High speed turnout injury identification method based on vibration signal wavelet threshold value denoising

Abstract

The present invention discloses a high speed turnout injury identification method based on vibration signal wavelet threshold value denoising, which is used for identifying turnout fracture, flaw, nuclear injuries and other working conditions in a high speed turnout injury monitoring system. The method comprises the following steps: acquiring turnout vibration signals, extracting turnout vibration signal feature, and classifying and identifying turnout injury. Before extracting the turnout vibration signal feature, automatic endpoint detection is performed for the turnout vibration signals based on short-term energy, and the active vibration signal is extracted according to detection endpoint. The wavelet threshold value denoising is used for the extracted effective vibration signal to get injury identification-needed valid data. The method of the invention has a significant effect on suppressing noise, can effectively improve recognition rate and instantaneity of turnout injury identification based on the vibration signal, and has a wide application prospect in high speed railway turnout injury monitoring.

Description

technical field [0001] The invention relates to a technical method for safety monitoring and protection of high-speed railway turnouts, in particular to a method for identifying damage of high-speed railway turnouts. Background technique [0002] High-speed turnouts are necessary basic equipment for high-speed railways. They are movable parts and vulnerable links in railway lines. They have the characteristics of large number, short life, complex structure, and low safety. points and difficulties. Due to the interaction between the vehicle and the track, the turnout is affected by external loads, the environment, disasters and other factors, and it will inevitably be damaged to varying degrees during long-term use, and the structural performance will decline, which will affect the safety of the turnout. Therefore, the high-speed turnout The real-time monitoring of damage status has become a problem to be solved urgently. [0003] Existing turnout safety monitoring systems,...

AI Technical Summary

Problems solved by technology

The object monitored by this method is a fixed rail. Due to the strong regularity of the vibration signal of the fixed rail, low noise, and high signal-to-noise ratio, this method can still be realized; however, it is more powerful and complex for the background interference noise of the movable and changing cross-section High-speed turnouts with much more severe vibrations cannot be used

In addition, the empirical mode decomposition method used in this method has problems such as modal interference and high computational complexity, and is not suitable for real-time monitoring systems for high-speed turnout damage.

Images