Standards Certification Education & Training · The information presented in this publication is...

Standards

Certification

Education & Training

Publishing

Conferences & Exhibits

Table of Contents

View Excerpt

Buy the Book

Table of Contents

View Excerpt

Buy the Book

https://www.isa.org/store/detecting-leaks-in-pipelines/52427846

Detecting Leaksin Pipelines

By Edward J. Farmer, PE

Copyrighted Material


NoticeThe information presented in this publication is for the general education of the reader. Because nei-

ther the author nor the publisher has any control over the use of the information by the reader, both the author and the publisher disclaim any and all liability of any kind arising out of such use. The reader is expected to exercise sound professional judgment in using any of the information presented in a particu-lar application.

Additionally, neither the author nor the publisher has investigated or considered the effect of any patents on the ability of the reader to use any of the information in a particular application. The reader is responsible for reviewing any possible patents that may affect any particular use of the information pre-sented.

Any references to commercial products in the work are cited as examples only. Neither the author nor the publisher endorses any referenced commercial product. Any trademarks or tradenames refer-enced belong to the respective owner of the mark or name. Neither the author nor the publisher makes any representation regarding the availability of any referenced commercial product at any time. The manufacturer’s instructions on the use of any commercial product must be followed at all times, even if in conflict with the information in this publication.

Copyright © 2017 International Society of Automation (ISA)All rights reserved.

Printed in the United States of America. 10 9 8 7 6 5 4 3 2

ISBN: 978-1-941546-48-2

No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written per-mission of the publisher.

ISA67 T. W. Alexander DriveP.O. Box 12277Research Triangle Park, NC 27709

Library of Congress Cataloging-in-Publication Data in process



vii

Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xv

Section I — Petroleum in Society . . . . . . . . . . . . . . . . . . . . . . . 1

Chapter 1 The Petroleum Age and Its Implications in Context. . . . . . . . . . . . . .3In the Beginning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Public Interest, Convenience, and Necessity . . . . . . . . . . . . . . . . . . . . . . . . 6The Effect of the Industrial Revolution on Population. . . . . . . . . . . . . . . . 8But, When It Is Over… . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10Changing Times, Changing Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Chapter 2 Evolution of the Technology That Makes Leak Detection Possible. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13Relevant Moments in the History of Computers . . . . . . . . . . . . . . . . . . . 14Developments in Advanced Process Control . . . . . . . . . . . . . . . . . . . . . . 17Looking Back—Looking Forward . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Chapter 3 Regulating an Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21The Public, Their Government, and Industry—All Striving

for a Workable Operating Environment . . . . . . . . . . . . . . . . . . . . . . . . . 21Regulators, Regulations, and Permits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

Successful Pipeline Regulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23Also at Issue: A General Lack of Expertise. . . . . . . . . . . . . . . . . . . . . 26

Identification of Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27Planning for Control and Mitigation of Leaks. . . . . . . . . . . . . . . . . . . . . . 30



viii Detecting Leaks in Pipelines

Chapter 4 Thoughts about Pipeline and Petroleum Issues and Regulation . .31Human Productivity and Energy Augmentation . . . . . . . . . . . . . . . . . . . 31The Economics of Convenience and Health . . . . . . . . . . . . . . . . . . . . . . . . 32Regulatory Concepts and Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34Operator’s Sources of Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34Operator’s Motivations: Step Back to See the Big Picture . . . . . . . . . . . . 35Can the DOT-FAA Model Provide Guidance?. . . . . . . . . . . . . . . . . . . . . . 36

Section II — The Science of Pipeline Leak Detection. . . . . . . 39

Chapter 5 Fluids and Their States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41Matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

Specific Gravity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43Weight Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

States of Matter: Solids, Liquid, Gases, and Beyond . . . . . . . . . . . . . . . . . 44Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44Enthalpy and the Pressure-Enthalpy Diagram . . . . . . . . . . . . . . . . . 47

Characterizing Fluids: Rheology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48Viscosity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49Phases in a Flowing Stream. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51“This Is Important Because…” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

Chapter 6 Fluidic Mixtures and Multiphase Flow. . . . . . . . . . . . . . . . . . . . . . . .57Multiphase Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58Composition, State, Homogeneity, Transitions, Overall Effect . . . . . . . . 59

Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59Flow Visualization and Computation on Multiphase Lines . . . . . . . . . . 61

Slug Flow and Mass Balance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63Meters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64Line Pack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64Conservation of Momentum in Multiphase Systems:

Negative Pressure (Expansion) Waves . . . . . . . . . . . . . . . . . . . . . . . 66Wave Frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

Special Multiphase or Discontinuous Situations . . . . . . . . . . . . . . . . . . . . 67Slack Flow and Slack Regions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67Process Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68Bends and Elbows. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69Small Obstructions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

Chapter 7 Pipe and Its Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .71The Nature of Pipe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71Sizes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72



Contents ix

Pressure Expansion: Strength, Elasticity, Bulk Modulus (Modulus of Elasticity), Wall Thickness, Area of Flow . . . . . . . . . . . . . 73

Pipe Schedule and Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74Resistance to Flow: Surface Conditions, Roughness, Friction. . . . . . . . . 75Moody Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78Expansibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78Temperature Expansion and Contraction in Pipe. . . . . . . . . . . . . . . . . . . 80Unusual Duct Geometries: Hydraulic and Equivalent Diameter—

The Difference between Hydraulic and Equivalent Diameter with Formulas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

Hydraulic Diameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82Leakage Flow from Pipelines: Flow through Leaks . . . . . . . . . . . . . 82

Testing to Find Real Volume Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85Temperature Is Harder to Determine . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

Chapter 8 What Happens at Leaks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .91Leaks: The Fundamental Chain of Events . . . . . . . . . . . . . . . . . . . . . . . . . 91Pipeline State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94Actual Leaks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

Friction in the Flow of Fluids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97Stability of Fluids and Pipes in Actual Operation. . . . . . . . . . . . . . . 97

What Happens at Leaks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98Acoustic Waves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102Acoustic Emission Monitoring. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

In Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

Chapter 9 Leaks with Compressible and Multistate Fluids. . . . . . . . . . . . . . .105Fluid State and Thermodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105Density, Velocity, Mass Flow, and Volume Flow with

Compressible Fluids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106The Basic Leak Evaluation Situation . . . . . . . . . . . . . . . . . . . . . . . . . 107Conditions At and Around a Leak Site. . . . . . . . . . . . . . . . . . . . . . . 107The Leak Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

Issues with Specific Fluid Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111Liquids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111Gas Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113Liquids with Substantial Vapor Pressure . . . . . . . . . . . . . . . . . . . . . 114Non-Newtonian Fluids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

