A Feature-Specific Probabilistic Assessment of Pipeline Defect Size From ILI MFL Signal Using Convolutional Neural Network

2020 ◽  
Author(s):  
Jenny Jing Chen ◽  
Stephen Westwood ◽  
David Heaney
Keyword(s):  
Risk Assessment ◽  
Standard Deviation ◽  
Wall Thickness ◽  
Error Tolerance ◽  
Burst Pressure ◽  
Metal Loss ◽  
Failure Pressure ◽  
Novel Approach ◽  
Pressure Estimation ◽  
Specific Error

Abstract When estimating pipeline burst pressure, one of the prevalent sources of uncertainty that needs to be factored into the calculation is the model error in the estimation of feature depth and length from the in-line inspection tool. Due to modeling technique limitation, as of today many ILI vendors have feature specific error bounds depending on the morphologies of the corrosion, this error can only be reported to operators as an overall error known as the ILI tool tolerance which is usually obtained from samples of excavation data or pull test data. At the most, the error is reported by classes based on corrosion morphologies specified by Pipeline Operators Forum. For example, a commonly reported corrosion depth sizing specification is ±10% of pipe wall thickness at 80% confidence for the General type of corrosion. This can be interpreted as that the error of each reported depth estimations is assumed to fall in a normal distribution with a mean equal to 0 and standard deviation equal to 7.8% of wall thickness. The shape of the distribution, the mean and standard deviation will then be used as constants to factor in the burst pressure calculation. However, these factors are never constant for a sample of defects in reality. In fact, they ought to be variables on an individual feature basis. An example of such an approach would be a feature specific error tolerance, this could be that the estimated depth of a feature is 36%wt in an interval of [30%, 48%] of wall thickness with 80% confidence. This is believed to greatly reduce the level of uncertainty when it comes to failure pressure estimation or other type of pipeline risk assessment. The advancement in Machine Learning today, deep learning with deep neural networks, allows feature-specific error tolerance to be obtained after analyzing visual imagery of MFL signal. In this paper we will describe a novel approach to predict the size of metal loss defects and more importantly the distribution associated with each prediction. We will then discuss the benefits of this approach has with respect to risk assessment such as failure pressure estimation.

Fitness for Service Assessment of a Metal Loss Using Combination of Small-Scale Testing and DNV RP F101/API579 Procedures

2007 ◽  
Author(s):  
A. Motarjemi
Keyword(s):  
Tensile Properties ◽  
Wall Thickness ◽  
Pressure Vessels ◽  
Small Scale ◽  
Metal Loss ◽  
Remaining Life ◽  
Test Technique ◽  
Failure Pressure ◽  
Destructive Test ◽  
Tensile Data

One of the major issues in the oil and gas industries is occurrence of corrosion on equipments in-service such as tanks, pressure vessels, piping, etc. Metal loss (general/localized) and pitting are amongst the typical corrosion damages. For assessing the significance of metal loss, information such as (a) geometry of the component, (b) a record of thickness measurements (point or profile readings) and (c) tensile properties such as Yield and Tensile strengths, preferably in the vicinity of the metal loss, are required. This information are usually fed into a Fitness for Service (FFS) assessment guideline/recommended practice such as DNV RP-F101 or API579, and a minimum required wall thickness (tmin), failure pressure or remaining life is derived. In the absence of actual tensile data (obtained from a conventional tensile test), specified minimum values (lower-bound), as suggested in the design codes, are currently the only other alternative. However, this paper is aimed at presenting two more alternative techniques; non-destructive test technique called Instrumented Indentation Testing (IIT) or Automated Ball Indenter (ABI) and a semi-destructive test technique, called Micro-Flat tensile (MFT). Both techniques are capable of determining the local tensile properties of the material in the vicinity of the metal loss. Values of the minimum required wall thickness (tmin), failure pressure and remaining life, using tensile data obtained from the IIT, MFT and specified minimum values are compared with the predictions based on the actual tensile data.

ILI-to-Field Data Comparisons: What Accuracy Can You Expect?

