Antonov A. Improving of the high current elements interaction on electric transport

The dissertation is devoted to the improvement interaction of high-current contact elements of electric transport, taking into account the physical and mechanical parameters of the high-current contact elements. Defined set of major factors influencing the process of wear high-current contact elements of electric transport. The analysis of the work of the current collectors shows that their renewal at the maximal wear of the current collectors elements is carried out only in 18% of cases. The main causes of failure are due to the uneven wear of the current collectors elements (28%), deepening and arson (24%). Established static relationship and described by the analytical expression of linear regression between hardness and specific electrical resistance of carbon current collection elements. An experimental test stand has been developed for experimental research. Experimental research have been carried out on the wear of the elements of a friction pair of high-current contact. Obtained dependences of the contact wire and current-collection elements wear from the set of influencing factors. Rational conditions for the interaction current collection elements of friction pair are established. The forecasting neural network model of wear current-collection friction pair "contact wire - current collection element" is developed. Forecast neural network model allows to determine the amount of wear, depending on the variables of influencing factors and parameters. The method of wear prediction of a sliding high-current contact is developed. It is aimed to achieving reduction of the contact wires and current collection elements wear, by determining the rational position on the pantograph. Taking into account the disadvantages of existing devices and the peculiarities of operating conditions, a device for determining the parameters of carbon recirculation elements was developed, in which the algorithm of the method of forecasting the wear of a sliding high-current contact is implemented. The analysis of the prediction wear results indicates that the application of the proposed method in operation will allow, in the first approximation, to achieve a reduction in the wear of the contact wire by 9-10% and increase the running of current collection elements pantographs by 25%. The feasibility study for the application of predictive neural network models and the use the forecasting method of the sliding high-current contact wear has been carried out. The economic effect from the use of prognostic neural network models instead of carrying out long-term experimental research is 33.4 thousand UAH at the performance of the volume of experimental research on the wear of high-current sliding contacts. The economic effect of using the method of forecasting the wear of a sliding high-current contact in a single locomotive depot may be 130 thousand UAH / year, and the estimated payback period is less than one year. The proposed method is an effective and flexible tool, which provides an opportunity for an adequate response to changes in the parameters of the system of current-collection.