【作者】 李大光；

【导师】 阙大顺；

【作者基本信息】 武汉理工大学 ， 信号与信息处理， 2005， 硕士

【摘要】 铁谱技术是以磨损磨粒分析为基础的故障诊断方法。铁谱磨粒分析因其高效、经济而在机械设备的监测、故障诊断和预防性维修等方面得到了广泛应用。磨粒识别是铁谱磨粒分析的核心。由于磨粒的多样性和复杂性,磨粒识别尚无成熟的理论方法来指导,目前的磨粒识别还主要由专家人工完成。计算机图像处理和人工智能的发展为磨粒分析的智能化提供了有力的条件。本文以铁谱故障诊断的基本理论为基础,结合图像处理的原理和方法,研究了基于图像处理的铁谱覆盖面积计算,通过对一组油样铁谱覆盖面积的回归分析验证了该方法在铁谱故障诊断和预测中的作用;结合模式识别的原理和方法研究了神经网络在磨粒类型识别中的应用,并以此为基础实现了磨粒的实时分析。本文研究对于推动磨粒分析在机械磨损监测中的应用有重要意义。 本文的主要内容有: 1.综合国内外有关文献,对铁谱磨粒分析技术的发展和现状进行综述,结合本课题研究的要求,阐述了本文的主要研究内容。 2.分析总结了磨损的产生机理与分类,与磨损对应的磨粒的分类及特征。阐述了基本磨粒类型、特征、成分和产生机理与设备磨损状态之间的内在联系。 3.基于数字图像处理技术,对图像的平滑、滤波、边缘检测等在磨粒图像预处理中的应用进行了分析、讨论。重点讨论了不同情况下的磨粒图像分割技术。 4.在大量实验的基础上,通过对一组油样的铁谱数据的回归分析研究了基于图像处理的铁谱定量分析的可行性。 5.分析计算了磨粒的形貌,纹理和颜色等几类特征参数,建立了一套较为完备的磨粒特征描述体系。 6.讨论了神经网络在磨粒类别分析中的应用。在此基础上将视频采集、图像处理和模式识别的知识结合起来,在VC++.net平台上实现了磨粒图像的实时获取、分割、特征参数提取和类型识别。更多还原

【Abstract】 The ferrography is a fault diagnosis method based on the analysis of wear debris. Ferrogram wear debris analysis technology has been applied broadly in many aspects, such as inspection of machine operation state, failure diagnose and preventive maintenance, because of its advantage of good effect and economical. The recognition of the wear particle is the core step of the ferrography. Because of the diversity and complicacy of the wear particle, the recognition procedure is carried out without guidance of mature theory. Currently, the recognition of the wear debris is carried out by experts. The development of computer image processing and artificial intelligence technologies would be helpful for the improvement of accuracy and automation of ferrography analysis. With the use of the basic theory and technique of image processing, the calculation of ferrogram cover area was studied. Then, based on the basic theory of ferrography fault diagnosis, regress analysis was applied to a set of oil sample ferrogram cover area. The feasibility of the method was than tested. With the use of the theory and technique of pattern recognition in the analysis of wear debris image, the use of neural network in the recognition of wear debris was studied. With all the studies above, a realization of the real time analysis of wear debris was given. The study will promote the use of wear debris analysis in the use of the machinery wear status monitor.The main contents of the thesis are as follows:1. The development and up-to-date status of wear particle analysis technology at home and abroad are evaluated synthetically. The study plan and main content are presented.2. Wear mechanism and classification including wear particle classification and characters are analyzed and discussed. The inner relations between each basic kind of wear particle and wear type, wear particle characters, wear mechanism, machine running status are analyzed and expatiated.3. According to digital image processing technologies, some methods such as smoothing, filtering are applied to process wear debris images. Wear debris segmentation of different background is also discussed as a main aspect.4. Lots of experiment was carried out and a set of oil samples ferrogram were studied to analyze the feasibility of quantitative analysis based on image processing.5. By analyzing and calculating the shape, color and texture characteristic parameters of wear particles, a complete set of wear particle quantification characterization comes into being.6. The use of neural network in the recognition of wear debris was discussed. With all the studies listed above and the combine of video collection, image processing and pattern recognition, a real time wear debris recognition system is designed and tested using the VC++.net programming platform. The system could give a real time collection, segmentation, characteristic parameters calculation and classification of wear debris image.更多还原