基于多种无损检测技术的锈蚀钢筋混凝土梁的失效机制与性能劣化规律研究

In this project, failure mechanism and deterioration characteristics of corroded reinforced concrete beams in marine corrosive environment will be studied based on a variety of nondestructive testing technology. Firstly, the morphology and time-varying evolution of the surface of the corroded steel bar will be observed by X-ray computed tomography (XCT), and the relationship between the morphological parameters of the surface and the corrosion rate will be established. The mechanical properties and stochastic constitutive models of corroded steel bars will be obtained by tensile test of corroded steel bars. The three-dimensional finite element model of the corroded steel bar will be constructed by the geometric inverse reconstruction of the surface morphology of the corroded steel bar and the constitutive model of the corroded reinforcement. Secondly, based on the acoustic emission test system and the pull-out test, the bond performance of the reinforced concrete under the different corrosion rate will be studied and its relationship with the characteristics of the acoustic emission signal will be obtained. Based on this, the bond-slip constitutive model between the corroded steel bar and the concrete will be established. Finally, the finite element model of the corroded reinforced concrete beam will be established, while the corroded reinforced concrete beam specimens will be designed and fabricated. Through the analysis of the vibration signal of the whole corrosion process of the test beams, the dynamic characteristics of the specimens with the increase of the corrosion rate will be studied by the numerical simulation and the experiment. And this study will provide the scientific basis for the reliability evaluation, maintenance and reinforcement of the reinforced concrete structure in the marine corrosive environment.

本项目基于多种无损检测技术开展海洋腐蚀环境中锈蚀钢筋混凝土梁的失效机制与性能劣化规律研究。首先，通过X射线计算机微断层扫描仪（XCT）观察锈蚀钢筋表面形貌分布形态和时变演化特征，建立其表面形貌特征参数与锈蚀率之间的关系；通过锈蚀钢筋试件的拉伸试验，获得锈蚀钢筋的力学性能和随机本构模型；通过锈蚀钢筋表面形貌三维几何逆向重构和已建立的锈蚀钢筋本构关系模型构建锈蚀钢筋三维有限元模型。其次，基于声发射测试系统和拉拔试验研究不同锈蚀率下钢筋混凝土的粘结性能及其与声发射信号特征的关系，并以此为依据建立锈蚀钢筋与混凝土的粘结滑移本构模型。最后，建立锈蚀钢筋混凝土梁的有限元计算模型，同时设计并制作锈蚀钢筋混凝土梁试件，通过对试验梁锈蚀全过程的振动信号分析，从数值模拟和试验两方面研究其动力特性随着钢筋锈蚀率增大的变化规律。为海洋腐蚀环境中的钢筋混凝土结构的可靠性评估、维修加固及全寿命分析等提供科学依据。

影响混凝土结构耐久性劣化的机制复杂多样，其中钢筋锈蚀被认为是最重要的动因。本项目结合多种无损检测技术对锈蚀钢筋混凝土构件服役性能劣化机理及损伤评估展开系统研究。通过本研究得到如下几条结论：（1）基于三维扫描技术获得了平均锈蚀率、临界锈蚀率、截面差异率、不均匀系数、特征轮廓参数、整体不均匀系数等锈蚀钢筋表面形貌特征量；建立了不同锈蚀率下多种锈蚀特征参数与锈蚀钢筋静力、动力性能参数间的关系；基于三维扫描数据建立了锈蚀钢筋有限元模型，研究了锈蚀钢筋的应力分布特征；基于随机森林算法，提出了通过钢筋锈蚀特征参数预测其力学性能的方法。（2）基于模态测试技术分析了锈蚀对钢筋混凝土构件动力特性（固有频率、振型、阻尼比等）的影响；基于振动信号和时间序列分析模型，提出了两种改进的钢筋混凝土构件非线性裂缝识别方法。（3）基于机器视觉和深度学习技术，提出了一种新的基于掩膜-区域卷积神经网络（Mask-RCNN）的多场景裂缝检测与表征方法，该方法可以在存在其他干扰项（如砖块接缝、藤蔓、污渍、人工痕迹等）的情况下，根据工程结构物表观照片准确识别定位裂缝。（4）建立了锈蚀钢筋混凝土梁的有限元模型；通过有限元模拟预测了荷载-挠度曲线、主裂缝发展分布和构件破坏形态等主要力学性能，并与实验结果进行了对比分析；分析了箍筋锈蚀和剪跨比对RC梁抗剪承载力的影响；研究了箍筋锈蚀对箍筋-混凝土之间粘结性能的影响；研究了粘结性能的改变对梁的最大裂缝宽度、抗剪承载力和主裂缝发展分布的影响。