Tuesday, July 7, 2015

Various forms of N-S equations

1) Lamb’s form of N-S equation:-
The N-S equation of motion for a viscous incompressible fluid is
                    $\frac{D\overrightarrow{q}}{Dt}=\overrightarrow{F}-\frac{\nabla p}{\rho }+\upsilon {{\nabla }^{2}}\overrightarrow{q}$
    or, $\frac{\partial \vec{q}}{\partial t}+\left( \vec{q}.\nabla  \right)\vec{q}=\vec{F}-\frac{\nabla p}{\rho }+\upsilon {{\nabla }^{2}}\overrightarrow{q}$……(1)               

Where  $\vec{q}=(u,v,w)$
                $\vec{F}=(X,Y,Z)$
                 $\nabla ={{(}^{\partial }}{{/}_{\partial x}}{{,}^{\partial }}{{/}_{\partial y}}{{,}^{\partial }}{{/}_{\partial z}})$
               r = Pressure
               u= Kinematic viscosity

$or,\frac{\partial \vec{q}}{\partial t}+\nabla \left( \frac{1}{2}{{\overrightarrow{q}}^{2}} \right)-\overrightarrow{q}\times \left( \nabla \times \overrightarrow{q} \right)=\vec{F}-\frac{\nabla p}{\rho }+\upsilon {{\nabla }^{2}}\overrightarrow{q}$  [ By Lagrange’s vector identity]
or, , $\frac{\partial \vec{q}}{\partial t}+\nabla \left( \frac{1}{2}{{\overrightarrow{q}}^{2}} \right)-\overrightarrow{q}\times \left( \widetilde{w} \right)=\vec{F}-\frac{\nabla p}{\rho }+\upsilon {{\nabla }^{2}}\overrightarrow{q}$ 

where  $\widetilde{w}=\nabla \times \overrightarrow{q}$ as vorticity vector
Or, \[\frac{\partial \vec{q}}{\partial t}+\nabla \left( \frac{1}{2}{{\overrightarrow{q}}^{2}} \right)+\widetilde{w}\times \overrightarrow{q}=\vec{F}-\frac{\nabla p}{\rho }-\upsilon {{\nabla }^{2}}\overrightarrow{q}\]  which is known as Lamb’s form of N-S equation                    
2) If external forces form a conservative field of force then F=-ÑW, where W is potential 
     function.
The lamb’s form of N-S equation now becomes
\[or,\frac{\partial \vec{q}}{\partial t}+\nabla \left( \frac{1}{2}{{\overrightarrow{q}}^{2}} \right)-\overrightarrow{q}\times \widetilde{w}=-\nabla \Omega -\frac{\nabla p}{\rho }-\upsilon {{\nabla }^{2}}\overrightarrow{q}\]
\[or,\frac{\partial \vec{q}}{\partial t}-\overrightarrow{q}\times \widetilde{w}=-\nabla \left( \frac{1}{2}{{\overrightarrow{q}}^{2}}+\Omega +\frac{p}{\rho } \right)-\upsilon {{\nabla }^{2}}\overrightarrow{q}\]$$
But,
\[{{\nabla }^{2}}\overrightarrow{q}=\nabla (\nabla .\overrightarrow{q})-\nabla \times \left( \nabla \times \overrightarrow{q} \right)=0-\nabla \times \left( \nabla \times \overrightarrow{q} \right)=-\nabla \times \widetilde{w}\]for incompressible fluid.
The above equation reduces to
\[\frac{\partial \vec{q}}{\partial t}-\overrightarrow{q}\times \widetilde{w}=-\nabla \left( \frac{p}{\rho }+\Omega +\frac{1}{2}{{\overrightarrow{q}}^{2}} \right)-\upsilon \nabla \times \widetilde{w}\]which is another form of Lamb’s form of N-S equation under the external force is conservative.
3) When there exist a functional relation between pressure and density, the N-S equation of motion for compressible viscous fluid is
$\frac{D\overrightarrow{q}}{Dt}=\overrightarrow{F}-\int{\frac{\partial \rho }{\rho }}+\upsilon {{\nabla }^{2}}\overrightarrow{q}+\frac{\upsilon }{3}\nabla (\nabla .\overrightarrow{q})$….(1)

