| Abstract [eng] |
Relevance of the research: Having adapted a railway section to electric traction and frequently exploiting it, the problems of maintenance of the infrastructure are faced. It is highly relevant to properly and periodically maintain the most important sites of electric traction equipment, which are of vital importance to the entire system of electric traction; these are the places which are significant in ceaseless operation of the entire system. Maintenance of the condition of the pantograph is one of such sites. It is relevant and highly important because the pantograph facilitates the supply to train engines. Since electricity supply to train engines proceeds also at a high speed of the train, the mechanical impact may inevitably damage the pantograph. This happens when the train moves at a high speed; also it is impacted by friction and other causes: meteorological phenomena, alteration of railway relief, angles of railway tilt etc. Regular examination is required to identify the condition of the pantograph. The aim of the research: grounding on analysis of theoretical and practical materials and the results of investigation of possibilities for diagnostics of the pantograph, to process the signal caused by oscillations of the pantograph, which is received from a testing stand, and to identify it by attributing to different classes of the condition of the pantograph. The object of the research: an automated system of identification of defects in electric traction pantographs. Objectives of the research: 1. to create a method of signal identification and classification, which is suitable for the automated system of detection of defects in pantographs; 2. to select an algorithm of signal transformation, which allows proper identification of characteristics of the signals of oscillations caused by pantographs; 3. to investigate precision of a suggested method of classification in identifying a defect and separating signals of defectless pantographs from signals of damaged pantographs. Content and structure of the thesis: The Master’s Thesis consists of 3 parts: analysis of literature sources, investigation of both possibilities for diagnostics of the pantograph and identification of defects in the pantograph. The volume of the Thesis is 67 pages, including 35 figures and 17 tables. The first part of the Thesis deals with analysis of literature describing the principles of electric traction, research on identification of the condition of the pantograph, presents the means of mathematical analysis of signals. The second part of the paper introduces the idea of structures of the diagnostic systems for the pantograph, selects the transformation of mathematical signals, which is suitable for identification of the characteristics of diagnostic systems. The third part describes an experiment during which signals of different classes of defects are attributed to the sets of characteristics instructed in advance. The conclusions proving the possibilities for automated diagnostic systems for pantographs are formulated. |
