Receiving Errors On Matrix Dimensions

Hello, I am unsure if I am making a simple mistake but I am receiving errors for matrix size within my equation code. Here is the code:

for 
Z = 0.1

W_n = 2.5;

B = sqrt(1 - Z^2);

Theta = atan(B/Z);

c(t) = 1 - (1/B) * exp(1) .^ -(Z * W_n * t) * sin(B * W_n * t + Theta);

end

I am receiving these error messages:

Error using * Inner matrix dimensions must agree.

Error in FeedbackControlSystem (line 65) c(t) = 1 - (1/B) * exp(1) .^ -(Z * W_n * t) * sin(B * W_n * t + Theta);

Error using ^ One argument must be a square matrix and the other must be a scalar. Use POWER (.^) for elementwise power.

Error in FeedbackControlSystem (line 65) c(t) = 1 - (1/B) * exp(1)^-(Z * W_n * t) * sin(B * W_n * t + Theta);

Any help would be greatly appreciated. Thank you!

[KSSV]

What is t here?

[Veera Kanmani]

c(t) = 1 - (1/B) * exp(1) .^ -(Z * W_n * t) * sin(B * W_n * t + Theta);.. i think the problem occurs, can you specify code in detail manner and what is t value ?

[Jacob Hammer]

This is the updated code: (Also I apologize for the response time)

for

t = linspace(1,3600,360000)

c = zeros(1,360000);

Z = 0.1;

W_n = 2.5;

B = sqrt(1 - Z^2);

Theta = atan(B/Z);

 c(t) = 1 - ((1/B) * (exp(1) .^ -(Z * W_n .* t)) * (sin(B * W_n .* t + Theta)));

end

I am trying to simulate a feedback control system and c(t) was a given equation.

These are the variables:

B = Beta % Beta value will change with each section

% B = sqrt(1 - Z^2)

% Z = Zeta % Zeta value will change with each section

% W_n = natural frequency % Natural frequency will change with each % section

% Theta = theta % Theta in radians

% Theta = atan(B/Z)

% tan(Theta) = B/Z

[Jacob Hammer]

Updated Error Prompt:

>> FeedbackControlSystem Subscript indices must either be real positive integers or logicals.

Error in FeedbackControlSystem (line 67) c(t) = 1 - ((1/B) * (exp(1) .^ -(Z * W_n .* t)) * (sin(B * W_n .* t + Theta)));

[Jacob Hammer]

[Jacob Hammer]

[Jacob Hammer]

Answer 1

[Ameer Hamza]

From your code in the comment section, It can be seen that you are making a couple of mistakes.

• t contain non-integers and thus can't be used to index c. In fact in this case you don't need to initialize c with zeros since it will not bring any speed up. Pre-allocation is only useful if your array size is changing in every loop iteration, here you can avoid that as given in below code snippet.
• Not taking care of matrix dimension when multiplying. In MATLAB * is for matrix multiplication and .* is for element-wise multiplication. (exp(1) .^ -(Z * W_n .* t)) is 1*360000 and (sin(B * W_n .* t + Theta))) is also 1*360000. Therefore you need .* operation between them.

t = linspace(1,3600,360000);
Z = 0.1;
W_n = 2.5;
B = sqrt(1 - Z^2);
Theta = atan(B/Z);
c = 1 - ((1/B) * (exp(1) .^ -(Z * W_n .* t)) .* (sin(B * W_n .* t + Theta)));

[Jacob Hammer]

When using this specific code, c yields 1. Shouldn't c yield a matrix of the same dimensions as t?

[Jacob Hammer]

Correction: When eliminating the for loop, c yields a matrix of the same dimensions as t, but consists of only 1's

[Jacob Hammer]

Pardon the second correction but I just realized my time vector is much too large to see the actual fluctuations within the plot. Thank you for your help.

[Ameer Hamza]

Yes, t and c have same dimensions. But the range of t is rather too large to see anything useful in plot. You might want to reduce the range of t.