This paper presents the development and testing of an Electro-Magnetic Acoustic Transducer (EMAT) sensor prototype to detect and quantify longitudinal cracks in small diameter and difficult to inspect or unpiggable gas pipelines. The development of the system was a collaborative and jointly-funded work between Quest Integrated, Gas Technology Institute, Operations Technology Development, and US DOT, Pipeline Hazardous Material Safety Admin (PHMSA). The initial focus for the project was to inspect 8-inch (200 mm) diameter pipes with robotic or tethered towing, with the eventual goal of a free-swimming tool. A bench scale lab prototype has been successfully completed and tested in Phase 1 of the project in 2016. The prototype demonstrated the basic approach of a EMAT tool for crack detection and sizing that could be packaged into a single module, had reasonable flaw depth sensitivity, was bidirectional, and could negotiate a 1.5 D bend. Phase 2 focused on identifying and solving additional implementation issues, developing a more hardened tool for field pull testing, improving flaw sizing, and the necessary internal electronics and processing algorithms. The prototype recently developed in Phase 2 was tested in an extended length of 8-inch diameter steel pipe with pre-set and controlled longitudinal cracks. The results demonstrated the applicability of the integrated prototype in locating and sizing multiple flaws in the axial direction. This paper discusses the EMAT sensor development and results of the laboratory testing program.

With the continuous development of offshore oil and gas resources, calculation software for multiphase flowing pipe network has become an important tool for the design and daily operation of multiphase flowing pipe network. Improved accuracy of hydraulic and thermal calculation is an engineering requirement for economic and efficient production. Therefore, a new program is developed for multiphase pipe network in this paper. This program contains a general data structure to describe the complex connection of a pipe network. The structure is based on the conception of the incidence matrix and the adjacency matrix in graph theory. Two processes, hydraulic equilibrium calculation and thermodynamic equilibrium calculation are successively taken in this program to gain the steady-state for a multiphase pipe network. For the hydraulic equilibrium calculation, applying flow equation to each pipe in the network gains a pipe flow vector. A nonlinear system of equations, which represent flow balance of each node, is obtained by multiplying the incidence matrix and the pipe flow vector. To solve these equations, the Newton-Raphson iterative algorithm is used and afterwards, the hydraulic parameters of the pipe network are obtained. For thermal equilibrium calculation, since all the temperature of source nodes is known, the key step is to find the solution order of other node temperature. The program obtains the order by transforming the adjacency matrix. Deng temperature drop formula is used to calculate the end temperature of each pipe. When a node has more than one inflow, an average temperature based on the heat capacity and mass flow is adopted after gaining each pipe’s outlet temperature. Combining hydraulic and thermal algorithms, a complete set of solution program for steady-state of multiphase pipe network is compiled. In the end, two cases are performed to check the accuracy of the program. In the first case, a pipe network is created by using the data collected from a condensate gas gathering network in the South China Sea. The result indicates that the program has a good agreement with the actual data. In the second case, the program is applied in a single-phase network and gains almost the same result calculated by PipePhase and PipeSim.

