A Full Information Expression Model for Track Irregularity Based on Stochastic Harmonic Functions in Vehicle–Turnout Structure Stochastic Vibration Analysis
Turnout irregularity governs the stochastic vibration response of vehicle–turnout coupling systems, yet frequency-domain models that preserve the statistical characteristics of each frequency point remain scarce. This study establishes a turnout irregularity full information expression model (TIFIEM) using stochastic harmonic functions (SHF) and applies it to vehicle–turnout stochastic vibration and reliability analysis. Spectral estimation trials identify the Hamming window with a 4096-point window length as optimal for turnout irregularity power spectral density (PSD) estimation, and a fifth-order polynomial is recommended for PSD fitting with minimal error. The TIFIEM reproduces randomness in amplitude, frequency, and phase; a sample size of 250 irregularity realizations minimizes the root-mean-square error against the target spectrum. PSD amplitudes at distinct frequency points follow a Chi-square distribution with 2 degrees of freedom. Application to a No. 18 turnout at 300 km/h identifies the straight switch rail 3–7 m from the switch rail tip and the point rail 53–54 m from the tip as the most wear-susceptible regions. The reliability of the vehicle–turnout structure at the crossing panel decreases to 95.8%, indicating that these zones warrant prioritized inspection.