基于机器学习的钢管混凝土构件腐蚀损伤评估与剩余承载力预测研究

机器学习凭借卓越的计算效率、强大的推理能力等优势，开始应用于结构工程领域的智能设计与运维，成为传统方法的有效补充。钢管混凝土具有承载能力强、施工便捷、经济性好等诸多优势，是桥梁、码头、塔架、平台等高性能海洋工程结构的优选形式之一。然而，海工钢管混凝土结构在服役期内不可避免地遭受海洋腐蚀和长期荷载的非线性耦合作用，导致其力学性能的时变劣化，为腐蚀损伤的识别与结构安全性的评估带来了严峻挑战。因此，开展基于机器学习与深度学习的钢管混凝土性能预测及腐蚀损伤评估研究，为该类结构的设计、监测及维护提供科学指导，在降低运维成本、辅助结构全寿命设计等方面具有重要意义。本文以海工钢管混凝土构件为主要研究对象，通过数据驱动算法与力学原理相结合的方式对其腐蚀损伤评估技术和剩余承载力预测方法进行了系统地研究，开展了如下四个方面的工作：
1. 通过广泛的数据收集，建立了轴压、受弯和偏压3种典型工况下共6000余个圆、矩形钢管混凝土构件的试验数据库。针对机器学习算法高度依赖数据规律、忽视问题本身潜在的物理机制等瓶颈，提出了一种基于力学原理优化的机器学习模型搭建方法，从输入参数选取、数据库划分、超参数确定等方面将力学原理与数据驱动的智能算法相结合。采用多种单一机器学习算法和集成算法对钢管混凝土构件在轴压、受弯与偏压荷载下的承载力进行预测，深入分析了模型的预测稳定性、不确定性、输入参数的影响规律。与国内外主流的设计规范和经验公式相比，本文所建立的模型具有更高的预测精度和更广的适用范围。
2. 提出了一种模拟钢管随机局部腐蚀的非线性有限元建模方法，建立了钢管混凝土在长期荷载和随机局部腐蚀耦合作用下的精细化有限元模型，明晰了腐蚀类型（均匀腐蚀、随机局部腐蚀）、体积腐蚀率、材料强度等参数对构件力学行为的影响规律。通过数值模拟和文献收集相结合的方式构建了包含无腐蚀、均匀腐蚀和随机局部腐蚀3种类型共4038个构件的数据库，采用7种机器学习算法建立了钢管混凝土柱剩余承载力的智能预测模型。针对腐蚀工况下力学性能的影响因素复杂且数据相对匮乏等问题，采用主动学习方法辅助机器学习模型训练，从初始数据的选取策略、集成学习袋装法的应用等方面优化了现有的主动学习方法，通过对比试验证明了该优化方法的有效性。
3. 开展了Q355、Q460和Q690三种共231个钢板试件（尺寸为280 mm×60 mm×6 mm）的盐雾腐蚀试验，采集了钢板除锈前后的宏观和微观腐蚀形貌，通过表面形貌参数定量研究了腐蚀程度和钢材强度对腐蚀形貌的影响规律。基于超景深显微镜采集到的钢材表面微观腐蚀形貌图像，提出了一种通过二重积分法计算钢材均匀腐蚀分量和局部腐蚀分量的方法。通过上述腐蚀试验结合互联网搜索引擎，建立了腐蚀等级评估和腐蚀位置识别两个图像数据库，并结合现行国家标准GB 50144-2019制定了钢管混凝土腐蚀程度的分类标准，为后续深度学习模型的建立提供了数据保障和分类依据。
4. 采用5种典型的语义分割算法，建立了可用于识别复杂背景下钢管混凝土结构腐蚀位置的模型，从注意力机制和多尺度特征融合两个方面对上述模型进行了改进，并通过搜集到的实际工程图像对该模型的识别准确性进行了验证。采用多种图像分类算法，建立了钢管混凝土腐蚀等级的评估模型，从注意力机制、颜色空间变换、多尺度输出等方面对模型进行优化。最后，基于本文所建立的多种机器学习模型，提出了一种构建海工钢管混凝土智能诊断系统的思路，为既有结构的安全服役和维修加固提供决策参考，以期降低腐蚀带来的经济损失与安全隐患。

Machine learning (ML), well recognized for its excellent computational efficiency and powerful reasoning ability, has just started to be applied in structural engineering for intelligent design and maintenance, serving as a valuable complement to traditional methods. Concrete-filled steel tube (CFST) exhibits exceptional capacity, convenient construction, and cost-effectiveness, which is one of the preferred forms for high-performance marine structures such as bridges, docks, towers, and platforms. However, marine CFST structures inevitably suffer from nonlinear coupling effects of corrosion and long-term loading during their service lifespans, resulting in time-dependent deterioration of mechanical properties. This poses great challenges for the detection of corrosion damage and the evaluation of structural safety. Therefore, predicting structural performances and assessing corrosion damage through ML and deep learning (DL) can provide scientific guidance for the design, monitoring, and maintenance of such structures, which is significant for reducing maintenance costs and evaluating life-cycle based structural performances. This paper focuses on the marine CFST members and systematically investigates their corrosion damage and residual capacity through data-driven algorithms combined with mechanical principals. The research achieved in this dissertation is fourfold:
1. Test databases are established through extensive literature collection, comprising over 6,000 circular and rectangular CFST members under axial compression, bending, and eccentric compression. In order to address the limitations of ML algorithms, e.g., they heavily rely on data patterns while tending to ignore the potential physical principals, an approach for developing ML models based on mechanical principals is proposed. This method integrates mechanical principals with data-driven intelligent algorithms, focusing on input parameter selection, database subdivision, and hyperparameter determination. Multiple single and ensemble ML algorithms are utilized to predict the capacity of CFSTs under three loading conditions, with prediction stability, uncertainty as well as the influence of input parameters comprehensively analyzed. Compared with mainstream standards and empirical formulas, the model established in this paper exhibits higher prediction accuracy and wider applicable range.
2. A modeling technique is proposed to simulate the steel tube with random localized corrosion, and a refined finite element model is established for CFST subjected to long-term loading and random localized corrosion. The influences of corrosion types (uniform corrosion, random localized corrosion), volume corrosion rates, and material properties on the mechanical behaviors of CFST are clarified. By combining numerical simulation and literature review, databases of CFST with non-corrosion, uniform corrosion, and random localized corrosion are established, comprising a total of 4,038 samples. Seven ML algorithms are adopted to develop intelligent prediction models for the residual capacity of CFST columns. Given the complex influencing factors and the scarcity of data under corrosion conditions, active learning methods are used to facilitate ML model training. Improvements are made to existing active learning methods, especially in the selection strategy of initial data and the incorporation of the bagging method. The effectiveness of this optimization method is verified through comparative experiments.
