Seismic Performance analysis based on existing quake record.

[JoeH78]

Dear All,

I've done performance analysis on SAP2000 for the exisiting RC structure which has undergone a quake, the purpose of that performance analysis was to determine the servicability, current damaged state of the structure and to determine whether the plastic mechanism will develop or not in any member.

Since it wasn't a design, I've applied the seismic ground motion as a time history acceleration in three principal direction of structure( at the attachement you can find a copy of seismograph record) which has peak ground acceleration of 479gal in U-D direction.

Material

Concrete C30​

Isotropic​

Takeda hysterisis cycles, Mander type of stress-strain curve, unconfined rectangular section​

Rebar​

A992Fy50, Isotropic​

Kinematic hysterisis cycles, Simple type of stress-strain curve​

Elements

Layered non linear thick shell element​

Two layer of vertical rebars​

Set "local redistrubition method" for nodes where elastic to plastic transition occrus.​

After analysis s/w reports the S11,S22, S12 etc.. stresses but not the Von misses.

Based on the description above how can I determine that plastic mechanism occurs and exact location of it in structure?

 

[ahmedhegazi]

Performance based design remains a vague part of structural engineering till the moment for most engineers ( i bet more than 90%) , if you need more information on that , you can read ASCE41-06 ( there i new edition 2013) and TBI 2010

I was looking for information about this topic in the forum but was not successful , I believe in next 5 or 10 years it will become the common way for seismic design of buildings to avoid design codes structural systems limitations

to answer your question , i presume your lateral force resisting system is shear wall , in this case you do not check stress level , what you look at is mainly Performance classification for wall segments according to rotation and drifts and you look at overall building drift as well.
SAP 2000 has a good tutorial video for non linear analysis of shear wall , you can find it on youtube

hope that helps

 

[MotorCity]

How are you accounting for and modeling damaged members in your model?

Depending on the number of damaged members and the degree of damage, it seems like it would be a nearly impossible task if you are trying to model each defect.

Is all of the damage visible?

If there is significant damage, the plastic hinge may have already occurred.

 

[JoeH78]

@ahmedhegazi
thanks for the comment and code entry-point I'll have a look at them. I completely agree with you that its bit of vague subject on itself, there is not exact criteria on how you will determine the collapsing occurs or not, steel may yield concrete may crush but that gives you a hint in theoritical level, not totally what will happen with it in pratic or in real case.

As you correctly guessed, resisting system is totally composed of shear walls, I'll check relative and overall story drift levels but I think that getting the response plot(stress/strain, force/displacement curve) of most strained(by guessign tentatively) nodes/members will give me more clue on what is happening at the internals. For example, if plastic state occurs simulateneously in rebar and concrete more than 3 times in same location then its inevitable that in that location collapsing will develop or at the verge of it.

@MotorCity
For the damaged elements, the residents and govermental auth. have taken the concrete samples from structure and I'm struggling to find that report but for now I only judge by the visual observation of the damaged element and I agree that its not quite engineering-wise approach. For the damaged elements I take the young modulus as half of the undamaged member with this I'm hoping that it will give some info on how mechanism occurs at least.

Regards,