2.1.19 · HinglishAnalytical Mechanics

Principle of least action — Hamilton's principle derivation

1,867 words8 min readRead in English

2.1.19 · Physics › Analytical Mechanics


Action kya hai?


Hamilton's Principle (statement)


se Euler–Lagrange equation derive karna

Yeh chapter ka dil hai. Hum sab kuch scratch se banate hain.

Setup (KYA). Asli path lo aur ise perturb karo: jahan ek arbitrary smooth "wiggle" hai aur ek chota number hai.

Step 1 — Varied path ka action likho. Yeh step kyun? Asli path se correspond karti hai. Stationarity ka matlab hai ka wahan flat tangent ho: .

Step 2 — Integral ke andar differentiate karo (chain rule). Yeh step kyun? sirf aur ke through par depend karta hai. Chain rule factors aur deta hai.

Step 3 — Doosre term ko parts se integrate karo. Yeh crucial move hai: yeh ko mein convert karta hai taaki hum ko factor out kar sakein. Yeh step kyun? Boundary term vanish hoti hai kyunki . Isliye endpoint-fixing matter karti thi.

Step 4 — Terms collect karo.

Step 5 — Calculus of Variations ka Fundamental Lemma apply karo. Kyunki arbitrary hai, bracket har par vanish honi chahiye (agar woh kahin nonzero hoti, toh wahan bumped ek choose karo aur integral zero nahi hoga).

Figure — Principle of least action — Hamilton's principle derivation

Check karo ki Newton deta hai (Feynman-style sanity test)

1D particle lo: , , toh .

Euler–Lagrange: . ✅ Newton's second law nikal aata hai. Yeh kyun matter karta hai: yeh prove karta hai ki sahi choice thi aur Hamilton's principle Newtonian mechanics ke equivalent hai.


Worked Examples


Common Mistakes (Steel-manned)


Recall Feynman: ek 12-saal ke bacche ko explain karo

Socho tum ghar se school ja rahe ho aur tumhare paas ek "thakaan score" hai jo tum raaste mein collect karte ho. Score tumhari running energy add karta hai lekin ek bonus subtract karta hai jo tumhe ache neeche wali jagahon par hone par milta hai. Har possible route aur pace mein se, tum naturally woh lete ho jiska total score kisi bhi choti tweak se nahi badh sakta — path ko thoda nudge karne se score bilkul nahi badalta (yeh "flat" hai bottom par, jaise ek valley ka bottom). Woh special "flat" route hi woh path hai jo nature choose karti hai, aur jab tum math karte ho, toh pata chalta hai ki yeh exactly push-and-pull forces obey karne jaisa hi hai. Ek lazy rule hazaaron force calculations ko replace kar deta hai.


Flashcards

Action kya hai?
Functional , Lagrangian ka time-integral; units J·s.
Hamilton's principle state karo.
Fixed endpoints ke beech asli path action ko stationary banati hai: .
Lagrangian kya hai?
(kinetic minus potential energy).
Euler–Lagrange equation likho.
.
Derivation mein boundary term kyun vanish hoti hai?
Kyunki variation fixed endpoints par vanish hoti hai: .
Kaun sa step ko mein convert karta hai?
Integration by parts.
Integrand bracket zero kyun hota hai?
Calculus of variations ka Fundamental Lemma: arbitrary hai, toh bracket har jagah vanish hona chahiye.
Dikhaao ki EL ke liye Newton deta hai.
.
Kya action hamesha minimum hoti hai?
Nahi — sirf stationary; minimum ya saddle ho sakti hai (rarely maximum).
EL ke liye kya deta hai?
, SHM with .
kyun aur kyun nahi?
Sirf EL ko reproduce karne deta hai; conserved energy (Hamiltonian) hai, ek alag object.

Connections

  • Lagrangian Mechanics — EL equation bahut saare coordinates tak generalize hoti hai.
  • Euler–Lagrange Equation — woh differential equation jo yeh principle produce karta hai.
  • Calculus of Variations — stationary functionals ka math; same lemma.
  • Noether's Theorem ki symmetries ⟹ conservation laws.
  • Hamiltonian Mechanics ka Legendre transform deta hai.
  • Fermat's Principle — optics analogue (stationary optical path/time).
  • Feynman Path Integral — quantum sum over paths weighted by .
  • Newton's Second Law — equivalent formulation jo check ke roop mein recover hoti hai.

Concept Map

integrated over time

must be

states

apply to

constrained by

leads to

then

kills boundary term in

yields

check with L=T-V

units J.s link to

Lagrangian L = T minus V

Action S functional

Hamilton's Principle

Stationary action delta S = 0

Path perturbation q + eps eta

Fixed endpoints eta = 0

Differentiate under integral

Integration by parts

Euler-Lagrange equation

Recovers F = ma

Feynman path integral