Intelligent parameter identification for bridge cables based on characteristic frequency equation of transverse dynamic stiffness

Abstract

Determining the cable force and other parameters of cables is important for condition assessment of cable-stayed structures. This study proposes a frequency characteristic equation of transverse dynamic stiffness for cables; this equation is suitable for measuring the vibrations to evaluate the primary factors that influence the accuracy of cable parameter identification. Further, a cable parameter identification method based on the particle swarm optimization algorithm is proposed. The method is suitable for a cable system of arbitrary length and with moderate sag especially when the measurement quality of the modal frequencies of cables is poor. Both numerical case studies and a cable vibration test proved that the proposed method can identify parameters with high accuracy for cables of any length and for cases requiring low-frequency measurement. Moreover, structural modal order information is not required. The extreme case is that only one order frequency can achieve highly accurate result in this way. The proposed method is suitable for parameter identification of short cables, hanger cables, and parallel strand cables, which are commonly applicable in engineering applications.

Accuracy Improvement of Mutual Integration Mechanism Driven Algorithms for Boom Cable Force Recognition

“…As a probabilistic search method that does not require gradient information of functions, heuristic algorithms are highly robust [15,16] and widely used in the civil engineering field for parameter identification problems [15]. Liu Hongbo et al [18] proposed a deep learning-based method for cable force identification in short cables, demonstrating its effectiveness by comparing various machine learning algorithms; Li Xiaozhang et al [19] used particle swarm optimization for cable force identification under fixed boundary conditions, validating its effectiveness with a combination of virtual and real data; Sun Liangfeng et al [20] used genetic algorithms and neural networks for parameter identification of short cables and stiffness, finding that uncertain complex boundary conditions can cause significant errors in heuristic algorithm-based cable force identification; Dan et al [21] also used particle swarm optimization, considering the effect of sag, to complete cable force identification under fixed boundary conditions.…”