Guidelines for predicting the remaining life of underground pipe networks that are subjected to the combined effects of external corrosion and internal pressure

Abstract

Underground pipelines are used in various process piping systems to transport gasses or fluids and are usually subjected to the effects of external corrosion. Corrosion can be defined as the deterioration of a material due to a reaction with its environment or the destruction of the material by means that are not mechanical (Fontana and Greene, 1967:2). External corrosion, due to the interaction between the pipe and the soil, is generally a slow process and the corrosion rate is influenced by a variety of external factors. Some of these factors include the ambient pH and salinity, the presence of moisture and bacteria, temperature, the electrical potential difference between the pipe and other structures and the implementation of preventative measures (such as cathodic protection and wrapping). Although the external corrosion of underground pipelines is generally a slow process in mild environments, pipe degradation as a result of external corrosion remains one of the prevalent reasons for the failure of underground pipelines. As with many mechanical systems that are prone to fail at one time or the other, the high costs involved with unforeseen failure necessitate some quantitative (or qualitative) indication of the condition of the pipe system. Some of the costs that can be expected as a result of unforeseen pipeline failure are, amongst others: • costs as a result of the failure of dependent systems; • costs as a result of the loss of production; • costs as a result of the loss of product (in distribution networks); • the cost of unscheduled maintenance (logistical costs); • costs as a result of damage to public property; • fines imposed by customers (in distribution networks); • costs related to pollution control, and • the loss of life The single most important parameter associated with the condition of a system is its profitable remaining life. This is the time during which a sub-system contributes to the well-being of a larger system and the organisation. Therefore, it is necessary to determine, with reasonable accuracy, the extent of the remaining life of a system so that managerial decisions (i.e. investments, cash-flow analyses, maintenance task scheduling and replacement programmes), based on this figure, can be made. Done correctly, this can directly lead to a decrease in maintenance costs and subsequently to an increase in profit. The extent of a corrosive attack on the pipeline might be highly localised or might be fairly uniform over the length of the installation. The fact of the matter is that, since the pipe is buried, it is very difficult to quantify the external damage caused by corrosion. A variety of techniques are in use to survey pipelines and detect anomalies. However, for large pipelines, most of these techniques are either inefficient or too expensive. There will always remain some uncertainty regarding the integrity of the pipeline. The work presented in this study is explained with valid generic examples and aims: 1. to provide the reader with sufficient background information so that the need for determining the integrity of a pipeline becomes apparent; 2. to indicate why a reliability-centred approach is necessary (Chapter 1); 3. to explain the basic principles of corrosion and the electrochemical nature of corrosion (Chapter 2); 4. to indicate areas, based on the basic principles of corrosion, where severe corrosion can be expected (Chapters 2 and 7); 5. to provide and elaborate on information regarding pipe surveillance techniques that are currently available (Chapter 3); 6. to establish the criteria for pipeline failure, in the form of a limit state Junction, for pipes that are subjected to near-constant internal pressures (static failure domain) as well as for pipes subjected to varying internal pressures (fatigue domain) (Chapters 5 and 6); 7. to indicate the sensitivity of the fatigue domain solution to changes in the system variables and to indicate that a significant reduction in the system variables does not necessarily reduce the solution accuracy (Chapter 6), and 8. to integrate the above-mentioned into a practical and workable guideline that can be used to determine the remaining life of an underground pipe network (Chapter 7).