机械结构早期微缺陷非线性特征辨识及三维检测研究

Nonlinear feature recognition for the micro-defect of mechanical structure is the difficult technique and problem that the nonlinear ultrasonic measurement is applied in the structural health monitoring. It not only extracts the features of the nonlinear signal in the time domain and frequency domain, but also ensures corresponding time-frequency-space relations between the input excitation and output response of the nonlinear features. The research is characteristic of the theoretical innovation and valuable applications with the challenges. In terms of the urgent requirements from the mechanical structure health monitoring, the research on the Nonlinear feature recognition and 3-D measurement for early micro-defect of mechanical structure is done. The mechanism about the generation of the nonlinear effect is done. The research about the single input single output nonlinear system identification theory is done. The model about frequency response and algorithm about the time domain measurement are established to improve the sensitivity and precision of the nonlinear ultrasonic feature recognition for the local defect..The research on the optimization algorithm about the physical process of the multi-input multi-output nonlinear system feature with the characteristic correspondence and consistency is done to solve the confliction among the higher efficiency, higher precision and calculation complexity. The reconstruction method about the 3-D space feature based on parallel factor decomposition theory is proposed to improve the quantity measurement. The 3-D measurement method based on the time reverse algorithm theory is done to improve the precision of the localization, promote the application of the nonlinear measurement in the mechanical structural monitoring and increase the accuracy and intelligent levels of the mechanical quantity and safety measurement.

机械结构微缺陷非线性特征辨识是非线性超声检测在结构健康监测中所遇到的技术瓶颈和难题，它不仅分离提取非线性信号时频域特征，而且要确保输入激励与系统响应之间非线性系统特征在时间、频率和空间上对应关系，它的研究具有理论创新和实用价值及挑战性。针对大型复杂机械结构健康监测的迫切需求，开展机械结构早期微缺陷非线性特征辨识及三维检测，分析微损伤微弱非线性效应形成机理；研究单输入单输出非线性系统识别理论，建立非线性系统频率响应模型和时域测量算法，改善局部缺陷非线性声学特征辨识敏感性和精确性；研究多输入多输出非线性系统特征物理过程具有对应性和整体一致性优化算法，调和特征提取高效、高精度与计算复杂性之间矛盾；提出基于平行因子分解理论三维空间特征重构方法，实现精确定量检测；进一步研究基于反演算法理论三维检测方法，改善定位精度，促进非线性检测在机械结构健康监测中的应用，提高机械质量安全检测的精准度和智能化水平。

针对非线性超声检测在机械结构早期健康监控中所遇到的技术瓶颈和难题，开展了机械结构早期缺陷非线性特征辨识及三维检测研究。研究复杂机械结构微缺陷非线性效应形成机理,分析缺陷区域激励的频谱非线性成份包含被sinc函数调制的高阶谱，探讨对于双波超声激励，调制刚度产生驱动频率混合，旁瓣调制幅值是裂纹缺陷非线性标量，混频产生高阶非线性和多个频率成份混合机理。分析在两个输入激励信号作用下局部缺陷结构非线性效应频率成份特征及其非线性变化规律，分析时域信号上缺陷引起畸变特征和微弱缺陷非线性信号时变过程。分析边界、换能器等系统非线性干扰对缺陷非线性效应影响，以及复杂结构对系统非线性声场作用机制。建立不同缺陷局部结构非线性与两个非共线超声波频率、幅值和空间位置之间最优化关系，揭示微缺陷非线性微弱特征信号频率转换变化机制。提出基于非线性系统频域响应分析理论、非线性自回归移动平均模型和去噪声理论的微缺陷非线性系统效应识别理论算法。优化非线性系统频域响应成分，定义局部缺陷结构声学特征增强敏感度评估标准和法则。分析探讨不同空间位置的系统输出响应的非线性时变特征相互内在关系，建立系统输入激励、微缺陷局部材料特征、非线性特征在时间、频率、空间的最佳优化非线性对应关系，建立基于多输入多输出非线性系统频域响应特性理论和非线性自回归移动平均模型的非线性特征提取，分析建立非线性时变信号在时域-频域-空间上的内在关系。大型复杂结构如大型压力容器、发电机、船舶螺旋浆、发动机等关键设备的智能化、远程式在线健康监测和故障诊断，是复杂制造信息化、智能化和自动化的前沿方法与技术，对于推动我国智能制造健康监控与质量检测的智能化、自动化和高精度发展，实现安全检测自动化，有着深远的科学理论意义和广泛的工程应用价值。