基于Volterra级数的旧零件内部损伤非线性检测机理研究

For shooting the hard problem on the internal damage detection and the nonlinear phenomenon in the used parts, a new detection and classification based on Volterra series have been researched in remanufacture. Through using the pulse hammer to impact the used parts on the free boundary condition, then its multiple response output signal in different position might been measured. A new test signal analysis method based on Volterra series, and its generalized frequency response function(GFRF) and nonlinear output frequency response function(NOFRF) will be proposed. The nonlinear characteristic index function which was sensitive to the internal damage and defects will be developed, the damage defects detection and the reuse classification will be realized. This method will be simple, high efficiency and low cost to detect the used parts damage and to make the classification. The main research contents: (1) According to single input multiple output system of pulse excitation, the generalized frequency response function and nonlinear output frequency response function will be derived, and then a nonlinear detection theory based on the impulse single input multiple output system will be constructed; (2) A indicator function which was sensitive to the internal damage of used parts will be optimized, and a nonlinear detection method to the internal damage of used parts would be built; (3) Through the fatigue test, the relationship between nonlinearity sensitive index function value and fatigue life will be studied, the classification evaluation method for used parts will be proposed. By this project research the failure prediction theory and the equipment health management method will be enriched, and the theoretical and technical basis to remanufacturing engineering will be provided also.

针对再制造工程中难以对旧零件内部损伤检测及内部损伤缺陷的非线性等问题，通过脉冲锤击处于自由边界条件下的旧零件，在不同位置测试其多个响应输出信号，提出基于Volterra级数及其广义频率响应函数、非线性输出频率响应函数等概念建立一种新的测试信号分析方法，构建出对内部损伤缺陷敏感的非线性特征指标函数，实现旧零件的内部损伤缺陷检测及其再用分类，建立一种简便、高效、低成本的旧零件内部损伤缺陷检测及再用分类评估方法。研究内容：1）针对脉冲激励单输入多输出系统推出广义频率响应函数和非线性输出频率响应函数，建立脉冲激励单输入多输出系统的非线性检测理论；2）构建对旧零件内部损伤敏感的指标函数，建立旧零件内部损伤非线性检测方法；3）通过疲劳试验，研究非线性特征敏感指标函数值与疲劳寿命关系，建立旧零件可否再制造的分类评估。这不仅丰富了故障预测的理论和方法，也为解决再制造工程的难题奠定理论和技术基础。

本项目研究主要完成了脉冲锤击旧零件的试验方法构建、基于Volterra级数的脉冲锤击响应信号的非线性检测理论、基于Volterra级数的旧零件内部损伤非线性检测方法、旧零件再用评估分类研究等四方面的内容，取得了如下重要结果。.推导出矩形脉冲激励信号的非线性输出频率响应函数（NOFRFs），分别针对单输入单输出（SISO）和单输入多输出（SIMO）系统建立了脉冲锤击激励下非线性输出频率响应函数（NOFRFs）两个估计方法。基于NOFRFs构建了NOFRF熵 、 NOFRF能谱熵 、NOFRF频谱复杂度熵 、NOFRFs频谱散度（divergence index, DI）等4个检测指标，实现对旧零件内部损伤的检测与辨识。由NOFＲFs检测指标以置信度 95% 实现了对旧零件连杆已使用时间进行了估计；建立旧零件内部缺陷测试得到的非线性特征敏感指标函数值及其使用寿命评估模型，探索建立旧零件再制造的分类准则。.通过混沌特性分析，深化了对金属疲劳损伤经历的裂纹萌生、裂纹缓慢扩展、裂纹快速扩展和临近破坏等四个阶段的声发射信号特征的认识。仿真及实验测试分析表明，高阶固有频率对损伤有很强的敏感性，且固有频率的反应变化是非线性的。.利用超声非线性研究零件的残余应力、疲劳裂纹和疲劳寿命。利用临界折射纵波的非线性特性对稳定状态下金属零件的应力状态进行表征，包括应力的大小与方向。利用共线混频技术对金属零件进行疲劳损伤检测，并构建基于超声非线性综合参数的疲劳寿命预测模型。.本研究结果将为再制造中旧零件疲劳累积损伤检测提供技术基础，随着再制造领域规模扩大及应用深化研究，该理论基础将会引导出更多应用技术。已在《Journal of Sound and Vibration》、《Advances in Materials Science and Engineering》等SCI杂志发表论文6篇，在《机械工程学报》、《振动与冲击》等EI杂志发表论文9篇；获得国家发明专利2件；培养硕士研究生9名、博士研究生1名。