2016 ◽  
Author(s):  
Matthew A. Ellinger ◽  
Thomas A. Bubenik ◽  
Pamela J. Moreno
Keyword(s):  
Wall Thickness ◽  
Signal Analysis ◽  
Metal Loss ◽  
Data Sets ◽  
Line Segments ◽  
Failure Pressure ◽  
General Field ◽  
Inspection Systems ◽  
Data Signal ◽  
Flux Leakage

Det Norske Veritas (U.S.A.), Inc. (DNV GL) prepared this paper in order to study the expected accuracy of in-line inspections (ILI) as a function of year, depth (both reported and field measured), and length, amongst other factors. DNV GL has access to a significant amount of data that span many different pipeline operators, ILI vendors, inspection years, and inspection technologies. DNV GL is well suited to complete this study as a result of our access to these various data sets. Over 3,000 individual comparisons of ILI and field depths and lengths spanning from 2010 through 2015 from 11 operators and 68 line segments were compiled to meet the objectives of this paper. Inspection technologies include axial magnetic flux leakage (MFL), ultrasonic wall thickness (UTWT), spiral MFL, and circumferential MFL. Based on the analyses conducted in this paper, the following conclusions were generated. • Axial MFL and UTWT inspections show significant improvements over the last several years. • Axial MFL inspection systems are capable of meeting a depth accuracy of +/−10% of the wall thickness with 80% certainty, but this has not always been the case. UTWT inspection systems are capable of meeting a higher depth accuracy. • Axial MFL inspection systems report more pits and circumferential grooves than UTWT systems. This could suggest UTWT systems are less sensitive to pits and circumferential grooves than axial MFL systems. • Both axial MFL and UTWT inspection systems routinely under call defects with field measured depths greater than 50 to 80% of the wall thickness. This is contrary to a widely held notion that ILI is conservative for deep defects. • ILI reported defect lengths do not correlate well to field measured defect lengths. In general, field measured defect lengths are greater than ILI reported defect lengths. • Depth accuracy tends to decrease slightly for very short defects (less than 1-inch) and for very long defects (greater than 40-inches). Based on these conclusions, the authors make the following recommendations: • Pipeline operators should dig more than the deepest reported defects to better understand the accuracy of the inspection tools being used and to determine whether deeper anomalies are being under called. • Pipeline operators should consider methods for evaluating change in corrosion depth from ILI survey to ILI survey to lessen the dependence on the accuracy of the ILI tools. This should include a raw data signal analysis in order to determine whether the general morphology (metal loss length and width) are changing between ILI surveys. • ILI reported defect lengths should be used in conjunction with field measured defect depths (if available) when performing failure pressure calculations. • Additional accuracy, especially for deeper defects, may only come with new tool developments. Industry support of such developments will be required to bring them to fruition.

Corrosion Metal Loss Interaction From In-Line Inspection and How it Can Affect the Calculated Failure Pressure

2012 ◽  
Author(s):  
Lucinda Smart ◽  
Richard McNealy ◽  
Harvey Haines
Keyword(s):  
Field Measurements ◽  
Metal Loss ◽  
Data Correlation ◽  
Failure Pressure ◽  
Industry Standards ◽  
Direct Measurements ◽  
Non Destructive ◽  
Flux Leakage

In-Line Inspection (ILI) is used to prioritize metal loss conditions based on predicted failure pressure in accordance with methods prescribed in industry standards such as ASME B31G-2009. Corrosion may occur in multiple areas of metal loss that interact and may result in a lower failure pressure than if flaws were analyzed separately. The B31G standard recommends a flaw interaction criterion for ILI metal loss predictions within a longitudinal and circumferential spacing of 3 times wall thickness, but cautions that methods employed for clustering of ILI anomalies should be validated with results from direct measurements in the ditch. Recent advances in non-destructive examination (NDE) and data correlation software have enabled reliable comparisons of ILI burst pressure predictions with the results from in-ditch examination. Data correlation using pattern matching algorithms allows the consideration of detection and reporting thresholds for both ILI and field measurements, and determination of error in the calculated failure pressure prediction attributable to the flaw interaction criterion. This paper presents a case study of magnetic flux leakage ILI failure pressure predictions compared with field results obtained during excavations. The effect of interaction criterion on calculated failure pressure and the probability of an ILI measurement underestimating failure pressure have been studied. We concluded a reason failure pressure specifications do not exist for ILI measurements is because of the variety of possible interaction criteria and data thresholds that can be employed, and demonstrate herein a method for their validation.

