Mindscrape
- 1,854
- 1
I am linearizing the function f(x,y,z) = tan^{-1}(xyz) at the point (1,1,1).
Since f(x_0,y_0,z_0)= \frac{\pi}{4} + \pi*n should I just take the first value or do I have to carry all the solutions through the linearization process?
Um, anybody remember this? I can put up some work if it helps.
L(x,y,z) = \Delta f = f_x \Delta x + f_y \Delta y + f_z \Delta z
So
L(x,y,z) = f(x_0,y_0,z_0) + f_x(x-x_0) + f_y (y-y_0) + f_z (z-z_0)
where f_x = \frac{\partial f}{\partial x} (and f_y and f_z)
So does L(x,y,z) = \frac{\pi}{4} +\pi n + \frac{1}{2}[(x-1)+(y-1)+(z-1)]
?
Since f(x_0,y_0,z_0)= \frac{\pi}{4} + \pi*n should I just take the first value or do I have to carry all the solutions through the linearization process?
Um, anybody remember this? I can put up some work if it helps.
L(x,y,z) = \Delta f = f_x \Delta x + f_y \Delta y + f_z \Delta z
So
L(x,y,z) = f(x_0,y_0,z_0) + f_x(x-x_0) + f_y (y-y_0) + f_z (z-z_0)
where f_x = \frac{\partial f}{\partial x} (and f_y and f_z)
So does L(x,y,z) = \frac{\pi}{4} +\pi n + \frac{1}{2}[(x-1)+(y-1)+(z-1)]
?
Last edited: