Foundations — Complex eigenvalues — rotation-scaling interpretation

4.5.32 · D1 · Maths › Linear Algebra (Full) › Complex eigenvalues — rotation-scaling interpretation

Is page pe assume kiya gaya hai ki tumne kuch nahi dekha. Parent note Complex eigenvalues — rotation-scaling interpretation padhne se pehle, usmein use hone wala har squiggle pehle ground up se banana hoga. Woh hum yahan karte hain, ek waqt mein ek symbol, har ek apne pehle wale pe tika hua.

0. Ek matrix ek aisi machine hai jo points ko move karti hai

Picture yeh hai: ek matrix ek movement machine hai. Tum ek arrow daalo, ek moved arrow bahar aata hai. Yeh plane ke har arrow ke saath karo aur poora plane warp ho jaata hai — stretch, squish, shear, flip, ya spin.

Do numbers ek $2\times 2$ machine ko ek nazar mein summarise karte hain (poori detail Trace and Determinant mein):

1. Vectors $v$, $x$, $y$ — arrows jinhe hum add kar sakte hain

Picture: arrows jinhe tum saath slide aur stretch kar sakte ho. Parent note mein, $x$ aur $y$ coordinates nahi hain — yeh do poore real arrows hain jinhe hum ek complex eigenvector banane ke liye jod te hain. Yeh seedha rakho: yahan $x,y$ arrows ko name karte hain, axis-positions ko nahi.

Nonzero matter karta hai: zero arrow $\left(\begin{array}{c}0\\ 0\end{array}\right)$ ki koi direction nahi hoti, isliye woh kabhi eigenvector nahi ho sakta. Har eigenvector definition secretly kehti hai "koi nonzero $v$."

2. Eigen-sawaal: kaun se arrows apni direction rakhte hain?

Neeche ki picture: ek eigenvector ek aisi line pe hota hai jise machine jagah pe chhod deti hai (points us line ke saath slide ho sakte hain, lekin line khud tilt nahi hoti). $\lambda$ batata hai kitna aage.

3. $\det (A-\lambda I)=0$ ko symbol by symbol padhna

$\lambda$ find karne ke liye hume teen naye marks chahiye.

Subtract kyun? Equation $Av=\lambda v$ rearrange hoti hai $Av-\lambda v=0$, yaani $(A-\lambda I)v=0$: machine $A-\lambda I$ arrow $v$ ko zero pe crush karti hai. Ek machine kisi nonzero arrow ko sirf tab crush kar sakti hai jab woh area ko kuch nahi kar de — yaani sirf tab jab uska determinant $0$ ho.

Symbol $\det (\cdots )=0$ isliye yeh sentence hai: "kin stretch amounts $\lambda$ ke liye machine kisi arrow ko kuch nahi kar deti?" Woh $\lambda$ eigenvalues hain.

4. Quadratic formula aur discriminant $\Delta$

Hamari equation ${\lambda }^2-(\mathrm{tr}A)\lambda +\det A=0$ ek quadratic hai — iska shape ${\lambda }^2-p\lambda +q=0$ hai. Woh tool jo har quadratic solve karta hai:

Wahi aakhri bullet point kyun hai ki agla section exist karta hai.

5. Imaginary unit $i$ aur complex numbers $\alpha +i\beta$

6. Modulus $r$ aur argument $\theta$ — polar view

Cartesian pair $(\alpha ,\beta )$ ka ek twin description hai distance aur angle ke terms mein. Yeh poore topic ki punchline hai, isliye ise ek right triangle pe carefully build karo.

Point $\lambda =\alpha +i\beta$ daalo aur ise origin se connect karo. Yeh ek right triangle banata hai: horizontal leg $\alpha$, vertical leg $\beta$, aur arrow khud sloped side ke roop mein (hypotenuse).

$\tan$ kyun, aur $atan2$ kyun? Triangle pe, angle ki steepness capture hoti hai $\tan \theta =\frac{\text{opposite}}{\text{adjacent}}=\frac{\beta }{\alpha }.$ $\alpha ,\beta$ se $\theta$ recover karne ke liye hum $\tan$ ko undo karna chahte hain. Plain inverse $\arctan (\beta /\alpha )$ ek point aur uske opposite mein farq nahi kar sakta (dono same ratio dete hain), toh woh plane ke aadhe hisse mein silently fail ho jaata hai. Fixed-up tool $atan2(\beta ,\alpha )$ $\beta$ aur $\alpha$ ke dono alag signs leta hai aur chaalon quadrants mein sahi angle return karta hai:

7. Similarity $A=PC{P}^{-1}$ aur rotation matrices

Do aakhri symbols jinhe parent note use karta hai.

Prerequisite map

Equipment checklist

Right side cover karo aur jawab do; har ek ek symbol hai jo parent note se pehle tumhare paas hona chahiye.