Seasonal frozen region railway subgrade settlement prediction method based on improved echo state network

By improving the method for predicting railway subgrade settlement in seasonally frozen areas using echo state networks, an IESN model with a small-world network topology was constructed using a multi-sensor monitoring platform and feature reconstruction technology. Hyperparameters were optimized, which solved the complexity and uncertainty problems in predicting railway subgrade deformation in seasonally frozen areas and achieved high-precision settlement prediction.

CN122389646APending Publication Date: 2026-07-14LANZHOU JIAOTONG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN Β· China
Patent Type
Applications(China)
Current Assignee / Owner
LANZHOU JIAOTONG UNIV
Filing Date
2026-05-25
Publication Date
2026-07-14

AI Technical Summary

Technical Problem

Predicting railway subgrade deformation in seasonally frozen areas faces challenges such as complex mechanisms, inconsistent data quality, insufficient model feature capture capabilities, and difficulties in hyperparameter optimization. Existing methods struggle to achieve high-precision, stable, and long-term predictions.

Method used

An improved echo state network (IESN) is adopted, which collects multi-source time-series data through a multi-sensor monitoring platform. It combines VMD and sample entropy to reconstruct features, builds a reserve pool of small-world network topology, introduces a configurable delay mechanism, and uses an improved Ivy algorithm (IIVYA) to optimize hyperparameters, thereby achieving efficient global optimization.

Benefits of technology

It improves data quality and physical interpretability, enhances the ability to capture complex time-series dynamics, achieves high-precision prediction of railway subgrade settlement, has strong engineering applicability, and its prediction accuracy and robustness are significantly better than traditional methods.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122389646A_ABST
    Figure CN122389646A_ABST
Patent Text Reader

Abstract

The application discloses a seasonal frozen region railway roadbed settlement prediction method based on an improved echo state network. First, a multi-sensor monitoring platform is built, railway roadbed settlement, air humidity and multi-depth soil humidity time series data are collected, and correlation analysis and feature screening are performed. Then, the VMD (Variational Mode Decomposition) and sample entropy reconstruction method are used to adaptively denoise and enhance the multi-scale features of the screened input feature sequence. Then, an improved echo state network (IESN) model is constructed. The model introduces a small-world network topology to generate a reserve pool connection weight matrix, and adds a configurable delay mechanism to enhance the ability to capture complex time series dynamic characteristics. Finally, an improved ivy algorithm (IIVYA) is proposed, which is a fusion of multi-elite reverse learning, Cauchy mutation and stable climbing strategy, which is used for global and efficient optimization of the hyperparameters of the IESN model to obtain the final prediction model.
Need to check novelty before this filing date? Find Prior Art