面向城轨列车走行安全的轴承在途故障诊断研究

发布时间：2018-05-04 08:10

本文选题：面向 + 城轨　；参考：《北京交通大学》2015年博士论文

【摘要】：城市轨道交通是我国城镇化和城市现代化的全局性和支撑性的基础设施,是城市综合交通的骨干交通方式。截止2012年底,全国城市轨道交通规划总里程超过14000公里,覆盖53个大中城市；截止2013年底,全国累计批复36个城市的轨道交通建设总里程约6000公里；累计建成开通运营总里程已达2266公里。如何保障城市轨道交通系统的运营安全,提升运营维护水平,降低全生命周期运营成本已成为我国城市轨道交通可持续健康发展的瓶颈问题,迫切需要研发适应我国国情和运营管理机制的包括城轨列车走行部轴承运行状态在途检测、故障诊断和预警技术在内的轨道交通安全保障技术与装备体系。本文以形成符合国情和自主知识产权的城轨列车走行部轴承运营状态在途监测及预警关键理论技术与相关系统为目标,形成了具有普适意义的如下理论方法和关键技术及装备：本文对城轨列车走行部轴承在途故障诊断展开以下研究： 1.深入研究了城轨列车走行部轴承结构、振动机理和故障形式及原因,提出了多因素(径向游隙、转速、载荷、波纹度等)综合作用下的城轨列车走行部轴承静、动力学模型。分析了不同因素对系统的影响,得出轴承内在结构以及外部原因与征兆表现之间的内在联系和映射关系,再结合城轨列车特定运营环境,确定走行部轴承的监测参数与监测部位,为后续城轨列车轴承疲劳寿命评估和在途故障辨识提供理论和技术支撑。 2.基于获取的实时动载荷数据,并在轴承疲劳寿命分析理论的基础上,构建了时变工况下城轨列车走行部轴承的疲劳寿命评估模型。首先系统分析了不同参数(转速、载荷、节径、滚动体数目)对轴承疲劳寿命的影响,在此基础上,结合轨道交通列车时变运营工况,建立了变工况下走行部轴承疲劳寿命模型,并利用广州地铁时变工况环境下的数据对模型进行了测试,验证了模型的合理性和有效性。 3.从基于实时数据特征提取方面考虑,提出了面向城轨列车走行部轴承多智能算法融合的在途故障辨识方法。在研究小波分析、包络分析、经验模态分解、神经网络、遗传算法等信号处理方法基础上,融合谐波小波良好的时频局部化特性和包络解调的优点,设计了基于谐波小波包络分析的城轨列车轴承故障辨识方法；基于小波包的时频性和神经网络的自学习、自适应性,构建了基于小波包神经网络的城轨列车轴承故障辨识方法：结合经验模态分解方法精细的时频解析度、神经网络的自学习、自适应性和遗传算法的全局搜索能力,建立了基于时频域多维特征参量和遗传神经网络的城轨列车走行部轴承在途故障辨识方法,并利用不同工况下的故障数据对算法辨识精度和实时性进行测试,诊断结果表明面向城轨列车走行部轴承多智能算法融合的在途故障辨识方法具有较高的辨识精度和较快的诊断效率,从而为在途故障诊断系统的研发奠定基础。 4.基于城轨列车走行部轴承多智能算法融合故障辨识方法的研究成果,并结合广州地铁现有安全监测装备,设计了城轨列车走行部轴承在途故障诊断系统,并通过试验台数据验证了该系统故障辨识的准确性和实时性。
[Abstract]:Urban rail transit is a global and supporting infrastructure for urbanization and urban modernization in China. It is the backbone of urban integrated transportation. By the end of 2012, the total mileage of urban rail transit planning in China was more than 14000 kilometers, covering 53 large and medium-sized cities. By the end of 2013, the whole country has approved the rail transit of 36 cities in China. The total mileage of construction is about 6000 km, and the total mileage of the total operation has reached 2266 km.
How to ensure the operation safety of urban rail transit system, improve the level of operation and maintenance and reduce the cost of life cycle has become the bottleneck of the sustainable and healthy development of urban rail transit in our country. It is urgent to develop the running state of the bearing of urban rail train, which is adapted to the national conditions and operation management mechanism of our country. The technology and equipment system of rail traffic safety, including fault diagnosis and early warning technology.
The aim of this paper is to form the key theory and technology and related system of the bearing operation of urban rail train, which is in line with the national conditions and independent intellectual property rights, and forms the following theoretical methods and key technology and equipment.
In this paper, the following research is carried out on the fault diagnosis of bearing on the way of urban rail train.
1. the bearing structure, the vibration mechanism and the fault form and the cause of the bearing of the rail train are studied in depth. The bearing static and dynamic models of the track train under the multiple factors (radial clearance, rotational speed, load and waviness) are put forward. The influence of different factors on the system is analyzed, and the internal structure and external reasons of the bearing are analyzed. The internal relation and mapping relation between the sign performance and the specific operating environment of the rail train will be combined to determine the monitoring parameters and monitoring parts of the bearing of the walking train, and provide the theoretical and technical support for the bearing fatigue life assessment and the fault identification of the following urban rail train bearings.
2. based on the real-time dynamic load data obtained, and on the basis of the theory of bearing fatigue life analysis, a fatigue life assessment model for the bearing of the rail train in time varying condition is constructed. First, the effects of different parameters (rotational speed, load, diameter and number of rolling body) on the fatigue life of the bearing are analyzed systematically, and on this basis, the track is combined with the track. The fatigue life model of the bearing is established under the variable operating conditions. The model is tested by the data of the time-varying working conditions of Guangzhou metro, which verifies the rationality and effectiveness of the model.
3. based on the feature extraction of real time data, a method of fault identification is proposed, which is based on the wavelet analysis, envelope analysis, empirical mode decomposition, neural network, genetic algorithm and other signal processing methods, which combines the good time frequency localization of the harmonic wavelets. As well as the advantages of envelope demodulation, a fault identification method for urban rail bearing based on the harmonic wavelet envelopment analysis is designed. Based on the time frequency of the wavelet packet and self-learning and self-adaptive of the neural network, a fault identification method for urban rail bearing based on the wavelet packet neural network is constructed, and the precise time frequency of the method is combined with the empirical mode decomposition method. Resolution, self-learning of neural network, self-adaptive and global searching ability of genetic algorithm, a fault identification method for bearing in the walk part of urban rail train based on multi-dimensional characteristic parameters of time frequency domain and genetic neural network is established. The identification accuracy and real-time performance of the algorithm are tested by using the fault data under different working conditions, and the diagnosis results are obtained. The results show that the fault identification method for the multi intelligent algorithm fusion for the bearing of urban rail trains has higher identification precision and faster diagnosis efficiency, thus laying the foundation for the research and development of the fault diagnosis system in the road.
4. based on the research results of the fault identification method of multi intelligent algorithm for bearing of urban rail train, combined with the existing safety monitoring equipment in Guangzhou metro, the fault diagnosis system of the bearing in the walk part of the rail train is designed, and the accuracy and real time of the fault identification of the system is verified by the data of the test bench.

【学位授予单位】：北京交通大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：U279

【参考文献】