Failure of gears in locomotive power transmission system causes interruption in service and may lead to derailment or other train related disasters. This creates a drastic problem with the trustiness of the operations. Therefore, for the sake of safe and cost-effective operations early detection of failures including gear faults is of great importance. The purpose of this research is to explor the possibility of early detection of gear pitting in locomotive poower transmission system. For this purpose, the dynamic response of the gear system with pitting defects in different scenarios is scrutenized. The primary excitation to the gear dynamic system comes from the time-variant mesh stiffness. The effect of pitting growth on the potential energy of the system in different scenarios is analytically estimated. Responses in time and frequency domains are examined. Some statistical features are used. Features most sensitive to the failure that enable the detection of gear tooth pitting and its propagation are introduced. The theoretical results are compared and validated with a laboratory test setup. It is also endeavored to uncover the earliest possible level at which the removed surface can be detected. Hence the effect of the pitting growth due to geometric parameters such as the radius of pitting, number of pits, and number of teeth with pits are examined.Failure of gears in locomotive power transmission system causes interruption in service and may lead to derailment or other train related disasters. This creates a drastic problem with the trustiness of the operations. Therefore, for the sake of safe and cost-effective operations early detection of failures including gear faults is of great importance. The purpose of this research is to explor the possibility of early detection of gear pitting in locomotive poower transmission system. For this purpose, the dynamic response of the gear system with pitting defects in different scenarios is scrutenized. The primary excitation to the gear dynamic system comes from the time-variant mesh stiffness. The effect of pitting growth on the potential energy of the system in different scenarios is analytically estimated. Responses in time and frequency domains are examined. Some statistical features are used. Features most sensitive to the failure that enable the detection of gear tooth pitting and its propagation are introduced. The theoretical results are compared and validated with a laboratory test setup. It is also endeavored to uncover the earliest possible level at which the removed surface can be detected. Hence the effect of the pitting growth due to geometric parameters such as the radius of pitting, number of pits, and number of teeth with pits are examined.
 

Keywords: Dynamic modelling, Locomotive gear transmission system, Statistical feature, Pitting growth

Full-Text [PDF 1009 kb]   (91 Downloads)