Full Well Corrosion Insight – Case Studies in the Added Value of Electromagnetic Thickness Measurements During Well Interventions

2021 ◽  
Author(s):  
Andrew Imrie ◽  
Maciej Kozlowski ◽  
Omar Torky ◽  
Aditya Arie Wijaya
Keyword(s):  
Case Studies ◽  
Wall Thickness ◽  
Added Value ◽  
Metal Loss ◽  
Outer Diameter ◽  
Metal Thickness ◽  
True Condition ◽  
Pass Through ◽  
External Corrosion ◽  
The Individual

AbstractMonitoring pipe corrosion is one of the critical aspects in the well intervention. Such analysis is used to evaluate and justify any remedial actions, to prolong the longevity of the well. Typical corrosion evaluation methods of tubulars consist of multifinger caliper tools that provide high-resolution measurements of the internal condition of the pipe. Routinely, this data is then analyzed and interpreted with respect to the manufacture's nominal specification for each tubular. However, this requires assumptions on the outer diameter of the tubular may add uncertainty, and incorrectly calculate the true metal thicknesses. This paper will highlight cases where the integration of such tool and electromagnetic (EM) thickness data adds value in discovering the true condition of both the first tubular and outer casings.These case studies demonstrate the use of a multireceiver, multitransmitter electromagnetic (EM) metal thickness tool operating at multiple simultaneous frequencies. It is used to measure the individual wall thickness across multiple strings (up to five) and operates continuously, making measurements in the frequency domain. This tool was combined with a multifinger caliper to provide a complete and efficient single-trip diagnosis of the tubing and casing integrity. The combination of multifinger caliper and EM metal thickness tool results gives both internal and external corrosion as well as metal thickness of first and outer tubular strings.The paper highlights multiple case studies including; i) successfully detecting several areas of metal loss (up to greater than 32%) on the outer string, which correlated to areas of the mobile salt formation, ii) overlapping defects in two tubulars and, iii) cases where a multifinger caliper alone doesn't provide an accurate indication of the true wall thickness. The final case highlights the advantages of integrating multiple tubular integrity tools when determining the condition of the casing wall.Metal thickness tools operating on EM principles benefit from a slim outer diameter design that allows the tools to pass through restrictions which typically would prevent ultrasonic scanning thickness tools. Additionally, EM tools are unaffected by the type of fluid in the wellbore and not affected by any non-ferrous scale buildup that may present in the inside of the tubular wall. Combinability between complementary multifinger caliper technology and EM thickness results in two independent sensors to provide a complete assessment of the well architecture.

scholarly journals Integrating public risk perception into formal natural hazard risk assessment

2006 ◽  
Vol 6 (3) ◽  
pp. 471-483 ◽  
Author(s):  
Th. Plattner ◽  
T. Plapp ◽  
B. Hebel
Keyword(s):  
Risk Assessment ◽  
Risk Perception ◽  
Risk Evaluation ◽  
Formal Model ◽  
Relevant Literature ◽  
Historical Context ◽  
Individual Risk ◽  
Affecting Factors ◽  
Plausibility Check ◽  
Novel Approach

Abstract. An urgent need to take perception into account for risk assessment has been pointed out by relevant literature, its impact in terms of risk-related behaviour by individuals is obvious. This study represents an effort to overcome the broadly discussed question of whether risk perception is quantifiable or not by proposing a still simple but applicable methodology. A novel approach is elaborated to obtain a more accurate and comprehensive quantification of risk in comparison to present formal risk evaluation practice. A consideration of relevant factors enables a explicit quantification of individual risk perception and evaluation. The model approach integrates the effective individual risk reff and a weighted mean of relevant perception affecting factors PAF. The relevant PAF cover voluntariness of risk-taking, individual reducibility of risk, knowledge and experience, endangerment, subjective damage rating and subjective recurrence frequency perception. The approach assigns an individual weight to each PAF to represent its impact magnitude. The quantification of these weights is target-group-dependent (e.g. experts, laypersons) and may be effected by psychometric methods. The novel approach is subject to a plausibility check using data from an expert-workshop. A first model application is conducted by means of data of an empirical risk perception study in Western Germany to deduce PAF and weight quantification as well as to confirm and evaluate model applicbility and flexibility. Main fields of application will be a quantification of risk perception by individual persons in a formal and technical way e.g. for the purpose of risk communication issues in illustrating differing perspectives of experts and non-experts. For decision making processes this model will have to be applied with caution, since it is by definition not designed to quantify risk acceptance or risk evaluation. The approach may well explain how risk perception differs, but not why it differs. The formal model generates only "snap shots" and considers neither the socio-cultural nor the historical context of risk perception, since it is a highly individualistic and non-contextual approach.