Understanding where, when, and how conditions are changing along the extent of an energy pipeline system, which can be vast, is a challenging task. The challenge can be even greater when natural disasters 1 create a condition where access to affected pipelines, qualified personnel, and equipment is limited. To address these challenges, pipeline operators are working directly with experts in satellite technology to develop innovative applications incorporating the use of satellite technology and analytical processes to improve natural disaster monitoring and response. Through recent experiences following Hurricane Harvey in the Gulf Coast region of the United States in August-September 2017 and the wildfires and mudslides in Southern California that occurred in December 2017 to January 2018, space-borne Synthetic Aperture Radar (SAR) satellite data was shown to be a useful tool for wide-area monitoring. Satellite-based SAR imagery has the unique advantage of penetrating through cloud cover and smoke and is capable of providing an early view of the extent of damage in both conditions. Satellite data and continuous improvements to their derived analytical products have resulted in significant benefits for pipeline operators preparing for and responding to the effects of potentially damaging natural processes, including river scour, erosion, avulsion, mudslides, and other threats to pipeline integrity and public safety. SAR change detection algorithms and processes can provide effective results in identifying areas affected by natural disasters that are not readily available by other means. These methods also provide timely information for allocating and directing resources to the most critical locations in support of post-disaster assessment and analysis. SAR satellite data and Amplitude Change Detection (ACD) algorithms provided the basis for confirming where flooding near pipeline infrastructure was most substantial following Hurricane Harvey. In the case of the Southern Californian forest fires and mudslides in Ventura and Santa Barbara counties, recent investigations into ACD and Coherence Change Detection (CCD) algorithms showed promising results, providing a detailed view of damaged areas in near-real time. This paper describes the process of collecting, analyzing, and applying satellite data for assessing the impacts of natural disasters on pipeline infrastructure, and the methods applied, consisting primarily of multiple change detection algorithms, that are used to process the large volume of satellite archive images to extract relevant changes. This paper also describes how these tools and products were practically applied to support decisions by pipeline operators to protect and ensure the integrity and safety of pipelines in the affected areas.

Excavation damage is one of the top causes of incidents in both the transmission and distribution pipeline sectors. Damage caused from insufficient notification of one-call centers or careless digging near gas pipelines can potentially result in property damage, significant injury, and/or loss of life. Pacific Gas and Electric Company (PG&E) and the Gas Technology Institute (GTI) have developed the framework for an Excavation Encroachment Notification (EEN) system to support damage prevention efforts to reduce damage from excavation activity. The research was funded by the California Energy Commission (CEC) and Operations Technology Development (OTD). The system utilizes real-time Geographic Information System (GIS) technology and cellular-connected location and motion sensors placed on excavation equipment. Controlled and field-based testing and training of machine learning algorithms were conducted to aid in characterization of excavation equipment. Additionally, a GIS system populated with pipeline information allowed operators of excavation equipment and utilities to receive an alarm and indication when equipment is adjacent to or excavating in the vicinity of a gas pipeline. More broadly, the utility stakeholder has increased situational awareness of excavation activities within its service territory with access to the real-time activity of excavation equipment through a mobile-compatible dashboard, reducing excavation damage risk and improving safety. Lastly, the project offers historical data archiving for data analysis and trend identification.

Based on process principles and the operational features of crude oil pipelines, this research developed mathematical models to optimize steady-state pipeline operation, to apportion monthly flow into daily or hourly flow rates, and to predict monthly energy consumption. Corresponding algorithms were also developed. Because these models and corresponding algorithms are process-based, they are suitable for predicting monthly energy consumption of existing isothermal and hot crude oil pipelines. The predicted monthly energy consumption of crude oil pipelines depends on which flow distribution method is used, which pumping operation scheme is used and which heating operation scheme is used, with different flow distributions, different pumping and heating operation scheme yielding a range of monthly energy consumption predictions for a given transportation volume. The minimum monthly energy consumption can be determined from these predictions, and the interval of the predictions can indicate the extent to which the flow rate fluctuation affects pipeline energy consumption. Both of these findings can be used by pipeline operators to reduce the amount of energy needed to operate crude oil pipelines.

This paper presents a tool for surface loading stress analysis that was developed in-house by TransCanada (TCPL). This tool utilizes fundamentals of the surface loading assessment method developed by Kiefner & Associates Inc. (KAI) for Canadian Energy Pipeline Association (CEPA), but incorporated many advanced functionalities to improve the accuracy, efficiency and transparency of the analysis. The new functions of the tool include the batch analysis, multiple angle analysis, generic/site-specific loading analysis, graphical display of stress distributions for refined assessment, user-defined impact factor and automated reporting for documentation of surface loading calculations. This tool also incorporated the improved numerical algorithm for longitudinal global bending stress considering the actual live load pressure distribution over a certain length of pipeline. The accuracy of the developed tool was validated by comparing it to the KAI tool. The improved algorithm for longitudinal global bending stress calculation reduces the conservatism of the longitudinal global bending stress compared to the original simplified method but does not sacrifice safety, which has been demonstrated by comparison with the experimental results. The new functionalities improved the business efficiency and maintains safety and regulatory compliance.

