PR719-243812-Z01 Addendum to API 1163 Performance Validation Guidelines Crack and Dent

The emphasis on pipeline safety through reliability analysis strategies in the pipeline industry has pushed the process of uncertainty quantification towards advanced statistical techniques. Probabilistic modeling of pipeline threats and inline inspection (ILI) validation/tool performance assessment are areas that have all benefited from advances in uncertainty quantification. More recently, the statistical calibration of measurement uncertainty has developed as a promising approach for reducing the effect of measurement errors that exist in both ILI and field non-destructive examination (NDE) measurement readings. Using traditional methods for estimating measurement error quantities in ILI data, calibration provides a robust and simple approach towards incorporating the error information towards obtaining an improved estimate of the true value. Since calibration is performed at the individual feature-level, its combination with probabilistic modeling using statistical distributions at the parameter-level can potentially provide an opportunity to understand its effect on the probabilistic analysis (leak or burst limit state). In this work, the impact of applying statistical calibration towards identifying/utilizing ILI and NDE measurement uncertainty in conjunction with probabilistic modelling is explored through an investigated problem.

A more result-driven, holistic view on the performance of an In-line inspection (ILI) system is desired by pipeline operators, namely an integrated view on the Probabilities of Detection, Identification and Sizing, i.e. POD, POI and POS respectively. It is self-evident to combine the product of the individual probabilities to a single value POX as proposed also by API 1163. The changes in material properties of existing pipelines related to the repurposing to Hydrogen service yields smaller acceptable anomaly dimension. Consequently the dependency of the POX vs. anomaly dimensions need to be considered as function. The current practice to document the performance of an ILI system is typically based only on the specific dimensions of minimum detectable crack-like anomaly. In the case of crack-like anomalies, the contribution of the dimensions (e.g. length, depth, orientation) and interactions of crack-like anomalies can be used to turn POX into a meaningful function/dependency depending on the anomaly dimensions instead of a single value. To establish a POX to anomaly dimension relation successfully for a specific ILI crack detection service, the relevance of field verifications needs to be taken into account as well. Previous publications as well as physics foundations of NDE demonstrate that an increase in anomaly dimensions typically yields a higher sensitivity of the underlying inspection system. This can be immediately linked to the POX function as product of POD, POI and POS. This paper provides access to an in-depth assessment of POX based on the results of recent field verification data. The paper focuses on the interaction between POX and different anomaly dimensions. This is supported by the most commonly applied standard for in-line inspection, API 1163, which implies the validation of inspection performance with use of field verifications based on the reported results. In addition, a performance specification can be derived exclusively based on a field verification campaign, in the absence of other comparable performance information. As an example, as part of a conversion of natural gas transmission pipelines to hydrogen service the sensitivity of inspection systems is frequently discussed. A potentially increased crack growth rate da/dN and a reduction in pipe steel toughness seem to require a reduction of minimum acceptable anomaly dimensions, in the absence of better statistical evidence of the actual ILI performance. The proposed POX assessment will be a credible means to establish optimal re-inspection intervals for existing in-line crack inspection technologies.

This study collected and analyzed field data to validate the probabilistic model developed previously for predicting internal corrosion threats resulting from condensed water from nominally dry natural gas. Model-predicted wall loss was compared to actual measurement to validate the model. One past failure case from the National Transportation Safety Board(1) due to internal corrosion and in-line inspection (ILI) data from five pipeline operators were collected and used to validate the Model. Two of the data sets indicate wall loss that is either evenly distributed at all locations around the circumference of the pipe or predominantly at the top of the pipe, which is in agreement with the negligible wall loss predicted by the Model. Another two sets of data show significant wall loss occurring at the bottom of the pipe, which is in agreement with non-negligible risk of internal corrosion predicted by the Model. The 5th data set confirmed the analysis of ILI data uncertainty and corroborated the validation results. For the failure case, the prediction is in good agreement with the post-failure examination. Although uncertainties exist in both Model predictions and ILI measurements, the wall loss predictions from the Model are in good agreement with trends seen in the ILI measurements.