Chapter 10 Searching for Leaks with Expert Systems: Origins and Progeny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .115In the Beginning... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115Inference Engines and Expert Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . 119Inference Algorithms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120




x Detecting Leaks in Pipelines

Performance of Inference Algorithms and Their Implementations: Sensitivity and Specificity. . . . . . . . . . . . . . . . . . . . . 122

Expectations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

Chapter 11 Mathematics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .127Calculus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128Necessary and Sufficient Conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131Set Theory, Boolean Logic, and Venn Diagrams . . . . . . . . . . . . . . . . . . . 133Event Bandwidth, Signal Bandwidth, and Jean-Baptiste

Joseph Fourier. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134Statistics and Understanding Data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

Statistical Tools and Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . 137Statistical Distributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138Statistical Means and Moments, and the Shape of

Distributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140Mean and Standard Deviation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

Probability Density Distributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142Confidence Intervals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143

The t-Statistic and the Student’s t. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143Correlation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144

Calculating and Using Correlation . . . . . . . . . . . . . . . . . . . . . . . . . . 146Cross-Correlation, Convolution, and Autocorrelation . . . . . . . . . . 149

Coherence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149Statistical Basics for Assessing Probable Outcomes . . . . . . . . . . . . . . . . 150Noise. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152

Chapter 12 Detection Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .155The Science of the Mechanics of Fluid Flow. . . . . . . . . . . . . . . . . . . . . . . 156Using Science to Detect and Locate Leaks . . . . . . . . . . . . . . . . . . . . . . . . 159Methodology for Pipeline Leak Detection . . . . . . . . . . . . . . . . . . . . . . . . 160

A Brief Review of Basic Nomenclature and Mathematical Ideas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160

Newton’s Fundamental Laws of Motion . . . . . . . . . . . . . . . . . . . . . 162The Idea of Head in Fluid Mechanics . . . . . . . . . . . . . . . . . . . . . . . . 163Conservation Equations in Fluid Mechanics . . . . . . . . . . . . . . . . . . 165Conservation of Energy: The Bernoulli Equation . . . . . . . . . . . . . . 165Conservation of Mass: The Continuity Equation. . . . . . . . . . . . . . . 166Conservation of Momentum: The Impulse and Momentum

Equation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167Coherence in Conservation Equations . . . . . . . . . . . . . . . . . . . . . . . 168Acoustic Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168Acoustic Signal Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170

Section III — Implementing Leak Detection Technology . . . 175



Contents xi

Chapter 13 Determining Performance Requirements . . . . . . . . . . . . . . . . . . . .177False Alarms, Missed Alarms, and Data Availability. . . . . . . . . . . . . . . 183There Are Often Special Design Factors and Considerations . . . . . . . 183

Chapter 14 Assessing Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .187In Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191

Chapter 15 Monitoring Technologies and Approaches. . . . . . . . . . . . . . . . . . .193Monitoring Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193

Computational Pipeline Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . 193Necessary Conditions for Identifying Leaks . . . . . . . . . . . . . . . . . . 197External Observation Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202Orthogonal Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204

Chapter 16 Supporting the Detection Technology— Flow Rate Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .207Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209Meter Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210Flow Computation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210Custody Transfer Metering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211Installing Metering. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212Selecting Vendors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212Common Meter Types for Measuring Liquid Flow . . . . . . . . . . . . . . . . 213

Positive Displacement Flow Meter . . . . . . . . . . . . . . . . . . . . . . . . . . 215Positive Displacement Meter Summary . . . . . . . . . . . . . . . . . . . . . . 216Full-Flow Turbine Meters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217Turbine Flow Meter Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218Coriolis Flow Meters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219Coriolis Meter Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221Ultrasonic Flow Meters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221Concluding Remarks about Meter Types. . . . . . . . . . . . . . . . . . . . . 223

Pipeline Metering Problems near Pig Launchers or Receivers . . . . . . . 224Typical Pig Launcher Operation and Flow Measurement. . . . . . . 225Typical Pig Receiver Operation and Flow Measurement . . . . . . . 226

Flow Computation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226Pressure Measurements for Line Pack Compensation . . . . . . . . . . . . . . 227Pressure Measurements for Pressure Wave Leak Detection . . . . . . . . . 227Leak Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228

Chapter 17 Instrumentation—Pressure, Temperature, Composition. . . . . . . .229Pressure Transmitters for Pipeline Leak Detection. . . . . . . . . . . . . . . . . 230

Range, Resolution, Bits, Pressure Increments . . . . . . . . . . . . . . . . . 231How Fast Is Fast Enough?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232



xii Detecting Leaks in Pipelines

Chapter 18 Leak Location Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .235Method of Intersecting Hydraulic Grade Lines . . . . . . . . . . . . . . . . . . . . 235Time-Velocity Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236

Data for Leak Location: Bandwidth Requirements . . . . . . . . . . . . . 238Data Collection Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239Data Format for Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240

Common Problems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240Linearization of Location Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . 241

How Much Accuracy Is Required? . . . . . . . . . . . . . . . . . . . . . . . . . . 241Yet Another Method: Acoustic Emission Monitoring. . . . . . . . . . . . . . . 243

Chapter 19 Information Technology Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . .245Monitored System Review and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 245System Security. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246Data Routing, Latency, Skew . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248

Chapter 20 Training and Experience Issues. . . . . . . . . . . . . . . . . . . . . . . . . . . .251

Chapter 21 Long-Term Management and Quality Assurance. . . . . . . . . . . . . .255Management for Stability and Change over Time. . . . . . . . . . . . . . . . . . 255Leak Detection: Periodic Testing, Performance Validation,

and Revalidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257Meter Performance Analysis/Meter Factor Tracking. . . . . . . . . . . . . . . 258

Attention to Changing Conditions, Fluids, and Flow Rates . . . . . 260Periodic Analysis of Leak Detection Instrumentation

Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262Monitored Data Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262

Preventive Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265Application Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265Server Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266

Appendix A. How to Do a Pipeline Leak Detection Project . . . . . . . . . . . . . . .269

Appendix B. Pipeline Profile Calculations (Liquid Lines) . . . . . . . . . . . . . . . .285

Appendix C. Friction and Energy: Ways to Calculate Friction—Darcy, Fanning, and Others . . . . . . . . . . . . . . . . . . . . . . . . . . . . .295

Appendix D. Modeling Concepts—Limitations and Practicality for Leak Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .303

Appendix E. Navier-Stokes Concepts and Fluid Flow: The Bigger Picture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .309

Appendix F. Heat Transfer—Laws, Mechanisms, Effect on Rheological Properties and State. . . . . . . . . . . . . . . . . . . . . . . . .315



Contents xiii

Appendix G. Bayesian Statistics—An Interesting Concept, but Does It Work? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .321

Appendix H. Risk Assessment Using Fault Tree Analysis . . . . . . . . . . . . . . .325

Appendix I. Pipeline Accident Causes, Damage, and Mitigation . . . . . . . . . .355

Bibliography. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .471

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .479

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .497



xv

Preface

Continuous pipeline monitoring became possible as process control technol-ogy developed and moved into the digital age. As measurement capability improved and data communication and processing capability exceeded what could even be imagined a few years earlier, a range of new possibilities emerged. These were not lost on regulators, and the U.S. Department of the Interior Minerals Management Service (MMS) began ramping up the leak detection performance requirements for operating permits. Public attention, motivated by some high-impact spills, raised the expectation that an operator should always be in control of their product.

