Analysis of high-speed train electromagnetic measurement data based on large data

Abstract

The electromagnetic environment of the train and the complex and variable train conditions are closely linked, which makes the analysis of the electromagnetic environment for the train is facing too many parameters, can not determine the causal relationship and other challenges. Based on the measurement theory of train electromagnetic signal, this paper obtains the common mode disturbance frequency domain signal of the antenna signal cable of balise transmission module (BTM) and Speed dynal, and preprocesses the data. Then, the classification and regression tree (CART) model and the radial basis function (RBF) neural network model of data mining (DM) are used to analyze the electromagnetic measurement data of the train, and the validity of the model is verified. Finally, the two data mining models were compared.