Grooved rail geometrical parameter trend prediction method and system

Abstract

The invention discloses a grooved rail geometrical parameter trend prediction method and system. The method comprises the following steps of subjecting detected grooved rail geometrical parameter values to data storage and batch processing; performing data preprocessing identification and correcting an abnormal value; constructing and training a radial basis function neural network, selecting left-right height, left-right rail direction, gauge and ultrahigh data of a groove-shaped rail setting detection section, and inputting average values of various parameters in the same detection time period into the radial basis function neural network for training; selecting the maximum value of the abrasion data of the set detection section of the grooved rail, and inputting the maximum value into aradial basis function neural network for training; iteratively updating the center and variance of the radial basis function neural network basis function and the weight between the hidden layer andthe output layer; and inputting detection data for prediction to obtain prediction data of the irregularity and the wear value of the grooved rail. According to the method, big data, the neural network and track geometrical parameter prediction are combined, and the generalization ability and convergence speed of the neural network are improved.

Description

technical field [0001] The invention relates to the technical field of grooved rail detection, in particular to a method and system for predicting the trend of geometric parameters of grooved rails. Background technique [0002] At present, the combination of tram projects and big data is mostly in pre-monitoring, processor-level response, data warehouse technology (data analysis, mining), Internet of Things technology (passenger service client) and embedded system driving technology (dynamic operation Evaluation), etc., but the application of the combination of big data analysis technology and the geometric parameters of tram grooved rail has not yet been deeply researched and expanded. [0003] In the existing technology, data mining technology is used to predict the internal damage trend of rails, ultrasonic flaw detection is used, and ultrasonic signals are used to achieve calculation position alignment. However, using ultrasonic waves as alignment signals requires high ...

AI Technical Summary

Problems solved by technology

[0002] At present, the combination of tram projects and big data is mostly in pre-monitoring, processor-level response, data warehouse technology (data analysis, mining), Internet of Things technology (passenger service client) and embedded system driving technology (dynamic operation evaluation), etc., but there are not many in-depth research and development on the application of combining big data analysis technology with the geometric parameters of tram grooved rails.

[0003] In the existing technology, data mining technology is used to predict the internal damage trend of rails, ultrasonic flaw detection is used, and ultrasonic signals are used to achieve calculation position alignment. However, using ultrasonic waves as alignment signals requires high detection equipment, and for special rails (groove rails) and In other words, there are many detection parameters, and there are many abnormal values ​​of parameters in the cutting section. This method is not suitable for the alignment of grooved rail track irregularity parameters.

Images