| 45 | 0 | 4 |
| 下载次数 | 被引频次 | 阅读次数 |
[目的]为解决当前航空发动机热障涂层孔探图像识别工作以人工目视为主、效率低且存在人为因素的问题，设计了一套基于图像识别与深度学习的发动机孔探图片识别系统。[方法]首先收集并整理了某航空公司一线飞机维修员拍摄的CFM56发动机涡轮叶片及燃烧室孔探图像，构建了发动机热端部件孔探图像数据集，接着对孔探图像进行预处理，最后用Blob分析对预处理后的孔探图像数据集进行特征提取与系统分析。[结果]该系统可有效识别发动机热障涂层的故障，运行快、准确率高且可对图像进行连续自动识别。[结论]利用图像识别技术对航空发动机热障涂层的孔探图像进行识别不仅可提高工作效率，而且可避免人为因素对航空安全的影响。
Abstract：[Objective] To address the inefficiency and human subjectivity in current visual inspection of borescope images of thermal barrier coatings on aero-engines, a borescope image recognition system was designed based on image processing and deep learning. [Method] The borescope images of CFM56 aero-engine turbine blades and combustion chamber were captured by aircraft maintenance technicians of an airline company were collected and sorted out to establish a borescope image data set of aero-engine hot-section components. After preprocessing, blob analysis was applied for feature extraction and systematic evaluation. [Result] The system achieved rapid and high-accuracy identification of thermal barrier coating defects and enabled continuous automated image analysis. [Conclusion]Automated borescope image recognition technology enhances inspection efficiency while eliminating human-error risks,significantly improving aviation safety.
[1]International Air Transport Association. IATA annual safety report—2022:executive summary and safety overview edition 59[R/OL].(2023–07–24)[2024–09–27]. https://www.iata.org/contentassets/a8e49941e8824a058fee3f5ae0c005d9/safety-report-executive-andsafety-overview.pdf.
[2]Airbus S.A.S. A statistical analysis of commercial aviation accidents1958–2022[R/OL].(2023–03–03)[2024–09–27]. https://www.airbus.com/sites/g/files/jlcbta136/files/2023-03/A-Statistical-Analysis-of-CommercialAviation-Accidents-2023.pdf.
[3]HARRIS D, LI W C. Using neural networks to predict HFACS unsafe acts from the pre-conditions of unsafe acts[J]. Ergonomics, 2019,62(2):181-191.
[4]贺佳伟.基于图像识别的飞机蒙皮损伤检测与评估[D].重庆:重庆交通大学, 2024.HE J W. Aircraft skin damage detection and assessment based on image recognition[D]. Chongqing:Chongqing Jiaotong University, 2024.
[5]樊成栋.基于机器视觉的地铁轨道障碍物检测系统设计[D].南京:南京理工大学, 2022.FAN C D. Design of subway track obstacle detection system based on machine vision[D]. Nanjing:Nanjing University of Science and Technology, 2022.
[6]孙毅.图像识别技术在供电所运行管理中的应用前景分析[J].产业创新研究, 2024(8):90-92.SUN Y. Application prospect analysis of image recognition technology in power supply office operation management[J].Industrial Innovation, 2024(8):90-92.
[7]邓丽,潘永杰,王琦,等.基于卷积LSTM的混凝土裂缝图像识别技术[J].铁道建筑, 2024, 64(4):11-17.DENG L, PAN Y J, WANG Q, et al. Concrete crack image recognition technology based on convolutional LSTM[J]. Railway Engineering, 2024, 64(4):11-17.
[8]袁社锋.计算机图像智能识别下的割草机器人设计研究[J].农机化研究, 2024, 46(11):136-139.YUAN S F. Research on design of mowing robot based on computer image intelligent recognition[J]. Journal of Agricultural Mechanization Research, 2024, 46(11):136-139.