3. Salt spray corrosion tests are conducted on 231 plates with dimensions of 280 mm×60 mm×6 mm, including Q355, Q460, and Q690 steel, with their macroscopic and microscopic morphologies collected before and after cleaning the corrosion products. The impacts of corrosion degree and steel strength on the corrosion morphology are quantitatively studied through surface morphology parameters. Based on the microscopic morphology images of steel surface obtained by super depth microscopy, a novel method for calculating the uniform corrosion component and localized corrosion component is proposed using the double integral. Through the above corrosion tests combined with the online searching engine, two image databases for corrosion evaluation and corrosion detection are established. Furthermore, the classification criterion for the corrosion degree of CFST is formulated based on the existing standard GB 50144-2019, providing valuable reference and data support for the subsequent development of DL models.
4. Five typical semantic segmentation algorithms are used to establish detection models for the corrosion location of CFST structures among complex backgrounds. The developed models are improved through attention mechanism and multi-scale feature fusion, with their effectiveness validated using collected actual engineering images. Evaluation models for the corrosion level of CFST are created through image classification algorithms and optimized by attention mechanism, color space, as well as multi-scale output. Based on various established ML models, a method is proposed to construct an intelligent diagnostic system for marine CFST, providing decision-making reference for the safe service and maintenance reinforcement of existing structures, thereby reducing economic losses and safety hazards caused by the corrosion.

[1] 樊健生, 王琛, 宋凌寒. 土木工程智能计算分析研究进展与应用[J]. 建筑结构学报, 2022, 43(9): 1-22.
[2] 刘红波, 张帆, 陈志华, 等. 人工智能在土木工程领域的应用研究现状及展望[J].土木与环境工程学报(中英文), 2024, 46(1): 14-32.
[3] 鲍跃全, 李惠. 人工智能时代的土木工程[J]. 土木工程学报, 2019, 52(5): 1-11.
[4] 庞玉兰, 王欣. “海洋强国”战略背景下海南海洋产业发展策略[J]. 改革与战略, 2016, 32(9): 82-84.
[5] 李乃胜. 从海洋大国到海洋强国[J]. 政工学刊, 2013, (2): 76-78.
[6] HAN L H, Li W, BJORHOVDE R. Developments and advanced applications of concrete-filled steel tubular (CFST) structures: Members[J]. Journal of Constructional Steel Research, 2014, 100: 211-228.
[7] ELLOBOFY E, YOUNG B, LAM D. Behaviour of normal and high strength concrete-filled compact steel tube circular stub columns[J]. Journal of Constructional Steel Research, 2006, 62(7): 706-715.
[8] CHEN P, WANG Y Y, ZHANG S M. Size effect prediction on axial compression strength of circular CFST columns[J]. Journal of Constructional Steel Research, 2020, 172: 106221.
[9] 干海子特大桥世界最长全钢管桁架梁公路桥[J]. 科学大观园, 2021, (15): 6.
[10] 日照-仪征原油管道及配套工程项目(日照港岚山港区30万吨级原油码头工程)[J]. 城乡建设, 2017, (18): 80-81.
[11] 侯保荣. 海洋钢结构浪花飞溅区腐蚀防护技术[J]. 中国材料进展, 2014, 33(1): 26-31.
[12] MELCHERS R E. Recent progress in the modeling of corrosion of structural steel immersed in seawaters[J]. Journal of Infrastructure Systems, 2006, 12(3): 154-162.
[13] HAN L H, HOU C C, WANG Q L. Behavior of circular CFST stub columns under sustained load and chloride corrosion[J]. Journal of Constructional Steel Research, 2014, 103: 23-36.
[14] HAN L H, HOU C, WANG Q L. Square concrete filled steel tubular (CFST) members under loading and chloride corrosion: Experiments[J]. Journal of Constructional Steel Research, 2012, 71: 11-25.
[15] HUANG H, GUO L H, QU B, et al. Tests of circular concrete-filled steel tubular stub columns with artificial notches representing local corrosions[J]. Engineering Structures, 2021, 242: 112598.
[16] HUANG H J, JIA C, ZHAO O, et al. Local corrosion morphology analysis and simplification of low carbon steel plates[J]. Ocean Engineering, 2023, 268: 113372.
[17] LI G, HOU C, SHEN L M, et al. Performance and strength calculation of CFST columns with localized pitting corrosion damage[J]. Journal of Constructional Steel Research, 2022, 188: 107011.
[18] LI G, HOU C, SHEN L M. Combined compression-bending performance and design of CFST with localised pitting corrosion[J]. Journal of Constructional Steel Research, 2022, 192: 107247.
[19] FRANCE E J. The Alexander L. Kielland disaster revisited: A review by an experienced welding engineer of the catastrophic north sea platform collapse[J]. Journal of Failure Analysis and Prevention, 2019, 19(4): 875-881.
[20] 侯保荣, 张盾, 王鹏. 海洋腐蚀防护的现状与未来[J]. 中国科学院院刊, 2016, 31(12): 1326-1331.
[21] 周志华. 机器学习[M]. 北京: 清华大学出版社, 2016.
[22] MCCULLOCH W S, PITTS W. A logical calculus of the ideas immanent in nervous activity[J]. Bulletin of Mathematical Biophysics, 1943, 5: 113-115.
[23] ROSENBLATT F. The perceptron: A probabilistic model for information storage and organization in the brain[J]. Psychological Review, 1958, 65(6): 386-408.
[24] 邓文杰. 基于聚粒子群算法的神经网络权值优化方法[J]. 计算机技术与发展, 2017, 27(10): 16-18+23.
[25] GOLDBERG D E. Genetic algorithms in search, optimization, and machine learning[M]. USA: Addison-Wesley Publishing Company, 1988.
[26] MOODY J, DARKEN C J. Fast learning in networks of locally-tuned processing units[J]. Neural Computation, 1989, 1: 281-294.
[27] JANG J S R. ANFIS: Adaptive-network-based fuzzy inference system[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1993, 23(3): 665-685.
[28] VAPNIK V N. The nature of statistical learning theory (Second edition)[M]. Springer, 2000.
[29] CHANG C C, LIN C J. LIBSVM: A library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology, 2011, 2(3): 1-27
[30] RASMUSSEN C E, WILLIAMS C K I. Gaussian processes for machine learning[M]. MIT Press, 2006.
[31] FRIEDMAN J H. Multivariate adaptive regression splines[J]. The Annals of Statistics, 1991, 19(1): 1-67.
[32] 刘晓, 王思迈, 卢磊, 等. 机器学习预测混凝土材料耐久性的研究进展[J]. 硅酸盐学报, 2023, 51(8): 2062-2073.
[33] 谢龙隆, 喻泽成, 余波. 基于类不平衡数据集的钢筋混凝土柱破坏模式判别方法[J]. 建筑结构学报, 2023, 44(5): 273-285.
