frequency range of matching
frequency range of matching
I have been trying to match with high-frequency content up to 100Hz, but I could not get the output more than 50Hz Even if the design spectrum has high-frequency content. Can anybody fix this? or Is it the maximum limit of frequency in SeismoMatch? OR Do we need to buy commercial SeismoMAtch to get the higher range of Frequency?
- seismosoft
- Posts: 1192
- Joined: 06 Jul 2007, 04:55
Re: frequency range of matching
These are standard limits of SeismoMatch both for the academic and the commercial releases.
Seismosoft Support
Seismosoft Support
- seismosoft
- Posts: 1192
- Joined: 06 Jul 2007, 04:55
Re: frequency range of matching
Note however that the minimum matching period in SeismoMatch is not 0.02 sec by default.
You get this limitation because the minimum period in Program Settings>Response Spectra is probably 0.02 sec. If you lower this value (setting the period step to 0.01), you will then be able to set the minimum matching period to 0.01sec.
Seismosoft Support
You get this limitation because the minimum period in Program Settings>Response Spectra is probably 0.02 sec. If you lower this value (setting the period step to 0.01), you will then be able to set the minimum matching period to 0.01sec.
Seismosoft Support
Re: frequency range of matching
actually, I have tried up to a min period of 0.005sec but still, the peak is at 50Hz. when I tried min period 0.001sec and max period 0.3sec the software gives a message saying more than 400 period -acceleration pairs of values. May I know any other solution or ideas?
Re: frequency range of matching
Bidhek, what is the actual range of matching you are trying to achieve? If you try a very narrow range centered at 0.01 seconds, then the peak of the seed accelerogram, if it has a peak beyond the range, may be relatively unaffected. So you could end up with a close match at 0.01 seconds, but not a peak at 0,01 seconds.
On the other hand, if you have a wide range of matching specified, then you may end up with a more relaxed match at 0.01 seconds, but a closer overall match in the range.
You may wish to investigate SeismoArtif to generate synthetics and make adjustments in the frequency domain instead of the time domain.
Finally, without a detailed knowledge of your goals, it is somewhat difficult to offer up additional suggestions. I will say that real accelerograms typically have peak spectral accelerations well beyond 0.01 seconds, even when they do possess significant high frequency content. That is my rationale for suggesting synthetics altered in the frequency domain. And the resulting records might be unrealistic. But you could give it a try.
I hope this helps. I'll be interested in hearing how it goes.
On the other hand, if you have a wide range of matching specified, then you may end up with a more relaxed match at 0.01 seconds, but a closer overall match in the range.
You may wish to investigate SeismoArtif to generate synthetics and make adjustments in the frequency domain instead of the time domain.
Finally, without a detailed knowledge of your goals, it is somewhat difficult to offer up additional suggestions. I will say that real accelerograms typically have peak spectral accelerations well beyond 0.01 seconds, even when they do possess significant high frequency content. That is my rationale for suggesting synthetics altered in the frequency domain. And the resulting records might be unrealistic. But you could give it a try.
I hope this helps. I'll be interested in hearing how it goes.
Tim Huff
Re: frequency range of matching
thank you Mr.Tim, I will try your suggestion and let you know. Actually, I am trying to generate high-frequency accelerogram which has a frequency higher than 100Hz, and I want to compare the properties with a real accelerogram, whether it contains similar properties of real accelerograms.
Re: frequency range of matching
Mr.Tim, my desired range of frequency is 70-150Hz.
Re: frequency range of matching
Mr.Tim, it couldn't give the desired frequency range of 100-150Hz, and when I upload the target spectrum at min period of 0.01 sec, even the original frequency domain response spectrum doesn't show the same spectrum which it has to be, but instead, it plots different the frequency domain response spectra of the target spectrum having a smaller frequency range for 30-60Hz.
Re: frequency range of matching
I am not surprised that you are having difficulties in trying to alter real earthquake accelerograms to have a peak in the 100-150Hz range. Which seed accelerograms are you using? You should probably be trying accelerograms recorded on hard rock stations as they will have higher frequency content to begin with (but most likely still significantly less than 100-150 Hz). Or, generate synthetics in SeismoArtif for hard rock conditions and try different magnitudes and distances. If you want to send me your Match and Artif files (or, at least your target spectrum) I will take a look as I find the time.
Tim Huff