This paper establishes a dynamic Bayesian network to model the growth of corrosion defects on energy pipelines. The integrated model characterizes the growth of defect depth by a homogeneous gamma process and considers the biases and random errors associated with the in-line inspection (ILI) tools. The distributions of the mean value and coefficient of variation of the annual growth of defect depth are learned from multiple ILI data using the parameter learning technique of Bayesian networks. With the same technique, the distributions of the biases and standard deviation of random errors associated with ILI tools are learned from ILI data and their corresponding field measurements. An example with real corrosion management data is used to illustrate the process of developing the model structure, learning model parameters and predicting the corrosion growth and time-dependent failure probability. The results indicate that the model can in general predict the growth of corrosion defects with reasonable accuracy and the ILI-reported and field-measured depth can be used to update the time-dependent failure probability in a near-real-time manner. In comparison with existing growth models, the graphical feature of Bayesian networks makes it more intuitive and transparent to users. The employment of parameter learning provides a semi-automated and convenient approach to elicit the probabilistic information from ILI and field measurement data. The above advantages will facilitate the application of the model in the practice of corrosion management in pipeline industry.

For many years, pipeline safety regulations in the US have defined prescriptive minimum requirements for integrity management combined with a clear expectation that operators should do more than the minimum where appropriate. The regulations have also provided operators with the flexibility to take a performance based integrity management approach leveraging as much information available to manage threats effectively. One the threats that must be managed is Selective Seam Weld Corrosion (SSWC). SSWC is an environmentally assisted mechanism in which there is increased degree of metal loss in the longitudinal weld in comparison to the surrounding pipe body. An appropriate definition is linear corrosion that is deeper in the longitudinal weld zone than the surrounding pipe body. In some cases, the surrounding pipe body may have limited or no corrosion present, and in other cases the pipe body corrosion may have occurred but at a slower rate than the local corrosion in the longitudinal weld zone. Conventional responses to potential or identified threats focus on in-situ investigations, often resulting in expensive and un-planned repairs for features reported by In-line Inspection (ILI) that when assessed properly demonstrate a remnant life well into the next inspection interval. When ILI identifies metal loss indications co-located with the longitudinal seam weld, the current prescribed response is often a blanket call for remediation. Such a response may not be appropriate if an ILI system is deployed to discriminate feature types and integrity assessment is exercised leveraging a sound understanding of the pipe’s material properties. This paper describes an approach that can be taken to manage the threat of SSWC. The foundation of the approach is deployment of an appropriate ILI system incorporating an effective ILI technology, an optimized evaluation process considering the specific threat morphology, material testing and a structured dig program. The evaluation process uses the ILI data and data from the field in combination material properties data and a susceptibility analysis to classify anomalies as “Likely”, “Possible” and “Unlikely” SSWC. This is aligned with the guidance in API RP 1176 “Assessment and Management of Cracking in Pipelines” for defining an appropriate response to ILI calls. Approaching the management of SSWC in this way allows operators to define a structured response for excavation activities to verify the process and remediate features as required. By using likelihood classification the risk to pipeline integrity can be reduced by acting on the most likely SSWC features as a priority, whilst collecting the data needed to make informed decisions on where to focus resources and efforts on what is a very complicated and difficult to manage threat. The output form this work, including a future plan for managing the remaining metal loss features, can be documented in a procedure and incorporated into an existing Integrity Management Plan.

