Visual walkthrough — Fluid kinematics — Eulerian vs Lagrangian description

2.2.9 · D2 · Physics › Fluid Mechanics › Fluid kinematics — Eulerian vs Lagrangian description

Hum ek number ke peeche hain: koi quantity kitni tezi se change hoti hai ek aisi fluid ki lump ke liye jo flow ke saath bahi ja rahi hai? Quantity ko $\varphi$ kaho (maano, temperature). Answer ko $\frac{D\varphi }{Dt}$ kaho. Bas yahi matlab hai $D/Dt$ ka — saath-saath bah te hue mehsoos ki gayi change ki rate.

---

Step 1 — Ek hi nadi ki do sachchi pictures

KYA. Koi bhi maths se pehle, fluid dekhne ke do tarike pakke karo, kyunki poori derivation unke beech ek conversion hai.

KYUN. Hamare instruments (thermometers, pressure gauges) ek jagah baith kar fluid ko guzarte dekhte hain — yeh Eulerian picture hai, ek field $\varphi (x,y,t)$ jo locations par tiki hui hai. Lekin Newton ke laws ek specific particle ki baat karte hain — yeh Lagrangian picture hai, ek tagged lump ko follow karna. Hume dono ke beech ek bridge chahiye, aur bridge banane ke liye pehle dono kinare draw karo.

PICTURE. Left side mein, ek insaan bridge par khada hai jo neeche se guzarte paani ko padhta hai (Eulerian, fixed dot). Right side mein, ek tagged red parcel downstream drift karta hai, apna khud ka history liye (Lagrangian, moving dot).

---

Step 2 — Ek field numbers ka landscape hai

KYA. Eulerian field $\varphi (x,y,t)$ ko ek frozen instant par ek coloured landscape ki tarah draw karo — har point par ek value.

KYUN. "Warmer region mein move karna" tab tak nahi bol sakte jab tak yeh dikh na jaaye ki regions alag hain. Ek field bas yeh hai: "har space ke point par ek number chipka hua." Isko terrain (ya shading) ki tarah sochna aane wale ideas — gradient, uphill chalna — ko literal bana deta hai.

PICTURE. Shading cool (light) se warm (dark) ki taraf jaati hai. Ek akela red dot hamara parcel hai jo is landscape par baitha hai. Abhi kuch move nahi kar raha; hum sirf map padhna seekh rahe hain.

---

Step 3 — Reason ①: landscape khud waqt ke saath badalta hai

KYA. Parcel ko ek point par freeze karo aur time chalne do. Poora landscape uth ya gir sakta hai — socho ek tap khula aur poora kamra garm ho gaya.

KYUN. Yeh woh change hai jo aap tab bhi measure karte agar kabhi hile nahi. Yeh ek fixed point par honest Eulerian rate hai. Hum ise symbol $\frac{\partial \varphi }{\partial t}$ dete hain.

PICTURE. Same location (red dot fixed), do snapshots: $t$ aur $t+\Delta t$ ke beech shading har jagah darker ho jaati hai. Parcel ki value bina hilay chadh jaati hai.

---

Step 4 — Reason ②: parcel nayi value tak chalta hai

KYA. Ab time freeze karo aur parcel ko move karne do. Bilkul unchanging landscape par bhi, low spot se high spot tak chalne se woh value badal jaati hai jis par aap baithe ho.

KYUN. Parcel khud rukne ka choice nahi karta — flow use leke jaata hai. Toh ek doosra contribution sirf displacement se aata hai. Yeh woh term hai jo beginners ko chaukaa deta hai, isliye hum ise chhote steps ke right triangle se banate hain.

PICTURE. Parcel ek tiny arrow $\Delta x$ right aur $\Delta y$ upar move karta hai. Raaste mein shading change hoti hai. Hum walk ko ek horizontal leg aur ek vertical leg mein tod dete hain.

