Solving Kepler's equation — Newton-Raphson iteration: Visual walkthrough

3.2.15 · D2 · Physics › Orbital Mechanics & Astrodynamics › Solving Kepler's equation — Newton-Raphson iteration

Hum is ek boxed sentence ka peecha kar rahe hain:

$M = E - e sin E$

Lekin uss box ka kuch bhi matlab hone se pehle, hume shapes draw karni hongi, angles ke naam rakhne honge, aur areas count karne honge. Chalte hain zero se.

Step 1 — Orbit aur uska shadow-circle draw karo

KYA. Hum elliptical orbit draw karte hain (woh squashed circle jis par planet actually travel karta hai) aur phir iske chaaron taraf ek perfect circle draw karte hain jo iske sabse wide points ko touch kare. Woh circle hai auxiliary circle.

KYUN. Ellipse par angles measure karna awkward hai — uski "radius" har jagah change hoti hai. Circle friendly hoti hai: har point centre se same distance par hota hai. Poori derivation ki trick yeh hai: geometry easy circle par karo, phir jawab ko ellipse par squash karo. (Yahi auxiliary circle eccentric anomaly ka ghar hai.)

PICTURE. Figure dekho. Mint ellipse mein hai:

semi-major axis $a$ — long width ka aadha (centre se door edge tak, wide direction mein),
semi-major axis label aur semi-minor axis $b$ — short height ka aadha.

Lavender circle ki radius exactly $a$ hai. Centre woh dot hai jise $C$ mark kiya gaya hai. Ek taraf thoda door focus $F$ baitha hai — woh special point jahan Sun rehta hai, centre nahi.

Step 2 — Steady clock-angle $M$ ka naam rakho

KYA. Hum ek pretend angle invent karte hain jo time ke saath perfectly constant rate se badhta hai. Ise $M$ kaho, yaani mean anomaly.

KYUN. Real planet constant angular speed se nahi chalta (Kepler's 2nd law: Sun ke paas hurry karta hai, door dawdle karta hai). Isse "time $t$ par woh kahan hai?" mushkil ho jaata hai. Toh hum pehle kuch aisa banate hain jo time mein uniform ho — ek mental clock-hand — aur baad mein ise real position se connect karte hain. $M$ is tarah define hai ki woh exactly ek orbital period $T$ mein poora $2 π$ sweep kare (dekho Mean Motion and Orbital Period).

PICTURE. Butter-coloured clock-hand steady rate se sweep karta hai. Equal time-slices mein woh equal angles kholta hai — chhote arcs $Δ M$ sab same size ke hain. Yahi uniformity hai poori wajah kyun $M$ exist karta hai.

Step 3 — Circle par geometric angle $E$ ka naam rakho

KYA. Ellipse par planet lo. Seedha upar slide karo (short axis ke parallel) jab tak auxiliary circle na mil jaye. Us circle-point tak centre $C$ par angle hai eccentric anomaly $E$ .

KYUN. Hume planet ki real position par ek handle chahiye, lekin friendly circle par measure kiya hua. $E$ wahi handle hai. Importantly $E$ centre se measure hota hai, jabki planet ka true angle focus se measure hota hai — yahi mismatch hai jahan se poori equation ka "correction term" aayega.

PICTURE. Vertical dashed line dhyaan se dekho: ellipse par planet $P$ aur circle par point $Q$ ka same horizontal position hai. Angle $∠ QC X$ (wide axis se, centre $C$ par) hai $E$ . $Q$ ki height hai $a sin E$ ; centre se uska horizontal offset hai $a cos E$ .

Step 4 — Equal areas means equal time

KYA. Kepler's 2nd law se, focus se planet tak ki line equal time mein equal areas sweep karti hai (dekho Kepler's Laws of Planetary Motion). Toh perihelion se sweep kiya gaya area total ellipse area ka ek fixed fraction hai — aur woh fraction exactly Step 2 se $M /2 π$ hai.

KYUN. Yeh ek time statement ko area statement mein convert karta hai. Time invisible hai; area hum geometry se compute kar sakte hain. Yeh poori derivation ka pivot hai.

PICTURE. Shaded coral wedge woh area hai jo focus-line ne perihelion se planet tak sweep kiya hai. Uska size, poore ellipse se divide karke, guzri hue period ka fraction equal karta hai.

Step 5 — Circle par swept area compute karo: pie slice minus triangle

KYA. Swept area circle par karo instead. Focus se measure ki gayi perihelion se swept region ek centre ke baare mein pie slice hai minus ek triangle jo focus ki taraf latka hua hai.

KYUN. Circle areas trivial hain: angle $E$ wala pie slice, radius $a$ ki circle mein, area hai $\frac{1}{2} a^{2} E$ . Lekin hamara wedge focus se measure hota hai, centre se nahi — focus $c = a e$ se shove hua hua hai. Woh offset ek triangle kaata hai. Subtract karo.

PICTURE. Lavender pie slice $\frac{1}{2} a^{2} E$ centre se full wedge hai. Coral triangle $C$ – $F$ – $Q$ remove karna zaroori hai kyunki hamara real reference point $F$ hai, jo $a e$ side mein baitha hai. Us triangle ki base $a e$ hai (centre-to-focus) aur height $a sin E$ hai ( $Q$ ki height).