Over the brief history of automatic leak detection, perhaps 40 years, there has been a great deal of experimentation and conjecture along with the applica-tion of real and meaningful science and technology. This is not unusual in a young field but has interfered with the development of a broad understand-ing of the underlying concepts. Leak detection by monitoring pipeline operat-ing parameters leverages the use of otherwise-necessary instrumentation, even though the objective of the monitoring is somewhat different. Instru-ments provide observability of leak-related process events that can be detected and monitored in various ways, including the Newtonian “conserva-tion equations” as well as other correlated events.

This book places the need for leak detection on pipelines in an industry con-text using both a regulatory and risk-based approach. It develops the applica-ble scientific principles, explores the technology available to implement them,




xvi Detecting Leaks in Pipelines

shows how to estimate and monitor performance, and discusses how to main-tain and assure the quality of a leak detection system. While most explana-tions begin with first principles, there are opportunities for a less science-oriented reader to understand the development of the key concepts. It pro-vides a highly understandable link between actual problems and the applica-tion of standards, practices, and regulations.

Leak-detection project design requires specialized knowledge and skills that are used infrequently with years of time between projects. The information in this book is designed to be a refresher for experienced technicians as well as a guide for new technicians who are assigned this complex task.

In Section I we look at the history of the petroleum age, its impact on society, and the evolution of regulation. It is not the goal of this book to discuss the pluses and minuses of regulation, the process of regulation, or how it can be improved. Suffice it to say, regulation design is a volatile process affected by politics, public opinion, and the oil industry itself. It is necessary in one form or another (legal or self-imposed) to protect the public, to ensure safety in all facets of the production and transportation processes, and to protect the environment.

The chapters in Section II contain discussions of the science and technology required to understand the fluids involved, the pipelines through which they flow, and the physical principles involved with fluid flow. These are the prin-ciples and technology that make leak detection possible and illustrate the lim-its of what can and cannot be done. Expert systems are discussed in some detail. Together, these discussions are useful for understanding the efficacy of a proposed detection methodology, and for understanding the basis of the system designs and technology discussed in Section III. They also provide the basis for assessing the range of usability of any particular leak detection technology.

Section III discusses, in general terms, the application of the available science and technology to leak detection issues. The discussion is focused on the use of the various technologies and not the benefits or issues of any particular product incorporating them. The issues discussed are directed toward the capabilities and limitations of the methodology employed to implement the various detection strategies presented in Section II. Methodology for perfor-



Preface xvii

mance assessment and monitoring, and for preserving consistency over time is discussed.

Appendices are provided to reinforce and expand coverage of some critical elements, including the risk exposure of pipelines. Accident statistics over time are presented and analyzed. There is also deeper treatment of some tech-nical issues.

Leak detection seems like a narrowly focused and highly technical field. Actu-ally, effective leak detection projects involve many disciplines including inter-actions with regulators; pipeline design, construction, and operation; and an understanding of economics, social dynamics, market sensitivity, and politics. When properly integrated and optimized, these factors work together to pro-vide an important component of pipeline safety.

Leak detection involves a substantial range of knowledge. Many readers will be intensely interested in some of the subjects while less interested in others. For any reader, the effort is made to provide the basic knowledge that is essential for productive analysis and discussion along with the context and perspective to understand why various issues are important. This provides an opportunity to gain a core understanding of the essential issues involved, their importance, and how they integrate to enhance pipeline safety.



1

Section IPetroleum in Society

• The petroleum age and its impact

• The evolution of computing and process control technology

• Regulating an industry

• Thoughts about regulation

“The further backward you look, the further forward you can see.”Sir Winston Churchill

Few events in the human time scape have affected people as profoundly as the age of petroleum. It began with industrialization in the mid-eighteenth century and will most likely endure well into the twenty-first. It spawned life-changing opportunities and challenges throughout the world. It changed people’s daily lives, their possibili-ties, and their dreams.

All events of such enormous consequence come with problems. People gained the capability to reshape and manage on the greatest scale in human history. The small and distributed population that sustained agriculture began consolidation and growth into cities where economic specialization and integration enabled hitherto unimaginable productivity. The simple infrastructure of agriculture expanded to include roads, seaways, dense housing, and a far more complex “commerce of ordi-



2 Detecting Leaks in Pipelines – SECTION I

nary life.” All of these changes came with the necessity to establish, maintain, improve, and defend them.

Events on such a scale motivate the observation that people embrace progress but decry change, however closely linked and inevitable. Societies organize to manage change and narrow its undesirable impacts. That has certainly been true with petro-leum. The issues are broad, ranging from the implications of petroleum usage to its safe production, handling, and coexistence with the people it serves. All of this is best done with understanding, promoting the synergistic and insightful community-focused efforts of all involved.



21

3Regulating an

Industry

“… but not in my backyard.”

The Public, Their Government, and Industry—All Striving for a Workable Operating Environment

In forming regulations, the regulatory body (government) must consider the following:

• Petroleum is a desirable product. The public has responded favorablyover the last 150 or so years to its availability in many useful forms atconvenient locations and favorable prices. Whether people should likepetroleum is a perfectly reasonable discussion that reasonable peoplemay have, but it is not a factor in the promulgation and enforcement ofregulations pertaining to the lawful operation of petroleum production,handling, manufacturing, and marketing facilities.

• This same public has not been tolerant of safety or environmentaltransgressions. Neither has it been tolerant of high prices or pooravailability. It has responded negatively to accidents, especially whenthey demonstrate a lack of care, competence, and safety coupled withthe potential for environmental damage. The petroleum industry hascome to recognize that preserving the tolerance and patronage of thesepeople is facilitated by them having a favorable impression of the





industry and its constituent companies. This is “the market” motivating safety.

• The industry that finds, produces, manufactures, delivers, and sellsthese products is legendary. John D. Rockefeller and Armand Hammerare icons of modern American history. The petroleum industry hasfacilitated the accumulation, loss, and reaccumulation of manyfortunes; provided employment for hundreds of thousands of people;and spurred and funded innovation on a nearly unequaled scale. Notonly does the industry include some of the largest companies in theworld, it involves thousands of smaller companies in roles rangingfrom full-fledged operations to esoteric technical and business services.

