The application of time series analysis to describe the dynamic movements of suspension bridges


Erdogan H., Guelal E.

NONLINEAR ANALYSIS-REAL WORLD APPLICATIONS, cilt.10, ss.910-927, 2009 (SCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 10 Konu: 2
  • Basım Tarihi: 2009
  • Doi Numarası: 10.1016/j.nonrwa.2007.11.013
  • Dergi Adı: NONLINEAR ANALYSIS-REAL WORLD APPLICATIONS
  • Sayfa Sayıları: ss.910-927

Özet

Operational safety and the operational status of engineering structures depend on the specifications given during the design stage of the structure, on the observations made during the construction and which continue when the structure is in use. Furthermore, it is important that comprehensive emergency situation plans exist in case irregular movements occur. The Istanbul Bosphorus Bridge was completed in 1973 and connects Asia to Europe. To determine the current operational safety of the bridge, its movements were observed under different stress factors such as wind speed, temperature change and traffic load. The observed lateral, longitudinal and vertical movements of bridge's deck and towers were studied and described using time series analyses. As a result of the analysis of the short-term observations, the linear trend components in the series originating from long-term periodical movements of the bridge were determined. In the analysis of the periodic components of the series, high-pass and low-pass filtering processes were applied to the series to determine the low and high frequencies of the bridge. The transformation of the series from the time domain to the frequency domain was ensured by Fast Fourier Transform (FFT), and power spectrums of the series were calculated. It was found that low and high frequencies obtained from power spectrums reflected the expected movements of the bridge. It was determined that the bridge deck had many significant periodical movements resulting from wind forces and traffic load. The calculated high frequency values for vertical movements of the bridge deck were found to be in harmony with natural frequency and periodic values of the first 6 modes of the Bosphorus Bridge at the 0.127-0.463 Hz interval. Finally, stochastic components of the time series were determined by the use of Autoregressive (AR) and Autoregressive Moving Average (ARMA) models. In terms of the operational safety and operational status of the Bosphorus Bridge, there were no unexpected results determined by the time series analysis in the observed movements of the bridge. (C) 2007 Elsevier Ltd. All rights reserved.