๋ฒกํ„ฐ์žฅ์„ ํ๋ฅด๋Š” ์œ ์ฒด๋กœ ๋ณด๊ณ , ํ•œ ์ ์— ๋Œ€ํ•ด ์œ ์ฒด๊ฐ€ ํŒฝ์ฐฝ or ์••์ถ• ํ•˜๋Š”์ง€, ์•„๋‹ˆ๋ฉด ํ•œ ์ ์„ ์ฃผ๋ณ€์œผ๋กœ ์œ ์ฒด๊ฐ€ ์–ผ๋งˆ๋‚˜ ๋น ๋ฅด๊ฒŒ ํšŒ์ „ํ•˜๋Š”์ง€๋ฅผ ์ •์˜ํ•œ ๋‘ ์ง€ํ‘œ. ์˜์—ญ ๋‚ด๋ถ€์˜ ํšŒ์ „(circulation)๊ณผ ์œ ์ถœ(flux)์„ ๊ทธ๋ฆฐ ์ •๋ฆฌ์˜ ์„ ์ ๋ถ„์œผ๋กœ ๊ณ„์‚ฐํ•˜๋Š” ๋ฐฉ๋ฒ•์— ๋Œ€ํ•ด์„œ ๐ŸŒŠ

20 minute read

๋ณต์ˆ˜์ „๊ณตํ•˜๊ณ  ์žˆ๋Š” ์ˆ˜ํ•™๊ณผ์˜ ์กธ์—…์‹œํ—˜์„ ์œ„ํ•ด ํ•™๋ถ€ ์ˆ˜ํ•™ ๊ณผ๋ชฉ๋“ค์„ ๋‹ค์‹œ ๊ณต๋ถ€ํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ๋ฏธ์ ๋ถ„ํ•™ ํฌ์ŠคํŠธ ์ „์ฒด ๋ณด๊ธฐ

์–ด์ฉŒ๋ฉด ์ง€๊ธˆ ์ด ์ƒํƒœ์ผ์ง€๋„โ€ฆ

๋“œ๋””์–ด!! ๋ฏธ์ 2์˜ ๊ฝƒ์ด๋ผ๊ณ  ํ•  ์ˆ˜ ์žˆ๋Š” ๋ฐœ์‚ฐ(Divergence)๊ณผ ํšŒ์ „(Curl)์ด๋‹ค!! ๋‘ ๊ฐœ๋… ๋ชจ๋‘ ๋ฒกํ„ฐ์žฅ์„ ์œ ์ฒด์˜ ํ๋ฆ„์ด๋ผ๊ณ  ๋ณด๊ณ , ๊ทธ ์œ ์ฒด๊ฐ€ ํ•œ ์ ์— ๋Œ€ํ•ด ํŒฝ์ฐฝ/์••์ถ• ํ•˜๊ณ  ์žˆ๋Š”์ง€, ์•„๋‹ˆ๋ฉด ํ•œ ์ ์—์„œ ์œ ์ฒด๊ฐ€ ์–ผ๋งˆ๋‚˜ ๋น ๋ฅด๊ฒŒ ํšŒ์ „ํ•˜๊ณ  ์žˆ๋Š”์ง€๋ฅผ ์ธก์ •ํ•˜๋Š” ์ง€ํ‘œ์ด๋‹ค. ๐ŸŒŠ

Divergence

If $\mathbf{F} = P \, \mathbf{i} + Q \, \mathbf{j} + R \, \mathbf{k}$ is a vector field on $\mathbb{R}^3$,

and the below partial derivatives exist

\[\frac{\partial P}{\partial x} \text{ , } \frac{\partial Q}{\partial y} \text{ , } \frac{\partial R}{\partial z}\]

then the โ€œdivergence of $\mathbf{F}$โ€ is a scalar value defined by

\[\text{div } \mathbf{F} = \frac{\partial P}{\partial x} + \frac{\partial Q}{\partial y} + \frac{\partial R}{\partial z}\]

๋ฐœ์‚ฐ(divergence)๋Š” ๋ฒกํ„ฐ์žฅ์— ๋Œ€ํ•ด ์ •์˜ํ•˜๋Š” ์Šค์นผ๋ผ ํ•จ์ˆ˜์ด๋‹ค. Gradient Vector $\nabla$์™€์˜ ๋‚ด์ ์œผ๋กœ ์‹์„ ๋” ์‰ฝ๊ฒŒ ํ‘œํ˜„ํ•  ์ˆ˜ ์žˆ๋‹ค.

\[\text{div } \mathbf{F} = \nabla \cdot \mathbf{F}\]

Geometric meaning

๋ฒกํ„ฐ์žฅ์„ ์œ ์ฒด์˜ ํ๋ฆ„ ์†์ด๋ผ๊ณ  ํ‘œํ˜„ํ•œ๋‹ค๋ฉด, ๋ฐœ์‚ฐ(div)๋Š” ํ•œ ์  $(x, y, z)$์— ๋ชจ์ด๋Š”(์••์ถ•) ๋ฐฉํ–ฅ์œผ๋กœ ์œ ์ฒด๊ฐ€ ํ๋ฅด๋Š”์ง€, ์•„๋‹ˆ๋ฉด ์ ์—์„œ ๋‚˜๊ฐ€๋Š”(ํŒฝ์ฐฝ)ํ•˜๋Š” ๋ฐฉํ–ฅ์œผ๋กœ ์œ ์ฒด๊ฐ€ ํ๋ฅด๋Š”์ง€๋ฅผ ์ˆ˜์น˜ํ™” ํ•˜๋Š” ๊ฐ’์ด๋‹ค.

  • $\text{div }\mathbf{F} > 0$
    • ํ•œ ์ ์—์„œ ์œ ์ฒด๊ฐ€ ํŒฝ์ฐฝํ•œ๋‹ค: outflowing-ness.
  • $\text{div }\mathbf{F} < 0$
    • ํ•œ ์ ์—์„œ ์œ ์ฒด๊ฐ€ ์••์ถ•๋œ๋‹ค: inflowing-ness.
  • $\text{div }\mathbf{F} = 0$
    • ํ•œ ์ ์ด ์œ ์ฒด์— ์ „ํ˜€ ์••๋ ฅ์„ ๋ฐ›๊ณ  ์žˆ์ง€ ์•Š๋‹ค: incompressible.


