建筑结构损伤智能检测与响应智能预测研究综述OA北大核心CSTPCD
State-of-the-art of intelligent damage detection and response prediction of building structures
随着计算机和人工智能技术的高速发展,越来越多的智能技术被引入土木工程学科交叉研究与应用中.相比于传统土木工程技术,结构智能防灾显著提升了工作效率与精度,因此成为学科交叉的重要发展方向之一.为系统梳理结构智能防灾前沿研究进展,从建筑结构局部损伤智能检测与评估、建筑结构整体损伤智能检测与评估、建筑结构响应智能预测等三个方面,对建筑结构损伤智能检测与响应智能预测研究开展综述.归纳了主流研究技术路线和常用智能算法,分析了已有研究方法的优越性和局限性,并给出了研究挑战.针对现阶段研究中构件局部损伤智能检测与评估的工程应用受限、建筑结构整体损伤智能检测精度较低、建筑结构响应智能预测的可解释性与可靠性较差等问题,针对性地提出未来研究方向的展望和建议,包括提升损伤智能检测算法泛化性、通过超分辨率技术提升建筑结构整体损伤智能检测精度以及将空间信息和物理信息融入建筑结构响应智能预测方法等方面.
With the rapid development of computer and artificial intelligence technology,intelligent technologies have been increasingly applied to the interdisciplinary research and applications in civil engineering.Compared with traditional civil engineering technology,intelligent structural disaster prevention,mitigation,and maintenance technology have significantly improved efficiency and accuracy,so it has become one important development directions of cross-disciplinary.In order to systematically sort out and display the research frontier progress,a comprehensive review of the state-of-the-art intelligent damage detection and response prediction of building structures is carried out.The review is divided into three aspects,namely intelligent local damage detection and evaluation,intelligent global damage detection and evaluation,and intelligent prediction of building structure response.Mainstream research technology routes and commonly used intelligent algorithms are summarized and analyzed,the advantages and limitations of existing research methods are discussed,and research challenges are identified based on current studies.The engineering application of intelligent detection and assessment of local damage in components in the current stage of research faces limitations.The overall accuracy of intelligent detection of damage in building structures is relatively low,and the interpretability and reliability of intelligent prediction of structural responses are poor.In response to these issues,targeted prospects and recommendations for future research directions are proposed.These include enhancing the generalization of damage detection algorithms,improving the overall accuracy of intelligent detection of damage in building structures through super-resolution techniques,and integrating spatial and physical information into intelligent prediction methods for structural responses.
周颖;孟诗乔;孔庆钊;翁渝峰
同济大学土木防灾减灾全国重点实验室,上海 200092
土木建筑
智能防灾研究综述人工智能机器学习损伤检测响应预测
intelligent disaster preventionresearch reviewartificial intelligencemachine learningdamage detectionresponse prediction
《建筑结构学报》 2024 (006)
107-132 / 26
国家自然科学基金杰出青年科学家基金项目(52025083),上海市"科技创新行动计划"社会发展科技攻关项目(22dz1201400),国家自然科学基金项目(U2139209).
评论