Double integral using integral2, too many input arguments - matlab

This is the code:
for Y=0:0.01:N,
eta(round(1+Y*100))=((i*alpha*lambda)^(1/3))*Y+eta0;
V2(round(1+Y*100))=((i*alpha*lambda)^(-2/3))*q*integral2(#(n) airy(n),eta0,eta,eta0,eta);
end
What's going on?
Error using profiles>#(n)airy(n)
Too many input arguments.

You do not want to calculate the volume under a plane z = f(x, y), which is the goal of integral2. You want to calculate the integral of the integral:
integral(#(n) integral(#(x) airy(x), xMin, n), xMin, xMax, 'ArrayValued', true)
Note that we need to add 'ArrayValued' = true for the outer integral, because the inner integral does not allow a vectorised evaluation for its upper limit.
This can be optimised (~10x faster) using integral2:
integral2(#(x, y) airy(y), xMin, xMax, xMin, #(x) x)
You can interpret it as follows: the inner integral is calculated by integrating over the y axis up to y = x and the outer integral is calculated by integrating the result of the y integration.
Validation of provided method
One can validate the method with the following test function:
airy = #(n) n; % linear test function
xMin = 0;
xMax = 10;
The second integral is: n^3/6 = 166.6667, which is indeed the result of both methods.

Related

MATLAB offset when plotting integration of sin

I have a question, because this work for many functions, but I have a trouble when trying to plot the integral of a sine (I am using matlab 2010):
clear all
close all
clc
x = linspace(-10, 10, 100);
f = #(x) sin(x);
I = arrayfun(#(x) quad(f, 0, x), x);
plot(x, f(x),'r', x, I, 'b')
I expect having a -cos(x), but instead I get something with an offset of 1, why is this happening? How should fix this problem?
The Fundamental Theorem of Calculus says that the indefinite integral of a nice function f(x) is equal to the function's antiderivative F(x), which is unique up-to an additive constant. Further, a definite integral has the form:
In this form, the constant of integration will cancel out, and the integral will exactly equal the desired antiderivative only if the lower bound evaluation vanishes. However, -cos(0) does not vanish and has a value of -1. So in order to calculate the desired antiderivative F(x), the lower bound evaluation should be added to the right-hand side.
plot(x, f(x),'r', x, I+ (-cos(0)), 'b');
This is the equivalent of assigning an initial value for the solution of ODEs a la ode45.
What you're trying to do can be achieved using the following:
x = linspace(-10, 10, 100);
syms y;
f = sin(y) %function
I =int(f,y) %integration of f
plot(x, subs(f,y,x),'r', x, subs(I,y,x), 'b')
Output:-
According to Matlab documentation, q = quad(fun,a,b)
Quadrature is a numerical method used to find the area under the graph of a function, that is, to compute a definite integral.
Integral of sin(x) equals -cos(x)
Definite integral of sin(x) from x = pi to x = 0:
-cos(pi) - (-cos(0)) = 2
Since quad compute a definite integral, I can't see any problem.
Same as:
figure;plot(-cos(x) - (-cos(0)))