๋ฐœ์‚ฐ์— ๋Œ€ํ•œ ์‹์„ 3์ฐจ์› ๊ณต๊ฐ„์— ๋Œ€ํ•œ ๊ฒƒ์œผ๋กœ ์ ์—ˆ์ง€๋งŒ, 2์ฐจ์› ํ‰๋ฉด์—์„œ ์ •์˜ํ•œ ๋ฒกํ„ฐ์žฅ์— ๋Œ€ํ•ด์„œ๋„ ๋ฐœ์‚ฐ(divergence) ๊ฐ’์„ ๊ณ„์‚ฐํ•  ์ˆ˜ ์žˆ๋‹ค.

Gilbert Strang - Calculus Vol 3.: Negative Divergence

์œ„์˜ ๋ฒกํ„ฐ์žฅ์€ $\mathbf{F}(x, y) = \left<-y, x\right>$๋กœ div๋ฅผ ๊ณ„์‚ฐํ•ด๋ณด๋ฉด, $-2$๊ฐ€ ๋œ๋‹ค. ๋”ฐ๋ผ์„œ ๋ชจ๋“  ์ ์ด ์••์ถ•ํ•˜๋Š” ๋ฐฉํ–ฅ์œผ๋กœ ์••๋ ฅ์„ ๋ฐ›๋Š”๋‹ค.


Gilbert Strang - Calculus Vol 3.: Zero Divergence

๋ฐ˜๋ฉด์— ์••์ถ•์„ ์ „ํ˜€ ๋ฐ›์ง€ ์•Š๋Š” ๋ฒกํ„ฐ์žฅ๋„ ์žˆ๋‹ค. $\mathbf{F}(x, y) = \left<-x, y\right>$์ธ ๋ฒกํ„ฐ์žฅ์€ div๋ฅผ ๊ณ„์‚ฐํ•ด๋ณด๋ฉด $0$์ด ๋œ๋‹ค. ๋”ฐ๋ผ์„œ ๋ชจ๋“  ์ ์—์„œ ์••์ด‰์„ ๋ฐ›์ง€ ์•Š๋Š”๋‹ค.


Gilbert Strang - Calculus Vol 3.: Zero Divergence

๋˜, ์œ„์™€ ๊ฐ™์ด ์œ ์ฒด์˜ ํ๋ฆ„์ด ์ƒ์ˆ˜์ธ ๊ฒฝ์šฐ๋„ ์œ ์ฒด๊ฐ€ ๋ฐ›๋Š” ์••๋ ฅ์ด $0$์œผ๋กœ ๊ณ„์‚ฐ๋œ๋‹ค.

๋ญ”๊ฐ€ $\text{div } \mathbf{F} = 0$์ธ ์ƒํ™ฉ์„ โ€œ์œ ์ฒด๊ฐ€ ๋ฐ›๋Š” ์••๋ ฅ์ด $0$โ€์ด๋‹ค ๋ผ๊ณ  ํ‘œํ˜„ํ•˜๋ฉด ์ข€ ๋ชจํ˜ธํ•œ ๊ฐ์ด ์žˆ๋‹ค. ๊ทธ๋ž˜์„œ ์ด๋ ‡๊ฒŒ ํ‘œํ˜„ํ•˜๋Š” ๊ฒƒ๋„ ๊ฐ€๋Šฅํ•˜๋‹ค.

ํ•œ ์ง€์ ์—์„œ ์œ ์ฒด์˜ ์œ ์ถœ์ด๋‚˜ ์œ ์ž…๋˜๋Š” ์œ ๋Ÿ‰์ด ์—†๋‹ค.

์ฆ‰, ์œ ์ฒด๊ฐ€ ๋‚˜๊ฐ€๋Š” ๋งŒํผ, ์–ด๋”˜๊ฐ€์—์„œ ์œ ์ฒด๊ฐ€ ๋“ค์–ด์™€ ๋‚˜๊ฐ€๋Š” ๋Ÿ‰์„ ์ƒ์‡„์‹œํ‚จ๋‹ค๋Š” ๊ฒƒ์ด๋‹ค.

Source-free Field

์œ„์—์„œ ์„ค๋ช…ํ•œ ๊ฒƒ๊ณผ ๊ฐ™์ด ์–ด๋–ค ์ง€์  ๋˜๋Š” ์˜์—ญ์—์„œ ์œ ์ถœ๋Ÿ‰์ด $0$์ธ ๋ฒกํ„ฐ์žฅ์„ โ€œsource-free Vector Fieldโ€œ๋ผ๊ณ  ํ•œ๋‹ค. ์ด๋Ÿฐ ๋ฒกํ„ฐ์žฅ์€ ์•„๋ฌด ๋‹ซํžŒ ๊ณก์„ ์„ ์žก๋”๋ผ๋„, ๊ทธ ๊ณก์„ ์— ๋Œ€ํ•ด ์œ ์ž…/์œ ์ถœํ•˜๋Š” ์œ ์ฒด๋Ÿ‰์ด ๋™์ผํ•˜๋‹ค.