[34] QUINLAN J R. C4. 5: Programs for machine learning[M]. Elsevier, 2014.
[35] BREIMAN L. Classification and regression trees[M]. Routledge, 2017.
[36] QUINLAN J R. Learning with continuous classes[C]. Proceedings of the 5th Australian Joint Conference on Artificial Intelligence, 1992: 343-348.
[37] WANG Y, WITTEN I H. Induction of model trees for predicting continuous classes[C]. Working Paper 96/23, University of Waikato, Hamilton, New Zealand, 1996: 1-12.
[38] ZHOU Z H. Ensemble methods: foundations and algorithms[M]. Boca Raton: CRC Press, 2012.
[39] BREIMAN L. Random forests[J]. Machine Learning, 2001, 45: 5-32.
[40] CHEN T Q, GUESTRIN C. XGBoost: A scalable tree boosting system[C]. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, 2016: 785-794.
[41] DESPOTOVIC M, NEDIC V, DESPOTOVIC D, et al. Review and statistical analysis of different global solar radiation sunshine models[J]. Renewable and Sustainable Energy Reviews, 2015, 52: 1869-1880.
[42] 惠存, 李永刚, 李克, 等. 多灾耦合作用下钢管混凝土柱承载性能研究进展[J]. 防灾减灾工程学报, 2022, 42(2): 259-268.
[43] 熊清清, 钱佳慧, 陈志华. 腐蚀后钢管混凝土力学性能研究进展[J]. 钢结构(中英文), 2022, 37(7): 1-19.
[44] HAN L H, HUA Y X, HOU C, et al. Circular concrete-filled steel tubes subjected to coupled tension and chloride corrosion[J]. Journal of Structural Engineering, 2017, 143(10): 04017134.
[45] 王庆利, 李清林, 屈绍娥. 长期荷载下圆钢管混凝土梁的耐腐蚀性能试验研究[J]. 建筑结构学报, 2015, 36: 50-55.
[46] 王庆利, 冯立明, 屈绍娥. 圆钢管混凝土轴压短柱在长期荷载-氯盐腐蚀耦合作用下的试验研究[J]. 土木工程学报, 2015, 48(S1): 48-52.
[47] HOU C C, HAN L H, WANG Q L, et al. Flexural behavior of circular concrete filled steel tubes (CFST) under sustained load and chloride corrosion[J]. Thin-Walled Structures, 2016, 107: 182-196.
[48] HUA Y X, HAN L H, WANG Q L, et al. Behaviour of square CFST beam-columns under combined sustained load and corrosion: Experiments[J]. Thin-Walled Structures, 2019, 136: 353-366.
[49] HOU C, HAN L H, ZHAO X L. Full-range analysis on square CFST stub columns and beams under loading and chloride corrosion[J]. Thin-Walled Structures, 2013, 68: 50-64.
[50] 陈梦成, 张凡孟, 黄宏, 等. 酸雨环境下钢管混凝土柱轴压性能研究[J]. 华东交通大学学报, 2015, 32(3): 86-90.
[51] 陈梦成, 林博洋, 黄宏. 锈蚀方钢管混凝土短柱轴压承载力研究[J]. 钢结构, 2017, 32(5): 110-116.
[52] 黄宏, 胡志慧, 朱琪, 等. 酸雨环境下方钢管再生混凝土偏心受压试验研究[J]. 铁道学报, 2018, 40(4): 90-97.
[53] 文成伍. 锈蚀方钢管再生混凝土长柱双向偏压承载性能研究[D]. 绵阳: 西南科技大学, 2021.
[54] GAO S, PENG Z, WANG X, et al. Compressive behavior of circular hollow and concrete-filled steel tubular stub columns under atmospheric corrosion[J]. Steel and Composite Structures, 2019, 33(4): 615-627.
[55] GAO S, GUO L H, ZHANG S M, et al. Performance degradation of circular thin-walled CFST stub columns in high-latitude offshore region[J]. Thin-Walled Structures, 2020, 154: 106906.
[56] YUAN F, CHEN M C, HUANG H, et al. Circular concrete filled steel tubular (CFST) columns under cyclic load and acid rain attack: Test simulation[J]. Thin-Walled Structures, 2018, 122: 90-101.
[57] 陈梦成, 方苇, 黄宏. 模拟酸雨腐蚀钢管混凝土构件静力性能研究[J]. 工程力学, 2020, 37(2): 34-43.
[58] ZHANG F, XIA J, Li G, et al. Degradation of axial ultimate load-bearing capacity of circular thin-walled concrete-filled steel tubular stub columns after corrosion[J]. Materials, 2020, 13(3): 795.
[59] 陈梦成, 李骐, 谢力. 锈蚀对钢管混凝土柱协同工作机理的影响[J]. 广西大学学报(自然科学版), 2017, 42(1): 60-68.
[60] 许开成, 陈梦成, 何小平. 氯离子腐蚀对钢管混凝土界面黏结性能影响的试验分析[J]. 工业建筑, 2013, 43(1): 71-74.
[61] CHANG X, FU L, ZHAO H B, et al. Behaviors of axially loaded circular concrete-filled steel tube (CFT) stub columns with notch in steel tubes[J]. Thin-Walled Structures, 2013, 73(12): 273-280.
[62] DING F X, FU L, YU Z W. Behaviors of axially loaded square concrete-filled steel tube (CFST) Stub columns with notch in steel tube[J]. Thin-Walled Structures, 2017, 115: 196-204.
[63] GUO L H, HUANG H J, JIA C, et al. Axial behavior of square CFST with local corrosion simulated by artificial notch[J]. Journal of Constructional Steel Research, 2020, 174: 106314.
[64] 李振林, 林思奇, 赵衍刚, 等. 局部腐蚀钢管混凝土短柱轴压性能试验研究[J]. 工程力学, 2023, 40(8): 170-180.
[65] 钱文磊. 局部腐蚀后圆钢管混凝土短柱轴压力学性能研究[D]. 南昌: 华东交通大学, 2021.
[66] WANG H, ZHANG Z, QIAN H, et al. Effect of local corrosion on the axial compression behavior of circular steel tubes[J]. Engineering Structures, 2020, 224: 111205.
[67] 钟磊. 局部腐蚀方钢管混凝土轴压短柱力学性能[D]. 哈尔滨: 哈尔滨工业大学, 2021.
[68] 陈梦成, 陈娜茹. 局部腐蚀对圆钢管混凝土柱轴向承载力的影响分析[J]. 华东交通大学学报, 2019, 36(3): 80-90.
[69] 王东锋, 邵永波, 欧佳灵. CFRP加固含腐蚀缺陷圆钢管混凝土短柱轴压承载力试验研究[J]. 工程力学, 2021, 38(10): 188-199.
