Jacobian Matrix set-up:

J=[โˆ‚Fโˆ‚xโˆ‚Fโˆ‚yโˆ‚Gโˆ‚xโˆ‚Gโˆ‚y]J=\begin{bmatrix} \dfrac{\partial F}{\partial x} & \dfrac{\partial F}{\partial y} \\ \dfrac{\partial G}{\partial x} & \dfrac{\partial G}{\partial y}\end{bmatrix} , where JJ is the coefficient of linearized equations. For this example let’s take the two equations F(x,y)=x+yโˆ’2y2=4F(x,y)=x+y-2y^2=4 and G(x,y)=x2+y2=8G(x,y)=x^2+y^2=8. (Use x0=1x_0=1 and y0=1y_0=1 for initial approximations.)

Let’s do the partial derivative calculate to plug into our JJ matrix.

  • โˆ‚F/โˆ‚x=1\partial F/\partial x=1
  • โˆ‚F/โˆ‚y=1โˆ’4y0\partial F/\partial y= 1-4y_0
  • โˆ‚G/โˆ‚x=2x0\partial G/\partial x= 2x_0
  • โˆ‚G/โˆ‚y=2y0\partial G/\partial y= 2y_0

Define our XX matrix as X=[dxdy]X= \begin{bmatrix} dx \\ dy \end{bmatrix}, and our KK matrix as K=[โˆ’4โˆ’(x0+y0โˆ’2y02)8โˆ’(x02+y02)]K= \begin{bmatrix} -4-(x_0+y_0-2y_0^2) \\ 8-(x_0^2+y_0^2) \end{bmatrix}. Remember that to solve for XX matrix, where X=Jโˆ’1KX=J^{\\-1}K. Plugging in x0x_0 and y0y_0 gives us the following JJ matrix: J=[1โˆ’322]J= \begin{bmatrix} 1 & -3 \\ 2 & 2 \end{bmatrix}, where Jโˆ’1=11โ‹…2โˆ’(โˆ’3)โ‹…2ร—[23โˆ’21]=โ…›ร—[23โˆ’21]{J}^{\\-1} = \frac{1}{1\cdot2-(-3)\cdot2} \times\begin{bmatrix} 2 & 3 \\ -2 & 1 \end{bmatrix} = โ…› \times \begin{bmatrix} 2 & 3 \\ -2 & 1 \end{bmatrix}. This would make Jโˆ’1=[ยผโ…œโˆ’ยผโ…›]J^{\\-1}=\begin{bmatrix} ยผ & โ…œ \\ -ยผ & โ…› \end{bmatrix} and K=[โˆ’46]K= \begin{bmatrix} -4\\ 6 \end{bmatrix}.

So this means that [ยผโ…œโˆ’ยผโ…›]ร—[โˆ’46]=X=[dxdy]=[1ยผ1ยพ]\begin{bmatrix} ยผ & โ…œ \\ -ยผ & โ…› \end{bmatrix}\times\begin{bmatrix} -4\\ 6 \end{bmatrix}=X= \begin{bmatrix} dx \\ dy \end{bmatrix}= \begin{bmatrix} 1ยผ \\ 1ยพ \end{bmatrix}


Leave a Reply

Your email address will not be published. Required fields are marked *