โ€œsourceโ€๋ผ๋Š” ํ‘œํ˜„์€ ํ•œ ์ ์—์„œ ๋‚˜๊ฐ€๋Š” ์œ ๋Ÿ‰์„ ๋งํ•œ๋‹ค. ๋ฐ˜๋Œ€๋กœ โ€œsinkโ€๋Š” ํ•œ ์ ์œผ๋กœ ๋“ค์–ด์˜ค๋Š” ์œ ๋Ÿ‰์„ ๋งํ•œ๋‹ค. wikipedia

๋˜, ์ด๋Ÿฐ $\text{div }\mathbf{F} = 0$์ธ ๋ฒกํ„ฐ์žฅ์„ โ€œSolenoidal Vector Fieldโ€œ๋ผ๊ณ ๋„ ๋ถ€๋ฅธ๋‹ค. ์†”๋ ˆ๋…ธ์ด๋“œ๋Š” ๊ณ ๋“ฑํ•™๊ต ๋ฌผ๋ฆฌ์‹œ๊ฐ„์— ๋ดค๋˜ ๊ทธ โ€˜์†”๋ ˆ๋…ธ์ด๋“œโ€™๋ฅผ ๋งํ•œ๋‹ค ใ…‹ใ…‹

Curl

If $\mathbf{F} = P \, \mathbf{i} + Q \, \mathbf{j} + R \, \mathbf{k}$ is a vector field on $\mathbb{R}^3$, and the partial derivatives of $P$, $Q$ and $R$ all exists,

then the โ€œcurl of $\mathbf{F}$โ€ is the vector field on $\mathbb{R}^3$ defined by

\[\text{curl } \mathbf{F} = \left( \frac{\partial R}{\partial y} - \frac{\partial Q}{\partial z} \right) \mathbf{i} + \left( \frac{\partial P}{\partial z} - \frac{\partial R}{\partial x} \right) \mathbf{j} + \left( \frac{\partial Q}{\partial x} - \frac{\partial P}{\partial y} \right) \mathbf{k}\]

์‹์ด ์กฐ๊ธˆ ๋ณต์žกํ•œ๋ฐ, ์œ„์˜ ์‹์„ ์™ธ์šฐ๊ธฐ ๋ณด๋‹ค๋Š” ์™ธ์ ์œผ๋กœ ํ‘œ๊ธฐํ•œ ์•„๋ž˜์˜ ์‹์œผ๋กœ ์ดํ•ดํ•˜๋Š”๊ฒŒ ๋” ์œ ์šฉํ•˜๋‹ค.

\[\text{curl } \mathbf{F} = \nabla \times \mathbf{F}\]

์ฆ‰, Gradient ๋ฒกํ„ฐ์™€ ๋ฒกํ„ฐ์žฅ์„ ์™ธ์ ํ•œ ๊ฒƒ์ด ํšŒ์ „(curl)์ด๋‹ค.

\[\nabla \times \mathbf{F} = \left| \begin{matrix} \mathbf{i} & \mathbf{j} & \mathbf{k} \\ \frac{\partial}{\partial x} & \frac{\partial}{\partial y} & \frac{\partial}{\partial z} \\ P & Q & R \end{matrix} \right|\]

Geometric meaning

๋ฒกํ„ฐ์žฅ์„ ์œ ์ฒด์˜ ํ๋ฆ„ ์†์ด๋ผ๊ณ  ํ‘œํ˜„ํ•œ๋‹ค๋ฉด, ํšŒ์ „(curl)์€ ํ•œ ์  $(x, y, z)$ ์ฃผ๋ณ€์„ ํšŒ์ „ํ•˜๋Š” ์œ ์ฒด์˜ ํšŒ์ „ ํ๋ฆ„์„ ํ‘œํ˜„ํ•˜๋Š” ๋ฒกํ„ฐ๋‹ค.

  • curl ๋ฒกํ„ฐ์˜ ๋ฐฉํ–ฅ = ํšŒ์ „์ถ•
  • curl ๋ฒกํ„ฐ์˜ ๋ถ€ํ˜ธ = ํšŒ์ „ ๋ฐฉํ–ฅ: CCW(+), CW(-)
  • curl ๋ฒกํ„ฐ์˜ ํฌ๊ธฐ = ํšŒ์ „ํ•˜๋Š” ์†๋„

Gilbert Strang - Calculus Vol 3.: Zero Divergence, Positive Curl

$\mathbf{F} = \left< -y, x, 0\right>$๋ผ๋Š” ๋ฒกํ„ฐ์žฅ์˜ ์›์  $O$์—์„œ์˜ curl์„ ๊ณ„์‚ฐํ•ด๋ณด๋ฉด,

\[\text{curl } \mathbf{F} = 0 \, \mathbf{i} + 0 \, \mathbf{j} + 2 \, \mathbf{k}\]

๋กœ, $+z$ ๋ฐฉํ–ฅ์˜ curl ๋ฒกํ„ฐ๋ฅผ ์–ป์œผ๋ฉฐ, ์œ ์ฒด๊ฐ€ ์›์ ์„ ๊ธฐ์ค€์œผ๋กœ ์‹œ๊ณ„ ๋ฐฉํ–ฅ์œผ๋กœ ํšŒ์ „ํ•˜๊ณ  ์žˆ์Œ์„ ํ•  ์ˆ˜ ์žˆ๋‹ค.


curl ๋ฒกํ„ฐ๊ฐ€ ์˜๋ฒกํ„ฐ์ธ ๊ฒฝ์šฐ๋ฅผ ์œ ์ฒด๊ฐ€ ํ•ด๋‹น ์  ์ฃผ๋ณ€์„ ํšŒ์ „ํ•˜์ง€ ์•Š๋Š”(irrotational)ํ•˜๋‹ค๊ณ  ๋งํ•œ๋‹ค. ์ด๊ฒƒ์€ ์œ ์ฒด๊ฐ€ ํšŒ์ „ ์šด๋™์—์„œ ์ž์œ ๋กญ๋‹ค๋Š” ๊ฒƒ์œผ๋กœ, ์œ ์ฒด๊ฐ€ ์ง์„  ์šด๋™์„ ํ•˜๊ณ  ์žˆ๋‹ค๋ฉด, ๊ทธ ๋ฐฉํ–ฅ์ด ๊ฑฐ์˜ ๋ฐ”๋€Œ์ง€ ์•Š์„ ๊ฒƒ์ž„์„ ๋งํ•œ๋‹ค.

