Smart Traction Power Supply Systems for High-Speed Railways / Shibin Gao and Zhengqing Han.

Format:

Book

Author/Creator:

Gao, Shibin, author.

Han, Zhengqing, author.

Language:

English

Subjects (All):

Electric railroads.

Physical Description:

1 online resource (494 pages)

Edition:

First edition.

Place of Publication:

Cambridge, MA : Elsevier Inc., [2025]

Summary:

Given the increasingly busy railway networks, increased axle loads, and railway speeds,modern railways and urban transport systems require smart and energy-efficient tractionpower supply products and solutions to ensure safe, reliable, and environmentallysustainable operations.

Contents:

Front Cover

Smart Traction Power Supply Systems for High-Speed Railways

Copyright Page

Contents

Preface

1 Introduction

1.1 Intelligence

1.1.1 Human intelligence

1.1.2 Artificial intelligence

1.1.3 Smart machine

1.1.4 Smart system

1.2 Smart railways

1.2.1 Development of foreign smart railways

1.2.1.1 Background to Shift2Rail

1.2.1.1.1 Comprehensive challenges

1.2.1.1.2 Challenges of service quality

1.2.1.1.3 Cost challenges

1.2.1.1.4 Challenge of competitiveness

1.2.1.2 Key objectives of Shift2Rail

1.2.1.3 Main contents of Shift2Rail

1.2.2 Development of smart railways in China

1.2.2.1 Concept of smart railway

1.2.2.2 Technical architecture and system of smart railway

1.2.2.2.1 Technical architecture

1.2.2.2.2 Technical system

1.2.2.3 Construction of smart railway

1.2.2.3.1 Test on Beijing-Shenyang smart high-speed railway

1.2.2.3.2 Construction of smart Beijing-Zhangjiakou high-speed railway

1.3 Smart traction power supply system

1.3.1 Definition

1.3.2 Structure and characteristics

1.3.2.1 System architecture

1.3.2.1.1 System architecture

1.3.2.1.2 Level division

1.3.2.2 System functions

1.3.2.3 Technical characteristics

1.3.2.3.1 Extrinsic characteristics

1.3.2.3.2 Intrinsic characteristics

Summary

References

2 Smart traction substations

2.1 Overall architecture

2.2 Smart primary equipment

2.2.1 Overview

2.2.2 Smart traction transformer

2.2.2.1 Online monitoring items

2.2.2.2 Online monitoring methods

2.2.2.2.1 Temperature of winding and core

2.2.2.2.2 Gases dissolved in oil

2.2.2.2.3 Core grounding current

2.2.2.3 Smart components for online monitoring

2.2.2.3.1 Smart components and layout of sensors

2.2.2.3.2 Smart component cabinet

2.2.3 220kV smart circuit breaker.

2.2.3.1 Online monitoring items

2.2.3.2 Smart components and monitoring sensors

2.2.3.2.1 Smart components

2.2.3.2.2 Monitoring sensors

2.2.4 27.5kV or 2×27.5kV smart GIS switchgear

2.2.4.1 Online monitoring items

2.2.4.2 Smart components and monitoring sensors

2.2.4.2.1 Smart components

2.2.4.2.2 Monitoring sensors

2.2.5 220kV smart disconnector

2.2.5.1 Online monitoring items

2.2.5.2 Smart components and monitoring sensors

2.2.5.2.1 Smart components

2.2.5.2.2 Monitoring sensors

2.2.5.2.3 Layout of monitoring unit for disconnector

2.3 Smart secondary equipment

2.3.1 Traction network protection and fault location

2.3.1.1 Three-level feeder protection: local area, substation area, and wide area

2.3.1.1.1 Local protection

2.3.1.1.2 Substation area protection

2.3.1.1.3 Wide-area protection

2.3.1.2 Fault location and type judgment in traction network

2.3.1.2.1 Definitions

2.3.1.2.2 Location of fault in traction network

Fault location principle based on the AT neutral point boosting current ratio

Fault location principle based on the up and down current ratio

Fault l006Fcation principle based on the cross-connection line current ratio

2.3.1.2.3 Enabling elements for fault location

Enabling by activation signal of feeder protection at traction substation

Enabling by low-voltage of traction network

2.3.1.2.4 Fault type judgment of traction network

2.3.2 Wide-area protection of traction network

2.3.2.1 Fault analysis of all-parallel autotransformer network

2.3.2.2 Wide-area protection for traction network based on current characteristics

2.3.2.3 Wide-area protection for traction network based on impedance characteristics

2.3.2.3.1 Setting values of protections 1 and 2 at traction substation

2.3.2.3.2 Setting values of protections 3 and 4 at AT post.

2.3.2.3.3 Setting values of protections 5 and 6 at section post

2.3.3 Self-healing reconfiguration of traction power supply system

2.3.3.1 Self-healing reconfiguration of traction substation

2.3.3.1.1 Automatic switching to standby incoming line

Automatic switching to the 2# incoming line upon voltage loss in the 1# incoming line to enable shifting from opera...

