Interactive Simulation of Gas Decompression and Crack Propagation in Natural Gas Transmission Pipelines

Author(s):  
Hiroyuki Makino ◽  
Yoshiaki Kawaguchi ◽  
Yoichiro Matsumoto ◽  
Shu Takagi ◽  
Shinobu Yoshimura

In this paper, the propagating shear fracture in natural gas transmission pipelines is simulated by an interactive method between gas decompression and crack propagation. A rich gas which contains heavier hydrocarbons than methane is highlighted and the relation between the crack velocity and the distance is simulated for varied condition of pipelines. The results of simulation are shown in the relation between the fracture arrest distance and the toughness of the pipes used, and the effects of the difference in gas compositions, increase of the operating pressure and the change of the initial gas temperature are discussed. The results of the simulation make it clear that the rich gas increases the risk for long running fracture, the simple increase of the operating pressure by increasing the design factor causes long crack propagation, increase of the operating pressure by using higher grade pipes not always invites long crack propagation and lower temperature increases the fracture arrest distance in relatively lower pressure but decreases the distance in relatively higher pressure. All the discussion in this study indicates that the analysis of the decompression behavior of the inner gas is essential for the interpretation of the phenomenon of the propagating shear fracture in pipelines. It is concluded that the fluid characteristics of the gas transmitted and material characteristics of the pipes used should be matched appropriately for the safety of the pipelines.

2009 ◽  
pp. 237-237-10 ◽  
Author(s):  
E Sugie ◽  
M Matsuoka ◽  
T Akiyama ◽  
K Tanaka ◽  
M Tsukamoto

Author(s):  
Michael R. Acton ◽  
Philip J. Baldwin

For most fuels transported by pipeline, whether or not ignition of an accidental release occurs is a critical factor in determining the extent of the resulting hazard. The probability of ignition is therefore a key input when undertaking pipeline risk assessments and the value chosen is a direct multiplier of the risk calculated. Typically, the ignition probability assigned is based on an analysis of historical data. However, the pipeline industry has a good safety record and major incidents are rare, sometimes resulting in widely differing values being used due to the scarcity of reliable data. For high pressure natural gas transmission pipelines, it is observed that ruptures of large diameter underground pipelines operating at high pressures can result in ignited releases even in remote areas with no obvious ignition sources present. Conversely, failures of small diameter pipelines operating at lower pressures rarely result in ignited releases, suggesting that ignition sources generated as a result of the failure event itself may be significant in causing ignition of high pressure natural gas releases from underground pipelines. The results of analysis previously reported at IPC2002 indicated a trend for the ignition probability to increase with pd2, with p the pipeline operating pressure (bar) and d the pipeline diameter (m). The relationship forms the basis of the default ignition probabilities recommended for use in the PIPESAFE package developed for risk assessment of gas transmission pipelines. Since the previous study was carried out, the number of pipeline rupture incidents in the dataset used has increased by about 20%, and following a recent review, the statistical analysis has been extended and refined. This paper reports the results of recent analysis of the most comprehensive incident dataset available to Advantica for natural gas transmission pipelines, presenting the correlation derived from a simple statistical analysis together with consideration of possible physical explanations for the trends observed based on an ongoing programme of research into the causes of ignition.


Author(s):  
Michael R. Acton ◽  
Neil W. Jackson ◽  
Eric E. R. Jager

Due to the increasing demand for natural gas in many locations, there is often a need to increase the capacity of existing and future gas transmission pipeline networks. In some situations, there may be a possibility of increasing the operating pressure (e.g. uprating), but in others there may be no alternative but to lay new pipelines, often along the same route as an existing pipeline. If one pipeline fails in this situation, it is possible that a second parallel pipeline may also fail as a result. However, there is also increasing pressure on the use of land and therefore the minimum separations with which pipelines may be laid and operated safely when in parallel to other pipelines need to be considered. This paper describes work carried out as a collaborative project supported by gas transmission pipeline operators to provide guidance on the likelihood of failure of a pipeline, for a range of different conditions, following failure of an adjacent pipeline. A framework has been developed that identifies the sequence of events that could lead to failure of a parallel pipeline, including the possibility of escalation from a leak (or puncture) to a full bore rupture. Work has been carried out including large scale experiments and CFD (Computational Fluid Dynamics) modelling to enable the critical processes in the framework to be quantified. This methodology has been used to produce general guidelines for parallel pipeline assessments, in order to support the design of new parallel pipeline installations. The methodology has been developed specifically for parallel natural gas transmission pipelines. However, the principles are relevant to parallel pipelines transporting other substances, and consideration is given to how the methodology may be adapted for such circumstances. The methodology provides input to any risk assessments of parallel pipeline installations, to quantify the possible contribution to the failure frequency from escalation. General guidance developed using the methodology presented in this paper, has recently been included in the recommendations for steel transmission pipelines, IGEM/TD/1 (Edition 5), published by the Institution of Gas Engineers and Managers. However, where general recommendations are not achievable, the methodology may be applied to take site and pipeline-specific factors into account.