[70] 黄海家, 徐炎, 郭兰慧, 等. CFRP布加固局部腐蚀圆钢管混凝土轴压短柱受力性能[J]. 建筑结构学报, 2020, 41(S1): 109-116.
[71] RIVAS D, CALEYO F, VALOR A, et al. Extreme value analysis applied to pitting corrosion experiments in low carbon steel: Comparison of block maxima and peak over threshold approaches[J]. Corrosion Science, 2008, 50(11): 3193-3204.
[72] WANG R H, GUO H C, SHENOI R A. Experimental and numerical study of localized pitting effect on compressive behavior of tubular members[J]. Marine Structures, 2020, 72: 102784.
[73] YUAN Y, ZHANG N, LIU H Q, et al. Influence of random pit corrosion on axial stiffness of thin-walled circular tubes[J]. Structures, 2020, 28: 2596-2604.
[74] WANG R H, SHENOI R A. Experimental and numerical study on ultimate strength of steel tubular members with pitting corrosion damage[J]. Marine Structures, 2019, 64: 124-137.
[75] 徐阳, 金晓威, 李惠. 土木工程智能科学与技术研究现状及展望[J]. 建筑结构学报, 2022, 43(9): 23-35.
[76] 金浏, 赵瑞, 杜修力. 混凝土抗压强度尺寸效应的神经网络预测模型[J]. 北京工业大学学报, 2021, 47(3): 260-268.
[77] 张研, 王鹏鹏, 吴哲康. 基于相关向量机模型的混凝土综合性能预测[J]. 硅酸盐通报, 2022, 41(1): 118-125.
[78] YAN K Z, SHI C J. Prediction of elastic modulus of normal and high strength concrete by support vector machine[J]. Construction and Building Materials, 2010, 24: 1479-1485.
[79] 郝艳娥, 翟振东. 矩形钢管混凝土短柱轴压承载力神经网络评估[J]. 长安大学学报(建筑与环境科学版), 2004, 21(3): 24-28.
[80] 朱美春, 王清湘, 冯秀峰. 方钢管混凝土短柱轴心受压承载力的神经网络模拟[J]. 计算力学学报, 2006, 23(3): 353-356.
[81] DU Y S, CHEN Z H, ZHANG C Q, et al. Research on axial bearing capacity of rectangular concrete-filled steel tubular columns based on artificial neural networks[J]. Frontiers of Computer Science, 2017, 11: 863-873.
[82] 张清允, 陈志华, 王小盾. 基于神经网络的矩形钢管高强混凝土计算方法[J]. 石家庄铁道大学学报(自然科学版), 2018, 31(1): 11-15.
[83] TRAN V L, THAI D K, KIM S E. Application of ANN in predicting ACC of SCFST column[J]. Composite Structures, 2019, 228: 111332.
[84] NGUYEN H Q, LY H B, TRAN V Q, et al. Optimization of artificial intelligence system by evolutionary algorithm for prediction of axial capacity of rectangular concrete filled steel tubes under compression[J]. Materials, 2020, 13: 1205.
[85] ZARRINGOL M, THAI H T, THAI S, et al. Application of ANN to the design of CFST columns[J]. Structures, 2020, 28: 2203-2220.
[86] MAI S H, SEGHIER M E A B, NGUYEN P L, et al. A hybrid model for predicting the axial compression capacity of square concrete-filled steel tubular columns[J]. Engineering with Computers, 2020, 38: 1205-1222.
[87] LE T T. Practical machine learning-based prediction model for axial capacity of square CFST columns[J]. Mechanics of Advanced Materials and Structures, 2020, 29(12): 1782-1797.
[88] BEN SEGHIER M E A, GAO X Z, JAFARI-ASL J, et al. Modeling the nonlinear behavior of ACC for SCFST columns using experimental-data and a novel evolutionary-algorithm[J]. Structures, 2021, 30: 692-709.
[89] NASER M Z, THAI S, THAI H T. Evaluating structural response of concrete-filled steel tubular columns through machine learning[J]. Journal of Building Engineering, 2021, 34: 101888.
[90] LEE S, VO T P, THAI H T, et al. Strength prediction of concrete-filled steel tubular columns using Categorical Gradient Boosting algorithm[J]. Engineering Structures, 2021, 238: 112109.
[91] KHEYRODDIN A, NADERPOUR H, AHMADI M. Performance of circular concrete filled steel tube members subjected to axial load[C]. 4th International Conference on Concrete and Development, 2013.
[92] AHMADI M, NADERPOUR H, KHEYRODDIN A. Utilization of artificial neural networks to prediction of the capacity of CCFT short columns subject to short term axial load[J]. Archives of Civil and Mechanical Engineering, 2014, 14: 510-517.
[93] GÜNEYISI E M, GÜLTEKIN A, MERMERDAŞ K. Ultimate capacity prediction of axially loaded CFST short columns[J]. International Journal of Steel Structures 2016, 16(1): 99-114.
[94] NOUR A I, GÜNEYISI E M. Prediction model on the compressive strength of recycled aggregate concrete filled steel tube columns[J]. Composites Part B: Engineering, 2019, 173: 106938.
[95] TRAN V L, THAI D K, NGUYEN D D. Practical artificial neural network tool for predicting the axial compression capacity of circular concrete-filled steel tube columns with ultra-high-strength concrete[J]. Thin-Walled Structures, 2020, 151: 106720.
[96] VU Q V, TRUONG V H, THAI H.T. Machine learning-based prediction of CFST columns using gradient tree boosting algorithm[J]. Composite Structures, 2021, 259: 113505.
[97] LYU F, FAN X Y, DING F X, et al. Prediction of the axial compressive strength of circular concrete-filled steel tube columns using sine cosine algorithm-support vector regression[J]. Composite structures, 2021, 273: 114282.
[98] 陈曦泽, 贾俊峰, 白玉磊, 等. 基于XGBoost-SHAP的钢管混凝土柱轴向承载力预测模型[J]. 浙江大学学报(工学版), 2023, 57(6): 1061-1070.
[99] XU J J, WANG Y M, REN R, et al. Performance evaluation of recycled aggregate concrete-filled steel tubes under different loading conditions: Database analysis and modelling[J]. Journal of Building Engineering, 2020, 30: 101308.
[100] LY H B, PHAM B T, LE L M, et al. Estimation of axial load-carrying capacity of concrete-filled steel tubes using surrogate models[J]. Neural Computing and Applications, 2021, 33: 3437-3458.
[101] LIU X R, WU Y Q, ZHOU Y S. Axial compression prediction and GUI design for CCFST column using machine learning and shapley additive explanation[J]. Buildings, 2022, 12: 698.
[102] BASARIR H, ELCHALAKANI M, KARRECH A. The prediction of ultimate pure bending moment of concrete-filled steel tubes by adaptive neuro-fuzzy inference system (ANFIS)[J]. Neural Computing and Applications, 2019, 31(S2): 1239-1252.
