Error analysis of the application of combined subspace identification to the modal analysis of railway vehicles

This study is part of the development of a methodology for the application of the combined deterministic-stochastic subspace identification method for the identification of the modal parameters of railway vehicles. The proposed methodology employs the vertical accelerations measured on the axle boxes of one wheelset as excitations and the accelerations measured in the carbody and bogies as responses, which are registered when the vehicle is in circulation. With this aim, first the parameters that may affect the modal identification accuracy are identified. Then, they are grouped into three categories: parameters associated with the vehicle, parameters related with the excitation that the vehicle receives and parameters associated with the test execution. Next, an index named Modal Parameter Error Index (MPEI) is introduced. This index allows estimating the error that will be obtained in the identification of the natural frequency and damping ratio of the modes based on the knowledge of the parameters of the test. The terms that compose the index have been obtained applying the Monte Carlo method in numerical simulations for simple models. The availability of MPEI is a very powerful tool in the application of the combined subspace identification method for identifying modal parameters of railway vehicles. Also, based on MPEI and the expected variation range of the influencing factors, useful application guidelines are given.