방향도함수와 기울기벡터

(a) We are told that $T(x,y,z)=\frac{k}{|\left< x,y,z\right>|} =\frac{k}{\sqrt{x^2+y^2+z^2}}$ for some constant $k$ and that

$$120=T(1,2,2)=\frac{k}{|\left< 1,2,2\right>|} \implies k=120\times\sqrt{1+2^2+2^2} = 360$$

(b) The (unit) direction from $(1,2,2)$ to $(2,1,3)$ is $\mathbf{d}=\frac{\left< 2,1,3\right>-\left< 1,2,2\right>}{|\left< 2,1,3\right>-\left< 1,2,2\right>|} =\frac{\left< 1,-1,1\right>}{|\left< 1,-1,1\right>|}=\frac{1}{\sqrt{3}}\left< 1,-1,1\right>$. The desired rate of change of temperature is

$$\begin{aligned} D_{\mathbf{d}} T(1,2,2) &={ \boldsymbol{\nabla} } T(1,2,2)\cdot\mathbf{d} =-360\frac{x\hat{\boldsymbol{\imath}}+y\hat{\boldsymbol{\jmath}}+z\hat{\mathbf{k}}}{{(x^2+y^2+z^2)}^{3/2}}\Big|_{\left< 1,2,2\right>}\cdot\mathbf{d} \\ &=-360\frac{\left< 1,2,2\right>}{27}\cdot\frac{\left< 1,-1,1\right>}{\sqrt{3}} =-\frac{40}{3\sqrt{3}}\approx-7.70 \end{aligned}$$

degrees per unit distance.

(c) At $(x,y,z)$, the direction of greatest increase is in the direction of the temperature gradient at $(x,y,z)$, which is ${ \boldsymbol{\nabla} } T(x,y,z) =-360\frac{x\hat{\boldsymbol{\imath}}+y\hat{\boldsymbol{\jmath}}+z\hat{\mathbf{k}}}{{(x^2+y^2+z^2)}^{3/2}}$ and which points opposite to the radius vector. That is, it points towards the origin. This argument only fails at $(x,y,z)=(0,0,0)$, where the gradient, and indeed $T(x,y,z)$, is not defined.