use Fixed Point iteration, to solve the nonlinear equations. u(x,y)=x^2+xy-10=0 v(x,y)=3xy^2+y-57=0 use initial guess x=1, y=2
we wont use jacobian matrix here ha? that was for newton..
i have no idea how to solve this . i can solve in single variable using Fixed point iteration.
\(x^2+xy-10=0\) isolate x --> \(x=\sqrt{10-xy}\) \(3xy^2+y-57=0\) isolate y --> \(y=\sqrt{\frac{57-y}{3x}}\) so forming fixed iteration equations like this \[x_{n+1}=\sqrt{10-x_ny_n}\]\[y_{n+1}=\sqrt{\frac{57-y_n}{3x_n}}\]solve and tell me if it converges or not?
\(x_0=1\) , \(y_0=2\)
oops its not working... @amistre64 will help us
are those 2 seperate equations to work out? or is this a system of equations?
if i recall correctly, we determine a sequence generated by a function such that:\[x_{n}=f(x_{n-1})\] if the sequence converges, we assume that to be our answer
its system.
yes .
mukushla seems to have worked up some functions to play with, my question is, do we have to form these into sqrts ? y = (10-x^2)/x x = (57-y)/(2y^2)
not necessarily sqrts...that one will work i think.
for Fixed point iteration xn=f(xn-1) it should be in such form that if derivative is taken then it comes out Less than 1for all values.
@mukushla i think you got it right i am getting convergence after 7th iteration x is approaching 2 and y is approaching 3.0459
but it turns to negatives... under radical !!
my book agrees with the answers i got above !. thanks for your help :)
my idea doesnt seem to converge :)
lol...im still wondering how she got it.. http://www.wolframalpha.com/input/?i=x%5E2%2Bxy-10%3D0+and+3xy%5E2%2By-57%3D0+
X0=1 Y0=2 X1=2.828 Y1=2.546 X2=1.673 Y2=3.293 X3=2.119 Y3=2.9066 X4=1.9598 Y4=3.033
\[x_1=\sqrt{10-x_0y_0}=\sqrt{10-2}=\sqrt{8}\]
if you want i can do more iteration.?
my y1 is 4.28
if i code in mukus equations i get
oh oh ... i think i got it ... u put x1 for calculating y1 instead of x0
|dw:1345855408848:dw|
Join our real-time social learning platform and learn together with your friends!