IDENTIFYING BALLAST FOULING USING STATISTICAL PATTERN RECOGNITION TECHNIQUES ON SMARTROCK DATA

Abstract

Railroad ballast serves different functions including draining water from track and distribution of the train loads. The ballast layer deteriorates and becomes fouled with time due to ballast particle abrasion and breakage as well as subgrade soil intrusion. Ballast fouling has become one of the most commonly seen track defects that can lead to inconsistent track performance. In the case of fouling, the ballast strength will decrease when it is wet (usually referred to as “mud-spot”) due to the lack of particle interlocking and lubrication effect of fine materials. However, both of the ballast strength and stiffness will increase dramatically when it is in dry condition as the ballast particles are well confined (Qian, 2016). This inconsistency in track behavior can cause higher deterioration rate of other track components such as rail and sleeper. Therefore, identifying mud spots in a timely manner is a critical issue in ballasted track maintenance.

 

The main purpose of this thesis is using advanced sensor networks and statistical pattern recognition techniques to identify ballast fouling by studying the relationship between ballast fouling condition and ballast particle movement. To that end, several field experiments were carried out with the aim of monitoring and recording the particle movements under different ballast conditions. In particular, two sections with the same traffic but different track conditions: one with clean ballast and the other with mud pumping, were chosen. The SmartRock (Liu, 2015) is used to obtain ballast particle movement information under traffic. The SmartRock is a wireless sensor device built using the 3D printing technology and resembles the real ballast particles in terms of shape, inter particle friction and specific gravity. This sensor device has the ability to record translational and rotational movement of a single ballast particle under dynamic loading and transfer the real-time data via Bluetooth to a base station. The autoregressive (AR) model was then applied to each of the acceleration and rotation time histories collected from the SmartRocks embedded in the two sections, during which the autoregressive coefficients will be obtained. Those coefficients will serve as damage indicators to identify ballast fouling severities.  The results and important findings are highlighted in this thesis.

 

Keywords: Ballast Fouling, Railway, SmartRock, Statistical Pattern Recognition Analysis

 

 

Table of Contents

 

List of Figures_____________________________________________ vi

List of Tables_____________________________________________ vii

Acknowledgments_________________________________________ viii

1 Introduction_____________________________________________ 1

1.1 Background_________________________________________________________ 1

1.2 Research Objective___________________________________________________ 3

1.3 Outline____________________________________________________________ 3

2 Literature Review_________________________________________ 5

2.1 Previous studies on ballast fouling________________________________________ 5

2.2 Previous studies on structural health monitoring (SHM)________________________ 14

2.2.1 The statistical analysis approaches for defect detection____________________________ 21

2.2.1.1 Statistical pattern recognition____________________________________________ 22

2.3 Ballast inspection___________________________________________________ 31

2.3.1 SmartRock_______________________________________________________ 34

3 Project descriptions_______________________________________ 44

3.2 Site Selection_______________________________________________________ 44

3.3 Instrumentation plan_________________________________________________ 46

4. Data analysis and results__________________________________ 49
5. Conclusions and recommendations___________________________ 68

5.1 Conclusion_________________________________________________________ 69

5.2 Future recommendations______________________________________________ 70

References______________________________________________ 72

1 Introduction

1.1 Background

Rail transportation is considered as one of the most economic and efficient modes of transportation for both passengers and freight. A typical railroad track (Figure 1-1) has two structural components: superstructure including rail, fasteners and sleeper and substructure consisting of ballast, sub-ballast, and subgrade.

 

The Ballast layer, depending on their locations within a track, can be divided into 3 main parts. The part between two adjacent ties which is called crib; the part outside the end of ties which are called shoulder ballast; and the part under the bottom of the tie (Zarembski, 2014).

 

The function of ballast layer includes but not limited to: distributing the train vertical load to protect the subgrade and providing drainage. In this regard, the ballast layer plays a key role in track performance. Various criteria including ballast layer thickness, particle size, and distribution can affect the ability of the ballast to fulfill its own function (Li, 2015).

 

Fouling is the condition when the voids in ballast become occupied by the smaller particles generated mainly from the ballast particle breakage. It is believed to significantly affect the stiffness and strength of the ballast. (Indraratna et al., 2013) Therefore, frequent evaluation of ballast and monitoring its condition are necessary from the efficiency and safety point of view.

 

Particle size distribution, normally obtained by sieve analysis, is one of the most common approaches to evaluate the ballast condition. A gap graded ballast assembly will create large enough voids to provide the track with proper drainage.  A typical ballast grain size distribution suggested by AREMA is depicted in the following figure.

 

Figure1- 2: #24 and #4 ballast particle gradation ranges based on AREMA (LI, 2015)

 

 

 

Although sieve analysis offers ground truth for the severity of ballast fouling, it is time consuming to conduct onsite sampling and sieving, not even mention the interruption of the daily traffic operation. A non-destructive and real time monitoring system with advanced statistical algorithms to accurately detect ballast fouling in a timely manner is in great need.

 

Nowadays, the inspection and monitoring of the structures are of a great interest in the railway area. A great number of advanced sensors and technologies have been developed in order to achieve more efficient inspection and monitoring system. These emerging technologies provide a vast amount of data from real-time monitoring of track infrastructure.  Extracting efficient knowledge required for the maintenance planning from these data is the next challenge.

 

 

1.2 Research Objective

The primary objective of this research is to develop an algorithm for evaluating the ballast condition through real-time particle movement data recorded using the SmartRock. This objective is reached by applying the statistical pattern recognition techniques on the data collected from two different structural conditions of ballasted track in order to find the best damage detection approach.

 

1.3 Outline

This thesis is presented in five chapters. Chapter 1 includes an introduction to the subject and research objectives. Chapter 2 presents a literature review of the current state of research in ballast fouling and structural health monitoring. Chapter 3 describes the procedures for site selection and the instrumentation plan of the project. Chapter 4 presents the analysis procedure in order to develop the damage detection algorithm and the result of applying these analyses on the data collected from the field. Chapter 5

provides suggestions for future research.

IDENTIFYING BALLAST FOULING USING STATISTICAL PATTERN RECOGNITION TECHNIQUES ON SMARTROCK DATA

Sharing is caring!