Government has seen both opportunity and necessity in regulating the petro-leum industry. It has been a wealthy industry with the initiative, entrepre-neurial spirit, and resources to press its issues and agenda. It has learned, albeit slowly in some cases, how to deal productively with both the public and government. This process has provided government with tremendous income from taxes, from both core operations and the multitude of related activities. The use of petroleum in the nineteenth and twentieth centuries has aug-mented human productivity beyond eighteenth-century imagination, further increasing the wealth of our society. Government has participated in these benefits.

On the other hand, this adventure has not been problem free; and the public has demanded a government regulatory role that ensures the public’s needs are met and quality of life and safety are preserved. Government organiza-tions are formed and empowered to perform various regulatory functions to accomplish these ends. While the goals are usually simple, the details involved in reaching them within everyone’s expectations and satisfaction are far more complex. Like most things involving government there is a process, usually an inclusionary one, which attempts to create rules that ensure the goals will be met. If this is successful, then technical issues, often well beyond the technical competence of those directly involved with the rule making, must be serviced.

Regulators, Regulations, and Permits

As petroleum use increased and population density grew, availability became an increasingly difficult issue. Petroleum products had to be available where



Chapter 3 – Regulating an Industry 23

the users were, and the proximity of the means of transporting and processing them inevitably found conflict.

Regulators faced the public’s ire along with operator companies. With a short-age of petroleum-handling consultants willing to work “against the industry” and few in the environmental lobby with real and useful experience in petro-leum handling, meaningful regulation had a very tough time. Permits and design requirements were accepted by industry as a cost of exploration and production. Those costs along with many others formed the cost base of the eventual production. Determining what was legally required for a specific sit-uation seemed harder and more expensive than providing it.

Successful Pipeline Regulation

There are many examples of intent and results. There are examples illustrat-ing economic consequences, as well as unexpected consequences. The impor-tant thing is what can be learned from them. When asked about successful regulations, most knowledgeable people would point to Alaska. Many other states have struggled with regulation issues with varying degrees of success, but California has probably addressed the broadest range of disparate situa-tions and has been proactive about both prevention and mitigation. Looking abroad—where politics, goals, and economics are different from the United States—is also useful for gaining perspective.

Alaska

The most successful of the regulators was the state of Alaska, with the Alaska Department of Environment and Conservation (ADEC) in the lead. While a bit more complex than this, they required regulated pipelines to demonstrate rapid detection of leaks amounting to a rate of 1% of the line’s daily through-put volume. While the specification of size and time was a bit loose, the intent was clear in most situations. The performance required for most applications was easily determined. Fundamentally, the intent was 1% leak detection when technically feasible.

ADEC has explained this requirement thusly:

Our expectation is that a leak detection system (LDS) needs to be capa-ble of continuously detecting 1% of daily throughput “promptly,” if technically feasible. We do not have a regulatory definition of “promptly,” although the longest acceptable timeframe would be 24




hours, and, in reality, it would likely require a much shorter period of time. The 1% threshold only pertains to the volume (not flow rate). The 1% threshold was explicitly called out in 1992 whereas the 24 hours (i.e., minimum flow verification requirement) detection time has been subsequently established during the evolution of the regulatory imple-mentation process. ADEC’s requirement could be considered to be similar to “sensitivity” as defined in API 1130.

ADEC’s approach for the oversight of pipeline leak detection issues has been based on the general concept of API 1130. Although the detection of “…1% daily throughput … and flow verification …within 24 hours…” are intended to be the norm, ADEC’s leak detection regu-lation, 18 AAC 75.055, also provides for exemptions (in recognizing the concept of “minimum detectable leak size” and “minimum attain-able response time” on any given pipeline). To be exempted from the requirements, operators need to submit justifications (i.e., technical feasibility) for ADEC review. Thus, the terms “if technically feasible” and “1% daily throughput within 24 hours” provide methodologies that work well for most pipelines as they provide specificities with a provision to address (pipeline) system-specific configurations.

ADEC monitors the development of leak detection technology and requires use of “best available technology,” which is interpreted as technology that has demonstrated capability consistent with the state of the art. ADEC can change requirements substantially within their authority.

California

Of all the states, California probably confronts the greatest scope of regulatory issues. Several state agencies exercise statutory authority over pipelines:

• In 1981, the California legislature established the Hazardous LiquidPipeline Safety Act with the intent that the Office of the State FireMarshal would exercise exclusive safety regulatory and enforcementauthority over intrastate hazardous liquid pipelines. Agency staffinspects pipeline operators to ensure compliance with federal and statepipeline safety laws and regulations. They also investigate all spills,ruptures, fires, and pipeline incidents to determine the cause and toassess probable violations.



Chapter 3 – Regulating an Industry 25

• California has considerable geothermal production, which confrontsdifferent issues. The Division of Oil, Gas, and Geothermal Resourceswas created to oversee the drilling, operation, maintenance, plugging,and abandonment of oil, natural gas, and geothermal wells, as well asgathering pipelines within facilities.

• Since the early 1980s, offshore operations have been regulated andmonitored by the California Department of Fish and Game’s Office ofOil Spill Prevention and Remediation. This organization operatedsuccessfully through several serious spills and developed a great dealof expertise with remediation technology.

• In 2014, the authority for the Office of Spill Prevention and Response(OSPR) was expanded to address oil spill prevention and response forall waters of the state, whereas previously its authority was limited tothe state’s tidal and marine waters. The OSPR is required to meet a bestachievable protection standard utilizing best available technology.

This patchwork of agencies promulgating and enforcing regulations can cre-ate complexities and the potential for overlapping regulatory schemes that can be difficult to navigate. Since the 1980s, oil and gas exploration in Califor-nia, along with investment in production, has been small despite the diversity of oil and gas resources. Fracking opportunities in California may eventually change this trend.

United States in General

Some states or jurisdictions have attempted to specify how the detection pro-cess was to work—mostly by hiring consultants to draft specifications and regulations. Others have emulated regulation from the petroleum-experi-enced states or federal regulations.

Federal regulatory efforts are presently (2015) divided based on a pipeline’s location, whether it is off shore or on land.

• Offshore pipelines within federal jurisdiction are regulated by theBureau of Ocean Energy Management, Regulation, and Enforcement.Within that organization, pipeline safety is managed by the Bureau ofSafety and Environmental Enforcement (BSEE). In the past, issues nowaddressed by this organization were handled by the U.S. Departmentof the Interior Minerals Management Service.



39

Section IIThe Science of Pipeline

Leak Detection

Pipelines move fluids of one kind or another over significant distances through what-ever environmental conditions lie along the path or present themselves on any given day. Perhaps the most interesting thing about pipelines is the diversity of disciplines involved in thoroughly understanding them.