Gilbert Strang - Calculus Vol 3.: Zero Divergence, Zero Curl

์š”๋Ÿฐ ๋ฒกํ„ฐ์žฅ๋„ curl ๋ฒกํ„ฐ๊ฐ€ ์˜๋ฒกํ„ฐ์ธ๋ฐ, ์œ ์ฒด๊ฐ€ ์›€์ง์ด๋Š” ๋ฐฉํ–ฅ๊ณผ ๋‹ค๋ฅธ ๋ฐฉํ–ฅ์œผ๋กœ ์›€์ง์ด๊ฒŒ ํ•˜๋ ค๋Š” ์™ธ๋ ฅ์ด ์กด์žฌํ•˜์ง€ ์•Š๊ธฐ ๋•Œ๋ฌธ์ด๋‹ค.

If Conservative field, curl is zero vector

If $f$ a function of three variables that has continuous second-order derivatives, then

\[\text{curl}(\nabla f) = \mathbf{0}\]

์ •๋ฆฌ์—์„œ๋Š” potential function $f(x, y, z)$๋ฅผ ๊ธฐ์ค€์œผ๋กœ ๋˜์–ด ์žˆ์ง€๋งŒ, Conservative field $\mathbf{F} = \nabla f$์˜ ๊ฒฝ์šฐ, curl์ด ์˜๋ฒกํ„ฐ์ž„์„ ๋งํ•˜๊ณ  ์žˆ๋‹ค.

์ฆ๋ช…์€ Conservative Field์˜ ์ •์˜์— ์ถฉ์‹คํ•˜๊ฒŒ ์‹์„ ์ „๊ฐœํ•˜๊ธฐ๋งŒ ํ•˜๋ฉด ๋œ๋‹ค.

\[\mathbf{F} = \nabla f = \frac{\partial f}{\partial x} \, \mathbf{i} + \frac{\partial f}{\partial y} \, \mathbf{j} + \frac{\partial f}{\partial z} \, \mathbf{k}\] \[\text{curl } \mathbf{F} = \left(\frac{\partial^2 f}{\partial y\partial z} - \frac{\partial^2 f}{\partial z\partial y}\right) \, \mathbf{i} + \left(\frac{\partial^2 f}{\partial z\partial x} - \frac{\partial^2 f}{\partial x\partial z}\right) \, \mathbf{j} + \left(\frac{\partial^2 f}{\partial x\partial y} - \frac{\partial^2 f}{\partial y\partial x}\right) \, \mathbf{k} = \mathbf{0}\]

EzEz $\blacksquare$ (์š”๊ฒŒ ๊ฐ€๋Šฅํ•œ ์ด์œ ๋Š” ํŽธ๋ฏธ๋ถ„ ์ˆœ์„œ๋ฅผ ๋ฐ”๊ฟ”๋„ ์ƒ๊ด€ ์—†๋‹ค๋Š” ํด๋ ˆ๋กœ์˜ ์ •๋ฆฌ(Clairautโ€™s Theorem) ๋•Œ๋ฌธ.)

์œ„์˜ ์ •๋ฆฌ๋Š” Conservative ๋ฒกํ„ฐ์žฅ์˜ curl ๋ฒกํ„ฐ๊ฐ€ ํ•ญ์ƒ ์˜๋ฒกํ„ฐ๋ผ ๊ณ„์‚ฐํ•  ํ•„์š” ์—†๋‹ค๋Š” ๊ฒƒ๋„ ๋งํ•ด์ฃผ์ง€๋งŒ, ์˜คํžˆ๋ ค ๋ฒกํ„ฐ์žฅ์˜ curl์ด ์˜๋ฒกํ„ฐ๊ฐ€ ์•„๋‹Œ ์ƒํ™ฉ์—์„œ ์œ ์šฉํ•˜๋‹ค. ์™œ๋ƒํ•˜๋ฉด, ๋Œ€์šฐ ๋ช…์ œ์— ๋”ฐ๋ผ $\text{curl }\mathbf{F} \ne 0$๋ผ๋ฉด, ๋ฒกํ„ฐ์žฅ์ด non-conservative์ด๊ธฐ ๋•Œ๋ฌธ์ด๋‹ค.

์œ„์˜ ์ •๋ฆฌ์˜ ์—ญ๋ช…์ œ๋Š” ์„ฑ๋ฆฝํ•˜์ง€ ์•Š๋Š”๋‹ค. $\text{curl } \mathbf{F} = \mathbf{0}$์ด๋”๋ผ๋„ ๊ทธ ๋ฒกํ„ฐ์žฅ์ด conservative๊ฐ€ ์•„๋‹Œ ๋ฐ˜๋ก€๊ฐ€ ์žˆ๊ธฐ ๋•Œ๋ฌธ.


๋‹จ, ๋ฒกํ„ฐ์žฅ์ด ์•„๋ž˜์˜ ์กฐ๊ฑด์„ ๋งŒ์กฑํ•œ๋‹ค๋ฉด ์—ญ๋ช…์ œ๋„ ์„ฑ๋ฆฝํ•œ๋‹ค๊ณ  ํ•œ๋‹ค.

If given vector field $\mathbf{F}$ is

  • defined on all of $\mathbb{R}^3$
  • and whose component functions have continuous partial derivatives
  • and $\text{curl } \mathbf{F} = \mathbf{0}$

then $\mathbf{F}$ is a conservative vector field.