Where  $\vec{q}=(u,v,w)$
                $\vec{F}=(X,Y,Z)$
                 $\nabla ={{(}^{\partial }}{{/}_{\partial x}}{{,}^{\partial }}{{/}_{\partial y}}{{,}^{\partial }}{{/}_{\partial z}})$

  $\frac{D\vec{q}}{Dt}=\frac{\partial \vec{q}}{\partial t}+\left( \vec{q}.\nabla  \right)\vec{q}$
                   $=\frac{\partial \overrightarrow{q}}{\partial t}+\nabla \left( \frac{1}{2}{{\overrightarrow{q}}^{2}} \right)-\overrightarrow{q}\times \left( \nabla \times \overrightarrow{q} \right)$   

   [∵$\left( \overrightarrow{q}.\nabla  \right)\overrightarrow{q}=\nabla \left( \frac{1}{2}{{\overrightarrow{q}}^{2}} \right)-\overrightarrow{q}\times \left( \nabla \times \overrightarrow{q} \right)$]     \[and\nabla \times \left( \nabla \times \overrightarrow{q} \right)=\nabla \left( \nabla .\overrightarrow{q} \right)-{{\nabla }^{2}}\overrightarrow{q}\Leftrightarrow {{\nabla }^{2}}\overrightarrow{q}=\nabla \left( \nabla .\overrightarrow{q} \right)-\nabla \times \left( \nabla \times \overrightarrow{q} \right)\]  
Then the equarion (1) reduces to
$\frac{\partial \overrightarrow{q}}{\partial t}+\nabla \left( \frac{1}{2}{{\overrightarrow{q}}^{2}} \right)-\overrightarrow{q}\times \left( \nabla \times \overrightarrow{q} \right)=\overrightarrow{F}-\int{\frac{\partial \rho }{\rho }}+\upsilon \left\{ \nabla \left( \nabla .\overrightarrow{q} \right)-\nabla \times \left( \nabla \times \overrightarrow{q} \right) \right\}+\frac{\upsilon }{3}\nabla (\nabla .\overrightarrow{q})$

$or,\frac{\partial \overrightarrow{q}}{\partial t}+\nabla \left( \frac{1}{2}{{\overrightarrow{q}}^{2}} \right)-\overrightarrow{q}\times \left( \nabla \times \overrightarrow{q} \right)=\overrightarrow{F}-\int{\frac{\partial \rho }{\rho }}-\upsilon \nabla \times \left( \nabla \times \overrightarrow{q} \right)+\frac{4\upsilon }{3}\nabla (\nabla .\overrightarrow{q})$

Wednesday, July 1, 2015

Relation between Cartesian (or rectangular) components of stress


Consider the motion of a small rectangular parallelepiped in viscous fluid with centre at P(x,y,z) and it’s edges of length dx, dy, dz parallel to the fixed rectangular axes.
The mass (density´volume) i.e rdxdydz of the element remains constant and the element is assumed to move along with fluid. In relation to the surface at right angle to the axis of x, the force components per unit area (or stress component) at P(x,y,z) are Pxx, Pxy, Pxz.

The corresponding force components at the centre
                                                         of the face of area dydz far from the origin are 
The corresponding force components at the centre
                                                           of the face of area dydz near from the origin are 
acting in the fluid in the opposite sense of former.


1)The force component on this pair of parallel faces of area dydz through P1 & P2 at right angle to the axis of x may be compounded into a single force having components
                                 

acting at P(x,y,z) in the direction parallel to ox, oy & oz respectively together with couple of moments -pxydxdydz  and +pxydxdydz about oy and oz respectively.


2) The force component on this pair of parallel faces of area dxdz through P1 & P2 at right angle to the axis of y may be compounded into a single force having components
                           
   
acting at P(x,y,z) in the direction parallel to ox, oy & oz respectively together with couple of moments -pyzdxdydz  and +pyzdxdydz about oz and ox respectively.