At a minimum, a capable pipeline engineer must have knowledge of:

• The science of the flow of fluids in pipes and ducts, what we call fluid mechanics

• The fluids involved and their characteristics

• The thermodynamics of the fluids in the pipeline environment

• The state of the fluids and the conditions that can exist in the pipeline

• The effect of various process equipment on fluid flow and state

• The material properties (strength, elasticity, thermal expansivity) of the linepipe

• The impact of topography and changing thermal environments over distance

Detecting leaks requires devising productive ways of looking for them. These meth-ods may be broadly categorized as ways to ascertain that the fluid is where it is sup-posed to be (computational pipeline monitoring or CPM), or ways to detect that it is somewhere it is not supposed to be (external monitoring). This treatment of the sub-ject is largely oriented toward CPM.




40 Detecting Leaks in Pipelines – SECTION II

Most of CPM is based on rudimentary science reinforced by some useful ideas and concepts. The science involves Newton’s laws of motion and the conservation laws: conservation of energy, mass, and momentum. In process monitoring we concern ourselves with the conservation issues but also with how the various parameters involved change over time. Often, change in a set of key variables is sufficient for the detection of the monitored condition. There is a lot of math that can be applied to change detection, much of which is described in Chapter 11, “Mathematics.” Using all these ideas to detect and locate leaks in pipelines is developed in Chapter 10, “Search-ing for Leaks with Expert Systems.”

External monitoring is discussed in Section III. It consists of external sensors, or arrays of sensors, arranged to detect the presence of fluid outside the pipeline. These sensors may be discrete-location devices with sensitivity to one or more of the com-ponents that should be inside the pipeline or they may be a distributed collection (an array) of sensors positioned to detect external presence or movement. A special case is fiber-optic cable routed along the pipeline that responds to the presence of certain kinds of contamination.

This section assumes the reader has a basic understanding at an engineering level of most of the broad issues involved with pipelines, and it attempts to provide both a narrower and deeper focus on those issues directly involved with leak detection. The concepts and references should provide some direction to help guide efforts at fur-ther exploration.

One could argue that a knowledge of the characteristics of pipe materials and failure mechanisms (e.g., corrosion) is important in understanding the initiation and nature of leakage from pipelines. Those topics are mentioned here but are not thoroughly presented. Strength-of-materials and pipe failure modes are big subjects, each of which are topics of interest to be studied in their own right. For some additional per-spective on pipe materials and failure mechanisms, see Appendix I which covers pipeline accident statistics.



71

7Pipe and Its

Characteristics

Pipeline accident statistics disclose that line pipe is involved in most pipeline accidents. The failure of the pipe itself is by far the most significant source of fluid loss and property damage. This is hardly surprising. The source of failure is commonly corrosion and outside force. There is a tremendous amount of data about failure mechanisms. Most pipe failure issues are man-aged by proper siting, design, material selection, and maintenance.

Leak detection system design includes a requirement to calculate and evaluate a number of flow-impacting parameters. Rarely, if ever, does a leak detection requirement dictate the selec-tion of the pipe. Determining the performance to be expected from a leak detection system, though, requires evaluating a few things about the pipe with which a pipeline has been built.

The Nature of Pipe

What we refer to as pipelines are really designed systems consisting of many components. These components may include pumps, compressors, valves, tanks, and pipe. To distinguish between the pipeline system and the compo-nent pipe, the pipe itself is referred to as line pipe while the system is referred to as a pipe line or pipeline.

Pipelines are usually constructed to move large amounts of fluid product from one place to another. They provide a function similar to railroads and tanker trucks. They are a very safe way to move these products.




There are other situations in which hazardous fluids move through pipe, such as process piping in refineries, chemical plants, and distribution terminals. Such piping is typically located on the operator’s property, is within view of the workforce, and is monitored by various external means such as atmo-spheric sensors. The density of the plant facilitates an organized monitoring effort. The impact of liquid spills can be minimized by containment, which often includes polymer barriers and earthen or concrete berms. Atmospheric releases can often be eliminated, controlled, or mitigated by conventional pro-cess safety technology such as pressure-sensitive valves and flares.

The use and interconnection of this in-plant piping changes with normal oper-ation of the process so it is difficult to design monitoring plans that smoothly and automatically transition from one quasi-normal operating condition to another. Various efforts have been made to extend leak detection from trans-mission pipelines to this sort of service piping but without much success. The quantity of sensors involved easily becomes huge and the data gathering sys-tems required to collect the data into usable form are complex, especially when a plant’s infrastructure comes from more than one design concept or era. This treatment of the subject of pipe is focused on transmission pipelines.

Sizes

It is uncommon for there to be much interest in leak detection in pipes smaller than 2 inches. Pipe smaller than that is typically used within process plants for ancillary purposes and its use generally involves frequent adjustments and reconfiguring. For reasons mostly based on equipment costs and operating expenses, most pipelines include pipe between 4 inches and 42 inches in diameter. Larger pipe, up to 48 inches in diameter, is easily available. Pipe sizes change in increments of 2 inches for nominal diameters of 4 inches and above.

Pipe is described by its nominal diameter. For pipe with nominal diameters of 12 inches and below, the nominal diameter is approximately the same as the inside diameter. For larger sizes, nominal diameter is the same as the outside diameter. (Remember, in converting between inside and outside diameter the wall thickness appears twice.)



Chapter 7 – Pipe and Its Characteristics 73

Pressure Expansion: Strength, Elasticity, Bulk Modulus (Mod-ulus of Elasticity), Wall Thickness, Area of Flow

The strength of pipe, defined in this context as its ability to resist expansion under pressure, depends on its wall thickness and the material of which it is made. Wall thickness is listed in pipe data tables. The material is usually selected with consideration of both strength and corrosion resistance. The key strength parameter is the modulus of elasticity, also known as Young’s modu-lus, the elastic modulus, or the tensile modulus. It is the ratio of stress (in pres-sure units) to strain as shown in Equation 7-1.

E = stress / strain = σ / ε (7-1)

where

E = is the elastic modulus

σ = is the applied stress

ε = is the stain, the deformation resulting from the applied stress

Strain is the change in size resulting from the application of force, which in pipelines is usually defined as a pressure acting on an area. If an object with an initial length (Lo) is subjected to tensile stress and increases in length by a small amount (dL), then the strain would be as shown in Equation 7-2.

ε = dL / Lo (7-2)

Since both dL and Lo have length units, ε has no units. To avoid potential con-fusion, many refer to strain in terms of “meters per meter” or “inches per inch.”

This stress/strain relationship is linear and reversible over most of the mate-rial’s useful strength range. An application of stress produces a proportional amount of strain. Removing (or reducing) that stress results in a proportional reduction in strain.

