A control algorithm has been developed for controlling Active Variable Stiffness (AVS) structures. This algorithm analyses information of an observed seismic excitation, estimates the future structural responses and determines how to alter the structure stiffness. An objective structure is assumed to possess N on-off elements whose states are controlled by the proposed algorithm. That is, at a given moment tk, (1) seismic excitation information is expressed by an Auto Regressive (AR) model as the identification model; (2) future excitation information is predicted using the AR model; (3) future responses due to predicted excitation are estimated; (4) based on the initial condition at tk, the responses of 2N possible structural states from tk, to tk+L are calculated; (5) the state which minimizes the input energy during tL is selected; and (6) immediately after tk, on-off elements are set up and subjected to the selected states. The effectiveness of the induced algorithm is confirmed by numerical experiments on a model of a three-storey building under sine and seismic excitations.
[1]
T. T. Soong,et al.
Experimental Study of Active Control for MDOF Seismic Structures
,
1989
.
[2]
J. Burg.
THE RELATIONSHIP BETWEEN MAXIMUM ENTROPY SPECTRA AND MAXIMUM LIKELIHOOD SPECTRA
,
1972
.
[3]
R. M. Oliver,et al.
Simulating and analyzing artificial nonstationary earthquake ground motions
,
1982
.
[4]
N. Andersen.
On the calculation of filter coefficients for maximum entropy spectral analysis
,
1974
.
[5]
M. Hoshiya,et al.
Adaptive Identification of Autoregressive Processes
,
1991
.
[6]
Masanobu Shinozuka,et al.
ARMA Representation of Random Processes
,
1985
.
[7]
Takuji Kobori,et al.
Seismic response controlled structure with Active Variable Stiffness system
,
1993
.
[8]
Takuji Kobori,et al.
Seismic‐response‐controlled structure with active mass driver system. Part 2: Verification
,
1991
.