
Trader: Scammer
Sep 20 2015 10:00am
The planes nearly come together on the upper right and lower left -- it's easier to see in MATLAB because you can rotate it, but you can probably see the upper right "almost" solution
Member
Posts: 996
Joined: Oct 18 2012
Gold: Locked
Trader: Scammer
Sep 20 2015 10:03am
If you've got matlab, put this in and play with the figure:

% setup the circle data
phi = linspace(0,pi,100)
theta = linspace(0,2*pi,100)
[phi,theta] = meshgrid(phi,theta)
Xs = sin(phi).*cos(theta)
Ys=sin(phi).*sin(theta)
Zs = cos(phi)

% draw the planes
[X,Y] = meshgrid(-5:1:5)
Z1 = (300.21*X-45*Y)/25
Z2 = (-45*X+40.21*Y)/(-65)
Z3 = (-25*X+65*Y)/(-114.21)
surf(X,Y,Z1)
hold on
surf(X,Y,Z2)
surf(X,Y,Z3)
plot3(Xs,Ys,Zs)
axis([-2,2,-2,2,-2,2])

This post was edited by Amaston on Sep 20 2015 10:04am
Member
Posts: 35,671
Joined: Aug 11 2008
Gold: 2,130.00
Sep 20 2015 10:12am
yea they are approximate solution cause he rounded it lol

but how can i solve this on matlab? can someone tell me the code for it
Go Back To Homework Help Topic List
Prev12345Next
Add Reply New Topic New Poll