Beyond a certain amount of stress, the relationship between stress and strain is no longer linear and the straight stress/strain curve bows—additional stress produces more than proportional strain. The onset of this nonlinear relationship is called the knee of the curve. Once the stress/strain relationship reaches this knee, the process is no longer reversible—the pipe is permanently deformed. This deformation results in various changes in the strength of the




pipe material and in its response to certain disturbances. Since the pipe may no longer behave as expected, it is considered to have failed; it may not be leaking but it may exhibit brittleness, strength changes, or a change in corro-sion resistance characteristics.

Much is made of this relationship in regulations, notably the Department of Transportation Pipeline Safety Act that identifies a Maximum Allowable Operating Pressure (MAOP) based on pipe specifications. Many leak detec-tion operations monitor a pipeline segment’s pressure history against MAOP limits.

Pipe Schedule and Strength

The short-form way of describing pipe strength is the pipe schedule. The schedule number is found by dividing the service pressure by the allowable stress and multiplying by 1,000. The result is usually coarsely rounded into one of the following categories: 5, 10, 20, 30, 40, STD, 60, XS, 80, 100, 120, 140, 160, and XXS (STD = standard, XS = extra strong, XXS = double extra strong). Low-pressure (low-strength) pipe has a low schedule number, perhaps as low as 5. Really strong pipe might have a schedule number of 160. Schedule 40 is the most common.

If the nominal diameter and schedule are known, it is simple to find a unique table entry for the subject pipe. This table entry will provide considerable data including the inside and outside diameters and the wall thickness. Area of flow is often tabulated as are other dimensions including the area of the metal wall in cross section (useful in strength and expansion calculations) and sur-face area (useful in heat gain or loss calculations).

Most leak detection work involves calculations of one kind or another. The key pipe parameters are typically the diameter of flow, the wall thickness, and the modulus of elasticity. Parameters related to geometry (diameter of flow, thermal expansion coefficient, elasticity) are directly involved in calculating flow in the pipeline. Strength parameters (modulus of elasticity and wall thickness) have a direct effect on the speed at which mechanical waves (e.g., pressure expansion waves or acoustic waves) travel in the pipeline.




175

Section IIIImplementing Leak

Detection Technology

• Determining and matching performance requirements

• Assessing risk

• Monitoring technology – selection and evaluation

• Flow meter performance

• Ancillary instruments

• Leak location

• IT issues

• Training and experience issues

• Long-term care

Great products are spawned by innovation but succeed only with dedication to and concentration on the plethora of issues necessary for them to provide value consis-tently. Whether one seeks to solve a mission-critical safety issue or develop a product to do so, the issues are the same. Execution varies with the deployment methodology but is unforgiving of short cuts, poor science, wishful thinking, or inept management. As the process industries have come to fully and intensely understand over the last few decades, there is absolutely no substitute for “doing it right.” Even the best of ideas cannot provide value if they are inappropriately deployed, inadequately man-aged, or poorly understood. Worse still, the best-sounding idea will not produce



176 Detecting Leaks in Pipelines – SECTION III

value unless it is based in applicable science and executed adequately for the application.

Leak detection is among the most demanding of engineering applications. Most user organizations do very few leak detection projects and thus do not develop or retain a fully capable and adequately funded support staff. In addition, there are few consul-tants with any substantive capability in this area.

Success in these projects requires careful analysis, viable technology, proper imple-mentation, and a commitment to measure and maintain proper operation. Doing this may sound onerous but companies that set out to do so succeed.



235

18Leak Location

Methods

There are two leak location methods in common use with process-monitoring leak location sys-tems. One is rapid, with good accuracy potential. It is referred to as the “time-velocity method.” The other provides accurate locations only after the pipeline comes to a stable steady state with the leak running. This method is often called (somewhat incorrectly) the “method of intersecting hydraulic grade lines.” Both will be covered here but the time-velocity method will be presented in more detail.

Method of Intersecting Hydraulic Grade Lines

The solid line in Figure 18-1 indicates the pressure or head profile of the pipe-line segment along its length. The slope of the line is proportional to the rate of flow. The upstream (left) side indicates the total head at the inlet to the seg-ment, often a point on a pump’s head-capacity curve. The downstream (right) side shows the head at the receiving end. If flow is delivered into a tank, for example, the right-hand head or pressure will be the same as that of the tank.

As shown in Figure 18-2, when there is a leak on the pipeline there is a discon-tinuity in flow rate, hence the pressure versus distance curve will be steeper upstream of the leak and less steep downstream of it. Since a leak makes flow “easier,” flow on the line increases. If the line is being supplied by a centrifu-gal pump, for example, this will load the pump more heavily and the supply head or pressure will move down the pump curve to indicate more flow at less head or pressure. If the line terminates into a tank, the head or pressure



236 Detecting Leaks in Pipelines – SECTION III

will be controlled by the tank and will remain more or less constant. Because the endpoints are known or can be determined, and because the head versus distance slope is proportional to flow on the segment it is possible to draw two profile lines and compute their point of intersection, which will indicate the leak location as demonstrated here. Consider carefully—the line has to run with the leak in place long enough to stabilize with this new profile. Depend-ing on distances, pressures, and fluid characteristics, it is not uncommon for this to take many minutes. Shutting in a leaking line may reduce product loss at the expense of being able to locate the leak accurately. Location accuracy depends heavily on the accuracy of the flow meters. In practice, it is difficult to achieve location accuracy better than within a couple of kilometers, and doing so requires expensive meters and considerable post-leak run-time.

Time-Velocity Method

When a leak occurs, a pressure expansion wave is generated that travels from the leak to all ends of the pipeline segment. As has been established earlier, that pressure wave moves at the speed of sound in the fluid in the pipe and it can be used to locate leaks. It is intuitively understandable that if a leak occurs

Figure 18-1. Hydraulic Profile of a Pipeline Operating Normally



Chapter 18 – Leak Location Methods 237

at the center of the segment, the pressure wave will arrive at each end at the same time (with some small and generally negligible obfuscation due to local flow velocity along the line). Using the arrival time difference of the pressure wave at the ends and an estimate or measured value of the wave velocity, the leak’s location can be computed.

This method is generally referred to as the time-velocity method. It depends on detecting the results of the occurrence of the leak so the monitoring system must be engaged when the leak occurs. There is but one chance to capture the data and determine the location. All of this happens rapidly and provides an accurate location within seconds to minutes of the leak’s occurrence, a period of time that depends on the wave travel time from the leak location to the most distant end.

Figure 18-2. Leaks Force Changes

New grade lines develop as the line comes to a steady state, with the new grade lines incorporating the effects of the leak. The upstream grade line steepens; the downstream line’s slope becomes flatter. The slope of each line becomes proportional to the flow rate in the pipe section with the leak running. The intersection of these two “leak included” grade lines indicates the position of the leak along the pipeline.