In the oil industry liquid pipelines are very important for the transport of liquids, particularly in long offshore pipelines. The operation of these oil pipelines is susceptible to the occurrence of leaks in the system. Localizing a leak in a very long oil pipeline is an important piece of information that needs to be obtained before mitigating actions can be taken. These pipelines are usually subject to the temperature gradients that exist in the bottom of the ocean, and the resulting heat transfer process may lead to wax formation and deposition. The single-phase flow that occurs in this type of offshore pipeline that presents one leak point and suffers the effects of an external temperature gradient is numerically simulated in this paper. We consider a one-dimensional mathematical model that includes conservation equations of mass, momentum and energy, and its associated numerical method to calculate the transient liquid flow inside the pipeline. We are particularly interested in testing a leak localization model based upon the intersection of the hydraulic grade lines emanating from the pipeline ends under the influence of a non-zero temperature distribution. This paper proposes to compare the results for a non-isothermal flow with the corresponding isothermal flow to study the influence of the temperature distribution upon the leak localization strategy. The flow that develops along the entire pipeline, upstream and downstream of the leak, strongly affects the pressure gradient and has a significant influence on the location of the leak. Our numerical simulations show results that allow the model sensitivity to be studied by changing the leak magnitude, for a given leak position. From this analysis, we may observe how these parameters affect the pressure gradients along the pipeline that develop upstream and downstream of the leak and the model’s ability to predict the leak location.

Accurate defect sizing is crucial for maintaining effective pipeline safety and operation. Under growing pressure from local, national and world organizations, pipeline operators demand improved magnetic flux leakage (MFL) metal-loss sizing accuracy and classification from in-line inspection (ILI) tools. The axial MFL field response in pipeline steel near a metal-loss defect is a very complex phenomenon. Although critical for proper sizing model development, the effects of tool speed due to product flow is very difficult to model during finite element analysis (FEA) and therefore is often overlooked. However, understanding the dynamic MFL response is crucial for proper ILI tool design and the development of accurate defect sizing algorithms. T.D. Williamson (TDW) utilizes dynamic computer simulation modeling, paired with laboratory testing, to develop the complex parametric relationships between metal loss geometry, pipeline material and ILI tool speed. The blend of simulation and physical test results allow for TDW to iterate more quickly across multiple physics variables with simulation models, while maintaining a firm footing in reality with physical test validation. Accurately simulating magnetic field responses of metal loss under dynamic conditions produces the data necessary to identify optimal magnetizer design, including optimizing sensor spacing and placement for metal-loss defect sizing and characterization. This paper will provide an overview of advances in the use of computer simulation modeling for predicting dynamic flux leakage field response. Besides increasing accuracy, results from this work will extend specifications beyond optimal speed ranges and provide the basis for general corrosion profilometry predictions from decomposition of the full MFL signal.

Pipeline integrity management commonly leverages nondestructive inspection of pipeline defects via inline inspection (ILI) and assessment of the resultant data. Key parameters for dent analysis include the feature geometry measured by caliper tools and the presence/severity of any interacting features (such as cracks or areas of corrosion) which can be measured with a variety of technologies (such as magnetic flux leakage or ultrasonic tools). Dent profile measurements can be especially susceptible to noise due to the measurement techniques employed, signal quality, and overall tool performance. Analytical methods for strain assessment of dents can employ curve/surface fitting techniques to estimate the curvature and calculate the strain of the dent based on the fitted curve/surface. Noise in the measured profile can result in local areas of high perceived strain, which could lead to misinterpretation of a dent’s true severity, especially when using automated or purely analytical assessment methods. A deterministic strain-based approach for evaluating the severity of dented pipelines has been presented previously which leverages multi-dimensional B-spline functions to more accurately apply the non-mandatory ASME B31.8 equations for dent assessment. The approach presented previously requires relatively smooth dent profile information to minimize the effects of signal noise. While low pass filters can effectively eliminate noise in the signal, they may also lead to loss of accuracy (e.g. excessive smoothing can reduce the depth and sharpness of a measured dent’s profile). This paper discusses how low pass filters can be optimally used to smooth the raw ILI signals to allow for analytical representation of the dent shape without underestimating its severity. The conclusion of this venture is a detailed workflow for the analytical assessment of dented pipelines for the rapid characterization of the severity of deformation in pipelines with limited computational demand. This type of assessment allows for initial ranking and assessment of large and complex pipeline systems to select features requiring more detailed assessment or mitigation.

