基于小波（包）理论的隧道初期支护质量诊断及其分析方法研究

The quality defect of primary support of tunnels is the most important reason which eventually leads to the tunnel disease. The widely-adopted tunnel quality nondestructive testing analysis based on the traditional Fourier analysis theory has greatly lagged behind the needs of engineering practices, for it has seriously influenced the actual application effect of various monitoring and detection means. Since the tunnel engineering structure has the nature of concealment and the quality nondestructive detection signals are generally complex, nonlinear and non-stationary, considering these characteristics, this project makes use of the unique time-frequency localization characteristics and zoom spectrum analysis techniques of wavelet and wavelet packet theory. Meanwhile, through the processing and analysis of the signals of primary support of tunnels acquired by different detectors under different conditions, it explores the signal features of common quality failings appeared during construction process, such as insufficient thickness of shotcrete, insufficient closeness of lining and surrounding rocks, cavity between lining and surrounding rocks and shortness of effective length of rock bolt. In addition, it studies many key scientific issues such as signals interpretation technology and advanced theories which can rapidly, easily and accurately test, diagnose and presume the construction quality of primary support of tunnels, providing theory basis and technical means for timely feedback analysis of dynamic information during tunnel construction, and reasonable adjustment of the construction method, supporting time and supporting parameters, as well as better improvement to the quality of primary support of tunnels.

隧道初期支护的质量缺陷是最终导致隧道产生病害的最主要的原因。目前普遍采用的基于传统傅里叶分析理论的有关隧道质量无损检测分析技术已大大滞后于工程实践的需要，严重影响了各种监控与检测手段的实际应用效果。本项目针对隧道工程结构具有隐蔽性及其质量无损检测信号普遍具有复杂多变、非线性、非平稳的特点，运用小波（包）理论独特的时频局部化能力和细化谱分析技术，通过对不同检测仪在不同条件下获取的隧道初期支护的信号进行处理与分析，探索隧道初期支护施工过程中出现的诸如初喷混凝土厚度不足、衬砌与围岩不密贴、衬砌背部与围岩间产生空洞以及锚杆有效长度不足等质量通病的信号特征，研究能快速、简捷、准确地对隧道初期支护施工质量进行检测、诊断、推定的先进理论与信号解释技术等关键科学问题，为隧道施工适时进行动态信息反馈分析，合理地对施工方法、支护时机、支护参数等加以调整，更好地提高隧道初期支护的质量提供理论依据与技术手段。

隧道初期支护的质量缺陷是导致隧道病害的主要原因。地质雷达无损检测技术是检测隧道支护质量的主要方法，现有的滤波、去偏移等信号处理技术能有效提高被测目标识别度，但无法实现雷达信号的定量分析。由于准确识别和精确量化隧道病害的位置和尺寸对隧道病害的评估和修复具有重要的指导意义，提出一种可以定量分析隧道病害的方法是十分必要的。本项目运用小波（包）理论独特的时频局部化能力和细化谱分析技术，提高了地质雷达和金属超声波检测仪的工程应用效果。针对工程信号固有特点，选择或构造最优小波基是小波分析的基础，为了提高小波构造的效率、降低小波构造的盲目性，基于小波构造理论和完全重构滤波器理论，本项目提出了一种新的双正交小波构造方法。根据地质雷达信号固有特点构造相应的双正交小波基（应用于地质雷达信号分析的小波基，简称GPR3.3小波），提出了基于GPR3.3小波的地质雷达信号定量分析方法（简称基于GPR3.3小波的QAGPR法）。在室内试验中被测目标的尺寸和埋深均是已知的，室内试验是研究和验证新小波、新方法和新技术的最有效手段之一。运用室内砂槽的空腔和钢筋检测试验分别模拟隧道衬砌脱空检测、锚杆质量检测，并将新的小波基和新的雷达数据定量分析方法用于试验数据分析处理。基于GPR3.3小波的QAGPR法能够准确识别雷达数据中被测目标的埋深和垂直尺寸，为无损检测研究提供了更深的理论基础和更科学的分析；运用金属超声波检测仪完成了螺纹钢的长度和损毁点检测试验，运用小波（包）分析可以进一步提高金属超声波检测仪的分析效果，丰富了隧道锚杆锚固质量的检测手段。