[103] 黄乐雯, 党文刚. 基于神经网络的钢管混凝土纯弯构件承载力预测[J]. 湖南科技学院学报, 2022, 43(3): 16-20.
[104] HANOON A N, ZAND A, YASEEN Z M. Designing new hybrid artificial intelligence model for CFST beam flexural performance prediction[J]. Engineering with Computers, 2022, 38: 3109-3135.
[105] GAO H G. Calculation of load-carrying capacity of square concrete filled tube columns based on neural network[J]. Applied Mechanics and Materials, 2011, 71-78: 847-850.
[106] SAADOON A S, NASSER K Z, MOHAMED I Q. A neural network model to predict ultimate strength of rectangular concrete filled steel tube beam-columns[J]. Engineering and Technology Journal, 2012, 30(19): 3328-3340.
[107] LUAT N V, LEE J, LEE D H, et al. GS-MARS method for predicting the ultimate load-carrying capacity of rectangular CFST columns under eccentric loading[J]. Computers and Concrete, 2020, 25(1): 1-14.
[108] 李帼昌, 陈博文, 杨志坚, 等. 高强方钢管高强混凝土长柱偏压性能[J]. 建筑钢结构进展, 2022, 24(6): 40-53+108.
[109] 刘坚, 彭林苗, 马宏伟, 等. 高强钢管高强混凝土短柱力学性能智能模型研究[J]. 建筑钢结构进展, 2023, 25(11): 12-20.
[110] NGUYEN M S T, TRINH M C, KIM S E. Uncertainty quantification of ultimate compressive strength of CCFST columns using hybrid machine learning model[J]. Engineering with Computers, 2022, 38: 2719-2738
[111] LU X Z, LIAO W J, ZHANG Y, et al. Intelligent structural design of shear wall residence using physics-enhanced generative adversarial networks[J]. Earthquake Engineering & Structural Dynamics, 2022, 51(7): 1657-1676.
[112] MANGALATHU S, JEON J S. Regional seismic risk assessment of infrastructure systems through machine learning: Active learning approach[J]. Journal of Structural Engineering, 2020, 146: 04020269.
[113] ZHANG H Y, CHENG X W, LI Y, et al. Rapid seismic damage state assessment of RC frames using machine learning methods[J]. Journal of Building Engineering, 2023, 65: 105797.
[114] DOUAK F, MELGANI F, BENOUDJIT N. Kernel ridge regression with active learning for wind speed prediction[J]. Applied Energy, 2013, 103: 328-340.
[115] 张函, 钱权, 武星. 面向材料数据的主动回归学习方法[J]. 工程科学学报, 2023, 45(7): 1232-1237.
[116] SAADALLAH A, EGOROV A, CAO B T, et al. Active learning for accurate settlement prediction using numerical simulations in mechanized tunneling[J]. Procedia CIRP, 2019, 81: 1052-1058.
[117] 岳清瑞, 徐刚, 刘晓刚. 桥梁裂缝智能识别与监测方法研究[J]. 中国公路学报, 2024, 37(2): 16-28.
[118] 杨娜, 王烁, 汪德佳. 基于计算机视觉的藏式古建筑石砌体壁画墙裂缝生长变形监测[J/OL]. 工程力学, 2024: 1-15.
[119] 黄永, 于建琦, 林旭川, 等. 基于深度学习的建筑破坏状态智能评估研究[J]. 自然灾害学报, 2023, 32(4): 148-158.
[120] 尹文博, 徐善华. 图像处理在表征大气酸腐蚀钢结构表面颜色特征中的应用[J]. 华北水利水电学院学报, 2011, 32(3): 21-24.
[121] 孔德英, 王守琰, 宋诗哲. 金属材料腐蚀形貌图像与实海挂片数据的相关性研究[J]. 中国腐蚀与防护学报, 2001, 21(6): 34-38.
[122] CHEN P H, CHANG L M. Artificial intelligence application to bridge painting assessment[J]. Automation in Construction, 2003, 12: 431-445.
[123] CHEN P H, YANG Y C, CHANG L M. Automated bridge coating defect recognition using adaptive ellipse approach[J]. Automation in Construction, 2009, 18: 632-643.
[124] SHEN H K, CHEN P H, CHANG L M. Automated steel bridge coating rust defect recognition method based on color and texture feature[J]. Automation in Construction, 2013, 31: 338-356.
[125] LIAO K W, LEE Y T. Detection of rust defects on steel bridge coatings via digital image recognition[J]. Automation in Construction, 2016, 71: 294-306.
[126] YAO Y, YANG Y, WANG Y P, et al. Artificial intelligence-based hull structural plate corrosion damage detection and recognition using convolutional neural network[J]. Applied Ocean Research, 2019, 90: 101823.
[127] 朱劲松, 李欢, 王世芳. 基于卷积神经网络和迁移学习的钢桥病害识别[J]. 长安大学学报(自然科学版), 2021, 41(3): 52-63.
[128] WANG Y, SHEN X L, WU K, et al. Corrosion grade recognition for weathering steel plate based on a convolutional neural network[J]. Measurement Science and Technology, 2022, 33: 095014.
[129] 陈爽. 基于深度神经网络的耐候桥梁钢锈蚀图像识别[D]. 北京: 北京交通大学, 2020.
[130] 陆廷杰, 刘东海, 齐志龙. 基于深度学习的水下钢结构锈蚀识别与评价[J]. 天津大学学报(自然科学与工程技术版), 2023, 56(7): 713-722.
[131] 王达磊, 彭博, 潘玥, 等. 基于深度神经网络的锈蚀图像分割与定量分析[J]. 华南理工大学学报(自然科学版), 2018, 46(12): 121-127.
[132] 郝善行. 基于计算机视觉的钢结构局部腐蚀检测及特征提取[D]. 哈尔滨: 哈尔滨工业大学, 2022.
[133] XU J, GUI C Q, HAN Q H. Recognition of rust grade and rust ratio of steel structures based on ensembled convolutional neural network[J]. Computer-Aided Civil and Infrastructure Engineering, 2020, 35: 1160-1174.
[134] 中华人民共和国住房和城乡建设部. GB 50010-2010, 混凝土结构设计规范[S]. 北京: 中国建筑工业出版社, 2015.
[135] GOODE C D, LAM D. Concrete-filled steel tube columns-tests compared with Eurocode 4[C]. Composite Construction VI. USA: Colorado, 2008.
[136] 中华人民共和国住房和城乡建设部. GB/T 50081-2019, 混凝土物理力学性能试验方法标准[S]. 北京: 中国建筑工业出版社, 2019.
[137] 中华人民共和国住房和城乡建设部. GB 50936-2014, 钢管混凝土结构技术规范[S]. 北京: 中国建筑工业出版社, 2014.