Natural gas pipeline network system is a critical infrastructure connecting gas resource and market, which is composed with the transmission pipeline system, underground gas storage (UGS) and liquefied natural gas (LNG) terminal demand. A methodology to assess the gas supply capacity and gas supply reliability of a natural gas pipeline network system is developed in this paper. Due to random failure and maintenance action of the components in the pipeline network system, the system can be in a number of operating states. The methodology is able to simulate the state transition process and the duration of each operating state based on a Monte Carlo approach. After the system transits to other states, the actual flow rate will change accordingly. The hydraulic analysis, which includes thermal-hydraulic simulation and maximum flow algorithm, is applied to analyze the change law of the actual flow rate. By combining the hydraulic analysis into the simulation of the state transition process, gas supply capacity of the pipeline network system is quantified. Furthermore, considering the uncertainty of market demand, the load duration curve (LDC) method is employed to predict the amount of demand for each consumer node. The gas supply reliability is then calculated by comparing the gas supply capacity with market demand. Finally, a detailed procedure for gas supply capacity and gas supply reliability assessment of a natural gas pipeline network system is presented, and its feasibility is confirmed with a case study. In the case study, the impact of market demand uncertainty on gas supply reliability is investigated in detail.

Risk assessment is an effective and commonly practiced process in industry, including oil and gas sector, as a basis for designing new pipeline terminals and stations and managing integrity of existing facilities. A holistic risk assessment method, which could be qualitative or quantitative, includes both the likelihood and consequence assessments of an undesired event. Prior to 2015, Enbridge Pipelines employed a qualitative risk assessment algorithm to assess the likelihood and consequence of a failure of liquids pipeline facilities. Over the past decade Enbridge has identified a number of shortcomings with the qualitative approach, necessitating the development and use of Quantitative Risk Assessment (QRA) to support consistency and defensibility in risk-informed decision making. A QRA requires rigorous quantitative algorithms to measure public and environmental safety, and potential business consequences of an undesired event at a facility. While significant literature has been produced, and considerable effort has been expended to quantify the potential impacts of a flammable product release on public safety, very limited work has been done on the quantitative measurement of environment related impacts. In particular, limited research has been successful in aggregating environmental consequences, public safety and business consequences to estimate the total consequence of a liquid hydrocarbon release within a pipeline facility. The consequence assessment of an unwanted event conducted through QRA can be combined with the associated likelihood to provide a quantitative measure of risk. This risk level may be used to support organizations in making risk informed decisions and in analyzing and treating facility risks, specifically in the: • Identification of top risk facilities and high consequence functional areas; • Identification of assets posing the most risk and worst case consequences; • Understanding of system reliability risk and opportunities to optimize facility operation; • Prioritization of facility maintenance projects in the capital and operating budget processes; • Supporting regulatory requirements and expectations; • Presentation of risk down to the equipment or component level; and • Understanding of residual risk and achieved risk reduction. This paper describes the development of a consequence model that monetizes the quantitative measure of public and environment safety, and potential business losses for a liquid product release at pipeline facilities. The proposed model characterizes the severity of impact of released product, expressed in dollars per event, as a function of system volume, proximity and category of receptors, asset location, and available controls.

