汽车起重机关键结构件损伤的声发射特性与源识别方法研究
发布时间：2018-08-08 17:25
【摘要】：汽车起重机作为一种被广泛应用的特种设备，在中国数量多、增速快。且由于其关键结构件长期受交变载荷的作用，所处工作环境恶劣，因此易产生裂纹、锈蚀等损伤缺陷，从而造成整机的灾难性事故，因此对其进行损伤检测与分析研究十分必要。由于声发射技术能够检测金属材料早期裂缝产生和扩展过程，使其非常适合作为结构件局部损伤的监测方法。现阶段对汽车起重机的关键结构件进行局部损伤监测还处于初级阶段，对其声发射源特性缺乏清晰的认识，未形成行之有效的方法与策略对关键结构件局部损伤及危险状态进行预警，而且对工作环境下汽车起重机局部损伤声发射源的识别有待于进一步的研究。因此结合国家重点高技术研究发展计划开展了相关课题的研究工作，具体研究内容包括： (1)通过汽车起重机常用Weldox960钢和HG70钢母材及焊接试件拉伸破坏的声发射监测实验，对材料损伤过程中声发射信号特征参数变化规律及其对损伤行为的敏感性进行了研究，并采用时差滤波方法实现了裂纹损伤声发射信号的有效分离。研究结果表明：有效值电压（RMS电压）对材料屈服变形损伤较为敏感，据此提出了一种基于RMS电压局部极小值的材料屈服变形损伤预警方法，其对HG70钢母材预警应力比为75%、对Weldox960钢母材为90%左右；而质心频率、峰值频率及幅值的变化可有效监测裂纹萌生及扩展过程。 (2)汽车起重机服役过程中其多个部件受弯曲载荷作用，，因此在实验室条件下，采用声发射技术监测了其常用钢母材及焊接缺陷试件的三点弯曲实验过程,研究了不同损伤阶段的声发射特征，并提出了一种基于最小改进b值（最小Ib值）的裂纹损伤预警方法。实验结果表明：最小Ib值可以实现对焊接裂纹扩展损伤的预警；幅值、振铃计数等多个特征参数值对母材损伤变形过程较为敏感；而信号能量计数值、质心频率等多个特征参数可作为焊接试件裂纹扩展的敏感特征指标。 (3)提出了一种声发射检测点优化设计及局部损伤定位方法，该方法首先进行汽车起重机结构件的建模和有限元分析，然后结合分析结果与加载应力实验确定应力集中及易损伤区，以此为依据确定声发射监测的重点部位，最后进行声发射检测实验方案及四探头阵列的声发射定位方法设计。实验结果表明：该方法准确确定了声发射监测的两个重点区域，断铅模拟实验及起重机加载实验均验证了该方法能够对吊臂等关键结构件上的损伤源实现准确定位。 (4)提出了一种基于混合特征向量与最小二乘支持向量机的汽车起重机典型声发射源识别方法。首先结合小波包分析，信号时域及频域特征分析生成声发射混合特征向量，然后结合类间距离评估因子方法对生成的特征向量进行降维处理，再将其作为输入值带入构造的最小二乘支持向量机识别模型完成识别分析。通过对工业现场常见6种典型声发射源数据识别结果表明：该模型对不同声发射源的识别率均超过了93%，而且相较于神经网络方法有更高的识别精度。 通过论文的研究，有助于加深对汽车起重机关键结构件损伤声发射源的全面了解，为相关声发射检测标准的制定提供理论及实验基础，对加速声发射技术在相关领域的推广应用具有重大的现实意义。
[Abstract]:As a kind of special equipment widely used, automobile crane is widely used in China, and its speed is fast. And because its key structural parts are subjected to alternating load for a long time, the working environment is bad, so it is easy to produce crack, corrosion and other damage defects, which cause catastrophic accidents of the whole machine, so the damage detection and analysis of it is ten. Because acoustic emission technology can detect the early crack generation and expansion process of metal materials, it is very suitable to be used as a monitoring method for local damage of structural parts. At the present stage, the local damage monitoring of the key structural parts of the truck crane is still in the primary stage, and the characteristics of the acoustic emission source are lack of clear understanding, and no line is formed. The effective methods and strategies are used to predict the local damage and dangerous state of the key structure, and the identification of the acoustic emission source of the local damage of the truck crane in the working environment needs further study.
