The Earthquake Engineering Online Archive

Investigation of the nonlinear characteristics of a three story steel frame using system identification

Kaya, Izak; McNiven, Hugh D.

UCB/EERC-78/25, Earthquake Engineering Research Center, University of California, Berkeley, 1978, 93 pages (555.3/K3/1978N)

This study describes two mathematical models designed to predict the responses to seismic disturbance of a three-story steel frame. This paper is an extension of previously reported work on the construction of mathematical models to predict the linear response of the frame. This study considers only the manner by which the previous models can be extended to accommodate nonlinear responses. This extension consists of accounting for changes in the damping coefficients when the amplitudes of motion are large, and for yielding by including the hysteretic behavior of the ends of the members. An important decision was to model the hysteretic relationship between the bending moments and the rotations at the ends of the members. It was decided to allow the relationship to be bilinear, partly because this form has not been used before and partly because it was considered simplest. With the introduction of hysteretic material behavior, three new parameters are added to the eight carried over from the linear model, so that the new nonlinear models contain eleven parameters. For economy, two of the parameters in each model were fixed, leaving the remaining nine to be determined from optimization. The difference between the two models is in the choice of which two of the parameters are fixed. In the first, the values of the damping parameters are fixed by giving them the values found after optimization for the linear model. In the second model, the parameters representing the yield moments in the bilinear model are fixed, one each for the columns and girders, and the damping parameters are found from optimization. Both of the models predict the experimental time histories of the floor translations and joint rotations very accurately.

Available online: http://nisee.berkeley.edu/documents/EERC/EERC-78-25.pdf (6 MB)