Fault Diagnosis Of Traction Gear Of HXD1C Electric Locomotive

With the rapid development of the railway industry in recent years,it has brought great convenience to people's travel and material transportation.The in increase in railway lines has increased the demand for electric locomotives,so the maintenance tasks of electric locomotives are more and more heavy.One of the most important tasks is to diagnose the faults of various parts of electric locomotives and carry out repairs based on the diagnosis results.Guarantee the normal operation of all parts of the locomotive is the premise of the normal operation of the entire electric locomotive.The gears in the gearbox are the key components of the transmission part,so the fault diagnosis of the gears is particularly important.If the gear failure cannot be found in time and repaired or replaced,it is likely to affect the normal operation of the train.Based on the in-depth understanding of the traction gear wear mechanism of HXD1C electric locomotive,this study obtains a variety of abrasive image from gears in a locomotive maintenance database,and uses image processing technology to process the abrasive image to provide data for judging the gear failure types.The MATLAB and VS software were used to develop the fault diagnosis system for the electric locomotive gearbox gear.The main research contents of this paper are as follows:(1)Based on the oil analysis technology,the common faults of the gears,the main abrasive types and the wear mechanism were studied.And the abrasive grains in the gearbox lubricating oil were qualitatively analyzed.The processing of colored abrasive image includes color image color space conversion,image enhancement and image segmentation,and extracts the color,texture and shape characteristic parameters of the seven kinds of abrasive grains,and provides data support for the paper.(2)The application of the reverse selection algorithm in gear fault diagnosis is studied.For the shortcomings of most existing fault diagnosis methods based on oil detection technology,the application of optimized reverse selection algorithm in the field of gear fault diagnosis is proposed to improve the accuracy of the fault diagnosis.The reverse selection algorithm trains a large number of gear data to generate detector sets of different abrasive types,and achieves the purpose of fault diagnosis through the detector set.This method is suitable for the processing of a large amount of data,and the accuracy of the fault diagnosis is greatly improved due to the large number of fault sample data.(3)Based on the premise of intelligent and modularization,a system for fault diagnosis of HXD1C electric locomotive gearbox gears was developed.The system mainly includes the processing of gear abrasive image,feature extraction,fault diagnosis results,maintenance recommendations and historical data query.The maintenance modules provide platform support for gear status detection and fault diagnosis.It is found that the accuracy of gear fault diagnosis using the method proposed in this paper is higher than that of the existing widely used methods,which provides a reliable basis for guiding workers to carry out maintenance and repair.