(1) through the acoustic emission monitoring experiment on the tensile failure of Weldox960 steel and HG70 steel and welding specimen, the characteristics of the characteristic parameters of acoustic emission signals and the sensitivity of the damage behavior in the process of material damage are studied, and the effective separation of the acoustic emission signals from the crack damage is realized by the time difference filter method. The results show that the effective value voltage (RMS voltage) is more sensitive to the damage of material yield deformation. Based on this, an early warning method for yield deformation damage of materials based on the local minimum of RMS voltage is proposed. The early warning stress ratio of HG70 steel is 75%, and the base material of Weldox960 steel is about 90%; and center frequency, peak frequency and amplitude The process of crack initiation and propagation can be monitored effectively.
(2) many parts of the truck crane are subjected to bending load during the service of the truck crane. So under the laboratory conditions, the acoustic emission technique is used to monitor the three point bending test process of its common steel material and welding defects, and the acoustic emission characteristics of different damage stages are studied, and a minimum improved b value (minimum Ib value) is proposed. The experimental results show that the minimum Ib value can realize the early warning of the weld crack propagation damage; the amplitude, the ringing count and other characteristic parameters are more sensitive to the damage and deformation process of the material, and the number of characteristic parameters such as the signal energy meter and the centroid frequency can be used as the sensitive characteristic of the crack propagation in the welding specimen. Indicators.
(3) an optimization design of acoustic emission detection point and a local damage location method are proposed. First, the modeling and finite element analysis of the structural parts of the truck crane are first carried out. Then the stress concentration and the vulnerable zone are determined by the analysis results and the loading stress experiments. In this way, the key parts of the acoustic emission monitoring are determined and the sound is sent to the sound. The experiment scheme and the acoustic emission positioning method of the four probe array are designed. The experimental results show that the method accurately determines the two key areas of acoustic emission monitoring. The simulation experiment of lead breaking and the loading experiment of the crane verify that the method can accurately locate the damage source on the key structural parts such as the crane arm.
(4) a typical acoustic emission source identification method based on hybrid eigenvector and least squares support vector machine is proposed. First, combined with wavelet packet analysis, the acoustic emission mixed eigenvectors are generated in time domain and frequency domain characteristic analysis, and then the generated eigenvector is reduced with the method of inter class distance assessment factor. The identification results of 6 typical acoustic emission sources in industrial field show that the recognition rate of the model is more than 93% for the different acoustic emission sources and has a higher recognition accuracy compared with the neural network method.
Through the research of this paper, it is helpful to deepen the comprehensive understanding of the damage acoustic emission source of the key structural parts of the automobile crane, and provide the theoretical and experimental basis for the formulation of the related acoustic emission testing standards. It has great practical meaning to the promotion and application of the accelerated acoustic emission technology in the related fields.
【学位授予单位】：北京理工大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：TH213.6
本文编号：2172491
[Abstract]:As a kind of special equipment widely used, automobile crane is widely used in China, and its speed is fast. And because its key structural parts are subjected to alternating load for a long time, the working environment is bad, so it is easy to produce crack, corrosion and other damage defects, which cause catastrophic accidents of the whole machine, so the damage detection and analysis of it is ten. Because acoustic emission technology can detect the early crack generation and expansion process of metal materials, it is very suitable to be used as a monitoring method for local damage of structural parts. At the present stage, the local damage monitoring of the key structural parts of the truck crane is still in the primary stage, and the characteristics of the acoustic emission source are lack of clear understanding, and no line is formed. The effective methods and strategies are used to predict the local damage and dangerous state of the key structure, and the identification of the acoustic emission source of the local damage of the truck crane in the working environment needs further study.
(1) through the acoustic emission monitoring experiment on the tensile failure of Weldox960 steel and HG70 steel and welding specimen, the characteristics of the characteristic parameters of acoustic emission signals and the sensitivity of the damage behavior in the process of material damage are studied, and the effective separation of the acoustic emission signals from the crack damage is realized by the time difference filter method. The results show that the effective value voltage (RMS voltage) is more sensitive to the damage of material yield deformation. Based on this, an early warning method for yield deformation damage of materials based on the local minimum of RMS voltage is proposed. The early warning stress ratio of HG70 steel is 75%, and the base material of Weldox960 steel is about 90%; and center frequency, peak frequency and amplitude The process of crack initiation and propagation can be monitored effectively.
