์น˜ํ™˜์„ ํ†ตํ•ด ํ•ด๊ฒฐํ•  ์ˆ˜ ์žˆ๋Š” ๋Œ€ํ‘œ์ ์ธ non-linear ODE ํŒจํ„ด

2 minute read

๋ณต์ˆ˜์ „๊ณตํ•˜๊ณ  ์žˆ๋Š” ์ˆ˜ํ•™๊ณผ์˜ ์กธ์—…์‹œํ—˜์„ ์œ„ํ•ด ํ•™๋ถ€ ์ˆ˜ํ•™ ๊ณผ๋ชฉ๋“ค์„ ๋‹ค์‹œ ๊ณต๋ถ€ํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค๋งŒโ€ฆ ๋ฏธ๋ถ„๋ฐฉ์ •์‹์€ ์กธ์—…์‹œํ—˜ ๋Œ€์ƒ ๊ณผ๋ชฉ์ด ์•„๋‹ˆ๋ผ๋Š” ๊ฑธ ๋‚˜์ค‘์— ์•Œ๊ฒŒ ๋˜์—ˆ์Šต๋‹ˆ๋‹คโ€ฆ OTLโ€ฆ ๊ทธ๋ž˜๋„ ์ด์™• ์‹œ์ž‘ํ•œ ๊ฑฐ ๋‹ค์‹œ ๋ณต์Šต ์ข€ ํ•ด๋ด…์‹œ๋‹ค! ๐Ÿƒ ๋ฏธ๋ถ„๋ฐฉ์ •์‹ ํฌ์ŠคํŠธ ์ „์ฒด ๋ณด๊ธฐ

๋“ค์–ด๊ฐ€๋ฉฐ

๋ฒ ๋ฅด๋ˆ„์ด ๋ฐฉ์ •์‹์€ ์•„๋ž˜์™€ ๊ฐ™์€ ํ˜•ํƒœ์˜ non-linear ODE์ด๋‹ค.

\[y' + p(x) y = g(x) y^a\]

$a$ is any real number.

If $a=0$ or $a = 1$, then above equation becomes linear. We will handle the non-linear case.

๋ฏธ๋ถ„๋ฐฉ์ •์‹์ด ์œ„์™€ ๊ฐ™์€ ํ˜•ํƒœ์˜ non-linear ODE๋ผ๋ฉด, ๋ฒ ๋ฅด๋ˆ„์ด๊ฐ€ ๊ฐœ๋ฐœํ•œ ์•„๋ž˜์˜ ์น˜ํ™˜๋ฒ•์„ ์ด์šฉํ•ด linear ODE๋กœ ๋ณ€ํ™˜ํ•˜๊ณ  ODE๋ฅผ ํ’€์–ด๋‚ผ ์ˆ˜ ์žˆ๋‹ค!!

Bernoulli Equationโ€™s Substitution

Let $u(x) = [y(x)]^{1 - a}$, then apply it to give ODE,

first, differentiate it.

\[u'(x) = y' \cdot (1-a) \cdot y^{-a}\]

then, we know $yโ€™ = g(x) y^a - p(x) y$, so

\[u' = (1-a) y^{-a} \cdot (g(x) y^a - p(x) y)\]

clean up right side

\[u' = (1 - a) g(x) - (1-a)p(x) y^{1-a}\]

in the begging, we set $u(x) = y^{1-a}$, so

\[u'(x) = (1-a) g(x) - (1-a) p(x) u(x)\]

Then, the above equation is linear ODE of $u(x)$

\[u' + (1-a)p(x) u = (1-a)g(x)\]

๋งบ์Œ๋ง

๋ฒ ๋ฅด๋ˆ„์ด ๋ฐฉ์ •์‹ ์ž์ฒด๋Š” non-linear ODE๋ฅผ linear ODE๋กœ ๋ณ€ํ™˜ํ•˜๋Š” ์น˜ํ™˜ ํ…Œํฌ๋‹‰์ด๋‹ค. ๋ญ”๊ฐ€ ์ด๊ฒƒ ์ž์ฒด๋กœ ๋” ์˜๋ฏธ์žˆ๋Š” ํ•ด์„์ด๋‚˜ ํ™œ์šฉ์„ ์ฐพ์„ ์ˆœ ์—†์„ ๊ฒƒ ๊ฐ™๋‹ค.

๋ฒ ๋ฅด๋ˆ„์ด์˜ ์น˜ํ™˜ ํ…Œํฌ๋‹‰์„ ํ™œ์šฉํ•  ์ˆ˜ ์žˆ๋Š” ๋Œ€ํ‘œ์ ์ธ ์‚ฌ๋ก€๊ฐ€ โ€œLogistic Population Modelโ€์— ๋Œ€ํ•œ ์‹์ด๋‹ค.

[Logistic Population Model]

\[y' = ky(M-y)\]

์š” ๋…€์„์˜ ๊ฒฝ์šฐ, ODE ์‹์—์„œ $y$์˜ ์ตœ๊ณ ์ฐจํ•ญ์ด $a=2$์ด๊ธฐ ๋•Œ๋ฌธ์—, $u = y^{-1}$๋กœ ๋ฐ”๊พธ๋Š” ์น˜ํ™˜ ํ…Œํฌ๋‹‰์œผ๋กœ linear ODE๋กœ ๋ณ€ํ™˜ํ•  ์ˆ˜ ์žˆ๋‹ค.

Logistic Population Model์— ๋Œ€ํ•œ ์ž์„ธํ•œ ๋‚ด์šฉ์€ ์•„๋ž˜์— ํฌ์ŠคํŠธ์— ์ž์„ธํžˆ ๊ธฐ์ˆ ํ•ด๋‘์—ˆ๋‹ค ๐Ÿ™‚

๐Ÿ‘‰ Logistic Population Model