Ploting streamline with vector's module on line.

% when using :

streamline(Velocity_x_1,Velocity_y_1,linspace(1,Nx,Ny),linspace(1,1,Ny));

hold on

streamline(Velocity_x_1,Velocity_y_1,linspace(1,Nx,Ny),linspace(Ny,Ny,Ny));

hold on

streamline(Velocity_x_1,Velocity_y_1,linspace(1,1,Ny),linspace(1,Ny,Ny));

hold on

streamline(Velocity_x_1,Velocity_y_1,linspace(Nx,Nx,Ny),linspace(1,Ny,Ny));

hold on

% I plot the streamline of the fluid field by 4 sides of the field.

% however, the result are not good of course.

% Since I have no knowledge of the field, I didn't know how to draw the

% effective streamline by this function. And I also want to know how to add

% the module of the vectors on streamlines just like the contourlines in

% geography textbooks. Thanks a lot!

% the result is:

% what I want is like:

% But with direction like vector lines.

[Dyuman Joshi]

Are you sure you want to plot streamlines and not contours?

[泽江]

Yes, streamlines with Direction, and the diffrent Magnititude presented along the line. Diffrent scales of arrows to describe the magnititude could be all right.

Answer 1

[Sam Chak]

Hi @泽江

Can you check if the following are what you want?

spacing = 0.2;
x       = -2:spacing:2;
y       = -2:spacing:2;
[X, Y]  = meshgrid(x,y);
Z       = X .* exp(-X.^2 - Y.^2);

figure(1)
level   = 10;
contour(X, Y, Z, level, 'ShowText', 'on')

figure(2)
[DX,DY] = gradient(Z, spacing);
quiver(X, Y, DX, DY), axis([-2 2 -2 2])

figure(3)
contour(X, Y, Z, level, 'ShowText', 'on'), hold on
quiver(X, Y, DX, DY), axis([-2 2 -2 2])

[泽江]

Thanks for answering, @Sam Chak, but it's maybe not what I want, actually I tried some method, this is my present result：

What I want to do now is adding the direction notation for lines I elected. I tried to control the streamline objects to change the properties of them so I can add markers in which place I want. But I didn't know how to manuplate the objects. Say:

s1=streamline(Velocity_x_4,Velocity_y_4,linspace(1,1,Ny/10),linspace(1,Ny,Ny/10));

hold on

s2=streamline(Velocity_x_4,Velocity_y_4,linspace(Nx,Nx,Ny/10),linspace(1,Ny,Ny/10));

hold on

for i=1:length(s1)

s1(i).Color='k';

end

for i=1:length(s2)

s2(i).Color='k';

end

when I create this two objects, I can't change the properties of the components of this two objects. For example, I want to change the markers of the mid point of streamline 1 from 'none' to '^' which means triangles, but I found I can only change all the points' properties like markers.

[Dyuman Joshi]

Could you please attach the data you are working with?

[泽江]

Sorry, it's too big even after zipping....

[Dyuman Joshi]

Maybe a sample of the data rather than all of it?

I assume you got this data from a simulation. Did you use Ansys Fluent?

[泽江]

Yes it's a simulation, but I got this by my C++ codes. And I found the compromised method. I used contour without number and lines to show magnititude only by colors. Then I use quiver to draw the vectors which are all normallized so the arrow's size could be the same. By this method, I seperate the direction and magnititude of vectors. Although this is not that good, it's all right. Thanks for your patient helping these days!

[Sam Chak]

No worries @泽江. You can show what you've plotted using the contour/quiver approach and then share your expectations with us. The plotting features in MATLAB are highly customizable. However, my plotting skills are only so-so. Perhaps @Dyuman Joshi can provide advice and a solution in his answer.

Considering the density of lines, I can imagine your data is quite big. Nevertheless, you can extract a portion of the data, say 20% to 30%. Here's an example of how to extract 10% of the data. We're looking forward to seeing Velocity_x_new and Velocity_y_new.

% Original Data with 101 points
x = linspace(0, 10, 101);
y = x.^2;
plot(x, y, '.'), hold on

% Extract 10% from the data
percent  = 10;
stepSize = percent/100*(numel(x) - 1)

stepSize = 10

xNew     = x(1:stepSize:end)

xNew = 1×11

0 1 2 3 4 5 6 7 8 9 10

c        = ismember(x, xNew);   % check if Set xNew is a subset of Set x
idx      = find(c);             % find indices of the subset ⊂
yNew     = y(idx)

yNew = 1×11

0 1 4 9 16 25 36 49 64 81 100

plot(xNew, yNew, 'o', 'MarkerSize', 11, 'LineWidth', 2), grid on
xlabel('x'), ylabel('y')

[泽江]

Thanks a lot！ This is also an useful technique for me! Sometimes I do need a rougher data with interpolation data. For example, the field may cost more time to converge if you increase the grid number. But with the rougher grid's data, I can set a good initial condition then make it conveged more quickly. That's very helpful!