3) The force component on this pair of parallel faces of area dxdy through P1 & P2 at right angle to the axis of z may be compounded into a single force having components
                             

acting at P(x,y,z) in the direction parallel to ox, oy & oz respectively together with couple of moments -pzxdxdydz  and +pzxdxdydz about oy and ox respectively.
* red indicate just changing the term

Thus taking into account, the surface forces on all six faces at parallelepiped, we see that they reduced into single force at P(x,y,z) having components
                                   
    together with a vector couple having components
                                          (pyz - pzy) dxdydz about ox
                                          (pzx - pxz) dxdydz about oy
                                          (pxy - pyx) dxdydz about oz
Note that the fact has been used here is that the moment of forces about an arbitrary axis must be zero when contracting the element to a point (i.e making edges shrink to zero), for instance
                                          (pyz - pzy) dxdydz =0
                                          (pzx - pxz) dxdydz =0
                                          (pxy - pyx) dxdydz =0
       Þ                               (pyz - pzy) =0
                                          (pzx - pxz) =0
                                          (pxy - pyx) =0
       Þ                               pyz = pzy
                                          pzx = pxz
                                          pxy = pyx

Hence nine rectangular components of stress at a point reduced to six.

Tuesday, June 30, 2015

Equation of continuity in euler' s form.

     

 let ρ be the density of the fluid at p(x,y,z)  and let q=(u,v,w)be the velocity. 

Construct a small parallelopiped taking the point P as centre with edges of length h,k and l parallel to  the coordinate axes which is occupied by fluid.
the mass of fluid that flows across the unit area parallel to yz plane through  P in time δt                                                                        =ρuδt
                                                  =f(x,y,z)δt….say

                                                                         

      
we know that ,mass=volume×density                      
 and here volume=Area×distance                  
  velocity=
→distance=velocity×δt=uδt and area=1 unit           
   volume=1×uδt=uδt ,
so mass=uδt×ρ=ρuδt     
since ρ and u depend upon the variable of x,y and z so they both can be considered as f(x,y,z)
                      
     let P1 and P2  be the projection of P at the faces kl near to and away from the origin respectively where the coordinate of
it' s because P is the centre of the parallelopiped and the length along x axis is h and following figure
             

   due to only projection of P ,P1&P2 do not change the coordinate of y and z it only change x coordinate here P1 lies on faces kl near to origin and P2  lies on faces kl but away from origin
                      the x coordinate of P1=ON-MN
                                                        
                 the x coordinate of P2=ON+NA
                                                
To find the flow across the face kl of the parallelopiped nearest to origin ,take a point          
 on the surface having area dη.dξ at this point 
then the mass of fluid that flows across the area dη.dξ in time δt 
   [using Taylor^' s expansion for several variables neglecting higher order]      
    The mass of the fluid that flows across the faces kl near to the origin in x↑direction in time δt, 
The mass of the fluid that flows across the faces kl away from the origin in x↑direction  in time δt, 
 
therefore the increase in mass inside the parallelopiped in time δt due to the flow across these two faces in x↑direction is

similarly the excess of mass of fluid that flow in across the hl faces near to the origin over the mass that flows out across the hl faces away from the origin in y↑direction in time δt 

 and the excess of mass of fluid that flow in across the kh faces near to origin over the mass flows out across the kh faces away from the origin in z↑direction in time δt,
  
hence the total mass of fliud that flows in the parallelopiped over the mass that flows out of it in time δt,


on the other hand ,the initial mass of the fluid inside the parallelopiped=ρhkl 
and the increase in mass of fluid inside the parallelopiped in time δt
 now by principal of continuity,(1)and (2)are equal. i.e


    which is the required equation of continuity in euler' s form.                                                              

Information about fluid

Fluid :-
          Fluid is a substance which is capable to flow.

Classification of fluid :


                                                                  
Isotropic substance
                    Fluid is treated as isotropic substance which implies that physical property(i.e pressure, density etc) is same in all direction.


Anisotropic substance
                    Fluid is treated as anisotropic substance if that  property is not same in all direction.

Newtonian fluid
The fluid which obeys Newtonian relationship between shearing stress and gradient of velocity given by
                        
where  
 is the velocity gradient and τ is shearing stress.

Ideal fluid
Ideal fluid is a Newtonian fluid in which there exist no shearing or tangential stress but exist only normal or direct stress between two contacting layers.
Note that in nature ideal fluid doesn’t occur. They are only as mathematical concept.

Real  fluid

Real fluid is a Newtonian fluid in which there exist both tangential and normal stress between two contacting layers.