Probabilistic Assessment of Minor Mechanical Damage

2006 ◽  
Author(s):  
Patrick H. Vieth ◽  
Clifford J. Maier ◽  
William V. Harper ◽  
Elden Johnson ◽  
Bhaskar Neogi ◽  
...  
Keyword(s):  
Risk Assessment ◽  
Mechanical Damage ◽  
Residual Deformation ◽  
Evaluation Process ◽  
Assessment Methods ◽  
Metal Loss ◽  
Probabilistic Assessment ◽  
Pipeline System ◽  
Pressure Cycle ◽  
Important Field

In-line inspection (ILI) of the Trans Alaska Pipeline System (TAPS) using high resolution metal loss tools indicated 77 locations with suspected minor mechanical damage features (MDF). The tools used are able to detect the presence of a suspected feature, and measure indented dimensions, but are insufficient to detect the presence of cracks or gouges needed to reliably assess feature severity based solely on the ILI data. Excavations of 42 sites deemed most severe provided important field data characterizing residual deformation dimensions, the occurrence of gouges or cracks, and allowing a reliable field assessment of defect severity. Upon completion of the excavations, 35 possible MDF locations remained unexcavated. An engineering evaluation was undertaken to assess whether or not these remaining minor MDF pose a threat that is significant enough to warrant excavation. Multiple assessment methods were utilized including deterministic, probabilistic, and risk assessment methods. The probabilistic assessment of 35 unexcavated MDFs was performed using PCFStat; or Pressure Cycle Fatigue Statistical Assessment, which uses Monte Carlo simulation to estimate remaining fatigue life. PCFStat performs 1,000’s of simulations for each case where the input parameters are randomly selected from expected distributions. Of particular importance is the fatigue environment of the location. The results of the probabilistic assessment were used to estimate the potential for failure of remaining MDFs. The results suggest that 25 of 35 unexpected damage features had a POF of less than 10−4 over the remaining expected pipeline life cycle and thus are unlikely to fail. Alyeska considered a combination of probabilistic, deterministic and risk assessment results to decide on the actual locations to be examined. The results of probabilistic analysis also were found to support the outcome of the operator’s risk-based evaluation process.

scholarly journals Examination of the Risk Assessment Levels of Athletes in Different Branches According to Some Variables

2018 ◽  
Vol 6 (10) ◽  
pp. 193
Author(s):  
Abdurrahman Kirtepe
Keyword(s):  
Risk Assessment ◽  
Data Analysis ◽  
Standard Deviation ◽  
Personal Information ◽  
Educational Status ◽  
Analysis Data ◽  
Descriptive Statistics ◽  
Income Status ◽  
Package Program ◽  
Sample Method

In this study, the risk assessment levels of athletes in different branches were examined in terms of various variables. Descriptive scanning model was used in the study. In the research, the survey was completed with a sample method of 105 people. The questionnaire was used as a data collection tool in the research. The questionnaire consists of questions about personal information and the Risk Assessment scale for athletes and coaches. Data analysis was performed in SPSS 21 package program. Descriptive statistics such as frequency, percent, and mean, standard deviation, minimum and maximum were used in data analysis. Data analysis was performed in SPSS 21 package program. Descriptive statistics such as frequency, percent, and mean, standard deviation, minimum and maximum were used in data analysis. As a result of the research, it was determined that the risk assessment perceptions of athletes according to their age, branches, educational status and income status did not differ. As a result of the research, it was determined that the risk assessment perceptions of athletes according to their age, branches, educational status and income status did not differ.