The Pipeline and Hazardous Materials Safety Administration (PHMSA) Notice of Proposed Rulemaking (NPRM), with Docket No. PHMSA-2011-0023, substantially revises 49 CFR Part 191 and 192. Notable among these changes was the addition of §192.607, verification of pipeline material. This section calls for the verification of material properties of pipe and fittings located in either high consequence areas, class 3, or class 4 locations where traceable, verifiable, and complete records do not exist. Material properties include grade (yield strength, YS, and ultimate tensile strength, UTS) and chemical composition. The proposed regulations include an independent third-party validation for non-destructive testing (NDT) methods to determine material strength and require an accuracy of within ±10% of an actual strength value. Among the NDT technologies currently available to pipeline operators to estimate material strength is instrumented indentation testing (IIT). IIT is based on the principal that there exists a relationship between the indentation response of a material and its stress-strain curve. The indentation response is measured during the IIT process whereby an indenter is sequentially forced into the material during testing. The link between the indentation response and the material stress-strain curve is established often through the use of iterative Finite Element Analysis (FEA). The IIT vendor’s proprietary software performs this calculation, converting force-displacement measurements into an estimate of YS and UTS. In this study we extracted force-displacement data from IIT performed using FEA on an idealized steel. This data was then coupled with literature algorithms developed at Seoul National University (Kwon et al.). Parametric sensitivity analysis was then performed on estimated YS with respect to the algorithm parameters. Preliminary results indicate that while variations in the indenter constant, ω, used to estimate surface deformation do not significantly alter the predicted UTS or YS, the sensitivity to deviations in the empirical constant, Ψ, relating normal load to representative stress was more pronounced due to an effect on the calculated power-law constant, K . PHMSA’s NPRM accuracy requirements for NDT to establish yield and tensile strength should be driven by a rigorous understanding of material inhomogeneities, uncertainties in actual tensile strength determination, experimental uncertainty, and modeling uncertainties. The analysis performed in this paper provides part of this rigorous framework to establish realistic accuracy requirements for NDT that must drive federal rulemaking. In addition, this research highlights the need for pipeline operators to establish controls on the algorithms adopted by commercial NDT vendors.

Equally important to choosing the correct pipeline risk modeling approach is ensuring the identified risks are properly managed through a fully integrated risk management program. Implementing the most sophisticated risk modeling algorithm on the market will not help a company manage their risks if the model and associated program are not usable to the company. The role of a framework is to provide specific guidance to support other company programs by referencing applicable management system elements and outlining additional elements specific to the program itself. A successful risk management framework sets the groundwork for effective risk management by ensuring the risk management program is integrated into the company in a way in which all relevant stakeholders benefit from its use. This paper outlines common pipeline risk management framework elements and their links to the Pipeline Safety Management System elements set forth in API RP 1173. As the development of such a framework is often an iterative process, prioritized aspects are identified for companies undertaking the development of such a system.

Coalbed methane (CBM) has attracted much attention as a kind of new energy sources, with increasing demand in energy consumption. As a lot of wells are being developed in CBM, pipeline networks of CBM fields are characterized by their complex topological structures and high investment costs. Therefore, the optimization of the gathering pipeline network topology structure will be very useful in lowering the production costs. Although previous research has been done on topological structure optimization, most of them are conducted on two-dimension geographical conditions, whose performances are not very satisfing. In this paper, first of all, the result of two-dimensional topological structure optimization is given based on four sorts of connecting structures using Genetic Algorithm (GA). Then, considering geographic factors, a three-dimensional topological structure is obtained through ant colony algorithm (ACO).Taking a CBM field with 38 wells as an example, the paper has optimized the topological structure. The result shows that by using this method the investment cost is reduced by 16.3% compared with that of the original structure. The study provides a guideline for designing pipeline network structure in CBM fields and is also applicable to the analogous problem for similar mineral source.

