| URT (Urban Rail Transit) is the global and fundamental infrastructure of China’s urbanization and urban modernization; it also plays a crucial role in city comprehensive transportation. By the end of2012, China’s planning mileage of URT has been over14,000km, covering53cities; By the end of2013, a total mileage of6000km of URT construction in36cities have been approved, and the total mileage available in service has achieved2266km.How to ensure the operation security of URT system, improve operation and maintenance levels, and reduce lifecycle operating costs has become a bottleneck of the sustainable and development of China’s URT. Therefore, it’s an urgent need to develop a rail transportation security technology and equipment system, which is adapted to China’s national conditions and the development of operational management mechanisms, including the acting bearing detecting of running gear, fault diagnosis and early-warning technology.This thesis’s objective is to formulate a system of acting bearing detecting of the running gear combined with early-warning technologies, which is in line with independent intellectual property rights and China’s national conditions:1. The deep study of the structure of running gear’s bearing, vibration mechanism, and bearing failure’s modes and causes. A bearing static and dynamic model of running gear under the combined effect caused by several factors (radial clearance, speed, load, waviness, etc.) is put forward. The analysis of influence of different factors on the system, lead to the internal relation and mappings between bearing internal structure, external causes and performance, combines with URT specific operating environment, determining the monitoring parameters and monitoring site of running gear bearing, which provide theoretical and technical support for the following rail train running gear bearing fatigue life assessment and identify in-transit of failure.2. Based on the data acquired in real-time dynamic load data and bearing fatigue life analysis theory, an assessment model of URT’s running gear bearing fatigue life is constructed. First, the thesis systematically analyzes the influence of different parameters (speed, load, pitch diameter, number of rolling elements) on bearing fatigue life. On this basis, combines with rail train changing operational conditions, the thesis establishes a running gear bearing life model under the changing conditions, and then using the data under changing operational conditions in Guangzhou Metro to test the model, verify its rationality and effectiveness.3. From the viewpoint of characteristics extraction of real-time data, a multi-intelligence fusion algorithm with real time state feature is presented. Based on the study of signal processing methods including wavelet analysis, envelope analysis, empirical mode decomposition, neural networks and genetic algorithms, also fuses with the advantage of frequency localization and envelope demodulation of harmonic wavelet, a URT’s running gear bearing fault identification method has been designed based on the envelope analysis of harmonic wavelet; based on the frequency and neural network self-learning, self-adaptive of wavelet packet, a URT’s running gear bearing fault identification method is constructed; combines with empirical mode decomposition method for fine time-frequency resolution, self-learning, self-adaptive global search ability and genetic algorithms, an in-transit running gear bearing fault identification methods based on multi-dimensional characteristic parameters in time-frequency domain and genetic neural network is established. Using data of fault under different conditions to test the identification accuracy and timeliness of the algorithm, the diagnostic results indicate that the in-transit fault identification with multi-intelligence algorithm fusion for URT’s running gear bearing has high recognition accuracy and rapid diagnostic efficiency, thereby it lays the foundation for the development of in-transit fault diagnosis system.4. Based on the research of fault identification with multi-intelligence algorithm fusion of URT’s running gear bearing, combines with Guangzhou Metro existing security monitoring equipment, an in-transit fault diagnosis system for URT’s running gear bearing has been designed, the accuracy and timeliness of the fault identification system also has been verified through test-bed data. |
PDF Full Text Request