Study of track irregularity time series calibration and variation pattern at unit section

Area of Science:

• Civil Engineering
• Data Science
• Signal Processing

Background:

• Track irregularity data quality is crucial for railway safety and maintenance.
• Existing data often suffers from anomalies, offsets, and noise, hindering accurate analysis.
• Time series analysis of track irregularities requires robust data preprocessing techniques.

Purpose of the Study:

• To address data quality issues in track irregularity time series.
• To develop and apply advanced algorithms for data cleaning and reconstruction.
• To analyze patterns, features, and trends in track irregularity standard deviation data.

Main Methods:

• Abnormal data identification and data offset correction algorithms.
• Local outlier data identification and noise cancellation techniques.
• Wavelet decomposition and reconstruction for time series analysis.
• Statistical study of track irregularity standard deviation in unit sections.

Main Results:

• Effective identification and correction of data anomalies and noise.
• Successful decomposition and reconstruction of track irregularity time series.
• Identification of distinct patterns and features in standard deviation data sequences.
• Discovery and description of changing trends in track irregularity.

Conclusions:

• The proposed methods significantly improve track irregularity time series data quality.
• Wavelet-based approaches are effective for processing and analyzing this type of data.
• Understanding trends in track irregularity is vital for predictive maintenance and safety.