To provide a more insightful and accurate feature description from Crack In-line Inspection (ILI) reporting as per the Fitness For Service analysis in API 1176, individual crack dimensions must be established to a given accuracy. PII Pipeline Solutions established an absolute depth sizing specification conforming to the dig verification processes of API 1163. This change represented a significant shift from a traditional reporting format for depth sizing in “bands” of 1–2 mm, 2–3 mm and > 3 mm depths within crack ILI inspection reporting. When assessing features with characteristics stated in a sizing band, the pipeline integrity assessment approach required a conservative assumptions that all of the features in that band must be treated as if they are in the deepest band value. The implication then meant that the specification created only 3 sizes of crack depths 1–2 mm, 2–3 mm, > 3 mm (± 0.5mm tolerance at 90% certainty). In practical terms a large quantity of features in the significant band of 2–3 mm must be treated as potential dig candidates with a depth of at least 3 mm, making length characteristics as the only severity ranking basis for any priority dig selection. Previous attempts at establishing absolute depth sizing for crack inspection required a series of calibration digs. The large sample size over multiple inspection runs and pipeline sections allowed for a statistical specification algorithm is developed as part of the analysis process, therefore no additional reporting time, or excavation cost was involved. The new absolute sizing algorithm has provided operators with a means of prioritizing digs, based upon individual feature length and depths. Replacing the traditional depth bands with individual feature specific peak depths and thereby providing a major step forward in achieving a cost effective process of prioritizing crack mitigation in pipelines. Following the dig verification process in API 1163, significant populations of infield NDE results were utilized on a variety of pipeline sections of different diameters. Predicted absolute depth estimation accuracy was determined for specific feature types and thereby created a depth tolerance, with statistical certainty levels established that match those available and recognized with metal loss ILI. This paper describes the process and the means by which an absolute depth crack ILI specification was established using characteristics from a significant set of real features. It also describes benefits realized within pipeline integrity engineering of moving to such a new reporting protocol.

In response to the National Transportation Safety Board (NTSB) Recommendation P-09-1, the Department of Transportation (DOT) Pipeline and Hazardous Material Safety Administration (PHMSA) initiated a comprehensive study to identify actions that could be implemented by pipeline operators to significantly reduce longitudinal seam failures in electric resistance weld (ERW) pipe. The purpose of this paper is to provide a review of Phase II of the project with focus on the study objectives and results. Phase II of the project consisted of five tasks with the following objectives relevant to the ERW and flash weld (FW) process: 1) develop and optimize viable hydrotest protocols for ERW/FW seam defects 2) improve the sensors, interpretive algorithms, and tool platforms in regard to In-Line-Inspection (ILI) and In-the-Ditch-Methods (ITDM) to better ensure structural integrity by developing and optimizing concepts to address problems in detecting and sizing, 3) bridge gaps in defect characterization in regard to types, sizes, geometries, and idealizations, to increase pipeline safety through improvements needed to implement both ILI and hydrotesting, 4) validate existing failure prediction models and, where gaps preclude validation, refine or develop these models needed to assess and quantify defect severity for cold welds, hook cracks, and selective seam weld corrosion (SSWC) (the primary ERW/FW seam threats) for failure subject to loadings that develop both during hydrotests and in service, and 5) develop software to support integrity management of seam welds with enough flexibility to benefit from the experience gained during this project. The reports generated during the course of the project are publically available and are located on following PHMSA website: http://primis.phmsa.dot.gov/matrix/PrjHome.rdm?prj=390.

Pipeline in-line inspections (ILI) are one of the primary methods used to assess the integrity of operating oil and gas pipelines. Conventional ILI technology is based on ultrasonic testing (UT) or magnetic flux leakage (MFL) sensors. Although these technologies are suitable for most pipeline inspections, there remains an opportunity to expand ILI technology and application. ExxonMobil and Innospection Ltd. are working to develop a new ILI sensor technology based on a combination of Magnetic Eddy Current (MEC) and multi-differential eddy current. This new technology provides the potential to detect small volumetric features, inspect heavy wall gas pipelines, and inspect pipelines with corrosion resistant alloy (CRA) or non-metallic liners. Initial feasibility trials were conducted with a prototype ILI MEC tool. Tests were conducted on an 8.625” (219 mm) X65 carbon steel pipe lined with 0.118” (3 mm) of Inconel 825 pipe. Four types of defects were machined into the pipe to represent natural defects anticipated in service: • Metal loss features from 3 to 24 mm in diameter on the external surface of the carbon steel base pipe • Erosion on the internal layer of the CRA liner • Internal girth weld crack-like defects • Metal loss defects at the interface of the CRA and carbon steel Over 80 pull tests were conducted to determine the detection capabilities and speed sensitivities of the tool. Defects were detected by the sensors including the very small (<10 mm) pinhole-type features. Signals were analyzed by a preliminary sizing algorithm to demonstrate proof of concept. Detection performance was not affected at speeds up to 0.75 m/s. Since detection capabilities exceeded expectations, future development will continue based on the current prototype.