[138] American Concrete Institute. Building code requirements for structural concrete (ACI 318-14) and commentary (ACI 318R-14), ACI 318R-14[S]. Detroit: American Concrete Institute, 2014.
[139] British Standards Institution. Eurocode 4: Design of composite steel and concrete structures-Part 1.1: General rules and rules for buildings, BS EN 1994-1-1: 2004[S]. London: British Standards Institution, 2004.
[140] American Institute of Steel Construction. Specification for Structural Steel Buildings, AISC 360-16[S]. Chicago: American Institute of Steel Construction, 2016.
[141] Standards Australia/New Zealand Committee. Composite structures-Composite steel-concrete construction in buildings, AS/NZS 2327: 2017[S]. Sydney/Wellington: Standards Australia/New Zealand Committee, 2017.
[142] 中华人民共和国住房和城乡建设部. GB/T 51446-2021钢管混凝土混合结构技术标准[S]. 北京: 中国建筑工业出版社, 2021.
[143] DENAVIT M. Steel-concrete composite column database[DB]. 2020.
[144] 肖从真, 蔡绍怀, 徐春丽. 钢管混凝土抗剪性能试验研究[J]. 土木工程学报, 2005, 38(4): 5-11.
[145] 钱稼茹, 崔瑶, 方小丹. 钢管混凝土柱受剪承载力试验[J]. 土木工程学报, 2007, 40(5): 1-9.
[146] KIM Y H, YOU S K, JUNG J H, et al. Strengthening effects of the shear key on the flexural behavior of concrete filled circular tube[J]. Steel Structures, 2006, 6: 183-190.
[147] 钱稼茹, 王刚, 赵作周, 等. 钢管高强混凝土构件截面弯矩-曲率全曲线研究[J]. 工业建筑, 2004, 34(8): 70-72+89.
[148] CAI Y C, CHAN T M, YOUNG B. Strength predictions of circular hollow section T-joints of steel grade 1100 MPa[J]. Journal of Constructional Steel Research, 2022, 188: 107003.
[149] LIU F Q, GARDNER L, YANG H. Post-fire behaviour of reinforced concrete stub columns confined by circular steel tubes[J]. Journal of Constructional Steel Research 2014, 102: 82-103.
[150] 韩林海. 钢管混凝土结构: 理论与实践(第三版)[M]. 北京: 科学出版社, 2016.
[151] EKMEKYAPAR T, AL-ELIWI B J M. Experimental behaviour of circular concrete filled steel tube columns and design specifications[J]. Thin-Walled Structures, 2016, 105: 220-230.
[152] SU M N, CAI Y C, CHEN X R, et al. Behaviour of concrete-filled cold-formed high strength steel circular stub columns[J]. Thin-Walled Structures, 2020, 157: 107078.
[153] ZEGHICHE J, CHAOUI K. An experimental behaviour of concrete-filled steel tubular columns[J]. Journal of Constructional Steel Research, 2005, 61: 53-66.
[154] 马丽盟, 李书文, 祝磊, 等. 高强圆钢管混凝土短柱轴心受压试验研究[J]. 工业建筑, 2016, 46(7): 16-21.
[155] PEARSON K. Note on regression and inheritance in the case of two parents[J]. Proceedings of the Royal Society of London, 1895, 58(347-352): 240-242.
[156] TRAN V L, THAI D K. A new empirical formula for prediction of the axial compression capacity of CCFT columns[J]. Steel and Composite Structures, 2019, 33(2): 181-194.
[157] 张乐均. 钢管混凝土短柱与中长柱的界限长细比研究[D]. 包头: 内蒙古科技大学, 2017.
[158] THAI S, THAI H T, UY B, et al. Concrete-filled steel tubular columns: Test database, design and calibration[J]. Journal of Constructional Steel Research, 2019, 157: 161-181.
[159] HAN L H. Tests on stub columns of concrete-filled RHS sections[J]. Journal of Constructional Steel Research, 2002, 58: 353-372.
[160] LIU D L, GHO W M, YUAN J. Ultimate capacity of high-strength rectangular concrete-filled steel hollow section stub columns[J]. Journal of Constructional Steel Research, 2003, 59: 1499-1515.
[161] DU Y S, CHEN Z H, XIONG M X. Experimental behavior and design method of rectangular concrete-filled tubular columns using Q460 high-strength steel[J]. Construction and Building Materials, 2016, 125: 856-872.
[162] HUANG Z C, UY B, LI D X, et al. Behaviour and design of ultra-high-strength CFST members subjected to compression and bending[J]. Journal of Constructional Steel Research, 2020, 175: 106351.
[163] HOSSAIN K M A, CHU K. Confinement of six different concretes in CFST columns having different shapes and slenderness[J]. International Journal of Advanced Structural Engineering, 2019, 11: 255-270.
[164] ISLAM M M, ALI R B, BEGUM M, et al. Experimental study of square concrete-filled welded cold-formed steel columns under concentric loading[J]. Arabian Journal for Science and Engineering, 46, 2021: 4225-4237.
[165] 福建省住房和城乡建设厅. DBJ/T13-51-2010, 钢管混凝土结构技术规程[S]. 福州: 福建省住房和城乡建设厅, 2010.
[166] HAN L H, LU H, YAO G H, et al. Further study on the flexural behaviour of concrete-filled steel tubes[J]. Journal of Constructional Steel Research, 2006, 62(6): 554-565.
[167] FISHER A, RUDIN C, DOMINICI F. Model class reliance: Variable importance measures for any machine learning model class, from the “rashomon” perspective[J/OL]. arXiv: 1801.01489, 2018.
[168] BARKHORDARI M S, TEHRANIZADEH M. Response estimation of reinforced concrete shear walls using artificial neural network and simulated annealing algorithm[J]. Structures, 2021, 34: 1155-1168.
[169] 丁发兴, 余志武. 圆钢管自密实混凝土纯弯力学性能[J]. 交通运输工程学报, 2006, 6(1): 63-68+79.
[170] 李斌, 闻洋. 矩形钢管混凝土梁受弯性能试验研究[J]. 建筑结构, 2010, 40(6): 51-52+80.
[171] WANG Y, LIN H, LAI Z, et al. Flexural behavior of high-strength rectangular concrete-filled steel tube members[J]. Journal of Structural Engineering, 2022, 148(1): 04021230.
[172] ZHANG R, CHEN S M, GU P, et al. Structural behavior of UHPC filled steel tubular columns under eccentric loading[J]. Thin-Walled Structures, 2020, 156: 106959.
[173] CHEN J, WANG Y Y, ROEDER C W, et al. Behavior of normal-strength recycled aggregate concrete filled steel tubes under combined loading[J]. Engineering Structures, 2017, 130: 23-40.