[9]李响,陆锋,陈钰婷,等.图像识别技术及其在架空线检查机器人中的运用[J].黑龙江科学, 2024, 15(4):144-147, 151.LI X, LU F, CHEN Y T, et al. Image recognition technology and its application of in overhead line inspection robot[J]. Heilongjiang Science, 2024, 15(4):144-147, 151.
[10]PADIYAR M J, ZANOTTI FRAGONARA L, PETRUNIN I, et al.Fast, accurate, and reliable detection of damage in aircraft composites by advanced synergistic infrared thermography and phased array techniques[J]. Applied Sciences, 2021, 11(6):2778.
[11]HUANG Z H, TANG J Y, SHAN G X, et al. An efficient passengerhunting recommendation framework with multitask deep learning[J].IEEE Internet of Things Journal, 2019, 6(5):7713-7721.
[12]LI H S. Research on fault detection algorithm of pantograph based on edge computing image processing[J]. IEEE Access, 2020, 8:84652-84659.
[13]袁忠大,王大伟.热障涂层在航空发动机涡轮叶片上的应用、失效与维护[J].上海涂料, 2024, 62(2):49-53.YUAN Z D, WANG D W. Application, failure and maintenance of thermal barrier coating on aero-engine turbine blades[J]. Shanghai Coatings, 2024, 62(2):49-53.
[14]宋静波,刘熊,田巍.波音737NG飞机动力装置(CFM56-7B&APU)[M].西安:西北工业大学出版社, 2018:78-79.SONG J B, LIU X, TIAN W. Boeing 737NG Aircraft Power Plant(CFM56-7B&APU)[M]. Xi’an:Northwestern Polytechnical University Press, 2018:78-79.
[15]袁忠大,刘傲天,张小锋,等.基于机器视觉的飞机电镀部件曲表面无损检测[J].电镀与涂饰, 2024, 43(7):43-50.YUAN Z D, LIU A T, ZHANG X F, et al. Nondestructive testing of curved surfaces of electroplated aircraft parts based on machine vision[J]. Electroplating&Finishing, 2024, 43(7):43-50.
[16]袁忠大,程秀全,王大伟.基于机器视觉的飞机故障检查系统[J].机床与液压, 2024, 52(6):196-200.YUAN Z D, CHENG X Q, WANG D W. Aircraft fault inspection system based on machine vision[J]. Machine Tool&Hydraulics,2024, 52(6):196-200.
[17]WANG H F, ZHAI L, HUANG H, et al. Measurement for cracks at the bottom of bridges based on tethered creeping unmanned aerial vehicle[J]. Automation in Construction, 2020, 119:103330.
[18]张言.基于代数信号处理的二维离散傅里叶变换算法研究[D].哈尔滨:哈尔滨工程大学, 2022.ZHANG Y. Research on two-dimensional Discrete Fourier Transform Algorithm based on algebraic signal processing[D]. Harbin:Harbin Engineering University, 2022.
[19]YU H M, LI Q Y, TAN Y Q, et al. A coarse-to-fine model for rail surface defect detection[J]. IEEE Transactions on Instrumentation and Measurement, 2019, 68(3):656-666.
[20]MIAO J W, YUAN H Y, LI L Z, et al. Line structured light vision online inspection of automotive shaft parts[J]. Advances in Intelligent Systems and Computing, 2019, 929:585-595.
[21]WEI D H, WEI X K, LIU Y X, et al. The identification and assessment of rail corrugation based on computer vision[J]. Applied Sciences, 2019, 9(18):3913.
基本信息:
DOI：10.19289/j.1004-227x.2025.06.020
中图分类号:TP391.41;V263
引用信息:
[1]袁忠大,胡能叶,龚晓峰等.基于机器视觉的航空发动机热障涂层损伤自动识别[J].电镀与涂饰,2025,44(06):145-152.DOI:10.19289/j.1004-227x.2025.06.020.
基金信息:
广东省高职院校产教融合创新平台（2024CJPT012）; 广东省普通高校特色创新项目（2023KTSCX238）; 广东省教育科学规划课题（2023GXJK696）