Traction Network Fault Determination Method Based On Mathematical Morphology_wavelets Combined Analysis

Relative to the highway, because high-speed railway covers the less of area,the higher significant advantages of land utilization,so high-speed railway has become the main tool of transportation in China.Construction of high-speed railway passenger dedicated line can meet the demand of national economic development and the situation at the present stage in China. Railway traction network is a special power supply circuit which is set up in the air, with a long line and exposures to the harsh outdoor environment conditions in a long time. It is unable to maintain the railway traction network effectively because the topography that the railway traction network go through is complex.And coupled with the influence of natural factors(such as weather), cause the railway traction network is the weakest part of the railway transportation system.So there are many difficulties in its operation and maintenance. Once the traction network fault, finding the fault location quickly and accurately is the key to repair the line in a short time.It can be seen that establishing a system with fault type judgment and measurement point of failure accurately has a direct impact to repair time on the improvement of transport efficiency and the shorten of repair time.After careful analysis research on high-speed railway traction network’s power supply system,according to the existing problems,this paper is combined with mathematical morphology and wavelet analysis to solve problems such as high-speed railway traction network fault line head identification and the fault type judgment by reference and summarizes the previous work experience.The main research work in the following aspects:(1)Designed and produced an active C type traveling wave signal sensor which combined with DC high voltage pulse source, pulse capacitor, traveling wave signal sensor, outdoor vacuum circuit breaker, data acquisition card and so on.The device simulation experiment was carried out in the laboratory for the environmental adaptability. The results show that the set device to capture the fault line wave signal characteristics meet for traveling wave ranging.(2)Through elaborating the basic principle of wavelet transform and mathematical morphology completely, compared the signal filter effect of both method quantitatively, put forward the integrated filter based on mathematical morphology and wavelet analysis considering the comprehensive advantage of the two.The results show that the integrated filter can effectively filter the impulse noise, white can effectively filter out noise again.(3)Through introducing morphological gradient completely, component fault feature extraction module 1by using morphological gradient.The simulation results show that the proposed algorithm can quickly extract traveling-wave range with simple calculation which is easy to implement.(4)Design algorithm to extraction and analysis failure properties,using the nature of the change of the reflection wave polarity judgment fault, can realize the recognition of the nature of open circuit, short circuit fault effectively.To sum up, the method and the whole device this thesis used can better meet the requirements of high speed railway traction network fault location, embody its own merit and has a certain reference value on the study of the field.