Modeling, simulation and development of fault diagnosis methods for pantograph-catenary system in railway systems

Nowadays, monitoring the interaction between the catenary line carrying the electrical energy and the pantograph system transmitting the energy to the locomotive is very important in terms of security in the electric railway that is a fast and reliable means of transport and uses clean energy. Especially with the spread of high-speed trains, the contact force between the pantograph and the catenary working under high current and voltage, the wear status of the pantographs surface and possible arcs must be checked periodically. Condition monitoring, active pantograph control and possible fault detection for the energy transmission quality between the pantograph and the catenary affected from environmental conditions, operating conditions and system parameters are compulsory process in terms of rail systems reliability. In this study, three new methods for modeling a pantograph-catenary system and analysis of the system parameters effects, the active control of the contact force between pantograph and catenary and fault diagnosis of the pantograph surface and the general pantograph-catenary system have been proposed and their validation results have been presented in the experimental data. First, transfer function of the pantograph-catenary system has been obtained by mathematical modeling. The system behavior according to the variation effects of the six parameters as mass, spring constant and the damping coefficient used in this model were analyzed for overall system. When the parameters of the system changed for any reason, it has been demonstrated that the oscillations occur in the system, pantograph-catenary interaction is adversely affected and the quality of energy transmission reduces as a results of the model simulation. The second study has been performed for active pantograph control that ensures the stable, reliable and efficient operation of the system in case of the system parameters variations in pantograph-catenary systems. For this purpose, a new and effective method is presented by adjusting the contact force of the pantograph according to the conditions constantly while the train is in motion in this thesis. Therefore, this new method based on the developed adaptive fuzzy controller and the image processing algorithm that finds and online checks the position information of the pantograph will make an important contribution to the literature. In the proposed approach, active control of the pantograph has been provided by using location information obtained from image processing techniques with an adaptive fuzzy controller in a pantograph-catenary image taken with a camera. Next study suggests an analysis approach in order to detect the regions of contact point and determine the wearing ratio of the pantograph regions and faults. Initially, the trajectory generated by catenary line when it is sliding on the pantograph has been modelled and then the fault analysis was carried out through the actual data by determining the point of the pantograph touched to catenary line with the camera and image processing algorithms. Thus, the possibility of the determination and prediction for the statistical usage status of each point and service life of the pantograph surface has been occurred. Recent study is to perform a method for detection of the arcs with fuzzy classifier by applying S-transformation to the current data received from pantograph-catenary system. Mayr arc model and current data obtained from simulation were used. A new method was developed by using actual current data of a train for detecting the arc via S-transformation and the corresponding properties of S-transformation and it was confirmed with various scenarios. Consequently new, efficient, highly accurate, robust, real-time applicable and non-contact condition monitoring, analysis, control and fault diagnosis methods for pantograph catenary systems that provide energy transmission to the locomotive from the electric line in the electrical trains have been presented. Results have been obtained for these new methods which have great importance in terms of security, energy transmission quality, efficiency and periodic monitoring in electrical trains and they will accelerate studies in this area. In addition, non-contact monitoring studies have been performed, while the train is in motion at high speeds, the advantage of the image processing techniques usage which allows obtaining effective results has been indicated. Additionally, the studies realized in this thesis are supported by the research project of 112E067 conducted at TUBITAK-1001 program. Keywords: Electrical railway systems, Pantograph-Catenary systems, Energy transmission, Active pantograph control, Condition monitoring, Fault diagnosis, Image processing, Arc detection.