爆炸复合结构结合状态动态响应机理与检测方法研究

Explosive cladding structures are wildly applied in fields such as chemical, petroleum and gas, aeronautics and astronautics, military, ships and automobile industry, which imposes urgent demands on their bonding states' on-line inspection and explosive cladding structures' overall performance assessment. Currently, sampling inspections of explosive cladding structures' bonding states are carried out in experimental environments by destructive means. Non-destructive and online inspection methods are still not available for explosive cladding structures. This research project aims to seek improvements in the following aspects. Customized wavelet finite element models of explosive cladding structure are developed to reveal its mechanism of dynamic response, whose corresponding association to the bonding state is also investigated. The incipient and coupling transient features of dynamic response signals from explosive cladding structures are extracted by using second generation wavelet transform and dual tree complex wavelet transform. Ensemble-enhanced incremental support vector machine based on wavelet kernel is introduced to form recognition model, realizing high precision inspection and assessment for explosive cladding structures' bonding state under small-sample condition. Theoretical fundamentals and key techniques of on-line nondestructive inspection and assessment for explosive cladding structures are investigated through experimental means and engineering applications. This research project are intend for finding feasible methods and essential techniques that can be used to non-destructive inspection and assessment of explosive cladding structures. This research project, focusing on the innovations of both academic research and engineering applications, is promising for the advancement in quality examination and service performance assessment of explosive cladding structures.

爆炸复合结构广泛应用于化工、石油天然气、航空航天、军工、船舶、汽车等领域，对其结合状态在线检测、整体评价的工程需求迫切。目前爆炸复合结构结合状态的检测一般采用破坏性抽检取样后在实验室检测的方法，无法实现无损在线检测和服役性能评估。本项目研究基于算子自定义小波有限元的爆炸复合结构动力学模型，揭示结合状态的动态响应机理，探求结合状态与其动态响应信号特征的映射关系；研究第二代小波和双树复小波信号特征提取方法，实现爆炸复合结构动态响应信号微弱、耦合特征信息的有效提取；研究并构建基于小波核函数的集成增强支持向量机识别模型，实现小样本条件下爆炸复合结构状态的高精度检测和评估。针对爆炸复合管进行实验研究与工程应用，为爆炸复合结构状态的在线无损检测和评估提供基础理论和关键技术，对其质量检测和服役性能评估提供有效手段，具有重要的学术意义和应用价值。

爆炸复合结构具有耐高温、耐高压、耐腐蚀等优良特性，已被广泛应用于石油天然气、化工、航空航天军工、船舶、汽车等行业和领域，对其结合状态在线检测、整体评价的工程需求迫切。目前爆炸复合结构结合状态的检测一般采用破坏性抽检取样后在实验室检测的方法，难以实现无损在线检测和服役性能评估。本项目研究了一种基于振动响应的爆炸复合结构结合状态的检测与评估方法，主要研究内容和研究成果如下：. （1）基于Euler-Bernoulli梁模型构建了爆炸复合管的解析动力学模型，基于Hamilton正则方程构建了爆炸复合管的有限元动力学模型，实验验证了模型的有效性。利用两种模型，探究了不同结合状态对结构动态响应和状态检测指标的影响关系，为状态检测指标的选择和确定提供了理论基础和先验知识。. （2）针对整体结合强度不足缺陷检测，研究提出了基于模态频率幅值比的检测指标。针对局部接触不良缺陷检测，研究提出了基于优化双树复小波时间熵的检测指标。研究构建了基于贝叶斯算法的多检测指标融合的结合状态检测模型，实验验证了检测指标和检测模型的有效性，实现了多种缺陷的在线检测。. （3）基于结合状态缺陷会使结构非线性特性增强的物理属性，研究了非线性输出频响函数对结合状态缺陷的表征形态，提出了基于非线性输出频响函数的检测方法，实验验证了其有效性。该方法检测精度高，但需要对结构进行多次激励，检测效率低不便于在线检测，可用于实现在线检测缺陷结构的离线检测确认。. （4）基于提出的检测指标和检测方法，制定了检测规范，开发了爆炸复合管结合状态检测与评估系统，可实现结合状态的在线检测和离线评估。该系统已应用于西安向阳航天材料股份公司的爆炸复合管生产线，初步验证了其有效性。. 本项目的研究成果为爆炸复合结构结合状态的在线、无损检测和评估提供了基础理论和关键技术，具有重要学术意义和工程应用前景。