Errors when using the Integral2 function in MATLAB

As far as I can tell, no one has asked this.
I've been asked to compute the double integral of a function, and also the same double integral but with the order of integration swapped (i.e: first integrate for dydx, then dxdy). Here is my code:
%Define function to be integrated
f = #(x,y) y^2*cos(x);
%First case. Integration order: dydx
ymin = #(x) cos(x);
I = integral2(f,ymin,1,0,2*pi)
%Second case. Integration order: dxdy
xmin = #(y) asin(y)+2*pi/2;
xmax = #(y) asin(y)-pi/2;
B = integral2(f,xmin,xmax,-1,1)
The error I'm getting is this:
Error using integral2 (line 71)
XMIN must be a floating point scalar.
Error in EngMathsA1Q1c (line 5)
I = integral2(f,ymin,1,0,2*pi)
I'm sure my mistake is something simple, but I've never used Integral2 before and I'm lost for answers. Thank you.
Per the integral2 documentation, the variable limits are given as the second pair of limits. So your first integral should be
% Define function to be integrated
f = #(x,y) y.^2.*cos(x);
% First case. Integration order: dydx
ymin = #(x) cos(x);
I = integral2(f,0,2*pi,ymin,1);
The set of constant limits always goes first, and Matlab assumes the first argument of f is associated with the first set of limits while the second argument of f is associated with the second set of limits, which may be a function of the first argument.
I point out that second part because if you wish to switch the order of integration, you also need to switch the order of the inputs of f accordingly. Consider the following example:
fun = #(x,y) 1./( sqrt(2*x + y) .* (1 + 2*x + y).^2 )
A nice little function that is not symmetric in its arguments (i.e., fun(x,y) ~= fun(y,x)). Let's integrate this over an elongated triangle in the first quadrant with vertices at (0,0), (2,0), and (0,1). Then integrating with dA == dy dx, we have
>> format('long');
>> ymax = #(x) 1 - x/2;
>> q = integral2(fun,0,2,0,ymax)
q =
0.220241017339352
Cool. Now let's integrate with dA == dx dy:
>> xmax = #(y) 2*(1-y);
>> q = integral2(fun,0,1,0,xmax)
q =
0.241956050772765
Oops, that's not equal to the first calculation! That's because fun is defined with x as the first argument and y as the second, but the previous call to integral2 is implying that y is the first argument to fun, and it has constant limits of 0 and 1. How do we fix this? Simply define a new function that flips the arguments:
>> fun2 = #(y,x) fun(x,y);
>> q = integral2(fun2,0,1,0,xmax)
q =
0.220241017706984
And all's right with the world. (Although you may notice small differences between the two correct answers due to the error tolerances of integral2, which can be adjusted via options per the documentation.)
The error states that you can't pass in a function for the limits of integration. You need to specify a scalar value for each limit of integration. Also, there are some errors in the dimensions/operations of the function. Try this:
%Define function to be integrated
f = #(x,y) y.^2.*cos(x);%changed to .^ and .*
%First case. Integration order: dydx
%ymin = #(x) cos(x);
I = integral2(f,-1,1,0,2*pi)%use scalar values for limits of integration
%Second case. Integration order: dxdy
%xmin = #(y) asin(y)+2*pi/2;
%xmax = #(y) asin(y)-pi/2;
B = integral2(f,0,2*pi,-1,1)% same issue, must use scalars

Monte Carlo integration of exp(-x^2/2) from x=-infinity to x=+infinity

I want to integrate
f(x) = exp(-x^2/2)
from x=-infinity to x=+infinity
by using the Monte Carlo method. I use the function randn() to generate all x_i for the function f(x_i) = exp(-x_i^2/2) I want to integrate to calculate afterwards the mean value of f([x_1,..x_n]). My problem is, that the result depends on what values I choose for my borders x1 and x2 (see below). My result is going far away from the real value by increasing the value of x1 and x2. Actually the result should be better and better by increasing x1 and x2.
Does someone see my mistake?
Here is my Matlab code
clear all;
b=10; % border
x1 = -b; % left border
x2 = b; % right border
n = 10^6; % number of random numbers
x = randn(n,1);
f = ones(n,1);
g = exp(-(x.^2)/2);
F = ((x2-x1)/n)*f'*g;
The right value should be ~2.5066.
Thanks
Try this:
clear all;
b=10; % border
x1 = -b; % left border
x2 = b; % right border
n = 10^6; % number of random numbers
x = sort(abs(x1 - x2) * rand(n,1) + x1);
f = exp(-x.^2/2);
F = trapz(x,f)
F =
2.5066
Ok, lets start with writing of general case of MC integration:
I = S f(x) * p(x) dx, x in [a...b]
S here is integral sign.
Usually, p(x) is normalized probability density function, f(x) you want to integrate, and algorithm is very simple one:
set accumulator s to zero
start loop of N events
sample x randomly from p(x)
given x, compute f(x) and add to accumulator
back to start loop if not done
if done, divide accumulator by N and return it
In simplest textbook case you have
I = S f(x) dx, x in [a...b]
where it means PDF is equal to uniformly distributed one
p(x) = 1/(b-a)
but what you have to sum is actually (b-a)*f(x), because your integral now looks like
I = S (b-a)*f(x) 1/(b-a) dx, x in [a...b]
In general, if both f(x) and p(x) could serve as PDF, then it is matter of choice whether you integrate f(x) over p(x), or p(x) over f(x). No difference! (Well, except maybe computation time)
So, back to particular integral (which is equal to \sqrt{2\pi}, i believe)
I = S exp(-x^2/2) dx, x in [-infinity...infinity]
You could use more traditional approach like #Agriculturist and write it
I = S exp(-x^2/2)*(2a) 1/(2a) dx, x in [-a...a]
and sample x from U(0,1) in [-a...a] interval, and for each x compute exp() and average it and get the result
From what I understand, you want to use exp() as PDF, so your integral looks like
I = S D * exp(-x^2/2)/D dx, x in [-infinity...infinity]
PDF to be normalized so it shall include normalization factor D, which is exactly equal to \sqrt{2 \pi} from gaussian integral.
Now f(x) is just a constant equal to D. It doesn't depend on x. It means that you for each sampled x should add to accumulator a CONSTANT value of D. After running N samples,
in accumulator you'll have exactly N*D. To find mean you'll divide by N and as a result you'll get perfect D, which is \sqrt{2 \pi}, which, in turn, is
2.5066.
Too rusty to write any matlab, and Happy New Year anyway