(2) many parts of the truck crane are subjected to bending load during the service of the truck crane. So under the laboratory conditions, the acoustic emission technique is used to monitor the three point bending test process of its common steel material and welding defects, and the acoustic emission characteristics of different damage stages are studied, and a minimum improved b value (minimum Ib value) is proposed. The experimental results show that the minimum Ib value can realize the early warning of the weld crack propagation damage; the amplitude, the ringing count and other characteristic parameters are more sensitive to the damage and deformation process of the material, and the number of characteristic parameters such as the signal energy meter and the centroid frequency can be used as the sensitive characteristic of the crack propagation in the welding specimen. Indicators.
(3) an optimization design of acoustic emission detection point and a local damage location method are proposed. First, the modeling and finite element analysis of the structural parts of the truck crane are first carried out. Then the stress concentration and the vulnerable zone are determined by the analysis results and the loading stress experiments. In this way, the key parts of the acoustic emission monitoring are determined and the sound is sent to the sound. The experiment scheme and the acoustic emission positioning method of the four probe array are designed. The experimental results show that the method accurately determines the two key areas of acoustic emission monitoring. The simulation experiment of lead breaking and the loading experiment of the crane verify that the method can accurately locate the damage source on the key structural parts such as the crane arm.
(4) a typical acoustic emission source identification method based on hybrid eigenvector and least squares support vector machine is proposed. First, combined with wavelet packet analysis, the acoustic emission mixed eigenvectors are generated in time domain and frequency domain characteristic analysis, and then the generated eigenvector is reduced with the method of inter class distance assessment factor. The identification results of 6 typical acoustic emission sources in industrial field show that the recognition rate of the model is more than 93% for the different acoustic emission sources and has a higher recognition accuracy compared with the neural network method.
Through the research of this paper, it is helpful to deepen the comprehensive understanding of the damage acoustic emission source of the key structural parts of the automobile crane, and provide the theoretical and experimental basis for the formulation of the related acoustic emission testing standards. It has great practical meaning to the promotion and application of the accelerated acoustic emission technology in the related fields.
【学位授予单位】：北京理工大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：TH213.6
【参考文献】
相关期刊论文 前10条
1 阳建红,赵光辉,余志刚,刘兵吉;HTPB复合固体推进剂材料声发射的特性[J];火炸药学报;2000年01期
2 王宏伟;骆红云;韩志远;钟群鹏;;基于声发射非线性模型的损伤状态转变的评估[J];北京航空航天大学学报;2010年10期
3 丁世飞;齐丙娟;谭红艳;;支持向量机理论与算法研究综述[J];电子科技大学学报;2011年01期
4 黄新民;彭跃社;;混凝土损伤的声发射特征分形分析[J];动力学与控制学报;2008年03期
5 常岩军;矫桂琼;张克实;王波;;3D C/SiC复合材料拉伸性能的声发射研究[J];复合材料学报;2010年06期
6 朱祖铭,尹万全;基于声发射测试的钢丝疲劳损伤过程[J];钢铁;2000年03期
7 王美成;100吨汽车起重机转台的有限元计算与分析[J];工程机械;2005年08期
8 阳建红,王芳文,覃世勇;HTPB复合固体推进剂的声发射特性及损伤模型的试验和理论研究[J];固体火箭技术;2000年03期
9 杜伟坊，杜海清，李玉书;陶瓷材料断裂源的声发射表征[J];硅酸盐学报;1994年05期
10 聂鹏;谌鑫;;基于最小二乘支持向量机对刀具切削状态的识别[J];北京工业大学学报;2012年08期
相关博士学位论文 前3条
1 阎玉芹;塔式起重机钢结构健康监测技术与实验研究[D];山东大学;2011年
2 赵静荣;声发射信号处理系统与源识别方法的研究[D];吉林大学;2010年
3 刘中星;轻量化汽车起重机安全性检测与评价技术研究[D];机械科学研究总院;2013年
本文编号：2172491
本文链接：https://www.wllwen.com/jixiegongchenglunwen/2172491.html
