Resumen de Pantograph-catenary dynamic models and their implementation in hardware-in-the-loop tests
There is an extensive network of electrified railway lines over the world. Most of them use overhead contact lines or catenaries to provide the trains with electrical power. Catenaries consist of electrified wires placed over the rail track, designed to contact the pantograph placed on the roof of the train. The proper operation of the system is very demanding, especially at high speed, when the continuity of the contact is compromised.
The most predominant tool for studying and designing the pantograph-catenary system is the use of numerical simulations. Notably, the Finite Element Method (FEM) is the most popular technique for modelling and simulating the dynamic interaction of the pantograph and the catenary. This method allows modelling catenaries with outstanding fidelity and without any loss of generality.
After the simulation stage, the pantograph and the catenaries have to be assessed by in-line experimental tests. However, there is an alternative that can replace those tests with a significant reduction in costs. The alternative method, called Hardware In the Loop (HIL), allows testing pantographs in the laboratory with a test rig that emulates the interaction with a virtual catenary. Different research groups have implemented HIL; however, in every attempt, a compromise solution has been adopted, demonstrating the challenging nature of HIL. This Thesis aims to advance in the field of HIL tests, pushing forward the capabilities of the technique and solving some of the limitations found in the literature. This Thesis proposes two different kinds of catenary models for their use in HIL tests.
The first is an analytical model based on a string of periodic geometric profile that accounts for the steady state. It reduces the complexity of the catenary but keeps the main features involved in the dynamic. The model has proven useful in explaining the fundamental dynamics of the catenary, helping understand the interference between two pantographs. This analytical model is suitable for HIL because of its low computational cost. An iterative algorithm is proposed to use the analytical model in HIL. The fact that the model is periodic permits a specific strategy to compensate the control loop delay. This strategy has excellent performance and accuracy, validated by comparing HIL tests with numerical simulations and getting an agreement. This agreement will not be possible if the pantograph model of the simulations is inaccurate. Therefore, the validation is carried out with a weight or mass model in place of the pantograph to eliminate potential differences. Even though the precision achieved is good, the analytical catenary model lacks fidelity, which has motivated the development of a more advanced periodic model.
The second catenary model for HIL tests is the Periodic Finite Element Model (PFEM), discretised with FEM to avoid further topological and structural simplifications. The model includes the periodicity condition, and the dynamics are solved by frequency analysis. Furthermore, the catenary non-linearities are considered in the formulation. An iterative algorithm, similar to the one used for the HIL tests with the analytical catenary, is used to realise HIL tests with PFEM catenaries. The previous strategy with a mass model is used to validate the test, confirming great precision. The results are gratifying due to the sophistication of the model, the accuracy of the tests and the cancellation of the delay. The tests simulate the response of realistic catenaries with the simplifying periodicity hypothesis. They are adequate for the dynamic of equal-span catenary at the central zone of every section, but future efforts have to be made to get rid of the periodicity condition while keeping the accuracy of the results.