In-line inspection (ILI) is often used to estimate failure pressure from anomalies as part of the integrity management of pipelines. The goal of this project is to improve failure pressure estimates of stress corrosion cracking (SCC) colonies by advancing field measurements and comparing ILI tool data. Electromagnetic acoustic transducer (EMAT) ILI technologies have made significant strides in providing crack depth measurements for SCC colonies, and emerging technologies such as ultrasonic testing (UT) imaging are providing more accurate in-ditch validation data. For new ILI technologies, feeding back accurate field measurements to vendors is an essential part of the process for improving ILI estimates of failure pressure. This project was designed to improve failure pressure estimates with three tasks spanning three years. Year 1 (Task 1) included the development of ultrasonic imaging processes and procedures for gathering accurate in-ditch measurements of depth, length and separation for SCC. Year 2 (Task 2) comprised field trials of the ultrasonic imaging and collection and analysis of EMAT data. Validation was performed and improvements to signal analysis for crack sizing and interaction criteria were made. Year 3 (Task 3) work included validation by means of burst tests to evaluate the performance of the model that predicts failure pressure based on ultrasonic imaging measurements. The main focus of Year 1 was comparing the dimensions of SCC colonies measured by X-ray computed tomography (XCT) to IWEX ultrasound imaging for thirty 4-inch (10 cm) wide plates containing SCC colonies. For truth data, two of these samples were destructively evaluated by breaking open cracks frozen in liquid nitrogen and sectioning through multiple parallel cracks with subsequent polishing of samples. Comparisons of the XCT and IWEX to the freeze break sample results are presented in this report across profiles of cracks in an SCC colony. The details of the work are documented in the first report for this project and summarized in this report. Year 2 focused on collecting data in the field with the IWEX system based on EMAT inspection results. A process was developed to facilitate comparisons between IWEX and EMAT data to allow an in-ditch verification of the axial cracking. Due to the technology differences between ILI and IWEX, the ILI recorded data/signals show less detail and do not allow for discrimination of the individual flaws the same way that IWEX does. Even with the differences in sampling and resolution between the techniques, it is possible to establish parameters for burst pressure calculations. A second goal of Year 2 was to select samples for Year 3 pressure testing. Year 3 focused on pressure testing of four samples to improve and verify failure pressure predictions calculated from field NDE and ILI measurements of SCC crack colony dimensions. Also examined was the potential of the recently developed intelligent interaction criteria (PRCI SIA-1-5) to predict failure pressure using IWEX results. The application of traditional interaction criteria produced mixed results. For the intelligent interaction criteria, the comparison between experimentally determined and predicted failure pressures was excellent, with a 3% average overestimation of predicted values and tight error distribution with a 3% standard deviation. Along with the specific results, this project successfully demonstrated a general approach for improving crack ILI technology using field NDE methods.

Corrosion growth rate (CGR) is one of the key elements in corrosion threat management. It is used to determine metal loss In-Line Inspection (ILI) re-assessment interval, identify high growth features that require excavation before the next ILI and predict probability of failure of an unmitigated feature. Overestimated CGR may result in unnecessary expenditures for excavations and remediation, while underestimated CGR could result in pipeline failure. Although there are many methods to derive CGR from successive ILIs, there is no particular approach widely accepted by industry. In this paper, several CGR methodologies used by Enbridge to manage corrosion threats on pipelines including CGR based on feature matching and signal matching are discussed and validated. CGR based on feature matching refers to the rate that is calculated based on ILI reported depths of two back-to-back ILIs. CGR based on signal matching refers to the rate that is determined based on ILI raw data review provided from the ILI vendor. Both feature level and joint level CGR have been used in corrosion threat management. A probabilistic CGR methodology based on Pipeline Research Council International (PRCI)(1) EC-1-2 project (Development of detailed procedures for comparing successive ILI runs to establish corrosion growth rate) that accounts for uncertainty of measurement tools is also included in this study. The ILI data and field non-destructive evaluation (NDE) results of six onshore pipelines are used for this study. Each of the six pipelines has at least three high resolution Magnetic Flux Leakage (MFL) inspections. The various CGRs are established based on the 1st and 2nd inspections. The CGR methodologies are validated based on comparing the predicted depth using CGR with the field measured depth as well as the depth reported by the 3rd ILI. Over 700 field measured feature depths are used in this study and over 60,000 ILI features are used to validate the estimated depth at the time of field measurements and at the time of the 3rd ILI. This paper demonstrates the validation results of feature level and joint level CGR based on both feature matching and signal matching. The probabilistic CGR methods are also validated using the same method. The CGR validation study enables pipeline operators to establish defect repair schedules and re-inspection intervals with increased confidence.

A Bayesian methodology was applied to use data from multiple inline inspection (ILI) runs and field measurements with non-destructive examination (NDE) tools to increase confidence in crack size estimates. Multiple crack depth measurements were used in two different ways — namely, to improve the characterization of ILI sizing error bias and to update the maximum depth distribution of individual crack features. This methodology was applied to selected datasets from an industrywide database for crack ILI data collected over a series of Pipeline Research Council International (PRCI) projects. The results of the approach are presented for two datasets, showing reduced variance in sizing error bias and improved confidence in crack depth estimates. In addition to the PRCI datasets, an additional dataset was collected and used to investigate the effect of multiple ILI runs on estimates of rate of detection and depth distribution of undetected features. The results of this analysis are also summarized.

