Data-driven evaluation of abnormal vibration of cables during typhoon In-Fa

Abstract

The input–output characteristics of cables of a long-span cable-stayed bridge in time and frequency domains were studied during the whole process of typhoon In-Fa landfall based on monitoring data. According to the signal characteristics of the time series of wind speed and direction, a new adaptive extraction method of quasi-steady wind speed and direction was developed, fusing wavelet transform and the certain-length time series averaging method. The central tendency and fluctuation about the wind speed and direction were assessed and the influences of wind and rainfall on cable abnormal vibration were analysed. Using the short-time Fourier transform, the characteristics of the evolution of vibrational energy in the time–frequency domain for abnormal cable vibration during the typhoon were reproduced. Slice analysis of the power spectrum in the frequency domain was performed at typical moments. The results of this work revealed the following. The extremely high-amplitude non-Gaussian section of abnormal cable vibrations in this typhoon event was caused by the simultaneous action of wind and rain. A high wind speed or a wind direction with low fluctuation will more likely excite high-amplitude abnormal cable vibration. Sufficient damping may help cables suppress rain-wind-induced high-amplitude vibration.