Research On Characteristics And Detection Methods Of DC Pantograph-catenary Arc

DC pantograph-catenary system is mainly used in urban rail transit.The appearance of catenary arc will affect the service life of catenary to a great extent,and may also cause interference to the normal running of subway trains.Therefore,the study of DC catenary arc is of great significance to the improvement of the service life of catenary and the stability of subway train operation.In addition,the current research results of pantograph-catenary arc are mainly concentrated in the field of AC,and the research of DC arc is still in the exploratory stage,so the research of DC arc is also of academic significance.Dissertation through the experimental platform,DC pantograph-catenary arc set for DC pantograph-catenary arc experiment,at the same time,the physical field of the DC pantographcatenary arc simulation,from two aspects of the electrical properties and physical properties of DC pantograph-catenary arc analyzed,by analyzing a variety of acquired greater degree of the influencing factors of DC pantograph-catenary arc existence state,as well as its influence on the DC pantograph-catenary arc.By analyzing the arc data obtained from the experiment,a method to detect dc arc is proposed.The thesis mainly obtains the following conclusions:(1)The influence of experimental conditions on the existence state of dc arc was analyzed.It was found that when other experimental conditions were constant,the arc firing time of dc arc was positively correlated with contact pressure,negatively correlated with rotating speed and positively correlated with working current level.The downflow level of DC pantograph-catenary arc is negatively correlated with contact pressure,positively correlated with rotating speed and positively correlated with working current level.The fluctuation level of DC pantograph-catenary arc is not affected by contact pressure and rotating speed.(2)The magnetohydrodynamic simulation of DC arc is carried out from two different aspects,and the variation law of arc temperature field distribution under various influences was studied.It was found that the maximum temperature of arc was positively correlated with the working current level,negatively correlated with the contact gap distance and negatively correlated with the flow velocity,and a reasonable explanation was given.At the same time,the flow field simulation of parallel train is carried out in the DC pantograph-catenary arc,and the distribution of temperature field under the flow field is explained.(3)Through to the DC pantograph-catenary arc line of randomness is analyzed by the fluctuation of the current signal through the phase space reconstruction and the maximum Lyapunov detection detection DC pantograph-catenary arc current signal has chaos characteristics,belong to the chaotic time series,and obtain the equivalent higher dimensional reconstruction sequences,by using manifold learning dimension reduction algorithm to reconstruct sequence of visual processing,can direct observation of low dimensional sequence,and the distribution of points in the sequence of vector as the basis,using the limit of machine learning algorithm to achieve the DC pantograph-catenary arc recognition,good recognition effect.This dissertation has 35 figures,4 tables,95 references.