Adaptive Identification of Time-Varying Cable Tension Based on Improved Variational Mode Decomposition

Abstract

Stay cables are crucial components in cable-stayed bridges. The time-varying cable tension is an important indicator for the condition assessment of cables because the variation in tension will cause fatigue damage. This paper proposes an adaptive identification method for time-varying cable tension based on an improved variational mode decomposition (VMD) algorithm. In the proposed method, the cable vibration signal is decomposed by improved VMD to identify the time-varying modal frequencies by the Hilbert transform, and then the time-varying cable tension is calculated by the vibration method. To realize adaptive and accurate identification, the VMD algorithm is improved by optimizing parameters: the number of decomposed modes is determined by the characteristic that frequency differences between the adjacent modal frequencies of the cable vibration signal are equal, and the balancing parameter is optimized by constructing an evaluation index and searching for the minimum of the index. A numerical simulation illustrates that the accuracy of the proposed method is satisfactory. To verify the practicability of the proposed method, the monitoring data of a single tower cable-stayed bridge are used. The results demonstrate that the proposed method is a practical and reliable method to identify the time-varying cable tension.