๊ทธ๋ž˜์„œ ๋ฒกํ„ฐ ํ•จ์ˆ˜๊ฐ€ ๋ถˆ์—ฐ์†์„ฑ์„ ๊ฐ€์ง€๊ฑฐ๋‚˜, ์„ฑ๋ถ„ ํ•จ์ˆ˜์˜ ํŽธ๋ฏธ๋ถ„์ด ๋ถˆ์—ฐ์†์„ฑ์„ ๊ฐ–๋Š” ๊ทธ๋Ÿฐ ํŠน์ˆ˜ํ•œ ์ƒํ™ฉ๋งŒ ์•„๋‹ˆ๋ผ๋ฉด, $\text{curl } \mathbf{F} = \mathbf{0}$์ธ์ง€ ํ™•์ธํ•˜์—ฌ conservative ์—ฌ๋ถ€๋ฅผ ํŒ๋‹จํ•  ์ˆ˜ ์žˆ์„ ๊ฒƒ์ด๋‹ค ใ…Žใ…Ž

์‚ฌ์‹ค ์ •์˜์—ญ์ด $\mathbb{R}^3$์ผ ํ•„์š˜ ์—†๊ณ , simply-connected region์ด๊ธฐ๋งŒ ํ•˜๋ฉด ๋œ๋‹ค๊ณ  ํ•œ๋‹ค. (์ฆ๋ช…์€ ์ดํ›„์— ์Šคํ† ์Šคํฌ ์ •๋ฆฌ ํ•  ๋•Œ ๋ฐฐ์šด๋‹ค๊ณ  ํ•จ.)

Curl and Div

์–ด๋–ค ๋ฒกํ„ฐ์žฅ์— Curl์„ ์ ์šฉํ•œ ํ›„, Divergence๋ฅผ ์ ์šฉํ•˜๋ฉด ์žฌ๋ฐŒ๋Š” ๊ฒฐ๊ณผ๊ฐ€ ๋‚˜์˜จ๋‹ค.

If $\mathbf{F} = P \, \mathbf{i} + Q \, \mathbf{j} + R \, \mathbf{k}$ is a vector field on $\mathbb{R}^3$ and $p$, $Q$, and $R$ have continuous second-order partial derivatives, then

\[\text{div } \text{curl } \mathbf{F} = 0\]

์ฆ๋ช…์€ curl์™€ div์˜ ์ •์˜์— ๋งž์— ์•„๋ž˜ ์‹์„ ์ „๊ฐœํ•˜๋ฉด ๋œ๋‹ค. Ez

\[\text{div } \text{curl } \mathbf{F} = \nabla \cdot (\nabla \times \mathbf{F}) = 0\]


์š” ์ •๋ฆฌ๋Š” ๋ฒกํ„ฐ์žฅ์ด curl ์—ฐ์‚ฐ์œผ๋กœ ์ธํ•ด ๋งŒ๋“ค์–ด์ง„ ๊ฒƒ์ด๋ผ๋ฉด, ๊ทธ๊ฒƒ์˜ div๊ฐ€ 0์ž„์„ ๋งํ•œ๋‹ค. ์ด๊ฒƒ์˜ ๋Œ€์šฐ ๋ช…์ œ๋ฅผ ํ™œ์šฉํ•˜๋ฉด, ์–ด๋–ค ๋ฒกํ„ฐ์žฅ $\mathbf{G}$๊ฐ€ $\text{div } \mathbf{G} \ne 0$๋ผ๋ฉด, ๊ทธ ๋ฒกํ„ฐ์žฅ์€ curl ๋ฒกํ„ฐ๋กœ ์œ ๋„๋œ ๊ฒƒ์ด ์•„๋‹ˆ๋ผ๊ณ  ํŒ๋‹จํ•  ์ˆ˜ ์žˆ๋‹ค.

Vector Potential

๋ฒกํ„ฐ์žฅ์— curl์„ ์ ์šฉํ•˜๋ฉด ํšŒ์ „ ๋ฒกํ„ฐ์žฅ์ด๋ผ๋Š” ๋˜๋‹ค๋ฅธ ๋ฒกํ„ฐ์žฅ์ด ๋›ฐ์ณ๋‚˜์˜จ๋‹ค. ๊ทธ๋ž˜์„œ ์ฃผ์–ด์ง„ ๋ฒกํ„ฐ์žฅ์ด ํšŒ์ „ ๋ฒกํ„ฐ์žฅ์ด๋ผ๋ฉด, curl์„ ์ ์šฉํ•œ ์›๋ณธ ๋ฒกํ„ฐ์žฅ์„ โ€œVector Potentialโ€๋ผ๊ณ  ๋ถ€๋ฅธ๋‹ค.

The vector field $\mathbf{A}$ is said to be a โ€œvector potentialโ€ for the vector field $\mathbf{B}$ if

\[\mathbf{B} = \nabla \times \mathbf{A}\]

์œ„์˜ Curl-Div ํ•ญ๋“ฑ์‹์„ Vector Potential๋กœ ๋‹ค์‹œ ๊ธฐ์ˆ ํ•˜๋ฉด ์•„๋ž˜์™€ ๊ฐ™๋‹ค.

If there exists a vector potential for the vector field $\mathbf{B}$, then

\[\nabla \cdot \mathbf{B} = 0\]

์•ž์— ๋‚˜์™”๋˜ $\nabla \times \mathbf{F} = 0$ ์ผ€์ด์Šค์™€ ์ข€ ํ—ท๊ฐˆ๋ฆด ์ˆ˜๋„ ์žˆ์–ด์„œ ์ •๋ฆฌ๋ฅผ ์ข€ ํ•ด๋ณด์ž.

For a vector field $\mathbf{F}$

  • If scalar Potential(= potential function) exists, then
    • the field is a conservative field
    • $\nabla \times \mathbf{F} = \mathbf{0}$
  • If vector Potential exists, then
    • the field is a curl vector field
    • $\nabla \cdot \mathbf{F} = 0$

Vector Potential์ด๋“  Scalar Potential์ด๋“  ์–ด๋–ค ๋ฒกํ„ฐ์žฅ์˜ ์›๋ณธ์ด ๋˜๋Š” vector/scalar function๋ผ๋Š”๊ฒŒ ๊ณตํ†ต์ !


์ด๋•Œ, ํšŒ์ „ ๋ฒกํ„ฐ์žฅ $\mathbf{B}$๋ฅผ ๋งŒ๋“œ๋Š” Vector Potential์€ ์œ ์ผํ•˜๊ฒŒ ๊ฒฐ์ •๋˜์ง€ ์•Š๋Š”๋‹ค.

