| 摘要: |
| 表孔弧形闸门局开泄洪流激振动问题直接关系到水利枢纽的运行安全，开展泄流激励下的弧形闸门运行模态参数分析是衡量闸门安全运行的重要内容。为此，本文基于协方差驱动的随机子空间框架，针对虚假极点干扰问题提出改进势能聚类的模态参数识别方法。通过构建六自由度剪切模型验证所提方法的识别精度与抗噪性，建立闸门支臂多自由度有限元模型论证水流脉动荷载代替白噪声开展模态参数辨识的可行性，基于此，以实际闸门局开泄洪振动观测为例建立两种闸门-水体流固耦合模型。研究结果表明：该方法可有效剔除由背景噪声和模型系统阶次设置过大引起的虚假模态，通过对比识别模态参数与仿真结果，除个别阶次外其余各阶识别频率误差均在10%以内，验证了方法的准确性与可靠性。流固耦合模型中的直接耦合法模型可有效反馈出闸门与水体的耦合作用，附加质量法需试算折减因子具有明显的局限性。 |
| 关键词: 弧形闸门 模态参数识别 改进聚类算法 有限元 直接耦合法 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金资助项目(52179060) |
|
| Modal Parameter Identification of Radial Gate Vibration under Discharge Excitation based on Improved Potential-based Hierarchical Agglomerative Clustering |
|
Ren Hengxin,Liu Yakun,Zhang Di,LU Yangliang
|
|
Dalian University of Technology Faculty of Infrastructure Engineering
|
| Abstract: |
| The vibration of radial gates in partial-opening flood discharge conditions is directly related to the operational safety of hydraulic structures, making modal parameter analysis under discharge-induced excitation a key aspect of gate safety assessment. Traditional identification methods often suffer from spurious poles and noise sensitivity, which hinder the reliable extraction of modal parameters. To overcome these challenges, this study proposes an improved potential-based hierarchical agglomerative (IPHA) clustering method within the covariance-driven stochastic subspace identification (COV-SSI) framework. First, a six-degree-of-freedom shear model is constructed to verify the accuracy and noise resistance of the proposed method. The results demonstrate that the IPHA-based approach effectively eliminates spurious modes induced by background noise and oversized model orders, thereby improving both identification stability and automation. Subsequently, a multi-degree-of-freedom finite element model of the gate supporting arms is developed to investigate the feasibility of replacing ideal white-noise excitation with flow-induced fluctuating loads for modal parameter identification. The analysis confirms that the fluctuating hydrodynamic pressure acting on the gate panels can be approximated as band-limited white noise in the low- to mid-frequency range, and is sufficient to excite the structural vibration modes. Building on this foundation, two gate–water fluid–structure interaction models are established using field vibration measurements of a radial gate under partial-opening discharge conditions, namely the added mass method and the direct coupling method. Comparative analysis reveals that the direct coupling model provides an accurate representation of gate–water coupling effects, while the added mass method requires repeated adjustment of the reduction factor ξ and therefore shows significant limitations. Overall, the comparison between identified modal parameters and simulation results indicates that, except for a few higher-order modes, the frequency identification errors remain within 10%.These findings verify the accuracy and robustness of the proposed IPHA-based COV-SSI method and highlight its potential for reliable modal parameter identification of radial gates under complex flow-induced vibration environments. |
| Key words: Radial gate modal parameter identification improved clustering algorithm finite element direct coupling method. |
