摘要
裂纹故障是影响生产生活和设备安全运行的一大因素,裂纹检测实验对于检测技术应用和裂纹故障信息获取具有重要意义,但领域内缺乏对检测实验的系统性归纳与剖析.该文对当前代表性的裂纹检测实验方法进行了分类比较,针对裂纹信息介绍了定位、形貌和深度的检测实验,从先进传感器、数据处理与机器学习应用等3个方面探究了裂纹检测实验方法进展.最后根据国内外裂纹检测实验的研究现状,提出了在裂纹扩展方向检测、误报漏报及实验与实际工况贴近程度方面所面临的挑战,并从实时动态、远程无线和多故障耦合检测等方面进行了展望,为今后裂纹检测实验方法的研究与发展提供了参考.
Abstract
[Significance]Crack failure considerably affects production stability and threatens life and property safety.Crack detection is important in maintaining safe and healthy structural operation and has become indispensable in structural health monitoring.Crack detection experiments are necessary links and important references for analyzing the effectiveness and practicality of detection methods.With advancements in detection technology and changes in crack detection conditions,experimental methods for crack detection have made tremendous progress.However,a systematic review of crack detection experiments is lacking.[Progress]Therefore,this article systematically analyzes experimental methods for crack detection from three aspects:crack detection technology,crack information detection,and crack detection experimental improvement methods.First,we compare the traditional and emerging classification techniques of the crack detection experimental methods and summarize them in terms of detection goals,application scenarios,and precision.Further,we provide insight into the localization,morphology,and depth detection experiments for crack information.Crack localization detection experiments include characteristic systems and data processing localization methods.Crack morphology detection experiments include length,morphology detection,and contour inversion experimental methods.The development of these information detection methods can determine the reliability of the health state of the structure,optimize the maintenance program,and provide an effective reference basis,thereby ensuring the safe operation of the structure and prolonging the service life.Finally,we analyze the progress in improving experimental methods,particularly the application of advanced sensors,data processing,and machine learning in crack detection.This section reveals that crack detection is moving toward higher accuracy and convenience,particularly in areas such as positioning,morphology,and depth detection.Researchers mainly focus on developing and applying advanced sensor technologies as well as improving data processing and analysis methods.Furthermore,the application of machine learning in crack detection,including automated detection systems and machine learning algorithms,has received considerable attention.[Challenge and Prospects]However,crack detection experimental methods still face some challenges,which affect their future development.Crack propagation direction detection,false and missed alarms,and the degree of closeness between experiments and actual working conditions are the current challenges that must be addressed.Hence,based on current international and domestic research status,three research prospects for crack detection experiments are proposed:real-time dynamic monitoring,remote wireless detection,and multifault coupling detection.Real-time dynamic crack monitoring can detect cracks early and even achieve a more meaningful early warning of crack initiation.Remote wireless detection can upload data,and cloud storage or manual monitoring platforms can achieve remote security monitoring.Multifault coupling detection can effectively solve the problem of various types and simultaneous occurrence of equipment faults.[Conclusions]In summary,this article presents a review of the research progress of crack detection experimental methods and prospects for future development directions.A complete experimental technology guidance system for crack detection is developed,which aids in controlling the law governing crack emergence and expansion,guiding accurate and rapid on-site detection of cracks,and providing a reference for the future research and development of crack detection experimental methods.
维普
万方数据