Remaining Local Buckling Resistance of Corroded Pipelines

2010 ◽  
Author(s):  
Qishi Chen ◽  
Heng Aik Khoo ◽  
Roger Cheng ◽  
Joe Zhou
Keyword(s):  
Finite Element ◽  
Wall Thickness ◽  
Phase 1 ◽  
Compressive Strain ◽  
Research Work ◽  
Local Buckling ◽  
Model Verification ◽  
Metal Loss ◽  
General Corrosion ◽  
Buckling Resistance

This paper describes a multi-year PRCI research program that investigated the local buckling (or wrinkling) of onshore pipelines with metal-loss corrosion. The dependence of local buckling resistance on wall thickness suggests that metal-loss defects will considerably reduce such resistance. Due to the lack of experimental data, overly conservative assumptions such as a uniform wall thickness reduction over the entire pipe circumference based on the defect depth have been used in practice. The objective of this research work was to develop local buckling criteria for pipelines with corrosion defects. The work related to local buckling was carried out in three phases by C-FER and the University of Alberta. The first phase included a comprehensive finite element analysis to evaluate the influence of various corrosion defect features and to rank key parameters. Based on the outcome of Phase 1 work, a test matrix was developed and ten full-scale tests were carried out in Phase 2 to collect data for model verification. In Phase 3, over 150 parametric cases were analyzed using finite element models to develop assessment criteria for maximum moment and compressive strain limit. Each criterion includes a set of partial safety factors that were calibrated to meet target reliabilities selected based on recent research related to pipeline code development. The proposed criteria were applied to in-service pipeline examples with general corrosion features to estimate the remaining load-carrying capacity and to assess the conservatism of current practice.

scholarly journals A novel approach to calibrating a photoacoustic absorption spectrometer using polydisperse absorbing aerosol

2019 ◽  
Vol 12 (6) ◽  
pp. 3351-3363 ◽  
Author(s):  
Katie Foster ◽  
Rudra Pokhrel ◽  
Matthew Burkhart ◽  
Shane Murphy
Keyword(s):  
Standard Deviation ◽  
Single Scattering ◽  
Single Scattering Albedo ◽  
Calibration Technique ◽  
Acoustic Cavity ◽  
Novel Approach ◽  
Carrier Gases ◽  
Reactive Gases ◽  
A New Technique ◽  
Calibration Approach

Abstract. A new technique for calibrating photoacoustic aerosol absorption spectrometers with multiple laser passes in the acoustic cavity (multi-pass PAS) has been developed utilizing polydisperse and highly absorbing aerosol. This is the first calibration technique for multi-pass PAS instruments that utilizes particles instead of reactive gases and does not require knowledge of the exact size or refractive index of the absorbing aerosol. In this new method, highly absorbing materials are aerosolized into a polydisperse distribution and measured simultaneously with a multi-pass PAS and a cavity-attenuated phase shift particulate matter single-scattering albedo (CAPS PMSSA, Aerodyne Inc.) instrument. The CAPS PMSSA measures the bulk absorption coefficient through the subtraction of the scattering coefficient from the extinction coefficient. While this approach can have significant errors in ambient aerosol, the accuracy and precision of the CAPS PMSSA are high when the measured aerosol has a low single-scattering albedo (SSA) and particles are less than 300 nm in size, in which case truncation errors are small. To confirm the precision and accuracy of the new calibration approach, a range of aerosol concentrations were sent to the multi-pass PAS and CAPS PMSSA instruments using three different absorbing substances: Aquadag, Regal Black, and Nigrosin. Six repetitions with each of the three substances produced stable calibrations, with the standard deviation of the calibration slopes being less than 2 % at 660 nm and less than 5 % at 405 nm for a given calibration substance. Calibrations were also consistent across the different calibration substances (standard deviation of 2 % at 660 nm and 10 % at 405 nm) except for Nigrosin at 405 nm. The accuracy of the calibration approach is dependent on the SSA of the calibration substance but is roughly 6 % for the calibration substances used here, which all have an SSA near 0.4 at 405 nm. This calibration technique is easily deployed in the field as it involves no toxic or reactive gases and it does not require generation of a monodisperse aerosol. Advantages to this particle-based calibration technique versus techniques based on ozone or nitrogen dioxide absorption include no reactive losses or impact from carrier gases and the broad absorption characteristics of the particles, which eliminate potentially significant errors in calibration that come with small errors in the peak wavelength of the laser light when utilizing gas-phase standards.

Sign in / Sign up

Export Citation Format

Share Document