Step 6 — Circle area ko ellipse par squash karo

KYA. Circle par har point ellipse par tab aata hai jab uski height $b / a$ factor se shrink hoti hai. Har height ko $b / a$ se shrink karne par har area bhi $b / a$ se shrink hota hai.

KYUN. Areas us same factor se scale hote hain jis direction ko tum squash karte ho (widths untouched rehti hain, heights sab $b / a$ se multiply hoti hain, toh product — area — $b / a$ se multiply hota hai). Yeh hume easy circle-area result borrow karke hard ellipse par land karne deta hai.

PICTURE. Baayi taraf tall lavender circle-wedge, aur wahi wedge right side par coral ellipse-wedge mein flat squash hua. Har vertical strip ne apni height ka same fraction $b / a$ khoya.

Step 7 — Assemble karo aur letters cancel hote dekho

KYA. Squashed area ko Step 4 ke area-law mein daalo aur simplify karo.

KYUN. Yeh final algebra hai — aur yeh beautiful hai, kyunki har $a$ aur $b$ cancel ho jaata hai, sirf angles ka ek pure relation bacha rehta hai.

PICTURE. Cancellation, drawn: upar $\frac{1}{2} ab$ aur neeche $π ab$ dono mein same $ab$ hai; inhe strike out karo aur sirf $(E - e sin E)$ aur $2 π$ bachte hain.

$\frac{\frac{1}{2} ab ( E - e sin E ) swept ellipse area}{whole ellipse π ab} = \frac{M}{2 π} ⟹ \frac{E - e s i n E}{2 π} = \frac{M}{2 π} ⟹ M = E - e sin E$

Step 8 — Edge cases: equation extremes par survive karni chahiye

KYA. Corners check karo: ek circle, dono apsides, aur near-parabolic limit.

KYUN. Jis law par tum trust karte ho woh $e = 0$ ya $e \to 1$ par break nahi hona chahiye. Ise stress-test karte hain.

PICTURE. Teen mini-panels: ek perfect circle (left), perihelion/aphelion line jahan correction mar jaata hai (middle), aur stretched near-parabola jahan correction rage karta hai (right).

Ek-picture summary

Upar sab kuch, compressed: circle pie-slice deta hai, focus triangle kaatta hai, $b / a$ se squash ellipse par drop karta hai, aur area-law ise clock fraction $M /2 π$ ke barabar karta hai — aur bacha rehta hai $M = E - e sin E$ .

Recall Feynman retelling — poora walkthrough plain words mein bolo

Hum chahte the ki planet ki journey mein time ko woh actually kahan hai se connect karein. Time smooth aur boring hai, toh humne ek pretend clock-hand invent kiya jise $M$ kaha — jo perfectly even speed se sweep karta hai — orbit ke period ka ek fraction guzarne ke baad, usne full turn ka wahi fraction sweep kiya hoga.

Lekin planet cheating karta hai: Sun ke paas race karta hai, door crawl karta hai, equal time mein equal areas sweep karta hai. Toh uske messy angle ko seedha chase karne ki bajaye, humne squashed orbit ke chaaron taraf ek bada helper circle draw kiya aur angle $E$ mark kiya centre se, planet ki shadow tak us circle par.

Area measure karne ke liye jo planet ne sweep kiya, humne circle se ek clean pie slice kaata (easy: $\frac{1}{2} a^{2} E$ ), phir ek triangle kaata kyunki Sun off-centre $a e$ se baitha hai — woh triangle hai $\frac{1}{2} a^{2} e sin E$ . Slice minus triangle deta hai $\frac{1}{2} a^{2} (E - e sin E)$ . Phir humne poori circle-area ko ellipse par $b / a$ factor se squash kiya, toh real swept area hai $\frac{1}{2} ab (E - e sin E)$ .

Finally humne kaha: (area swept)/(poora ellipse $π ab$ ) must equal (clock fraction) $M /2 π$ . Saare $a$ 's aur $b$ 's cancel ho gaye, aur nikla woh jewel: $M = E - e sin E$ . $- e sin E$ sirf us off-centre triangle ki yaad hai. Aur kyunki tum algebra se us equation mein se $E$ nahi nikal sakte, hume guess-and-correct karna padta hai — yahan Newton-Raphson aage aata hai.

Recall Quick self-test

Correction term perihelion aur aphelion par kyun gayab ho jaata hai? ::: Kyunki wahan $sin E = 0$ hota hai ( $E = 0$ aur $E = π$ par), toh off-centre triangle ki height aur area zero hai — subtract karne ke liye kuch nahi. $- e sin E$ term geometrically kya hai? ::: Triangle from centre $C$ to focus $F$ to circle-point $Q$ , base $a e$ aur height $a sin E$ ke saath. Circle par directly ellipse ki bajaye kaam kyun karte hain? ::: Circle sectors ka trivial area hota hai $\frac{1}{2} a^{2} E$ ; hum wahan compute karte hain aur $b / a$ se ellipse par squash karte hain.

Parent: Solving Kepler's equation — Newton-Raphson iteration · Related: Eccentric Anomaly and the Auxiliary Circle · True Anomaly and the Orbit Equation · Mean Motion and Orbital Period