I am trying to do a time history analysis of a Buckling Restrained Braced Frame (BRBF) using SAP2000. For time history analysis SAP2000 provides two options namely modal i.e. Fast Non Linear Analysis (FNA) and Direct Integration. I am in a dilemma regarding which one to use for the analysis. On one hand, CSI Knowledge Base claims FNA to be "generally more accurate and efficient than direct-integration time-history analysis" and on the other, on another page for Hinge FAQ it says that "FNA is based upon the mode-superposition method, intended for primarily linear-elastic systems which may have a limited number of predefined nonlinear elements."
Now I find these statements to be quite ambiguous.
https://wiki.csiamerica.com/display/kb/Comparison+between+FNA+and+direct-integration+time-history+analyses
https://wiki.csiamerica.com/display/kb/Hinge+FAQ#HingeFAQ-Whyareplastic-hingecolorsnotdisplayedwhenviewingFNAdeformation?
Also I have analysed the same frame employing both the methods, thereby getting different results. Please help.
FNA is faster (hence efficient), but only accurate for a specific range of problems - for example simple material non-linearity modeled via concentrated plasticity.
If you are introducing any other type of non-linearity (such as geometric) or more realistic (therefore more complicated) hysteretic rules you should perform direct integration.
I have successfully modeled isolation devices using FNA and it is much, much faster than direct integration. If you can limit the inelastic behavior to link-type elements as models for the BRB's, then FNA holds promise as a good choice, provided you verify results with a limited number of direct integration runs.
FNA, is generally more accurate and efficient than direct-integration time history analysis, using the default Convergence tolerance (1e-4) during FNA application, and for direct integration, a tolerance equal to or less than (1e-3).
Energy plots are available only during modal time history analysis, and not direct-integration time history and during FNA application, analytical models should:
1.Be primarily linear-elastic
2.Have a limited number of predefined nonlinear members.
3.Lump nonlinear behavior within link objects.
Dear Researchers Sushovan Dutta Georgios Baltzopoulos Timothy Huff Yasser Khademi @Ali A. Rostam Alilou
In the below link our published paper shows that for the research models you are looking for accurate results, your desire will not be met.
Article Seismic Fragility Assessment of RC Frame Equipped by Visco-E...
As I stated in my original answer, Majid Pouraminian, FNA is most effective when all of the nonlinear behavior is limited to link-type elements. The referenced paper involves hinging RC beams and columns.
Timothy Huff Sir , So incase of RCC framed Buildings , while performing NTH analysis to find the seismic demands for fragility purpose, shall we perform DI or FNA can also be done?
RCC structures typically are modeled with inelastic frame elements, not link elements. FNA may be difficult to apply in such cases as RCC structures with that modeling strategy. There may have been advances in FNA. So you may wish to do a detailed literature review to confirm my suspicions.
- Badges
- Science topic
- Similar topics
- Biological Science
- Physiology
- Physiological Processes
- Stress