短索基于修正有效振动长度的实时索力识别

Real-time cable tension identification for short cables based on corrected effective vibration length

  • 摘要: 基于有效振动长度(effective vibration length, EVL)的频率法虽能规避短索边界效应,但现有实时索力识别方法假设EVL恒定可能导致误差显著。本文通过构建复杂边界短索有限元模型,发现索力变化明显时EVL存在波动,为此提出基于多工况索力-EVL数据的样条插值修正方法。数值实验表明:通过理论实时频率识别实时索力,未修正时相对误差为‒0.98%~‒9.57%,修正后绝对误差缩减至‒2.34%~‒5.08%,误差峰值降低近50%。通过对比4种时频分析方法,证实同步压缩变换(synchro-squeezing transform, SST)具有较好时频分辨率,其与EVL修正方法协同作用下,实时索力在波峰波谷处的识别误差从20%降至10%以内。研究验证了结合动态修正EVL与SST时频分析对提升短索索力实时监测精度的有效性。

     

    Abstract: The frequency-based method utilizing Effective Vibration Length (EVL) can mitigate boundary effects in short cables, yet existing real-time cable tension identification methods assuming constant EVL may introduce significant errors. This study constructs a complex boundary model of finite element method for short cables and reveals that EVL fluctuates when cable tension varies. To address this, a spline interpolation correction method based on cable tension-EVL data is proposed. Numerical experiments demonstrate that when identifying real-time cable tension using theoretical real-time frequency, the uncorrected relative error ranges from ‒0.98% to ‒9.57%, whereas after correction, the relative error is reduced to ‒2.34%~‒5.08%, with peak errors nearly halved. By comparing four time-frequency analysis methods, it is confirmed that the Synchro-squeezing Transform (SST) offers superior time-frequency resolution. Under the combined application of the EVL correction method and SST, the real-time cable tension identification error at peak and valley positions is reduced from 20% to within 10%. This research validates the effectiveness of integrating dynamic EVL correction with SST time-frequency analysis in enhancing the accuracy of real-time monitoring for short cable tensions.

     

/

返回文章
返回