基于伪随机信号的锚杆动态响应与智能检测研究

项目名称： 基于伪随机信号的锚杆动态响应与智能检测研究

项目编号： No.50874035

项目类型： 面上项目

立项/批准年度： 2009

项目学科： 数理科学和化学

项目作者： 孙晓云

作者单位： 河北科技大学

项目金额： 36万元

中文摘要： 锚杆锚固技术是地下工程及边坡治理的重要支护手段，锚固质量的好坏关系着工程质量和安全，目前锚固质量的高精度智能检测是该领域面临的世界性难题，因此，本研究对保障地下工程安全等具有重要的理论意义和应用价值。项目以系统辨识、现代信息处理技术、智能控制理论为工具，解决基于伪随机信号的锚杆智能检测相关基础科学问题。首次提出采用具有最佳自相关特性的低频余弦型伪随机信号作为激励信号，进行锚杆动态响应特性的参数辨识；建立锚杆中导波传播的数值模型，计算低频情况下锚杆中导波传播的频散特性，研究频散波组特征与锚杆受力状态的响应特征；利用傅里叶变换方法实现信号的数字正交采样，准确计算反射信号与激励信号的相位差，进行相位差与锚杆受力状态的响应关系计算与识别；利用小波变换进行反射信号的特征值提取和滤波，提出利用改进的Fisher ratio方法优化神经网络结构，为建立锚杆锚固质量的高精度智能检测系统奠定理论基础。

中文关键词： 伪随机信号；动态响应；导波；锚固质量；智能检测

英文摘要： Rock bolts technology belongs to one of the most important methods for underground engineering and highwall reinforceing, which has tightly relation to engineering quality and safety, at present, it is a world-wide problem for high precision intelligent detecting. So, the research of rock bolts dynamic response and intelligent detecting based on pseudo random signal has important theoretical and realistic significance. The key of this project is to carry out the research in several ways and methods which include system identification, modern signal processing technology, intelligent control theory to resolve the relevant basic science problems of intelligent detecting for rock bolts. A pseudo random signal of low frequency cosine with the best autocorrelation characteristics used to identify the dynamic response parameters of audio frequency stress wave propagation in rock bolts is presented firstly; A finite element model of guided wave propagation in the bolts is established to study the frequency dispersion of guided wave in the bolts under low frequency, dispersion wave group characteristics and the corresponding characteristic under forced state of bolts. Digital quadrature sampling technique is used to calculate the phase-difference of reflected signal and incentive signal, and calculate the phase-difference of reflected signal and incentive signal. A kind of wavelet basis is choosed to remove noise and extract the characteristic values for reflected signal, an improved Fisher ratio method is proposed to optimize the structure of neural network. Which would establish the theory basis for setting up a kind of bolts high precision intelligent detecting system.

英文关键词： Pseudo random signal; Dynamic response; Guided wave; Rock bolts; Intelligent detecting