์ž„์˜์˜ ์Šค์นผ๋ผ ํ•จ์ˆ˜ $\psi$์— ๋Œ€ํ•ด ์•„๋ž˜์˜ ๋“ฑ์‹์ด ์„ฑ๋ฆฝํ•œ๋‹ค.

\[\mathbf{B} = \nabla \times \mathbf{A} = \nabla \times (\mathbf{A} + \nabla \psi)\]

์ด๊ฒƒ์ด ๊ฐ€๋Šฅํ•œ ์ด์œ ๋Š” $\nabla \times \nabla \psi = \mathbf{0}$์ด๊ธฐ ๋•Œ๋ฌธ์ด๋‹ค. $\mathbf{A}$์™€ $\mathbf{A} + \nabla \psi$ ๋‘˜๋‹ค ํšŒ์ „ ๋ฒกํ„ฐ์žฅ $\mathbf{B}$๋ฅผ ์œ ๋„ํ•˜๋ฏ€๋กœ, ๋‘˜๋‹ค Vector Potential ์ด๊ณ , Vector Potential์€ ์œ ์ผํ•˜๊ฒŒ ๊ฒฐ์ •๋˜์ง€ ์•Š๊ณ  ๋ฌดํ•œํžˆ ๋งŽ๋‹ค.

๋”ฐ๋ผ์„œ, ์•„๋ž˜์˜ ๋”ฐ๋ฆ„ ์ •๋ฆฌ๊ฐ€ ์„ฑ๋ฆฝํ•˜๋Š”๋ฐ,

If the vector field $\mathbf{B}$ has a vector potential,

then, there is a vector potential $\mathbf{A}$ for $\mathbf{B}$ with $\mathbf{A}_3 = 0$.

์ฆ‰, ํšŒ์ „ ๋ฒกํ„ฐ์žฅ์˜ Vector Potential๋กœ $\mathbf{k}$ ์ปดํฌ๋„ŒํŠธ๊ฐ€ 0์ธ ๋ฒกํ„ฐ์žฅ์ด ํ•ญ์ƒ ์กด์žฌํ•œ๋‹ค๋Š” ๊ฒƒ์ด๋‹ค. ์ด ๊ฒฐ๊ณผ๋Š” $\mathbf{k}$ ์ปดํฌ๋„ŒํŠธ๊ฐ€ ์•„๋‹ˆ๋ผ $\mathbf{i}$๋‚˜ $\mathbf{j}$๊ฐ€ 0์ด์–ด๋„ ์„ฑ๋ฆฝํ•œ๋‹ค.

Laplace Operator

๋ฒกํ„ฐ์žฅ $\mathbf{F}$๊ฐ€ Conservative Field๋ผ๋ฉด,

\[\mathbf{F} = \nabla f\]

์ธ๋ฐ, ์—ฌ๊ธฐ์— div ์—ฐ์‚ฐ์„ ํ•œ๋ฒˆ๋” ์ˆ˜ํ–‰ํ•ด๋ณด์ž.[* curl ์—ฐ์‚ฐ์„ ํ•˜๋ฉด, ์˜๋ฒกํ„ฐ๊ฐ€ ๋˜์—ˆ๋‹ค: $\nabla \times \nabla f = \mathbf{0}$] ๊ทธ๋Ÿฌ๋ฉด ์‹์€ ์•„๋ž˜์™€ ๊ฐ™์ด potential function $f$์— ์„ฑ๋ถ„๋ณ„๋กœ ํŽธ๋ฏธ๋ถ„์„ ๋‘ ๋ฒˆ ์ ์šฉํ•œ ๊ฒƒ์˜ ํ•ฉ์œผ๋กœ ํ‘œํ˜„๋œ๋‹ค.

\[\text{div}(\nabla f) = \nabla \cdot (\nabla f) = \frac{\partial^2 f}{\partial x^2} + \frac{\partial^2 f}{\partial y^2} + \frac{\partial^2 f}{\partial z^2}\]

์š”๋ ‡๊ฒŒ potential function์— Gradient ์—ฐ์‚ฐ๊ณผ Div ์—ฐ์‚ฐ์„ ์—ฐ์†ํ•ด ์ˆ˜ํ–‰ํ•˜๋Š” ๊ฒƒ์„ ๊ฐ„๋‹จํ•˜๊ฒŒ ์•„๋ž˜์˜ ํ‘œ๊ธฐ๋กœ ํ‘œํ˜„ํ•œ๋‹ค.

\[\nabla^2 f = \nabla \cdot (\nabla f) = \text{div}(\nabla f)\]

์š” ์—ฐ์‚ฐ์„ ์ˆ˜ํ–‰ํ•˜๋Š” ์—ฐ์‚ฐ์ž๋ฅผ โ€œLaplace Operatorโ€œ๋ผ๊ณ  ๋ถ€๋ฅธ๋‹ค. ์™œ๋ƒํ•˜๋ฉด, Laplace Equation์—์„œ ์œ ๋ž˜ํ•œ ๊ฒƒ์ด๊ธฐ ๋•Œ๋ฌธ.

\[\nabla^2 f = \frac{\partial^2 f}{\partial x^2} + \frac{\partial^2 f}{\partial y^2} + \frac{\partial^2 f}{\partial z^2} = 0\]

(์ฐธ๊ณ ๋กœ ๋ผํ”Œ๋ผ์Šค ๋ฐฉ์ •์‹์€ ๋ฏธ๋ฐฉ ๋•Œ ๋งŽ์ด ๋ณผ ์˜ˆ์ •โ€ฆ ^^)

Vector form of Greenโ€™s Theorem

๊ทธ๋ฆฐ ์ •๋ฆฌ(Greenโ€™s Theorem)๋Š” 2์ฐจ์› ํ‰๋ฉด ์œ„์˜ ๋ฒกํ„ฐ์žฅ $\mathbf{F}(x, y)$์— ๋Œ€ํ•œ ์„ ์ ๋ถ„์ด ๊ทธ๊ฒƒ์˜ ์„ฑ๋ถ„ ๋ฒกํ„ฐ์˜ ํŽธ๋ฏธ๋ถ„์„ ์กฐํ•ฉํ•œ ์–ด๋–ค ์ด์ค‘ ์ ๋ถ„๊ณผ ์—ฐ๊ฒฐํ•˜๋Š” ์ •๋ฆฌ์˜€๋‹ค.

