融合机器视觉与ESMD-ApEn算法的大坝新型安全监测方法研究

Research on Novel Dam Safety Monitoring Method Integrating Machine Vision and ESMD-ApEn Algorithm

为实现水利水电工程大坝变形监测自动化,引入一种新型机器视觉测量系统对大坝外观变形进行实时测量和分析。首先对机器视觉系统原理和架构进行详细说明,接着构建基于极点对称模态分解算法(ESMD)和近似熵理论(ApEn)相结合的ESMD-ApEn算法,用于对机器视觉原型监测序列进行多模态分解和重构,并基于拉依达和IQR准则,实现对ESMD-ApEn算法重构序列与机器视觉原型监测序列之间的重构残差进行异常识别和剔除。案例分析表明,构建的融合监测方法可以实现大坝外观变形观测自动化和实时化,通过引入红外补光系统,在夜间及雨雪等恶劣气候天气,也能实时准确获取大坝变形监测数据,从而确保大坝安全稳定运行。

To achieve the automation of deformation monitoring in hydroelectric engineering,this paper introduces a novel machine vision measurement system for real-time measurement and analysis of the external deformation of dams.Firstly,a detailed explanation for the principles of the machine vision system is provided.Subsequently,a new algo-rithm,named the Extreme-Point Symmetric Mode Decomposition Method combined with Approximate Entropy(ESMD-ApEn)algorithm,is constructed for multi-modal decomposition and reconstruction of machine vision monitoring se-quences.Based on the Lyddane-Shindo and the Interquartile Range(IQR)criterion,the ESMD-ApEn algorithm is used to identify and eliminate anomalies in the reconstruction residuals between the ESMD-ApEn algorithm and the machine vi-sion prototype monitoring sequences.Case studies demonstrate that the integrated monitoring method can achieve the au-tomation and real-time observation of external deformation in dams.Additionally,with the introduction of an infrared supplementary lighting system,it can accurately acquire dam deformation monitoring data even during nighttime and ad-verse weather conditions such as rain and snow,ensuring the safe and stable operation of the dam.