Plot symbolic equation using standard plot function in Matlab

In order to obtain a graphical representation of the behaviour of a fluid it is common practice to plot its streamlines.
For a given two-dimensional fluid with speed components u = Kx and v = -Ky (where K is a constant, for example: K = 5), the streamline equation can be obtained integrating the flow velocity field components as follows:
Streamline equation: ∫dx/u = ∫dy/v
The solved equation looks like this: A = B + C (where A is the solution of the first integral, B is the solution of the second integral and C is an integration constant).
Once we have achieved this, we can start plotting a streamline by simply assigning a value to C, for example: C = 1, and plotting the resulting equation. That would generate a single streamline, so in order to get more of them you need to iterate this last step assigning a different value of C each time.
I have successfully plotted the streamlines of this particular flow by letting matlab integrate the equation symbolically and using ezplot to produce a graphic as follows:
syms x y
K = 5; %Constant.
u = K*x; %Velocity component in x direction.
v = -K*y; %Velocity component in y direction.
A = int(1/u,x); %First integral.
B = int(1/v,y); %Second integral.
for C = -10:0.1:10; %Loop. C is assigned a different value in each iteration.
eqn = A == B + C; %Solved streamline equation.
ezplot(eqn,[-1,1]); %Plot streamline.
hold on;
end
axis equal;
axis([-1 1 -1 1]);
This is the result:
The problem is that for some particular regions of the flow ezplot is not accurate enough and doesn't handle singularities very well (asymptotes, etc.). That's why a standard "numeric" plot seems desirable, in order to obtain a better visual output.
The challenge here is to convert the symbolic streamline solution into an explicit expression that would be compatible with the standard plot function.
I have tried to do it like this, using subs and solve with no success at all (Matlab throws an error).
syms x y
K = 5; %Constant.
u = K*x; %Velocity component in x direction.
v = -K*y; %Velocity component in y direction.
A = int(1/u,x); %First integral.
B = int(1/v,y); %Second integral.
X = -1:0.1:1; %Array of x values for plotting.
for C = -10:0.1:10; %Loop. C is assigned a different value in each iteration.
eqn = A == B + C; %Solved streamline equation.
Y = subs(solve(eqn,y),x,X); %Explicit streamline expression for Y.
plot(X,Y); %Standard plot call.
hold on;
end
This is the error that is displayed on the command window:
Error using mupadmex
Error in MuPAD command: Division by zero.
[_power]
Evaluating: symobj::trysubs
Error in sym/subs>mupadsubs (line 139)
G =
mupadmex('symobj::fullsubs',F.s,X2,Y2);
Error in sym/subs (line 124)
G = mupadsubs(F,X,Y);
Error in Flow_Streamlines (line 18)
Y = subs(solve(eqn,y),x,X); %Explicit
streamline expression for Y.
So, how should this be done?
Since you are using subs many times, matlabFunction is more efficient. You can use C as a parameter, and solve for y in terms of both x and C. Then the for loop is very much faster:
syms x y
K = 5; %Constant.
u = K*x; %Velocity component in x direction.
v = -K*y; %Velocity component in y direction.
A = int(1/u,x); %First integral.
B = int(1/v,y); %Second integral.
X = -1:0.1:1; %Array of x values for plotting.
syms C % C is treated as a parameter
eqn = A == B + C; %Solved streamline equation.
% Now solve the eqn for y, and make it into a function of `x` and `C`
Y=matlabFunction(solve(eqn,y),'vars',{'x','C'})
for C = -10:0.1:10; %Loop. C is assigned a different value in each iteration.
plot(X,Y(X,C)); %Standard plot call, but using the function for `Y`
hold on;
end