[๊ทธ๋ฆฐ ์ •๋ฆฌ]

\[\int_C \mathbf{F} \cdot d\mathbf{r} = \int_C P \, dx + Q \, dy = \iint_D \left( \frac{\partial Q}{\partial x} - \frac{\partial P}{\partial y}\right) \, dA\]

๊ทธ๋Ÿฐ๋ฐ ์ด ์ •๋ฆฌ๋ฅผ 3์ฐจ์› ๊ณต๊ฐ„์—์„œ $z$ ์ขŒํ‘œ๊ฐ€ 0์ธ 3์ฐจ์› ๋ฒกํ„ฐ์žฅ $\mathbf{F}(x, y, 0)$์œผ๋กœ ๋ฐ”๊พธ์–ด ์‚ดํŽด๋ณด๋ฉด, ๊ทธ๋ฆฐ ์ •๋ฆฌ๋ฅผ ๋ฐœ์‚ฐ(div)๊ณผ ํšŒ์ „(curl)๊ณผ ์—ฐ๊ฒฐ ํ•ด๋ณผ ์ˆ˜ ์žˆ๋‹ค. ๐Ÿ˜ฎ


๋‚ด์šฉ์„ ์ •๋ฆฌํ•˜๊ธฐ ์ „์— dimenchoi๋‹˜์˜ ๊ทธ๋ฆฐ ์ •๋ฆฌ์˜ ์ง๊ด€์ ์ธ ์ดํ•ด์™€ ์ฆ๋ช…(Greenโ€™s Theorem) ํฌ์ŠคํŠธ๊ฐ€ ์ด ๋ถ€๋ถ„์„ ์ดํ•ดํ•˜๋Š”๋ฐ ๋งŽ์€ ๋„์›€์ด ๋˜์—ˆ์Œ์„ ๋ฐํžŒ๋‹ค. ์•„๋ž˜ ๊ธ€์„ ์ฝ๊ธฐ ์ „์— ์œ„์˜ ํฌ์ŠคํŠธ๋ฅผ ๋จผ์ € ์ฝ๊ณ  ์˜ค๊ธธ ๊ฐ•์ถ” ํ•œ๋‹ค!!

Tangential Form

2์ฐจ์›์˜ ๋ฒกํ„ฐ์žฅ์— $z=0$์ธ $z$ ์„ฑ๋ถ„์„ ์ถ”๊ฐ€ํ•˜์—ฌ 3์ฐจ์› ๋ฒกํ„ฐ์žฅ $\mathbf{F} = \left< x, y, 0\right>$์„ ์ƒ๊ฐํ•ด๋ณด์ž. ์ด๋•Œ, ์ด ๋ฒกํ„ฐ์žฅ์˜ curl ๋ฒกํ„ฐ๋Š” ์•„๋ž˜์™€ ๊ฐ™๋‹ค.

\[\text{curl } \mathbf{F} = \left|\begin{matrix} \mathbf{i} & \mathbf{j} & \mathbf{k} \\ \frac{\partial}{\partial x} & \frac{\partial}{\partial y} & \frac{\partial}{\partial z} \\ P(x, y) & Q(x, y) & 0 \end{matrix}\right| = \left( \frac{\partial Q}{\partial x} - \frac{\partial P}{\partial y}\right) \mathbf{k}\]

์š”๊ธฐ์— $z$ ์„ฑ๋ถ„๋งŒ ์žˆ๋Š” ๋ฒกํ„ฐ์— unit vector $\mathbf{k}$๋ฅผ ๋‚ด์ ํ•˜๋ฉด, ์ต์ˆ™ํ•œ ์‹์ด ๋‚˜์˜จ๋‹ค.

\[(\text{curl } \mathbf{F}) \cdot \mathbf{k} = \left( \frac{\partial Q}{\partial x} - \frac{\partial P}{\partial y}\right)\]

์š”๊ฑด ๊ทธ๋ฆฐ ์ •๋ฆฌ์—์„œ ์ด์ค‘ ์ ๋ถ„์— ๋“ค์–ด๊ฐ€๋Š” ์•„์ฃผ ์ต์ˆ™ํ•œ ํ˜•ํƒœ๋‹ค!! ๊ทธ๋ž˜์„œ ์‹์„ ์ •๋ฆฌํ•˜๋ฉดโ€ฆ

\[\int_C \mathbf{F} \cdot d\mathbf{r} = \iint_D \left( \frac{\partial Q}{\partial x} - \frac{\partial P}{\partial y}\right) \, dA = \iint_D \left( \text{curl } \mathbf{F} \right) \cdot \mathbf{k} \, dA\]

Gilbert Strang - Calculus Vol 3.

Tangential Form์€ ์˜์—ญ $D$ ๋‚ด๋ถ€์—์„œ์˜ ํšŒ์ „(curl)์˜ ์ดํ•ฉ์ด ๊ฒฝ๊ณ„ ๊ณก์„  $C$ ์œ„์—์„œ์˜ tangential integral๋กœ ๋Œ€์‹  ๊ตฌํ•  ์ˆ˜ ์žˆ์Œ์„ ๋งํ•˜๋Š” ๊ฒƒ์ด๋‹ค.

Normal Form

๊ทธ๋ฆฐ ์ •๋ฆฌ์˜ Normal Form์€ ์„ ์ ๋ถ„์„ ๊ณก์„ ์˜ ์ง„ํ–‰ ๋ฐฉํ–ฅ $d\mathbf{r}$๊ณผ ์ˆ˜์ง์ธ ๋ฒกํ„ฐ์— ๋Œ€ํ•ด์„œ ์„ ์ ๋ถ„์„ ์ˆ˜ํ–‰ํ•˜๋Š” ๊ฒƒ์ด๋‹ค.