[174] TU C L, SHI Y J, LIU D, et al. Behavior and general design method of concrete-filled high-strength steel tube (CFHST) columns[J]. Engineering Structures, 2021, 243: 112506.
[175] SNOEK J, LAROCHELLE H, ADAMS R P. Practical Bayesian optimization of machine learning algorithms[C]. Proceedings of the 25th International Conference on Neural Information Processing Systems, 2012: 2951-2959.
[176] YAN J B, CHEN A Z, ZHU J S. Tests and analysis on circular UHPFRC-filled steel tubular stub columns under eccentric compression[J]. Journal of Constructional Steel Research, 2021, 178: 106501.
[177] ELZIEN A, JI B H, FU Z Q, et al. The behavior of lightweight aggregate concrete filled steel tube columns under eccentric loading[J]. Steel and Composite Structures, 2011, 11(6): 469-488.
[178] TAM V W Y, XIAO J Z, LIU S, et al. Behaviors of recycled aggregate concrete-filled steel tubular columns under eccentric loadings[J]. Frontiers of Structural and Civil Engineering, 2019, 13(3): 628-639
[179] 钟善桐. 钢管混凝土偏心受压构件紧箍力的研究[J]. 哈尔滨建筑工程学院学报, 1983, (3): 1-18.
[180] 余志武, 丁发兴. 圆钢管混凝土偏压柱的力学性能[J]. 中国公路学报, 2008, 21(1): 40-46.
[181] GOPAL S R, MANOHARAN P D. Tests on fiber reinforced concrete filled steel tubular columns[J]. Steel and Composite Structures, 2004, 4(1): 37-48.
[182] REN Q X, HAN L H, HOU C, et al. Experimental behaviour of tapered CFST columns under combined compression and bending[J]. Journal of Constructional Steel Research, 2017, 128: 39-52.
[183] MATSUI C, TSUDA K, OZAKI I, et al. Strength of slender concrete filled steel tubular columns[J]. Journal of Structural and Construction Engineering (Transactions of AIJ), 1997, (494): 137-144.
[184] O’SHEA M D, BRIDGE R Q. Design of circular thin-walled concrete filled steel tubes[J]. Journal of Structural Engineering, 2000, 126(11): 1295-1303.
[185] PORTOLÉS J, ROMERO M L, BONET J L, et al. Experimental study of high strength concrete-filled circular tubular columns under eccentric loading[J]. Journal of Constructional Steel Research, 2011, 67(4): 623-633.
[186] LUNDBERG S M, LEE S I. A Unified Approach to Interpreting Model Predictions[C]. 31st Conference on Neural Information Processing Systems (NIPS 2017), 2017: 1-10.
[187] IBAÑEZ C, ROMERO M L, ESPINÓS A, et al. Ultra-high strength concrete on eccentrically loaded slender circular concrete-filled dual steel columns[J]. Structures, 2017, 12: 64-74.
[188] 何益斌, 肖阿林, 郭健, 等. 钢骨-钢管自密实高强混凝土偏压柱力学性能试验研究[J]. 建筑结构学报, 2010, 31(4): 102-109.
[189] ASTERIS P G, LEMONIS M E, LE T T, et al. Evaluation of the ultimate eccentric load of rectangular CFSTs using advanced neural network modeling[J]. Engineering Structures, 2021, 248: 113297.
[190] 张素梅, 郭兰慧, 王玉银, 等. 方钢管高强混凝土偏压构件的试验研究与理论分析[J]. 建筑结构学报, 2004, 25(1): 17-24+70.
[191] XIONG M X, XIONG D X, LIEW J Y R. Behaviour of steel tubular members infilled with ultra high strength concrete[J]. Journal of Constructional Steel Research, 2017, 138: 168-183.
[192] XIONG M X, XIONG D X, LIEW J Y R. Flexural performance of concrete filled tubes with high tensile steel and ultra-high strength concrete[J]. Journal of Constructional Steel Research, 2017, 132: 191-202.
[193] YANG Y F, HAN L H. Experimental behaviour of recycled aggregate concrete filled steel tubular columns[J]. Journal of Constructional Steel Research, 2006, 62(12): 1310-1324.
[194] YANG Y F, HAN L H. Compressive and flexural behaviour of recycled aggregate concrete filled steel tubes (RACFST) under short-term loadings[J]. Steel and Composite Structures, 2006, 6(3): 257-284.
[195] 朱骏球. 圆钢管UHPC压弯构件N-M曲线研究[D]. 福州: 福州大学, 2018.
[196] FANG H, VISINTIN P. Structural performance of geopolymer-concrete-filled steel tube members subjected to compression and bending[J]. Journal of Constructional Steel Research, 2022, 188: 107026.
[197] WARDENIER J. Hollow sections in structural applications[M]. England: CIDECT Publications, 2002.
[198] HAN L H, TAO Z, LIU W. Effects of sustained load on concrete-filled hollow structural steel columns[J]. Journal of Structural Engineering, 2004, 130(9): 1392-1404.
[199] HAN L H, HOU C, HUA Y X. Concrete-filled steel tubes subjected to axial compression: Life-cycle based performance[J]. Journal of Constructional Steel Research, 2020, 169: 106063.
[200] NAKAI T, MATSUSHITA H, YAMAMOTO N, et al. Effect of pitting corrosion on local strength of hold frames of bulk carriers (1st report)[J]. Marine Structures, 2004, 17: 403-432
[201] BHANDARI J, KHAN F, ABBASSI R, et al. Modelling of pitting corrosion in marine and offshore steel structures-A technical review[J]. Journal of Loss Prevention in the Process Industries, 2015, 37: 39-62.
[202] 陈梦成, 方苇, 黄宏, 等. 锈蚀圆钢管再生混凝土轴压短柱受力性能研究[J]. 建筑结构学报, 2019, 40(12): 138-146.
[203] LI M, YAO L B, HE L S, et al. Experimental study on the compressive behavior of concrete filled steel tubular columns with regional corrosion[J]. Structures, 2022, 35: 882-892.
[204] LI W, HAN L H, ZHAO X L. Behavior of CFDST stub columns under preload, sustained load and chloride corrosion[J]. Journal of Constructional Steel Research 2015, 107: 12-23.
[205] XIA R L, JIA C, LIU C Y, et al. Non-uniform corrosion characteristics of the steel pipe pile exposed to marine environments[J]. Ocean Engineering, 2023, 272: 113873.
[206] ZHOU X G, HOU C, PENG J H, et al. Structural mechanism-based intelligent capacity prediction methods for concrete-encased CFST columns[J]. Journal of Constructional Steel Research, 2023, 202: 107769.
[207] RAYCHAUDHURI T, HAMEY L G C. Minimisation of data collection by active learning[C]. International Conference on Neural Networks (ICNN), 1995: 1338-1341.
