Time series data problem

I'm trying to do vibration order analysis I save my data in time series from my simulation.

I want to take delta T for my waveform so when I try to do it using trigger block in a new simulation my data some how distorted.

You can see both waveform in the attachment.

How we can fix it!!!

[Mathieu NOE]

hello

in order analysis , we use a trigger to split the raw signal in buffers of equal length and do averaging, so it reduces the effect of noise and non synchronous signals

but in your case I don't understand what you are doing

do you have a simulink file to share ?

tx

[jessupj]

i've had trouble with simulink solver taking timesteps that were too large, which generated artifacts like this.

try perhaps: changing simulink solver tolerance, setting the maximum timestep to be much smalle, or toy with the timestep in a fixed increment solver until you balance step size with your tolerance for waveform irregularity.

[Adeel Ashraf]

Hello

Mathieu I'm trying to do same thing but unable to do it. You can see my simulink file in the attachment.

[Adeel Ashraf]

Jess I tried but it does not work.

[Mathieu NOE]

Hello Adeel

I tested your mdl file but it did not produce the distorded waves you showed in first instance

I just recommend to not use variable time solvers when you want to deal with triggered systems ;

I like the idea of having my simulink file functionning at fixed and deterministic time stamps

so I changed the parameters of the solver t "fixed time stamps" even I could not really notice the effect on your simulation (again, it did not reproduce the distorted waves)

[Adeel Ashraf]

Hi Methieu,

I really appreciate for your help. I tried accordingly but still the reults are not according to the expectation. I took that pulstrain signal from the incremental shaft encoder. You can see the attachment how I did it.