混沌初值敏感性在应力波检测管道损伤参数中的应用研究

Ultrasonic guided waves have been become an important detective method for long-distance pipelines, because of their superior abilities of fast speed and wide range, to the traditional nondestructive ultrasonic inspection technology. However, the detection precision, available detection length and parameter identification of defects are the main difficult problems of ultrasonic guided wave inspection technology at present. The sensitivity of the chaotic system to initial conditions will be used to establish the relationship among the damage characteristics (crack, corrosion pit, flange and weld joint, etc.) of pipeline structures and their local positions, propagation characteristics of guided wave signals (mode transformation, frequency dispersion, attenuation, wave number and signal array etc.), and chaos damage factor in this proposal; this will achieve the identification of the type and degree of defects/damages. Based on the study on basic NDT data of typical pipe structures, the DAC curves and damage identification curves of pipes in the ultrasonic guided wave inspections will be programmed, in order to improve the operability of the proposed method. The systematic theoretical and experimental studies will be performed, attempting to achieve some valuable results in the chaos inspection aspects using ultrasonic guided wave signals. The innovation of this proposal is to develop the detection index using the sensitivity of the chaotic system to initial conditions, which is helpful to improve the detection sensitivity of micro defects and extend the effective detection distance of pipelines.

与传统超声波无损检测相比，超声导波具有检测速度快，检测范围广等优势，已成为长距离管线检测的重要方法。目前，检测精度和检测距离以及缺陷参数识别问题已成为超声导波检测技术的难点问题。本项目拟利用混沌系统对初值敏感性，将导波信号作为混沌系统的初始输入，建立管道结构与局部损伤特征（裂纹、腐蚀坑、法兰、焊缝等），导波传播特征（在模态转换、频散、衰减、波数和信号阵列等）、以及混沌损伤因子三者之间的关系，可实现对缺陷损伤程度和损伤类型的识别；研究典型管道结构无损检测的基础数据，编制管道超声导波检测的DAC曲线和破损识别曲线，提高超声导波无损检测的操作性。本研究将通过系统的理论和实验研究，力争在超声导波检测信号的混沌检测方面获得有价值的研究成果。创新之处在于，以混沌系统初值敏感性构造导波检测指标，可提高小缺陷的检测灵敏度和延长管道有效检测距离。

本项目基于前期对超声导波混沌检测的研究基础，继续进行混沌系统管道导波无损检测中的适用性的研究，项目主要的研究内容及成果包括：.1、提出了基于Duffing振子的分维数进行弱导波信号识别的方法，利用Lyapunov指数理论推导了Duffing系统对噪声的免疫，并提出了时移Duffing振子检测方法，不仅可对任意频率、任意初相位的微弱信号进行识别，还可定位缺陷位置,评估缺陷大小。.2、提出了利用互相关、小波变换等理论与Duffing振子系统相结合的混合检测方法。将待测信号分别通过互相关或小波变换降噪后再输入Duffing系统进行检测，可实现更低的信噪比的混合信号的检测。.3、提出了基于三维自治Lorenz混沌检测系统的超声导波检测方法，利用相轨迹及Lyapunov指数作为混沌判据，实现了小缺陷管道弱导波信号的识别。.4、根据二分法研究了Duffing混沌系统和Lorenz混沌系统的定位效果，通过对待测信号逐级二分检测，最终确定缺陷回波信号的时域位置。.5、建立了基于双稳态系统和Duffing自由系统的弱导波信号的检测模型，提出了基于随机共振的弱超声导波检测方法。