The paper describes essentials, features and possibilities of a statistical tool that permits assessing the possible location (log distance coordinate) of defects that were missed-out during inline inspection (ILI). The method is based on complete analysis of the results of the latest ILI, using the basic set of ILI tool performance metrics, introduced and substantiated by the first author. The method is based on analysis of a full set of possible outcomes of any ILI measurement, namely: true detection/non-detection; false detection/non-detection. The developed statistical tool uses combined data from the ILI tool specifications, ILI run results and verification digs. This method permits quantitative assessment of the number of falsely detected and undetected actual defects and also the distribution of their sizes for each segment of an inspected pipeline that was not verified by digs. Application of the method is described for use when deciding how many defects of different categories have been unobserved. A decision rule is described with which it is possible to be practically sure that no dangerous/reportable defects were omitted. A combined (actual and computer simulated) case of analysis of ILI and dig data accumulated for a one-kilometer long pipeline segment is presented. Using existing strength codes it is shown how to locate possible dangerous spots on pipeline segments, where additional digs or monitoring would be advised/recommended. The described method is recommended for use by pipeline companies in order to minimize the risk of leak/rupture and enhance the culture of pipeline maintenance in general.

The purpose of this paper is the investigation of the source of errors in magnetic field measurements using magnetic probes in high-temperature plasmas and the correction or elimination of these errors. The particular sources of error considered are 1) errors in the magnetic field measurements caused by the finite size of the element which yields an average rather than a point value, 2) errors in the calibration caused by the heating of the probe element, 3) errors in the determination of probe position, and 4) errors associated with the disturbance of the current density resulting from the insertion of the probe into the plasma. In order to assure accurate current density calculations from the experimental data, it is necessary to reduce or correct for these errors. The magnitude of these errors and the compensation methods are discussed.

Corrosion is still one of the major threats to the integrity of onshore and offshore pipelines. Realistic corrosion growth rates are essential inputs to safe and effective pipeline integrity management decisions. For example, corrosion rates are needed to predict pipeline reliability as a function of time, to identify the need for and timing of field investigations and/or repairs and to determine optimum re-inspection intervals to name just a few applications. The consequences associated with using wrong corrosion growth rates range from the inefficient use of resources (time, people and money) on unnecessary repairs and/or inspections to unexpected pipeline releases. The identification of where corrosion is active on a pipeline and how fast it is growing is a complex process which is understood in the general sense but is highly variable. Corrosion is therefore difficult to predict due to the very localised nature of its behaviour and the many parameters that influence the corrosion reaction. Running an in-line inspection (ILI) tool in a pipeline identifies the internal and/or external corrosion located along the full length of the pipeline. The ILI inspection also determines the depth, length and width measurements for each corrosion site and for the overall feature. The use of repeat ILI data to match and compare metal loss sites in order to estimate the corrosion growth rates at individual defects along a pipeline is a well-used and established practice in the industry. The use of such corrosion rates to make predictions of the future integrity of a pipeline started in earnest approximately 5 to 10 years ago and over that time considerable experience has been gained. Now that we are starting to collect 3, 4 or even 5 or more ILI data sets for the same pipelines we are able to test and validate our earlier ILI based growth rate predictions versus what actually occurred in the pipeline over time. With the benefit of this hindsight, the methodologies employed for evaluating and applying ILI based corrosion rates can be further improved and refined to give more accurate predictions of the future pipeline condition, the response schedule and for setting the timing of re-inspections. This paper shares the experience gained and the improvements that can be made to the determination of corrosion rates and application of these rates in a pipeline integrity assessment. These topics are illustrated and investigated via the use of case studies on real ILI data sets.

