摘要
针对滑油中磨粒形状复杂且尺寸大小不一,传统滑油磨粒检测方法时效性差、检测尺度小、精度低、非铁磁性磨粒不能检测等问题。设计并研究了一种航空发动机滑油磨粒检测系统及方法。分析航空发动机磨损模式与形成机理,将连续流微流控技术与图像处理相结合,搭建磨粒检测系统,设计油液磨粒检测系统整体架构与检测流程。采用最佳算法处理滑油磨粒图像,提取图像尺寸特征和形状特征参数,构建磨粒样本特征数据集。基于极限学习算法根据磨粒特征参数建立极限学习分类器对磨粒识别分类,验证测试结果,识别准确率高达92%。证明航空发动机滑油磨粒检测技术的可行性以及识别算法的有效性。
In view of the complex shapes and different sizes of abrasive particles in lubricating oil,the traditional detection methods of lubricating oil abrasive particles have poor timeliness,small detection scale,low precision,and inability to detect non⁃ferromagnetic abrasive particles.Design and research an aero⁃engine lubricating oil abrasive particle detection system and method.Analyze the wear mode and formation mechanism of aero⁃engine,combine continuous flow microfluidic technology with image processing,build an abrasive particle detection system,and design the overall architecture and detection process of the oil abrasive particle detection system.The optimal algorithm is used to process the lubricating oil abrasive grain image,extract the image size feature and shape feature parameter,and construct the abrasive grain sample feature data set.Based on the extreme learning algorithm,an extreme learning classifier is established to identify and classify the abrasive particles according to the characteristic parameters of the abrasive particles,and the test results are verified,and the recognition accuracy rate is as high as 92%.It proves the feasibility of the aero⁃engine lubricating oil wear particle detection technology and the effectiveness of the identification algorithm.
作者
侯媛媛
李江红
张朝
HOU Yuanyuan;LI Jianghong;ZHANG Chao(School of Computer Science,Xi’an Aeronautical University,Xi’an 710077,China;School of Power and Energy,Northwestern Polytechnical University,Xi’an 710129,China)
出处
《电子设计工程》
2023年第12期180-184,共5页
Electronic Design Engineering
基金
西安航空学院科研立项项目(2020KY0204)。
关键词
航空发动机
油液检测
连续流微流控
磨粒识别
aeroengine
oil detection
continuous flow microfluidics
lubricant oil wear debris