Author(s):  
Marcus McCallum ◽  
Rafael G. Mora ◽  
Graham Emmerson ◽  
Thushanthi Senadheera ◽  
Andrew Francis

In Canada, a great deal of effort has been invested into the use of reliability-based techniques for the design and assessment of non-sour natural gas transmission pipelines. This led to the inclusion of Annex O in the Canadian onshore pipeline code CSA Z662 in 2007, which gives detailed descriptions of all of the key components of reliability-based approaches. However, the annex does not and is not intended to provide recipes for using the reliability-based techniques for particular fields of application such as evaluating the acceptability of changes to location class, service or increasing maximum operating pressure. Consequently, the onus is on the reliability/integrity engineer to tailor the approach to the particular field of application and the specifics of the pipeline. This means that even working in accordance with the code, the approach and optimizing techniques adopted by one engineer may be very different to that adopted by another. This presents a challenge for those reviewing reliability based plans, designs and alternatives for approval. The National Energy Board (NEB) engaged Andrew Francis & Associates Ltd (AFAA) to assist them with constructing a set of supporting guidelines to assess the comprehensiveness and safety of reliability based submissions. Unlike customary design reviews, the guidelines are geared towards provoking a reviewer into asking delving questions rather than into going through a ‘box-checking’ questionnaire. Indeed, asking the case-specific and clarification questions is regarded as a crucial step towards determining the adequacy and effectiveness of the measures proposed in the content and conclusions of a particular filing. Simply questioning whether Annex O has been followed is not encouraged and, even when safety criteria appear to have been met (i.e. box-checking), a reviewer is prompted to challenge the reasonableness of assumptions and ask whether safety levels are providing the lowest practicable risk to the Canadian public. One line of inquiring might be: are sufficient data available; are the data reliable; are the data relevant to the case under consideration; or have the data been analyzed using a valid method applicable to the case. Other typical questions would be have the consequences been properly assessed and are the mitigative and preventative measures providing the lowest practicable risk compared to pressure reduction and pipe replacement. The purpose of this paper is to present an overview of the assessment guidelines and the approach and key considerations for conducting efficient, consistent and fair reviews of reliability based assessments of hazardous material pipelines. In doing so, the paper also identifies some of the pitfalls that engineers conducting reliability based integrity assessments should seek to avoid.


Author(s):  
Jason P. Finley ◽  
David L. Slayter ◽  
Chris S. Hitchcock ◽  
Chih-Hung Lee

Landslides related to heavy rainfall can cause extensive damage to natural gas transmission pipelines. We have developed and implemented a geographic information system (GIS) model that evaluates near real-time precipitation-induced landslide susceptibility. This model incorporates state-wide precipitation data and geologically-based landslide classifications to produce rapid landslide risk evaluation for Pacific Gas & Electric Company’s (PG&E) gas transmission system during winter rain storms in California. The precipitation data include pre-storm event quantitative precipitation forecasts (QPF) and post-storm event quantitative precipitation estimates (QPE) from the United States National Oceanic and Atmospheric Administration (NOAA). The geologic classifications are based on slope, susceptible geologic formations, and the locations of historic or known landslide occurrences. Currently the model is calibrated using qualitative measures. Various scientists have developed large landslide databases with associated rainfall statistics to determine rainfall thresholds that trigger landslides. With a sufficient number of landslides, we can more precisely determine minimum rainfall thresholds using similar methods.


Sign in / Sign up

Export Citation Format

Share Document