PGD Fragility Fuctions

[liliana.ol...@gmail.com]

Hello everyone!

I have a question. How can I compute Classical Damage Calculator with Fragility Functions in PGD and with BooreAtkinson2008 GSIM? This GSIM only accept PGA data, I guess. I want to compute it with PGD data. How can I do that?

Thank you,
Liliana Oliveira.

[Vitor Silva]

Dear Liliana, 

Thank you for your message. 

By PGD do you mean peak ground displacement or permanent ground deformation? If the former, then BooreAtkinson2008 cannot be used (it supports PGV, PGA and SA - http://docs.openquake.org/oq-hazardlib/master/openquake.hazardlib.gsim.html#module-openquake.hazardlib.gsim.boore_atkinson_2008). You can consider CampbellBozorgnia2008 which supports peak ground displacement (http://docs.openquake.org/oq-hazardlib/master/openquake.hazardlib.gsim.html#module-openquake.hazardlib.gsim.campbell_bozorgnia_2008).

If you want permanent ground deformation (for example, for the assessment of damage or losses due to liquefaction or landslides), then the problem is much more complex and it is not yet supported by OpenQuake.

Best regards,

Vitor

--

You received this message because you are subscribed to the Google Groups "OpenQuake Users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to openquake-use...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Liliana Oliveira]

What I mean is that my fragility curves dependes on SD [cm]. My fragility curves are defined by a lognormal distribution function conditional on the maximum response SDmax of the building and the damages states are defined accordingly.

But my seismic action is defined using attenuation laws in SA (cm/s^2).

I am missing a link between hazard output in acceleration and the calculation of my damage states. I can only run openquake properly when I use attenuation laws that support ground displacement, which I really wanted to avoid.
Mostrar citação

[Vitor Silva]

Dear Liliana,

Thank you for your message.

If your fragility functions are in terms of inelastic spectral
displacement (thus defined in terms of an EDP and not IM), you would need
to use a GMPE expressed in this parameter, which is not currently
available in OpenQuake. Moreover, note that you would need to perform an
iterative process since the level of ductility is not known a-priori (it
depends on the expected damage).

If on the other hand your functions are expressed in terms of damped
elastic spectral displacement, then you can easily cover them to spectral
acceleration knowing the period of vibration of the structure (and use the
usual GMPEs for your region of interest).

Best regards,
Vitor


On 09/05/2017, 12:32, "Liliana Oliveira" <openqua...@googlegroups.com

on behalf of liliana.ol...@gmail.com> wrote: