Advanced Fluid Mechanics Problems And Solutions -

Step 1: Simplify the Navier-Stokes Equations We start with the incompressible Navier-Stokes equation for the x-momentum: $$ \rho \left( \frac\partial u\partial t + u \frac\partial u\partial x + v \frac\partial u\partial y \right) = -\frac\partial P\partial x + \mu \left( \frac\partial^2 u\partial x^2 + \frac\partial^2 u\partial y^2 \right) $$

Given the assumptions:

The equation reduces to a simple balance between pressure and viscous forces: $$ 0 = -\fracdPdx + \mu \fracd^2 udy^2 $$ (Note: Partial derivatives become total derivatives as $u$ depends only on $y$.)

Step 2: Integrate the Differential Equation Rearranging gives: $$ \fracd^2 udy^2 = \frac1\mu \fracdPdx $$

Integrate once with respect to $y$: $$ \fracdudy = \frac1\mu \fracdPdx y + C_1 $$

Integrate a second time: $$ u(y) = \frac12\mu \fracdPdx y^2 + C_1 y + C_2 $$

Step 3: Apply Boundary Conditions

Step 4: Final Velocity Profile Substitute $C_1$ and $C_2$ back into the equation: $$ u(y) = \fracU yB - \frac12\mu \left(-\fracdPdx\right) (By - y^2) $$ (Here, we typically define a favorable pressure gradient as negative, so we swap signs for clarity). advanced fluid mechanics problems and solutions

Step 5: Condition for Zero Net Flow The flow rate per unit width is $Q = \int_0^B u(y) dy$. $$ Q = \int_0^B \left[ \fracU yB + \frac12\mu \fracdPdx (By - y^2) \right] dy $$ $$ Q = \fracU B2 + \frac12\mu \fracdPdx \left[ \fracB y^22 - \fracy^33 \right]_0^B $$ $$ Q = \fracUB2 + \frac12\mu \fracdPdx \left( \fracB^32 - \fracB^33 \right) $$ $$ Q = \fracUB2 + \fracB^312\mu \fracdPdx $$

For $Q = 0$: $$ \fracUB2 = - \fracB^312\mu \fracdPdx $$ $$ \fracdPdx = \frac6\mu UB^2 $$ This implies an adverse pressure gradient is required to exactly counteract the shear-driven flow from the moving plate.

Linear stability is formulated via eigenvalue problems (Orr–Sommerfeld, Squire equations) or non-modal analysis using singular value decomposition of the evolution operator.

Turbulence modeling: RANS (eddy-viscosity, Reynolds-stress models), LES (filtered equations with subgrid-scale models), DNS (full Navier–Stokes resolution).

Kinetic descriptions for rarefied flows: Boltzmann or BGK models; moment methods (Grad’s 13-moment) as approximations.

Multi-phase: Level-set, Volume-of-Fluid (VOF), front-tracking, and phase-field methods for interface dynamics; jump conditions for discontinuities.

The Problem: A viscous jet impinges normally on an infinite flat plate. The external potential flow is ( u_e = a x ), ( w_e = -2a z ) (axisymmetric). Determine the exact velocity profile.

The Advanced Solution Method: Use similarity transformation. For axisymmetric stagnation flow, the stream function ( \psi = r^2 f(z) ). The radial velocity ( u_r = (1/r) \partial\psi/\partial z = r f'(z) ). The vertical velocity ( u_z = -(1/r)\partial\psi/\partial r = -2 f(z) ).

Substituting into the Navier-Stokes equations reduces the PDE to an ODE (the axisymmetric Hiemenz equation): [ f''' + 2f f'' - (f')^2 + a^2 = 0 ] with boundary conditions: ( f(0)=0, f'(0)=0, f'(\infty)=a ).

This is solved numerically to find the wall shear stress ( \tau_w = \mu r f''(0) ). The value ( f''(0) \approx 1.312 ) is a universal constant.

Application: This solution models cooling of turbine blades by impinging jets and chemical vapor deposition reactors. Step 1: Simplify the Navier-Stokes Equations We start

Velocity components:
( u = \frac\partial\psi\partial y = U f'(\eta) ), ( v = -\frac\partial\psi\partial x = \frac12 \sqrt\frac\nu Ux (\eta f' - f) ).

Wall shear stress: ( \tau_w = \mu \left. \frac\partial u\partial y \right|_y=0 = \mu U \sqrt\fracU\nu x f''(0) ).
Substitute ( f''(0)=0.332 ):
[ \tau_w = 0.332 \rho U^2 \sqrt\frac\nuU x ] Local skin friction coefficient: ( C_f = \frac\tau_w\frac12 \rho U^2 = 0.664 \sqrt\frac\nuU x = \frac0.664\sqrtRe_x ).

Boundary layer thickness ( \delta_99 ) where ( u/U=0.99 ) corresponds to ( \eta \approx 5.0 ) (from Blasius table).
[ \delta_99 = 5.0 \sqrt\frac\nu xU = \frac5.0 x\sqrtRe_x ]

Problem:
For a fully developed turbulent pipe flow, derive the log-law velocity profile using Prandtl’s mixing length theory with ( \ell = \kappa y ). Show that ( u^+ = \frac1\kappa \ln y^+ + B ).

Solution:

Near-wall balance: ( \tau_w = \rho \kappa^2 y^2 \left( \fracdudy \right)^2 ).

Take square root: ( u_\tau = \kappa y \fracdudy ). The equation reduces to a simple balance between

Rearrange: ( \fracdudy = \fracu_\tau\kappa y ).

Integrate: ( u = \fracu_\tau\kappa \ln y + C ).

Introduce viscous sublayer matching: Let ( y^+ = \fracy u_\tau\nu ), ( u^+ = \fracuu_\tau ).
Then
[ u^+ = \frac1\kappa \ln y^+ + B ]
Experimentally: ( \kappa \approx 0.41 ), ( B \approx 5.0 ) for smooth walls.

Problem A — Linear instability of a boundary layer (Orr–Sommerfeld)

Problem B — Shock–boundary layer interaction (compressible flow)

Problem C — Multiphase droplet breakup in turbulence

Problem D — Rarefied gas flow in microchannels (slip/transition regime)

Problem E — Fluid–structure interaction causing flutter

Fluid mechanics is often introduced via the pristine, orderly world of potential flow, laminar boundary layers, and simple pipe networks. But the "advanced" realm is where the discipline becomes both beautiful and bewildering. It is a world where vortices scream, interfaces rupture, and the continuum approximation itself is pushed to its limits. This essay explores three advanced problems that reveal the profound depth of fluid dynamics: the breakdown of Stokes flow due to inertial correction, the singular nature of free-surface cusp formation, and the paradoxical drag on a sphere in a confined channel.