2.3.3.1.2 Automatic switching to standby traction transformer

Automatic switching to T2 when a fault occurs in the main transformer T1 to enable shifting from operation mode 1 to operat...

Automatic switching to T1 when a fault occurs in the main transformer T2 to enable shifting from operation mode 2 to operat...

2.3.3.2 Self-healing reconfiguration of autotransformer AT post

2.3.3.2.1 Automatic switching between ATs in the connection mode through circuit breakers

2.3.3.2.2 Automatic switching between autotransformers in the connection mode through disconnectors

2.3.3.3 Post-fault self-healing reconfiguration of traction network

2.3.3.3.1 Case 1 - Self-healing reconfiguration of traction network after feeder circuit breaker at traction substation bec...

2.3.3.3.2 Case 2 - Self-healing reconfiguration of traction network with uninterrupted power supply to up and down feeding ...

2.3.3.3.3 Case 3 - Self-healing reconfiguration of traction network with partial power losses of up and down feeding sections

2.3.4 Application mode of smart traction substation

2.4 Summary

3 Smart OCS

3.1 Introduction

3.2 Overview

3.3 Design philosophy of comprehensive detection and monitoring

3.3.1 Characteristic Parameters of OCS

3.3.1.1 Geometric parameters

3.3.1.1.1 Static geometric parameters.

3.3.1.1.2 Dynamic geometric parameters

3.3.1.2 Electrical parameters

3.3.1.3 Mechanical parameters

3.3.1.4 Limits of characteristic parameters

3.3.2 Technical status of parts and components and equipment in OCS

3.3.3 Top-level design of comprehensive detection and monitoring

3.4 OCS detection and monitoring method

3.4.1 Detection method for characteristic parameters of OCS

3.4.1.1 Detection method for geometric parameters

3.4.1.1.1 Stereo-vision measurement method

Model 1: Pinhole camera model

Model 2: Line-scan camera model

Model 3: Measurement model of binocular line-scan camera

3.4.1.1.2 Detection methods for geometric parameters based on stereo vision

3.4.1.2 Electrical parameters-Detection method for pantograph-catenary arcing

3.4.1.3 Mechanical parameters-Detection method for pantograph-catenary contact forces

3.4.2 Inspection method for technical status of parts and components of OCS

3.4.2.1 Object detection based on deep convolutional neural network

3.4.2.1.1 Architecture of deep convolutional neural network

Fully connected layer

Convolutional layer

Activation function

Pooling layer

3.4.2.1.2 Deep convolutional neural network training

Loss function

Backward propagation

3.4.2.1.3 OCS components localization based on deep convolutional neural network

Region proposal-based object detection

Regression-based object detection

3.4.2.1.4 Object detection based on cascade network

3.4.2.1.5 Experimental Verification

Key components localization

Cascade localization of split pins

3.4.2.2 Defect detection of OCS components based on deep ensemble learning

3.4.2.2.1 Image feature extraction

3.4.2.2.2 Transfer learning

Basic methods

Transfer learning based on deep convolutional neural networks

3.4.2.2.3 Ensemble learning.

The role of ensemble learning

Error analysis of ensemble learning

Ensemble learning with diversity

3.4.2.2.4 Empirical risk and structural risk

Structural Risk Minimization principle

Support vector machine

3.4.2.2.5 Pin-missing detection based on deep ensemble classification

Architecture of the deep ensemble classifier

Training and prediction of the deep ensemble classifier

3.4.2.2.6 Experimental verification

3.4.2.3 Defect detection of OCS components based on deep unsupervised learning

3.4.2.3.1 Autoencoder

Shallow autoencoder

Deep autoencoder

3.4.2.3.2 Multi-task Learning

3.4.2.3.3 Insulator defect detection based on deep denoising autoencoder

Insulator defect detection framework

Deep multi-task convolutional neural networks

Criteria for defect detection

3.4.2.3.4 Experimental verification

3.4.2.4 Defect detection of OCS components based on deep bayesian segmentation network

3.4.2.4.1 Bayesian neural network

3.4.2.4.2 Deep bayesian neural network

Training of deep neural networks

Stochastic regularization of deep neural networks

3.4.2.4.3 Defect detection of OCS contact wire support based on deep bayesian segmentation network

Defect detection framework for OCS contact wire supports

Segmentation of contact wire supports components

Defect detection of CWS components

3.4.2.4.4 Experimental verification

3.4.2.5 Swivel clevis defect detection based on deep adaptive learning

3.4.2.5.1 Adaptive learning

Reliability evaluation of the model

Update of the model

3.4.2.5.2 Swivel clevis defect detection based on deep adaptive learning

Adaptive defect detection framework for swivel clevises

Adaptive Swivel Clevis segmentation

Swivel clevis defect detection

3.4.2.5.3 Experimental verification

Swivel clevis segmentation.

Swivel clevis defect detection.

Notes:

Includes bibliographical references and index.

Description based on publisher supplied metadata and other sources.

Description based on print version record.

ISBN:

9780443333248

0443333246

OCLC:

1492704305