Fluid Dynamics using the Computer

Friday 23 October 2015

Pipe Flow Simulation

Just ran another simulation related to HMT, this problem became steady state after about 36 seconds.

Water at 318 K starts flowing (0.00035 m^3/s) through a steel pipe initially at 298 K. The steel pipe had convection to air at 298 K at 3,000 W/m^2.K. A simple simulation yielded inner and outer wall temperatures of the pipe to be 309.07 K and 311.26 K respectively. Then I ran a transient simulation, to find out the time taken by the pipe’s walls to reach these temperatures (f...rom 298 K) as water flows through it. It came out to be around 36 seconds.

Then I ran a FEA. To calculate stresses induced in the pipe due to water pressure, thermal effects, gravity etc. The pipe’s diameter increased by 0.005866 mm and von-mises stress induced was 117,016,056 N/m^2 with a factor of safety of 5.302.

Then I ran fatigue study to see if the pipe will survive under these loads for 20 years or not. It will I think. The fatigue S-N curves were not available so I used the ones for carbon steel (slightly different from the ones I used for CFD analysis and FEA); so will it last for 20 years I am not sure yet (searching for curves).

Temperatures at inner wall surface

Temperature at outer wall surface

displacement and stress animation

LU Decomposition MATLAB

%Please don't mess around with my code
clc;
R=input('How many variables are there in your system of linear equations? ');

clc;
disp ('Please enter the elements of the Coefficient matrix "A" and the Constant/Known matrix "B" starting from first equation');

A=zeros(R,R);

B=zeros(R,1);
for H=1:R

for i=1:R

fprintf ('Coefficient Matrix Column \b'), disp(i), fprintf ('\b\b Row \b'), disp(H);

I=input(' ');

A(H,i)=A(H,i)+I;
end

fprintf ('Known Matrix Row \b'), disp(H);

J=input(' ');

B(H,1)=B(H,1)+J;
end
clc;

[L,U] = lu(A);

Y = L\B;

X = U\Y;
fprintf ('The Coefficient Matrix "A" is \n'), disp(A);

fprintf ('The Constant/Known Matrix "B" is \n'), disp(B);

fprintf ('The Solution Matrix "X" is \n'), disp(X);

Sunday 5 July 2015

Canal Turbine Concept

It's a concept I am currently working on, so far I gave made a CAD model (renderings attached) of it in SolidWorks and analyzed it using its built in CFD module.

There are many advantages of canal turbines over wind turbines, prominent one's being:

Unidirectional flow

Water flows in one direction in a canal so we don't need pitch and yaw control surfaces. That simplifies the design process and reduces weight.

Constant flow rate

We (humans) control water flow rate through canals and it's almost same all year, so we don't have to worry about blade aero foil design to suit variable/abruptly variable flow rate, that makes design process further straight forward.

Large Electricity potential

Canals are 100s of km long, imagine the electricity potential in the canals. You can put these turbines in irrigation canals and it'll power nearby villages and all the irrigation equipment etc.

Higher Power/Discharge Ratio

Water is ~816 times dense (powerful) than air, so for the same discharge (flow) rate we get potentially 816 times more power. Which means more we can make designs that are lighter, smaller and easier to manage and maintain.

Easy maintenance

Fitted less than ~1 m deep inside the canal and can be retracted for maintenance at ground level, making maintenance very easy or better yet, we can maintain them while canals are being cleaned.