Gilbert Strang - Calculus Vol 3.

\[\oint_C \mathbf{F} \cdot \mathbf{N} \, ds\]

๊ทธ๋ฆฌ๊ณ  ์ด๊ฒƒ์€ ๊ณก์„  $C$๊ฐ€ ๋งŒ๋“œ๋Š” ์˜์—ญ $D$๋ฅผ ์ถœ์ž…ํ•˜๋Š” ์œ ์ฒด์˜ ํ๋ฆ„์ธ ๋ฐœ์‚ฐ(divergence)์˜ ์ดํ•ฉ์˜ ๊ฐ’๊ณผ ๋™์ผํ•˜๋‹ค.

\[\oint_C \mathbf{F} \cdot \mathbf{N} \, ds = \iint_D \left( \nabla \cdot \mathbf{F} \right) \, dA = \iint_D \left( \frac{\partial P}{\partial x} + \frac{\partial Q}{\partial y} \right) \, dA\]

์œ„์˜ ์‹์ด ์–ด๋–ป๊ฒŒ ์œ ๋„๋˜๋Š”์ง€๋ฅผ ์ข€๋” ์‚ดํŽด๋ณด์ž.

๋จผ์ €, ๊ณก์„  $C$๊ฐ€ ์•„๋ž˜์™€ ๊ฐ™์€ ๋ฒกํ„ฐ ๋งค๊ฐœ๋ฐฉ์ •์‹์ด๋ผ๊ณ  ์ƒ๊ฐํ•ด๋ณด์ž.

\[\mathbf{r}(t) = x(t) \, \mathbf{i} + y(t) \, \mathbf{j}\]

์ด๋•Œ, ๊ณก์„ ์— ์ ‘ํ•˜๋Š” Unit Tangent Vector $\mathbf{T}(t)$๋ฅผ ๊ตฌํ•˜๋ฉด ์•„๋ž˜์™€ ๊ฐ™๋‹ค.

\[\mathbf{T}(t) = \frac{x'(t)}{\left| \mathbf{r}'(t) \right|} \, \mathbf{i} + \frac{y'(t)}{\left| \mathbf{r}'(t) \right|} \, \mathbf{j}\]

๊ทธ๋ฆฌ๊ณ  ์ด์— ๋Œ€ํ•œ ๋…ธ๋ฉ€ ๋ฒกํ„ฐ $\mathbf{N}(t)$๋Š” $\mathbf{T}(t) \cdot \mathbf{N}(t) = 0$์ž„์„ ์ƒ๊ฐํ•˜๋ฉด ์•„๋ž˜์™€ ๊ฐ™์ด ์œ ๋„๋œ๋‹ค.

\[\mathbf{N}(t) = \frac{y'(t)}{\left| \mathbf{r}'(t) \right|} \, \mathbf{i} - \frac{x'(t)}{\left| \mathbf{r}'(t) \right|} \, \mathbf{j}\]

์ด์ œ ๋‹ค์‹œ ์ ๋ถ„์‹์œผ๋กœ ๋Œ์•„์˜ค์ž. ์ ๋ถ„์‹์—์„œ ๋ฏธ์†Œ๊ธธ์ด๋Ÿ‰์„ ๊ณก์„ ์˜ ๋งค๊ฐœ ๋ณ€์ˆ˜๋กœ ๋‹ค์‹œ ์“ฐ๋ฉด ์•„๋ž˜์™€ ๊ฐ™๋‹ค.

\[\oint_C \mathbf{F} \cdot \mathbf{N} \, ds = \oint_C (\mathbf{F} \cdot \mathbf{N}) (t) \, \left| \mathbf{r}'(t) \right| \, dt\]

๊ทธ๋ฆฌ๊ณ  ์ด ์‹์„ ์ž˜ ์ •๋ฆฌํ•˜๋ฉดโ€ฆ

\[\begin{aligned} \oint_C \mathbf{F} \cdot \mathbf{N} \, ds &= \oint_C (\mathbf{F} \cdot \mathbf{N}) (t) \, \left| \mathbf{r}'(t) \right| \, dt \\ &= \oint_C \left( \frac{P(x, y) \cdot y'(t)}{\left| \mathbf{r}'(t) \right|} - \frac{Q(x, y) \cdot x'(t)}{\left| \mathbf{r}'(t) \right|} \right) \, \left| \mathbf{r}'(t) \right| \, dt \\ &= \oint_C P(x, y) \cdot y'(t) \, dt - Q(x, y) \cdot x'(t) \, dt \\ &= \oint_C P \, dy - Q \, dx \end{aligned}\]

๋งˆ์ง€๋ง‰ ์‹์„ $dx$, $dy$ ์ˆœ์„œ๋ฅผ ๋‹ค์‹œ ๋งž์ถ”๊ณ , ๊ทธ๋ฆฐ ์ •๋ฆฌ์˜ ํ˜•์‹์— ๋งž์ถฐ ํŽธ๋ฏธ๋ถ„์œผ๋กœ ๋‹ค์‹œ ์“ฐ๋ฉด ์•„๋ž˜์™€ ๊ฐ™๋‹ค.

\[\begin{aligned} &\oint_C P \, dy - Q \, dx \\ &= \oint_C \left( - Q \, dx + P \, dy \right) \\ &= \iint_D \left( \frac{\partial P}{\partial x} + \frac{\partial Q}{\partial y} \right) \, dA \end{aligned}\]

๊ทธ๋ฆฌ๊ณ  ์œ„์˜ ๋งˆ์ง€๋ง‰ ์‹์€ ๋ฒกํ„ฐ์žฅ $\mathbf{F}$์— ๋ฐœ์‚ฐ(div) ์—ฐ์‚ฐ์„ ์ทจํ•œ $\text{div } \mathbf{F} = \nabla \cdot \mathbf{F}$์™€ ๊ฐ™๋‹ค.

$\blacksquare$

Categories:

Updated: