Journal of Petroleum & Environmental Biotechnology

The probabilistic assessment of pipeline strength reliability

4^th International Conference on Petroleum Engineering

August 15-17, 2016 London, UK

S L Golofast

Transneft Research and Development Institute for Oil and Oil Products Transportation, Russia

Posters & Accepted Abstracts: J Pet Environ Biotechnol

Abstract:

The assessment of pipeline strength reliability is one of the key focus areas at both design and operation stages of a pipeline system. The correct and verified pipeline strength assessment influences operating costs, economic and ecological risks and pipeline system efficiency. Investigation of the issue becomes harder due to the strength probability and non-failure operation of pipelines which depend on a vast number of factors, some of them are independent and could not be controlled and others are defined as random. This paper aims to specify the most appropriate method for evaluating non-failure operation by using a probabilistic approach. The problem is stated as follows: How to assess the non-failure operation probability in case the applied stress exceeds the admissible stress? The probability of failure operation is determined as an overlap area of distribution density functions of stresses and admissible stresses. The reconstruction of probability distribution�??s density is calculated on random variable sample of the ultimate stress limit s and the �?-stress, which are defined on various sections of pipelines as random variables. Due to the large number of laws of random variables distribution in the framework of the parametric statistics; as well as impossibility of formulating the conditions for attributing the sample (during the results processing of a certain value measurement) to a particular distribution law in some cases, the most effective approach is based on the nonparametric statistics methods. Thus, the software and methodological support was developed, aiming to restore the density distribution function of the random variables. The methodology allows assessing the strength reliability of objects under conditions of real spectrums of external load impact and the natural laws of change of ultimate material state.

Biography :

Email: trasser@inbox.ru