Recognition that measurement error in the independent (predictor) variable of data used for regression modeling necessitates special methods of model fitting dates back to the late 1800s. In the ensuing 100 years attention was devoted almost entirely to measurement error in linear regression models, at which time advances in computing opened the door to the nonlinear statistical modeling generally, and to the study of measurement error in nonlinear regression modeling in particular. This article contains an introduction to the problem of measurement error in regression using linear regression measurement error models to illustrate key ideas. It includes a discussion of the types of measurement error commonly studied (classical and Berkson, differential and nondifferential) and how those types arise in practice. A summary of the consequences of ignoring measurement error when analyzing data measured with error is presented, with detailed results given for simple and multiple linear measurement error models. Pointers to the literature on both linear and nonlinear measurement error models are provided.

Majority of the pipeline operators manage corrosion using inline inspection (ILI). ILIs enable operators to choose which anomalies should be excavated and, on the flip side, which anomalies are safe to remain on the pipe. If the remaining anomalies grow to a critical size before the next ILI cycle there will be in-service failure. This is avoided by forecasting the growth of the measured defects appropriately and assessing them for future integrity. When an ILI is performed on a pipeline the ILI vendor reports the anomaly sizes, burst pressures and the Failure Pressure Ratios for each anomaly. Failure Pressure Ratio (FPR) for an anomaly is defined as the burst pressure divided by the Maximum allowable operation pressure (MAOP). In order to avoid ruptures, operators will excavate anomalies that have a limiting value of FPR. These limiting values are also referred to as response criteria, excavation criteria, safe failure pressure ratios, or safety factors. As the FPR depends heavily on the operating stress (or %SMYS) the limiting FPR value, below which we respond, should also depend on %SMYS. However, currently there is inconsistency in the use of response criteria in the industry. Some utilize the same response criteria for all pipe irrespective of %SMYS. This study shows that using the same limiting FPR does not provide the same level of safety for different %SMYS scenarios. It shows that the size of remaining anomalies in the lower %SMYS pipe would be significantly larger than in higher %SMYS pipe, leading to lower reliability in pipelines that have lower %SMYS. For gas pipelines, these lower %SMYS pipelines are often in higher location class pipelines with higher consequences of failure. In other words, even though the FPR value is the same for two anomalies in two different %SMYS pipes, the two anomalies would have very different probabilities of failure, where one anomaly could be safe while the other is not. This paper examines the ranges of FPR and the safe response criteria as a function of %SMYS. It examines the remaining anomaly sizes in different %SMYS pipe as allowed by current standards. The effect of the uncertainties on the response criteria due to measurement errors, material and geometric properties and model errors are also examined. It examines the response criteria in different jurisdictions. The effect of using different assessment equations, such as Modified B31G or RSTRENG, on the response criteria is also discussed. In order to obtain consistent safety, the consideration of %SMYS and consequences in defining a response criterion is discussed. Many sets of ILI data, which have pipe in different %SMYS, have been assessed in order to examine practical ranges of FPR with %SMYS. The practical implications of different response criteria are studied and discussed.

Measurement errors are very often come upon in data analysis. Classical statistical methods become disadvantageous and ordinary least squares estimator of parameters turns into inconsistent and biased, in the existence of measurement errors in the data. Although some methods are used, the performances of them are not good enough in the presence of multicollinearity and measurement errors in the data, simultaneously. That’s why researchers have been inquiring about the estimation of the parameters if the measurement error models have multicollinearity, lately. Especially, biased estimation techniques have been researched in the existence of multicollinearity for measurement error models recently. In this paper, the ridge and Liu estimation approaches to the measurement error models in the existence of multicollinearity are investigated. The comparisons of the biased estimators’ performances are analyzed theoretically and numerically.

Analyses of variables measured with error are often flawed when measurement error is ignored. An extensive literature on linear measurement error models is available by [1], and more recently, for non-linear measurement error models by [2] and [3]. We investigate problems encountered in the analysis of longitudinal change, where the difference in some outcome variable is modeled on a set of regressor variables in a linear regression model. This type of analysis has become quite popular in biomedical research, and the results are often biased when the outcome variable is measured with error and its observed baseline value is included as a covariate in the fitted model. [4] demonstrated the effects of measurement error in the analysis of change in wall thickness of the common carotid artery. They showed the naive analysis led to erroneous findings due to the measurement error bias, even when the regressor variables were assumed to be measured precisely. In this paper, we present a method-of-moments correction for measurement error bias, provided the measurement error variance is known or can be estimated. This work extends the work of [4] to include regressor variables that are measured with error.