基于SOM的滚动轴承故障状态识别方法的研究

发布时间：2018-06-15 22:31

本文选题：滚动轴承 + 故障状态识别　；参考：《沈阳理工大学》2017年硕士论文

【摘要】：作为旋转机械中重要零件的滚动轴承,在现今的各个行业有着非常广泛的应用,但由于经常在较为复杂的环境下工作,滚动轴承也是各类机械中最易损坏的零件之一,滚动轴承的损坏往往会导致整个机械设备的故障,引发更为严重的后果,而滚动轴承的寿命通常又具有离散性。因此,对滚动轴承的状态以及故障类型进行准确的识别就显得尤为重要了。本论文采用一种改进的集合经验模态分解(Ensemble Empirical Mode Decomposition,EEMD)对滚动轴承的振动信号进行降噪和分解,得到若干个固有模态函数(Intrinsic mode function,IMF),再计算包含主要故障信息的IMF分量的能量分布作为滚动轴承的故障特征。该方法根据在分解过程中的信号和所加入的白噪声的特点来选择各项参数,这样就避免了传统EEMD方法根据经验人为主观的设置参数导致分解结果不够准确的问题。由于传统的自组织映射神经网络(Self-organizing Feature Map,SOM)网络在滚动轴承的故障状态识别的过程中存在训练时间较长,识别准确率有待提高等不足。因此,本论文采用免疫遗传算法(Immune Genetic Algorithm , IGA)对SOM的权值更新过程进行优化。IGA通过模仿生物的免疫机制以及遗传的规则来优化SOM的权值更新过程,使权值的更新更为快速且全面,这样也就能够进一步提高SOM网络对滚动轴故障类型识别的速度以及准确性。最后将提取到的滚动轴承的故障特征输入到经过免疫遗传算法优化后的自组织映射(IGA-SOM)网络当中,通过IGA-SOM对滚动轴承的故障类型进行识别。通过实际的滚动轴承的故障数据验证表明,本论文的方法可以实现对于滚动轴承故障类型的高效且准确地识别。
[Abstract]:As an important part of rotating machinery, rolling bearings are widely used in various industries nowadays. However, because they often work in complex environments, rolling bearings are also one of the most easily damaged parts in all kinds of machinery. The failure of rolling bearings often leads to the failure of the whole mechanical equipment, which leads to more serious consequences, and the life of rolling bearings is usually discrete. Therefore, it is very important to identify the status and fault types of rolling bearings. In this paper, an improved set empirical mode decomposition (EMD) is used to reduce the noise and decompose the vibration signals of rolling bearings. Some intrinsic mode functions are obtained, and the energy distribution of the components containing the main fault information is calculated as the fault characteristics of the rolling bearing. This method selects the parameters according to the characteristics of the signal and the white noise added in the decomposition process, thus avoiding the problem that the traditional EEMD method sets the parameters artificially and empirically, which results in inaccurate decomposition results. Because the traditional self-organizing mapping neural network self-organizing feature Map-SOM network has a long training time in the process of rolling bearing fault state recognition, the recognition accuracy needs to be improved. Therefore, in this paper, the immune genetic algorithm (IGA) is used to optimize the weight updating process of SOM. IGA optimizes the weight updating process by imitating the immune mechanism of organisms and genetic rules, so that the weight updating process is faster and more comprehensive. This will further improve the speed and accuracy of SOM network in identifying rolling shaft fault types. Finally, the extracted fault features of rolling bearings are input into the self-organizing mapping IGA-SOM network optimized by immune genetic algorithm, and the fault types of rolling bearings are identified by IGA-SOM. The actual fault data verification of rolling bearing shows that the method of this paper can be used to identify the fault type of rolling bearing efficiently and accurately.
【学位授予单位】：沈阳理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH133.33;TP18

【参考文献】