Making a function in terms of a sum from 1 to n in Matlab

I'm trying to get Matlab to take this as a function of x_1 through x_n and y_1 through y_n, where k_i and r_i are all constants.
So far my idea was to take n from the user and make two 1×n vectors called x and y, and for the x_i just pull out x(i). But I don't know how to make an arbitrary sum in MATLAB.
I also need to get the gradient of this function, which I don't know how to do either. I was thinking maybe I could make a loop and add that to the function each time, but MATLAB doesn't like that.
I don't believe a loop is necessary for this calculation. MATLAB excels at vectorized operations, so would something like this work for you?
l = 10; % how large these vectors are
k = rand(l,1); % random junk values to work with
r = rand(l,1);
x = rand(l,1);
y = rand(l,1);
vals = k(1:end-1) .* (sqrt(diff(x).^2 + diff(y).^2) - r(1:end-1)).^2;
sum(vals)
EDIT: Thanks to #Amro for correcting the formula and simplifying it with diff.
You can solve for the gradient symbolically with:
n = 10;
k = sym('k',[1 n]); % Create n variables k1, k2, ..., kn
x = sym('x',[1 n]); % Create n variables x1, x2, ..., xn
y = sym('y',[1 n]); % Create n variables y1, y2, ..., yn
r = sym('r',[1 n]); % Create n variables r1, r2, ..., rn
% Symbolically sum equation
s = sum((k(1:end-1).*sqrt((x(2:end)-x(1:end-1)).^2+(y(2:end)-y(1:end-1)).^2)-r(1:end-1)).^2)
grad_x = gradient(s,x) % Gradient with respect to x vector
grad_y = gradient(s,y) % Gradient with respect to y vector
The symbolic sum and gradients can be evaluated and converted to floating point with:
% n random data values for k, x, y, and r
K = rand(1,n);
X = rand(1,n);
Y = rand(1,n);
R = rand(1,n);
% Substitute in data for symbolic variables
S = double(subs(s,{[k,x,y,r]},{[K,X,Y,R]}))
GRAD_X = double(subs(grad_x,{[k,x,y,r]},{[K,X,Y,R]}))
GRAD_Y = double(subs(grad_y,{[k,x,y,r]},{[K,X,Y,R]}))
The gradient function is the one overloaded for symbolic variables (type help sym/gradient) or see the more detailed documentation online).
Yes, you could indeed do this with a loop, considering that x, y, k, and r are already defined.
n = length(x);
s = 0;
for j = 2 : n
s = s + k(j-1) * (sqrt((x(j) - x(j-1)).^2 + (y(j) - y(j-1)).^2) - r(j-1)).^2
end
You should derive the gradient analytically and then plug in numbers. It should not be too hard to expand these terms and then find derivatives of the resulting polynomial.
Vectorized solution is something like (I wonder why do you use sqrt().^2):
is = 2:n;
result = sum( k(is - 1) .* abs((x(is) - x(is-1)).^2 + (y(is) - y(is-1)).^2 - r(is-1)));
You can either compute gradient symbolically or rewrite this code as a function and make a standard +-eps calculation. If you need a gradient to run optimization (you code looks like a fitness function) you could use algorithms that calculate them themselves, for example, fminsearch can do this