497

Index

49 CFR Parts 192 and 195 26

absolute roughness 75absolute viscosity 50absorption 171accuracy curve 216acoustic 158, 168

sensors 203waves 100, 102

acoustic emission 158monitoring 103, 243

active acoustic monitoring 170advection 318age of petroleum 1air age 5alarm margin 275algorithmic methods 127aliasing 101American Gas Association (AGA) 47American Petroleum Institute (API) 49, 277–278American Society for Testing of Materials

(ASTM) 49anti-derivative 128application window 133arrays of sensors 40assisted 318assumed distribution 321atmospheric 288atmospheric sensors 203attenuation 69, 305audible 158autocorrelation 149

Aziz, K. 61

Ballad of John Henry 32bandwidth 135, 238, 263, 280, 304, 306Bayes’ work 321Bayesian analysis 321Bayesian probability 137Bayesian statistics 137, 321–322Beggs, H. 61bends 68Bernoulli equation 165, 277, 289, 295–297, 316Bernoulli principle 156Bernoulli, Daniel 156beta 112bias errors 138biases 211bins 142black body 317Bletchley Park 14, 116Boltzmann’s work 318Boolean algebra 133, 343, 345, 353booster 287boundary conditions 95, 285, 306–308, 315, 330Brill, J. 61British gravitational system 42bubble 64bump test 86

calculus 127, 161causality chain 187cavitation 287celerity 169



498 Detecting Leaks in Pipelines

centipoise 50, 293centistoke 50, 293change point detection 137charge pump 287choked flow 114clamp-on 221coefficient of determination 149coherence 149coherency 149Colebrook-White friction factor 76, 296, 299Colossus 14, 116commercial 229, 301compressibility 48, 289, 307computational pipeline monitoring (CPM) 39, 193,

274, 303, 457computers 14, 299, 304–306condensation 48conduction 86, 316confidence intervals 143, 361, 376conservation laws 40, 200conservation of energy 40, 200conservation of mass 40, 84, 200conservation of momentum 40, 200containment 72, 274continuity equation 84, 166, 277contraction 80convection 86, 316, 318convolution 149coriolis flow meters 219coriolis meter

accuracy 221flow computer 221technical issues 221

correlation 144, 146correlation coefficient 146CPM 39, 274, 280, 303critical flow 68critical point 45–46critical pressure 46critical temperature 45–46cross-correlation 149curl 130custody transfer metering 211

Darcy equation 76, 164, 290, 297, 300data

availability 183, 303collection 239, 279, 322–323latency 249skew 250variability 274

decay 11deductive 187–188, 325Deepwater Horizon spill 19

degrees of freedom 52, 61, 277, 281, 304, 323del operator 128Deming’s work 322dense energy 5dense phase 45density 41, 43, 48, 106, 278, 285–287, 294, 296–298,

301, 304, 307, 309Department of Transportation (DOT) 7, 270, 355,

359, 363, 395, 415, 459dependence 144Dieck, Ronald H. 138differential equations 14, 128, 285, 304, 307differential head 287differential operators 128differentiation 128Digital Equipment Corporation (DEC) 15digital filtering 135direction 162, 301directional derivative 130distillation 319distribution 140, 295, 305, 321–323, 349, 362, 373,

381, 402, 419, 435, 440, 443, 447, 455, 467disturbance 100, 288, 290, 292dither 240divergence 95, 129dynamic viscosity 50

economic specialization 1elastic modulus 73elasticity 48elbows 68, 404electronic instruments 13elevation 289elevation head 97, 297emulsion 51, 60energy density 4energy, conservation of 40, 200, 277, 316energy-augmented productivity 31energy-enhanced productivity 5ENIAC 14Enigma 117Enigma cipher 116enthalpy 47Environmental Protection Agency (EPA) 7equilibrium pressure 46equivalent diameter 82error analysis 264Euler, Leonhard 157Euler’s equations 157exact sampling theory 144expansibility 78expansion 80, 345

wave 66, 288, 292expert systems 52, 116, 119, 303–304



Index 499

external monitoring 39external sensors 40Exxon Valdez 19, 256

false alarm frequency 257false alarm rates 183fault tree analysis 189, 276, 328, 337fault-tolerance 190Federal Aviation Administration (FAA) 36fiber-optic cable 40, 202fingerprint 124finite element/finite difference 306first order lag 305flashing 48, 278flow computation 210flow computer 210flow correlation 263fluid 105, 270, 278, 281, 286–289, 291–292, 295–298,

301, 303, 307, 309, 315, 318, 330mechanics 39, 277, 298, 309, 315, 323

force 160, 277, 286–287, 337, 360–361, 364, 416forced 318Fourier transform 135Fourier, Jean-Baptiste Joseph 102, 134fractional distillation 105frequency domain 134friction 75, 97, 290–291, 295–300full-flow turbine flow meters 217function blocks 305

gas 41, 45gas flow 113gas-like liquids 45Gauss, Johann Carl 128Gibbs phase rule 52government 22, 272, 359, 417, 460Govier, G. 61gradient 95, 129gravitational constant 42gravity 43, 160, 286Guzman-Andrade equation 50

Hagen-Poiseuille flow 77head 96, 163, 286, 296

elevation 97hydrostatic 97internal energy 97velocity 97

head-capacity curve 287–288heat conduction 86heat transfer 316Heisenberg, Werner 283Henry’s law 62high-accuracy metering 211

histogram 140homogeneous 52horsepower-augmented productivity 10, 31hydraulic

diameter 82, 300grade line 292jump 323radius 83

hydraulically inconsistent 293hydrocarbon sensors 202hydrostatic head 97

ideal gas law 106impulse 167, 277impulse piping 239inductive 187, 325industrial 229Industrial Revolution 4, 8industry 22, 271–272, 301, 355, 358, 376inference algorithms 120inference engine 116, 277information, transmission, modulation,

and noise 153infrastructure 5installed accuracy 210, 281, 292Institute of Electrical and Electronic Engineers

(IEEE) 190instruments

electronic 13pneumatic 13, 305

integration 128Intel 16interfaces 69internal energy 47, 97, 295, 297, 316, 319internet 18internet protocol (IP) 19intrinsically safe 13isothermal 88

joules per kilogram 296Joule-Thompson effect 99

K factor 217kinematic viscosity 50kurtosis 140–141

laminar 77, 299Laplacian 130laws of motion 162leak location 228leak path 109leak signatures 123leakage flow 82legal requirements 177




leptokurtic 141line pack 63–64, 412linearization 241liquid 41, 45, 111, 278, 289–290, 358–359, 364, 395,

417, 441location accuracy testing 258low-pass filter 67, 279, 305low-pass filtering 136

magnitude 134, 298, 349maneuver warfare 115manufacturing information 135market 7, 34mass 42, 290, 293, 295–296, 318

balance 63, 201, 270, 292, 412conservation 200

mass energy density 157mass, conservation of 40, 277master meter 212mathematics 127, 323matter 41–43maximum allowable operating pressure