Yeh tool kyun? Kyunki "jab main kisi direction mein chhota qadam rakhta hun toh $\varphi$ kitni change hoti hai" — yeh dono steepnesses ko milake answer hota hai. $(\Delta x,\Delta y)$ ka ek qadam $\varphi$ mein yeh change karta hai: $\Delta {\varphi }_{\text{from moving}}\approx \frac{\partial \varphi }{\partial x}\Delta x+\frac{\partial \varphi }{\partial y}\Delta y.$ Har product yeh kehta hai: "us direction mein steepness $\times$ us taraf kitna gaya."

---

Step 5 — Dono reasons ek saath hote hain: tiny changes jodo

KYA. Ek real tick $\Delta t$ mein, parcel dono experience karta hai: landscape shift hoti hai (Step 3) aur woh chalta hai (Step 4). Dono tiny changes jodo.

KYUN. Yeh independent causes hain jo same interval mein kaam kar rahi hain, isliye unke chhote effects simply add ho jaate hain. Yeh chain rule ka dil hai, plain arithmetic mein, kisi bhi symbol soup se pehle.

PICTURE. Do arrows tip-to-tail stack hote hain: ek vertical "landscape rose" arrow ($\partial \varphi /\partial t\Delta t$) aur Step 4 ka "walked uphill" arrow. Unka total woh change hai jo red parcel sach mein mehsoos karta hai.

$$\Delta {\varphi }_{\text{felt}}\approx \underset{\text{landscape badla}}{\underbrace{\frac{\partial \varphi }{\partial t}\Delta t}}+\underset{\text{parcel chala}}{\underbrace{\frac{\partial \varphi }{\partial x}\Delta x+\frac{\partial \varphi }{\partial y}\Delta y}}$$

---

Step 6 — $\Delta t$ se divide karo: displacements ko velocities mein badlo

KYA. Har term ko elapsed time $\Delta t$ se divide karo aur $\Delta t\to 0$ jaane do taaki "$\approx$" "$=$" ban jaaye.

KYUN. Hume ek rate chahiye (per second), na ki ek lump change. Displacement $\Delta x$ ko $\Delta t$ se divide karna parcel ki velocity ki exact definition hai — isliye velocity components aate hain, kisi farmaan se nahi balki division se.

PICTURE. Step 5 ke stacked arrows ko "per second" tags mil jaate hain. $\frac{\Delta x}{\Delta t}$ shrink hoke horizontal speed $u$ ban jaata hai; $\frac{\Delta y}{\Delta t}$ shrink hoke vertical speed $v$ ban jaata hai.

Sab milake:

$$\frac{D\varphi }{Dt}=\frac{\partial \varphi }{\partial t}+\frac{\partial \varphi }{\partial x}u+\frac{\partial \varphi }{\partial y}v.$$

---

Step 7 — Result ko naam do aur har symbol padho

KYA. Moving terms ko ek compact package mein collect karo.

KYUN. Do "walk" terms dono mein velocity ko steepness se multiply kiya gaya hai — yeh pattern velocity arrow aur gradient arrow ka dot product hai, likha $(\mathbf{v}\cdot \nabla )\varphi$. Dot product sahi tool hai kyunki yeh jawab deta hai "main uphill direction mein se actually kitna chal raha hun?"

Yeh wohi formula hai jo parent note ne likha tha — ab har symbol ek picture se kamaya gaya hai.

---

Step 8 — Edge cases (koi scenario bina dikhaye mat chhodna)

Char situations decide karti hain ki kaunse terms bachenge. Neeche har ek ka apna panel hai.