[208] BURBIDGE R, ROWLAND J J, KING R D. Active learning for regression based on query by committee[C]. International Conference on Intelligent Data Engineering and Automated Learning, 2007: 209-218.
[209] WU D R, LIN C T, HUANG J. Active learning for regression using greedy sampling[J]. Information Sciences, 2018, 474: 90-105.
[210] STEINLEY D. K-means clustering: A half-century synthesis[J]. British Journal of Mathematical and Statistical Psychology, 2006, 59: 1-34.
[211] 王振尧, 于国才, 韩薇. 钢的大气暴露腐蚀与室内模拟加速腐蚀的相关性[J]. 腐蚀科学与防护技术, 2004, 2: 70-73.
[212] MELCHERS R E. Modelling immersion corrosion of structural steels in natural fresh and brackish waters[J]. Corrosion Science, 2006, 48: 4174-4201.
[213] 欧阳志恒, 崔光茫, 赵巨峰, 等. 基于分层级各向异性滤波的图像景深渲染算法[J]. 光学技术, 2018, 44(4): 469-475.
[214] YAMASHITA M, MAEDA A, UCHIDA H, et al. Crystalline rust compositions and weathering properties of steels exposed in nation-wide atmospheres for 17 Years[J]. Journal of the Japan Institute of Metals and Materials, 2001, 65(11): 967-971.
[215] 张素梅, 陈广锐, 李爱东, 等. Q420qD钢材应力腐蚀特征试验研究[J]. 建筑科学与工程学报, 2022, 39(1): 55-64.
[216] 贾晨. Q690E钢板腐蚀损伤及其腐蚀后力学性能研究[D]. 哈尔滨: 哈尔滨工业大学, 2020.
[217] 何建新, 秦晓洲, 易平, 等. Q235钢海洋大气腐蚀暴露试验研究[J]. 表面技术, 2006, 35(4): 21-23.
[218] 中华人民共和国国家质量监督检验检疫总局. GB/T 3505-2009, 产品几何技术规范(GPS) 表面结构轮廓法术语、定义及表面结构参数[S]. 北京: 中华人民共和国国家质量监督检验检疫总局, 2009.
[219] 中华人民共和国国家质量监督检验检疫总局. GB/T 33523.2-2017, 产品几何技术规范(GPS) 表面结构区域法第2部分: 术语、定义及表面结构参数[S]. 北京: 中华人民共和国国家质量监督检验检疫总局, 2017.
[220] 徐善华, 何羽玲, 秦广冲, 等. 钢材锈蚀率与表面三维粗糙度参数的关系[J]. 材料科学与工程学报, 2016, 34(2): 292-295.
[221] 中华人民共和国国家质量监督检验检疫总局. GB/T 8923.1-2011, 涂覆涂料前钢材表面处理表面清洁度的目视评定第1部分: 未涂覆过的钢材表面和全面清除原有涂层后的钢材表面的锈蚀等级和处理等级[S]. 北京: 中华人民共和国国家质量监督检验检疫总局, 2011.
[222] ASTM International. ASTM D610-08 (2019), Standard practice for evaluating degree of rusting on painted steel surfaces[S]. West Conshohocken: ASTM International, 2019.
[223] International Organization for Standardization (ISO). ISO4628-3, Paints and varnishes-Evaluation of degradation of coatings-Designation of quantity and size of defects, and of intensity of uniform changes in appearance-Part 3: Assessment of degree of rusting[S]. Switzerland: International Organization for Standardization (ISO), 2016.
[224] 中华人民共和国住房和城乡建设部. GB 50144-2019, 工业建筑可靠性鉴定标准[S]. 北京: 中国建筑工业出版社, 2019.
[225] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
[226] LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521(7553): 436-444.
[227] GOODFELLOW I, BENGIO Y, COURVILLE A. Deep learning[M]. MIT Press, 2016.
[228] BENGIO Y. Practical recommendations for gradient-based training of deep architectures[M]. Springer, 2012.
[229] SRIVASTAVA N, HINTON G, KRIZHEVSKY A, et al. Dropout: a simple way to prevent neural networks from overfitting[J]. The Journal of Machine Learning Research, 2014, 15(1): 1929-1958.
[230] IOFFE S, SZEGEDY C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[J/OL]. arXiv: 1502.03167, 2015.
[231] LIN T Y, MAIRE M, BELONGIE S, et al. Microsoft coco: Common objects in context[J]. arXiv: 1405.0312, 2014.
[232] SHELHAMER E, LONG J, DARRELLAR T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4): 640-651.
[233] RONNEBERGER O, FISCHER P, BROX T. U-Net: Convolutional networks for biomedical image segmentation[C]. Medical Image Computing and Computer-Assisted Intervention, 2015.
[234] QIN X, ZHANG Z, HUANG C, et al. U2-Net: Going deeper with nested U-structure for salient object detection[J]. Pattern Recognition, 2020, 106: 107404.
[235] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. Semantic image segmentation with deep convolutional nets and fully connected CRFs[J/OL]. arXiv: 1412.7062, 2014.
[236] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. DeepLab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 40(4): 834-848.
[237] CHEN L C, PAPANDREOU G, SCHROFF F, et al. Rethinking atrous convolution for semantic image segmentation[J/OL]. arXiv: 1706.05587, 2017.
[238] HOWARD A, SANDLER M, CHU G, et al. Searching for MobileNetV3[C]. International Conference on Computer Vision (ICCV), 2019.
[239] WOO S, PARK J, LEE J Y, et al. CBAM: Convolutional block attention module[C]. Proceedings of the European Conference on Computer Vision (ECCV), 2018.
[240] 徐阳. 基于计算机视觉的桥梁结构局部损伤识别方法研究[D]. 哈尔滨: 哈尔滨工业大学, 2019.
[241] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[C]. Proceedings of the 25th International Conference on Neural Information Processing Systems, 2012.
[242] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[J/OL]. arXiv: 1409.1556, 2014.
[243] SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions[C]. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015.
[244] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016.
[245] HOWARD A G, ZHU M, CHEN B, et al. MobileNets: Efficient convolutional neural networks for mobile vision applications[J/OL]. arXiv: 1704.04861, 2017.
[246] SANDLER M, HOWARD A, ZHU M, et al. MobileNetV2: Inverted residuals and linear bottlenecks[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018.
[247] PAN S J, YANG Q. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2009, 22(10): 1345-1359.
[248] CHEN W, WANG G, WU B, et al. A state-of-the-art survey of transfer learning in structural health monitoring[C]. 7th International Conference on Systems and Informatics (ICSAI), 2021.
[249] 郭增伟, 李龙景, 姚国文. 交变荷载与腐蚀环境耦合作用下拉索钢丝腐蚀行为特征及预测[J]. 重庆大学学报, 2018, 41(7): 48-57.