(MAOP) 74mean 140–141, 361–362, 366, 396mean time between failure (MTBF) 189mechanical wave 93, 100median 141meters 64, 281–282, 292, 296, 409, 412

accuracy 210factor 259factor calculation 264factor tracking 258prover 211run 210skid 210

metering skid 208method of intersecting hydraulic grade lines 235microprocessor 15minicomputers 15MIPS 16mission-critical equipment 246mode 141Modisette, Jerry L. 61modulus of elasticity 73momentum 167

conservation 200momentum, conservation of 40, 277monitored data tests 262monitored segment 224monitoring plan 224, 274monitoring zone 224Moody diagram 78multicore processors 16multipath 68

multiphase flow 58multiple observer algorithm 125

natural gas 51natural gas liquids (NGLs) 106Navier-Stokes 92, 95, 108–109, 128, 130, 306, 309necessary condition 132, 198net accuracy 282net flow 226, 281network topology 249networking 15networks 15Newton’s laws of motion 40Newtonian 50, 277noise 66, 152, 293non-Newtonian 50, 114Nyquist-Shannon theorem 67

observability 58, 60, 131, 152, 208, 253, 278, 286, 306offsets 211, 258open channel flow 290operators 6, 271–273, 275–276, 303, 408–409, 412,

418, 420, 435optical sensors 203orifices 83orthogonal monitoring 123orthogonality 125, 151, 204outliers 143, 377

passive acoustic monitoring 170path latency 150per unit volume 296permeable tubing 202permits 23, 281petroleum age 1, 5phase 45, 51, 59, 329, 395, 423

diagram 45speed 169velocity 169

pig launchers 224pigging operations 226pipe schedule 74pipeline 273, 290Pipeline and Hazardous Materials Safety

Administration (PHMSA) 7pipeline engineer 39Pipeline Safety Act 74Pipeline Safety Regulations (Title 49 CFR Parts

190-199) 26Pipeline Safety Regulations Part 195 395Pipeline Safety Statutes (49 U.S.C. Chapters 601

and 603) 26plasma 41, 45, 120platykurtic 141



Index 501

PLC 15, 279pneumatic instruments 13, 305point 296point of production 6point of use 6positive displacement flow meter 215

accuracy 216flow computer 216technical issues 217

prescriptive 19, 272pressure 41, 160, 278, 285–289, 292, 297, 301, 304,

315, 319, 323correlation 264differential 84, 277expansion wave 93, 100, 288lost 291rarefaction wave 93trough 93waves 100, 307

pressure-enthalpy diagram 47pressure-flow correlation 264preventive maintenance 265prior distribution 321–322probability density 142

diagram 140probability of failure 150, 348, 417probability of success 150process control 13, 305, 308, 322process equipment 68productivity 4

energy-augmented 31energy-enhanced 5horsepower-augmented 10, 31

profile 59, 285, 295, 303–304, 306–307, 315, 408programmable logic controllers (PLCs) 15, 279proving 211public 21, 34, 271–272, 358, 380, 382, 463pulser 215pump curves 287

quality of life 32

radiation 86, 316–318random errors 138randomization 139real time 88, 232, 280receivers 224recommended practices 20region 100, 274, 290, 307, 316, 318regularize 83regulation 7, 34, 272–273, 325, 355, 417, 460regulators 7, 34, 272regulatory process 6relative roughness 75

remote terminal units (RTUs) 279resolution 230, 279, 293, 322, 332, 336response time 230, 423results-oriented 19Reynolds number 76, 298–299rheology 48, 60risk assessment 187, 325robustness 151, 190root-sum-square 209, 282, 292roughness 75, 299, 301

absolute 75relative 75

SAGE 14SCADA (supervisory control and data

acquisition) 279scattering 171Schrödinger’s cat 65Schwartz, Mischa 153segment 273sensitivity 122–123, 204, 269, 273–274, 278, 285,

292, 321, 412testing 257

set 133, 278, 282, 322, 343settling time 195shear viscosity 50shutoff head 287SI system 42signal processing 137signatures 152skewness 140slack flow 67, 290slack region 67slip joints 80slug 63

flow 63of solids 64of water 64

society 6, 272solid 41, 44sonic 168, 170

velocity 169, 307sound 169, 307

waves 169specialization 4specific gravity 43–44, 287, 291, 294specific heat 48specificity 122–123, 204, 269speed of sound 169, 292, 307spools 212square wave 134squared coherence 149stakeholders 6

market 7




operators 6regulators 7society 6

Starling, Kenneth 47state function 48state machine 277state of a thermodynamic system 48state of the fluids 39states 277states of matter 44statistical distribution 139statistical moments 140statistics 127, 283, 321–322, 355, 357–358, 424Std 500

reliability data 190steady state 52, 96, 290, 292Stefan-Boltzmann law 318strain 73stream 149, 279, 301, 319streamlines 107Streeter and Wylie 101, 306strength 73, 289stress 73, 309, 404substantial vapor pressure 114suction head 287sufficient condition 132supercritical fluid 45SUS (Saybolt Universal Seconds) 50system curve 291systematic errors 211, 258

temperature 41, 278, 285, 295, 301–302, 304, 307, 315–316, 318–319expansion 80

tensile modulus 73tensor derivative 130the next big thing 10theft detection 184thermal environments 87thermal expansivity 48thermodynamics 39, 61, 315–316Three Mile Island 17time domain 134time skew 150timestamp 240time-steps 232time-velocity method 235–236timing skew 239tracers 202transient 91, 288, 292, 307, 323transient model 201transit time 221, 248transition 77, 299, 305, 422, 445, 462t-statistic 143, 376

turbine flow meteraccuracy 218flow computer 219technical issues 219

turbulent flow 77turbulent region 63Turing, Alan 116

ultrasonic 158, 168–169ultrasonic flow meters 221uniform 91, 287, 322unit volume 296utility 31, 399, 402, 422, 462

vane meter 215vapor pressure 46, 106variability 140, 349variance 140, 142, 282variance-based 209VAX 15vectors 162, 309velocity head 97vena contracta 110Venn diagram 124, 134virtualize 83viscoelastic properties 48viscosity 48–49, 293, 298, 315

absolute 50dynamic 50kinematic 50shear 50

voluntary standards 20

wave 100, 307, 317celerity 169frequency 67speed 169velocity 169

wavelength 67weight 160, 293

density 44weight-and-spring 290wetted 221wetted part 229Woelflin, William 60work 160, 281, 293

Young’s modulus 73




Standards Certification Education & Training · The information presented in this publication is...

Documents

Transcript of Standards Certification Education & Training · The information presented in this publication is...