Neural Network-Based Impedance Identification and Stability Analysis for Double-Sided Feeding Railway Systems

Jul 16, 2026

Double-sided power supply railway systems enable simultaneous vehicle operation but increase the likelihood of instability and oscillation-related overvoltage due to interconnected power supply sections. Because vehicles frequently change operating points, stability must be evaluated across wide conditions, requiring accurate black-box impedance identification of vehicle converters. Traditional approaches demand extensive data and offer limited interpretability. This study proposes an interpretable residual feedforward neural network (ResFNN) with SHAP to reduce data while preserving accuracy, and a component connection method to derive the multivehicle impedance matrix, validated using real operational data.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages25

Reading Time1 h 3 min

Introduction

Impedance-based stability analysis

Converter impedance identification (gray-box vs black-box)

Abstract

arXiv :2607 . 11072v1 [ ee ss . SY] 13 Jul 2026 1 Neural Network-Based Impedance Identification and Stability Analysis for Double-Sided Feeding Railway Systems Xiangyu Meng, Guiyang Hu, Zhigang Liu, Hui Wang, Guinan Zhang, Hongjian Lin, Mahdieh S. Sadabadi Abstract The double-sided power supply railway system increases the simultaneous operation of vehicles on the grid, potentially causing system instability and oscillation overvoltage issues. As vehicles frequently switch operating points during operation, it is essential to analyze system stability across a wide range of conditions. Therefore, accurately identifying the black-box impedance of vehicle converters at multiple operating points is crucial for studying railway vehicle-grid system stability. However, traditional impedance identification methods require extensive data and lack interpretability, leading to significant computational and data burdens. This study introduces an interpretable residual feedforward neural network (ResFNN) combined with SHapley Additive exPlanations for training vehicle impedance models, reducing data requirements while maintaining accuracy. Additionally, a component connection method is proposed for deriving the impedance matrix of a multivehicle railway system under the double-sided feeding mode. This method incorporates the dynamic mobility of vehicles and their positional distribution, and it utilizes the ResFNN to identify impedance for stability analysis. Real operational data from actual railway lines is used as case study to analyze the stability of the double-sided power supply railway system. The results demonstrate that this approach accurately assesses both lowfrequency and high-frequency instability issues. Index Terms-Railway, vehicle, oscillation, stability, doublesided feeding, impedance identification. NOMENCLATURE ANN ARTIFICIAL NEURAL NETWORKS CCM Component Connection Method GNSC Generalized Nyquist Stability Criterion MSE Mean Squared Error NARX Nonlinear Auto-regressive with eXogenous inputs PP Parallel Point ResFNN Residual Feedforward Neural Network RNN Recurrent Neural Network SHAP SHapley Additive exPlanations I. INTRODUCTION THE demand for reliable, efficient, and high-capacity railway transportation is rapidly increasing. Traditional single-way power feeding modes in railway traction power systems suffer from neutral zones that can cause power interruptions and create barriers to power exchange [1] . However, the double-sided power feeding mode offers a more promising solution by eliminating neutral zones, thereby increasing power capacity and extending supply distance [2] . Despite the benefits of double-sided power feeding, the absence of neutral zones leads to electrical interconnection between multiple power supply sections. This interconnected environment results in a significant increase in the number of vehicles operating within the same power supply network. Current incident reports indicate that simultaneous operation of multiple vehicles in the grid can cause low-frequency oscillation (LFO) [3], harmonic instability (HIS) [4], and other instability issues [5], shown in Fig. 1. These factors can compromise system safety and stability. Given the heightened risk of instability associated with double-sided power feeding, it is critical to investigate the stability of the railway vehicle-grid system under this configuration. The impedance-based analysis method is extensively employed to assess the stability of railway vehicle-grid system. This approach aligns with the EN 50388-2022 standard [5], which mandates that the real part of the vehicle input admittance remains positive within a specific frequency range to prevent harmonic instability. However, practical challenges hinder the creation of accurate vehicle impedance models. On the one hand, vehicle manufacturers are typically reluctant to disclose the internal control structures and parameters of their vehicles due to intel

FAQs

Why is system stability more challenging in double-sided power supply railway systems?

Double-sided feeding removes neutral zones and electrically interconnects multiple power supply sections, increasing the number of simultaneously operating vehicles. This environment can trigger low-frequency oscillation and harmonic instability.

What problem do traditional impedance identification methods have?

They require extensive data and provide limited interpretability, which creates significant computational and data burdens when identifying vehicle black-box impedance across many operating points.

How does the proposed method improve impedance modeling and stability assessment?

It uses an interpretable residual feedforward neural network (ResFNN) combined with SHAP to train vehicle impedance models with fewer data. It also introduces a component connection method to derive the multivehicle impedance matrix under double-sided feeding, then uses identified impedance for stability analysis with real operational data.

Jul 16, 2026

Upload by Theodora