Running fracture control is a very important technology for gas transmission pipelines with large diameter and high pressure. The Battelle two-curve (BTC) model developed in the early 1970s has been widely used in pipeline industry to determine arrest toughness in terms of the Charpy energy. Because of its semi-empirical nature and calibration with test data only for grades up to X65, the BTC does not work for higher grades. Simple corrections were thus proposed to extend the BTC model to higher grades, but limited to those grades considered. Moreover, the BTC model only predicts the minimum arrest toughness, but not arrest distance. To fill the technical gaps, this paper proposes a modified two-curve (MTC) model and a fracture arrest distance model in reference to the Charpy energy. The MTC model coupling with an arrest distance algorithm can predict fracture arrest toughness and arrest distance in one simulation of numerical integration for a single pipe or a set of multiple pipes with given toughness. Two sets of full-scale burst test data for X70 and X80 are used to validate the proposed model, and the results show good agreements between the predictions and full-scale test data of arrest toughness and arrest distance as well. The MTC model is then applied to optimize a design of pipe segment arrangements for a mockup full-scale burst test on a high-strength pipeline steel. The MTC simulation results confirm the experimental observation that different pipe arrangements determine different arrest toughness and arrest distance for the same grade pipes.

Gas pipeline internal surface typically undergoes degradation for a variety of reasons such as fouling of the pipe inner surface, erosion, corrosion and deposits of objectionable materials that occasionally enter the gas stream at receipt points. Accurate monitoring of the pipe internal surface condition can hugely benefit the planning of cleaning activities. Theoretically the pipe wall roughness for a given pipe segment can be extracted based on measured flow data and other system parameters. The challenge lies in the fact that measured data all contain varying degrees of uncertainty, and the system becomes more complex to analyze when it contains different segments connected in series or parallel like many typical gas gathering and lateral networks. This paper demonstrates the application of the Error-in-Variable Model (EVM) using the Markov Chain Monte Carlo (MCMC) solution method in analyzing a complex pipeline network on the TransCanada NGTL System. EVM, a well-established Bayesian parameter estimation technique, accounts for uncertainties in the measured variables, such as flow and pressure data, when determining the most probable estimates of unknown parameters such as pipe internal wall surface roughness. In this work, the EVM problem is solved using the MCMC Metropolis-Hastings algorithm. The MCMC approach is demonstrated to be robust, easy to implement and capable of handling large quantities of data. It has the potential to analyze complex networks and monitor the pipe wall surface condition on-line with SCADA data. Using this method, the internal wall surface roughness for the segments of interest in this network were extracted from measured data collected before and after the pigging operation. Results demonstrate the model’s capability in estimating the degradation of the pipe wall internal surface and the effectiveness of pigging. Details on implementation and challenges in applying such methodology to analyze complex gas networks are discussed.

In this paper, a new leak detection system based on a pattern recognition algorithm in a shut-in condition of a pipeline is presented. For a fully shut-in pipeline, the governing fluid dynamic equations are simplified to the thermodynamic state postulates. Hence, the shut-in section can be treated as a closed thermodynamic system with no mass flow in or out of the system boundaries. The system always contains the same amount of matter, but heat and work can be exchanged across the boundaries of the system. The pattern recognition algorithm presented in this paper automatically monitors the pressure drop patterns and generates an alarm when the pattern of pressure gradients matches the leak signatures. The algorithm takes into account the effect of thermal cooling and other operational complexities to enhance the reliability performance of the scheme. Results of the performance of the shut-in leak detection system are presented and discussed in this paper. Both simulated and historical leak scenarios during shut-in state are used to investigate the performance of the shut-in leak detection scheme.