DEVELOPMENT OF SAND PRODUCTION PREDICTION MODELS TO SUPPORT THE RISK MITIGATION DURING OIL AND GAS PRODUCTION

Loading...
Thumbnail Image

Date

2021-05

Authors

Imankulova, Kamshat

Journal Title

Journal ISSN

Volume Title

Publisher

Nazarbayev University School of Engineering and Digital Sciences

Abstract

For the normal operation of the equipment, it is necessary to deal with sand phenomena at all stages of production. Because the particles of sands can fall on the bottom of the well, can destroy the layers, cause traffic jams and collapses in trunk. These damages can lead to a complete failure of the equipment, losses in profit and in some cases can provide risk even to the life of employees. As a result, any sand occurrences must be eliminated as soon as possible. The main purpose of this work is to study the main geomechanical parameters of the formation of sanding, to identify their relationship with failure occurrence in oil-wells and maximally accurate predict it to prevent the sand phenomena. According to the above, a simulation method using the results of real oil field observation was developed. A detailed mathematical calculation is provided for estimating the sand failure reference to the multivariate linear regression method. Based on the priority impact of parameters for high sanding occurrence a template of Model in GeNIe software according to the Bayesian probability to predict the appearance of failure is provided. As a result, conclusions are drawn and recommendations are given to ensure probability studies. In future, the model of GeNIe can be updated efficiently for the widely used in real oil field for engineers as a guide to know the risk of sand occurrences.

Description

Keywords

GeNIe, sand phenomena, Research Subject Categories::TECHNOLOGY, Type of access: Gated Access

Citation

Imankulova, K. (2021). Development of Sand Production Prediction Models to Support the Risk Mitigation During Oil and Gas Production (Unpublished master's thesis). Nazarbayev University, Nur-Sultan, Kazakhstan