I have read the scaling criteria of ASCE 7-10, which recommends that the group of selected acceleration records must be above the design response spectrum from 0.2T to 1.5T, where T is the fundamental period of the structure. However, as I know design spectra are made up to ensure the structure's seismic behavior reaches life safety performance level and not to reach collapse prevention.
So do you know a method to scale a record for collapse prevention damage level? If this method is in a standard, can you tell me which standard?
The hazard define in ACSE 7-10 is risk-targeted maximum earthquake level (MCER) and not the design earthquake (DE) level any more. The DE = 2/3 (MCER).....
Please make sure to scale your ground motions to MCER level spectrum to achieve your goals unless your doing IDA analysis.
According to Performance based design matrix (SEAOC 1999) , under MCER motions structure should be in the near collapse state.
Hope this will be useful
The scaling process is in general the same, i.e. you have to compute the spetrum of the record and compare the spectral accelerations within the given period range. I believe your problem is rather to define what is the target spectrum, as you are interested in a different return period than the one defined in the code (500 years) . The collapse prevention performance is related to the very rare event, i.e. a return period of 2500 years, hence you should increment the spectrum of the code. In the ACSE type of spectrum this is obtained by substituting the coefficient 2/3 by 1 in the spectrum expressions (i.e. multiply by 1.5).
However, if you are interested in different return periods you should look for the relation between hazard intensity (say acceleration) and the return period. In Eurocode 8-part 1 you may find an expression relating the acceleration for a different return period with the one for a reference return period as follows:
(a(T)/a(T_reference)=(T_reference/T)^(-1/k)
Here k depends on the local seismicity, and it is stated that is range of 3.
If you eliminate the 2/3 term from the ACSE spectrum it will be like considering k=4.
The only possible way is tu generate an satatistical model, in which you take in consideration te probable magnitude, the distance to the epicenter, and the
direction beteen the epicenter and the construction you want to project. The model shoul be based on measured acceleration records, obtained at the site.
Scaling an earthquake record to take it from DBE to MCE is not advisable, as there are some qualitative differences as the magnitude increases (e.g. frequency content and duration). In addition, if it is a near-fault record, the frequency of any velocity pulse will also reduce. Scaling can be carried out in different ways, and each will affect the result in different ways. Your best advice is to consult a geoseismic specialist. Alternatively, do a web-search for RSPmatch if you want to find out more about how many people are carrying out spectrum matching.
You should also note that the two-thirds relationship going from MCE to DBE is not an absolute relationship. For example, in India the code defines it as being half. In general, in more seismic areas, such as California, the DBE will be a larger proportion of the MCE. This means that if you are designing for seismic in New York and you apply the two-thirds factor as required by code, your DBE is actually for a less frequent earthquake than the DBE in California (i.e. more than 500 year return period).
Finally - you have not provided any information on why you want to do this. The answer will depend on what you are trying to do (and where you a trying to do it). If this is for a real design of an important building, then you should carry out the full seismic hazard studies. If it is just an exercise, then you can simply scale the magnitude according to the other suggestions, but you should be aware that the results will not be rigorous. An intermediate approach would be to use RSPmatch using seed ground motions that match the seismicity of your site (similar magnitude and characteristics).
Dear Andrew Mole,
First thank you for your generous answering. I want to be sure that under all earthquakes that i have my structure will undergo damages corresponding to collapse prevention performance level.
Dear Arman,
I suggest that you take a look at the following article in Earthquake Spectra (EERI), Vol 28, No. 3, "Modal Pushover-Based Scaling of Two Components of Ground Motion Records for Nonlinear RHA of Structures", by Juan Reyes and Anil Chopra. It starts off with a good overview, and it has some good references that you can follow for more information.
You can use ASCE7-10 for MCE, but the problem is that it gives vague (not specific) criteria for selecting the seed ground motions and no information on how to modify them. This is an area of active research and not at all settled. RSPmatch is one of the leading tools for this, though.
You said that you want to be "sure". This is not an option for earthquake engineering (and not really anywhere in life). All you can do is choose a level of confidence in your results, and the level of confidence depends upon the amount of work you are prepared to put into the exercise and this will depend upon the value and complexity of the project. This is why my answers are general and not specific. I do not have enough information from you to provide anything else. Also, the ground motions that you select will depend upon the location of your project and the type of structure. If it is a tall structure, then you need to watch out for the effect of the ground motion on the higher modes. If you have a near-fault condition, then you will need to think about including velocity pulses and making sure that they are a suitable duration.
In conclusion, if you really want to do this at the highest level, then you will need to either get an expert to help you, or learn a lot more about ground motions and their effects on buildings. The article I referenced above should help you in the right direction.
This report may also help you, which is available online: http://pubs.usgs.gov/of/2010/1068/ "Practical Guidelines to Select and Scale Earthquake Records for Nonlinear Response History Analysis of Structures" (USGS).
One other report, which is probably the most comprehensive that I have come across: NIST GCR 11-917-15, "Selecting and Scaling Earthquake Ground Motions for Performing Response-History Analyses" (https://www.atcouncil.org/files/NEHRPcjv/ATC-82_NIST_GCR_11-917-15_GM_Selection.pdf).
- Badges
- Science topic
- Similar topics
- Biological Science
- Histology
- Sectioning