Matching Time History

03-Unexpected behaviour/errors
Post Reply
cagatay
Posts: 3
Joined: 21 Feb 2023, 09:10

Matching Time History

Post by cagatay »

Hello,

I am trying to match selected ground motions from PEER NGA West 2. I am using a user defined target spectrum for the matching process and every time I change the scale factor for different ground motions, program scales the target spectrum with the scale factor. I did not understand why this is happening. Can you please explain the problem here?
huffte
Posts: 987
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

Re: Matching Time History

Post by huffte »

This is happening because it is precisely what is supposed to happen when you change the scale factor. If you are simply scaling records to your target, why do you want to change the scale factor from 1.0? The scale factor option enables you, for example, to scale one component of a record pair to 1.3 times the target and the other component of a record pair to 0.8 times the target, just to give example numbers. You might do this if you were trying to preserve the relative amplitude between two components of the same station record for a 3D analytical model. Again, if you simply wish to match individual records to your input target spectrum, use a scale factor of 1.0 for all analyses.
Tim Huff
cagatay
Posts: 3
Joined: 21 Feb 2023, 09:10

Re: Matching Time History

Post by cagatay »

I'm not talking about the response spectrums of scaled ground motions, of course their response spectrums should differ from my target spectrum. What I meant is that my target spectrum doesn't stay the same, it scales too. What's the point of scaling a gorund motion if the target spectrum will change with it? I'm going to give you an example based on your numbers, I want to scale the vertical component of a ground motion with 2.1 but my target spectrum (not the response spectrum of a selected ground motion) scales with 2.1 too and the result becomes ridiculous.
huffte
Posts: 987
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

Re: Matching Time History

Post by huffte »

Cagatay, I never claimed that you were talking about the spectra of scaled motions. As I said, there are reasons one may want to scale the target design response spectrum. I won't repeat the possible example I previously gave. Perhaps you are assuming that SeismoMatch modifies ground motion by scaling when you perform the matching process. SeismoMatch does NOT scale the accelerograms (unless you specify a scale factor when you load the accelerogram). SeismoMatch matches the response spectrum of ground motions to a target (or a multiple of the target) through spectral matching using wavelets in the time domain, not by amplitude scaling.

If you want to scale an input accelerogram by 2.1 and then match it to a user defined target response spectrum, then the following should do it:

1. specify the factor of 2.1 when you load the accelerogram
2. define the target spectrum
3. keep the target spectrum scale factor equal to 1.0
4. perform the matching

I am running SeismoMatch 2021.
Tim Huff
cagatay
Posts: 3
Joined: 21 Feb 2023, 09:10

Re: Matching Time History

Post by cagatay »

Thank you for your answers. I will try it today.
Post Reply

Return to “03-Unexpected behaviour/errors”