000021800 001__ 21800
000021800 005__ 20170622131305.0
000021800 04107 $$aeng
000021800 046__ $$k2017-06-15
000021800 100__ $$aChang, Hakjong
000021800 24500 $$aSEISMIC PERFORMANCE ASSESSMENT OF STEEL MOMENT FRAME WITH VARIOUS TYPE OF STEEL CONNECTION DESIGNED BY ARTIFICIAL NEURAL NETWORK

000021800 24630 $$n6.$$pComputational Methods in Structural Dynamics and Earhquake Engineering
000021800 260__ $$bNational Technical University of Athens, 2017
000021800 506__ $$arestricted
000021800 520__ $$2eng$$aThere are many kinds of steel joints used for steel moment resisting frames such as angle connection and endplate connection. When constructing the steel structure, one of these steel joints is selected to implement the beam-column joint. The behaviors of structures with different types of joints are different. The purpose of this study is to investigate the effect if various types of steel connections on seismic performance of steel structures. Three joints, single-angle connection, single-plate, endplate connection, were applied to the steel structure. The design of the joints can be done through various models such as Morris' s bilinear model and Panel zone model. But in this study according to the type of joints, the design of steel connections was carried out by ANN(artificial neural network) using the physical properties (angle thickness, plate thickness, etc.) and actual behavior data of the joints.If the joint is designed using actual behavior data, it can be considered that it is hard to consider theoretically. The joint model through artificial neural network function derives different results depending on the type of joint. Seismic performance of steel structures was evaluated using different artificial neural network joint models. The effect of steel joint type on the seismic performance of steel structures was considered. Seismic vulnerability curves through cumulative distribution function are used for seismic performance evaluation. First, we selected 22 pairs of earthquake records presented by FEMA P695D. Normalization was performed using the earthquake data PGV (Peak Ground Velocity) to remove the uncertainty factor of the seismic data. The time incremental analysis was performed using the 22 pairs of normalized seismic data presented in FEMA P695. The performance level of the steel structure is defined as three levels(IO(Immediate Occupancy) level, LS(Life Safety) level and CP(Collapse Prevention) Level), and the fragility curve for each level is derived. The method of deriving the vulnerability curve was provided by the MAE Center. The seismic performance evaluation was carried out by calculating the fragility curve with maximum inter story displacement ratio of the steel structure derived from the time incremental analysis. In the case of endplate connection, the probability of receiving three levels of damage was the lowest. The single plate and single angle joints showed no significant difference in the probability of IO level damage, but the probability of receiving LS and CP level damage was higher than that of the single angle joint.

000021800 540__ $$aText je chráněný podle autorského zákona č. 121/2000 Sb.
000021800 653__ $$a

000021800 7112_ $$aCOMPDYN 2017 - 6th International Thematic Conference$$cRhodes Island (GR)$$d2017-06-15 / 2017-06-17$$gCOMPDYN2017
000021800 720__ $$aChang, Hakjong$$iKim, Junhee
000021800 8560_ $$ffischerc@itam.cas.cz
000021800 8564_ $$s304716$$uhttps://invenio.itam.cas.cz/record/21800/files/17976.pdf$$yOriginal version of the author's contribution as presented on CD, section: [RS19] Seismic risk and reliability analysis
.
000021800 962__ $$r21500
000021800 980__ $$aPAPER