• ① Steady field mein khada rehna ($\mathbf{v}=0$ aur $\partial /\partial t=0$): dono terms zero, $\frac{D\varphi }{Dt}=0$. Parcel kuch mehsoos nahi karta. (Trivial, lekin baaki sab ko anchor karta hai.)
• ② Unsteady, lekin flat field ($\nabla \varphi =0$, tap khul raha hai): sirf local term. Landscape har jagah level hai, isliye chalna kuch nahi karta; sirf waqt ke saath uthna matter karta hai. $\frac{D\varphi }{Dt}=\frac{\partial \varphi }{\partial t}$.
• ③ Steady, lekin sloped field ($\partial /\partial t=0$, $\mathbf{v}\ne 0$): sirf convective term. Yeh headline case hai — kisi bhi fixed point par kuch nahi badlata, phir bhi parcel accelerate karta hai kyunki woh steeper/faster region mein walk kar jaata hai. $\frac{D\varphi }{Dt}=(\mathbf{v}\cdot \nabla )\varphi \ne 0$.
• ④ Landscape ke level ke saath-saath move karna ($\mathbf{v}\perp \nabla \varphi$): dot product zero hai chahe parcel move kare aur field slope kare — kyunki woh pahadi ke across chalta hai, kabhi upar nahi. Convective term $=0$.

---

Step 9 — $\varphi =\mathbf{v}$ set karo: fluid particle ka acceleration

KYA. Velocity khud ko machine mein daalo.

KYUN. Newton ko chahiye acceleration = parcel ki velocity ki change ki rate, jo exactly $D\mathbf{v}/Dt$ hai.

$$\mathbf{a}=\frac{D\mathbf{v}}{Dt}=\frac{\partial \mathbf{v}}{\partial t}+(\mathbf{v}\cdot \nabla )\mathbf{v}$$

Convective term mein ab $\mathbf{v}$ derivatives of $\mathbf{v}$ ko multiply kar raha hai — yeh nonlinear hai. Yeh ek akela fact turbulence ka beej hai aur yahi reason hai ki Navier-Stokes equations itni mushkil hain. Yahi acceleration exactly Euler's equation of motion mein plug hota hai.

---

Ek picture mein summary

Sab kuch ek canvas par: red parcel ek sloped, time-changing landscape par; do arrows (landscape ka uthna = local; upar chalna = convective) milke woh total change banate hain jo woh mehsoos karta hai.

Recall Feynman retelling (symbols ke bina, zor se bolo)

Socho tum ek chhoti si cork ho jo nadi mein bah rahi ho. Tumhare paas ek thermometer hai. Teri reading ko do cheezein change kar sakti hain. Ek: koi upstream hot spring on kar deta hai aur poori nadi dheere-dheere garm hoti hai — tum yeh tab bhi mehsoos karte chahe ek jagah chipke raho. Yeh local change hai. Do: kinare ka paani thanda hai aur beech ka garam, aur current tumhe kinare se beech le jaata hai — tum sirf move hone se garm ho jaate ho, chahe nadi ka temperature map kabhi na badle. Yeh convective change hai. Tumhari total mehsoos ki gayi change simply in dono ka addition hai. Likha jaaye toh yeh sum material derivative $D/Dt$ hai. Aur chhupa hua sabak yeh hai: ek nadi bilkul "steady" ho sakti hai — har fixed point ki reading hamesha constant — aur phir bhi tum, cork, speed up aur warm up karte ho, kyunki tum woh cheez ho jo move karti hai.

---

Recall Term-by-term self test

Big-$D$ derivative matlab kiska rate? ::: Moving parcel ka (Lagrangian). Curly-$\partial t$ derivative kya fixed rakhta hai? ::: Position (Eulerian, ek fixed spot). Steady flow mein kaunsa term mar jaata hai? ::: Local term $\partial /\partial t$. Steady flow mein acceleration kaunsa case deta hai? ::: Convective, parcel faster region mein walk karta hai. Convective term zero kab hota hai chahe parcel move kare? ::: Jab woh field ke level ke saath-saath move kare, yaani $\mathbf{v}\perp \nabla \varphi$. Fluid acceleration mathematically mushkil kyun hai? ::: $(\mathbf{v}\cdot \nabla )\mathbf{v}$ $\mathbf{v}$ mein nonlinear hai.

Related: Read this in Hinglish → · Steady vs unsteady flow · Streamlines, pathlines and streaklines · Continuity equation · Reynolds transport theorem