I integrated fault current which is sine wave with its time period as an interval. But, i am getting value as an answer instead of getting 0.

F=0.05
ix = (Phasecurrent.Time >= F) & (Phasecurrent.Time <=F+0.020047); 
Int = cumtrapz(Phasecurrent.Data(ix))
vpa(Int(end)) 

ans =

137031976.93982961773872375488281

[Paul]

What does this command show:

Int = cumtrapz(PhaseCurrent.Time(idx),Phasecurrent.Data(ix))

[Dyuman Joshi]

In addition to Madhan's answer, also note that cumtrapz() is a numerical approximation instead of a proper integral and the data points you have do not seem to be smooth enough - there are slight unsmoothness (for a lack of a better phrase) near the maximas and minimas.

So even after using correct syntax, the answer might not be exactly 0.

[A Lumbini]

Can't we implement proper integral in matlab

[Paul]

Why not just implement it in Simulink? Use a Clock, two Compare to Constant blocks, and an AND to output signal that is 1 (or true) over the desired interval and 0 otherwise. Multiply that signal with signal to be integrated and send that product into an Integrator. Make sure to use a variable step solver with zero crossing detection turned on, at least for the Compare to Constant and AND blocks, if not for the whole model.

Answer 1

[madhan ravi]

Not sure what you are doing with your code. When using cursor measurements the 2 cursor points should measure same value as 0. DeltaY should be 0. But in the screenshot you shared it’s not 0. As you can see at .05 the lower half sine wave is not exactly a sine wave , the shape of the u curve is not the same as the others.

As from your previous question, it was mentioned that continuous method is being used. Hence the correct syntax is:

 Int = cumtrapz(PhaseCurrent.Time(ix),Phasecurrent.Data(ix))

[A Lumbini]

Sorry maybe i didn't select proper zero crossings to find time period

[A Lumbini]

ix = (Phasecurrent.Time >= F) & (Phasecurrent.Time <=F+0.02);

Int = cumtrapz(Phasecurrent.Time(ix),Phasecurrent.Data(ix))

vpa(Int(end))

command window:

Int =

1.0e+06 *

0

-0.0000

-0.0668

-0.1525

-0.2532

-0.3646

-0.4778

-0.4778

-0.5955

-0.7086

-0.8122

-0.9017

-0.9731

-1.0232

-1.0499

-1.0532

-1.0532

-1.0541

-1.0541

-1.0532

-1.0351

-0.9941

-0.9320

-0.8514

-0.7556

-0.6488

-0.5354

-0.5085

-0.5085

-0.3936

-0.2829

-0.1811

-0.0924

-0.0206

0.0312

0.0609

0.0670

0.0670

0.0679

0.0679

0.0670

0.0489

0.0079

ans =

7936.6168438080712803639471530914

from cursor measurements i have found that the time period is 20ms

i am integrating from 0.05 to (0.05+20ms) the value i am getting is not even close to zero is it just because i am using numerical integration due to which it will not be pure sinewave

[madhan ravi]

"Sorry maybe i didn't select proper zero crossings to find time period"

Yes.

"i am integrating from 0.05 to (0.05+20ms) the value i am getting is not even close to zero is it just because i am using numerical integration due to which it will not be pure sinewave"

Like I said

"But in the screenshot you shared it’s not 0. As you can see at .05 the lower half sine wave is not exactly a sine wave , the shape of the u curve is not the same as the others."

Try choosing the next interval. Beware in the y scale axis, it is 10^8

.

Sorry you need to figure it out by yourself. I don't have time to analyze in detail.

[A Lumbini]

k thank you sir