Sand Management and Erosion Prediction Models for Oil and Gas Applications - Experimental and Numerical Validation Studies

Abstract

Abstract Due to the recent drop of the oil prices, the oil and gas industry is experiencing tremendous pressures to reduce costs. This is especially true for both oil and gas production systems (OGPS) development and operations to ensure profitability while meeting increasing regulatory requirements. Sand production and erosion can present rigorous challenges during the design and production phases of the OGPS and lead to significant increase in the CAPEX and OPEX. Thus, the accurate prediction of the erosion rates, location of erosion hotspots and expected sand accumulation zones in production system is a key task during the design, as well as for the development of sand monitoring systems used during the production phase. In the recent years, simulation of multiphase flows including sand management and erosion prediction has become a powerful engineering tool in the oil and gas industry. It can be applied to conduct virtual experiments to help designing the OGPS during the project design phase. There have been many significant improvements to the modelling approaches including more details of the flow physics, rapid development of the computational resources, and the adaption of modelling methodologies from other industries such as aviation and automobile. However, the generic simulation models and tools must be validated and tested before they can be used in the design, product development or monitoring systems of the oil and gas specific applications. In addition, high fidelity flow and erosion simulations can minimize the cost of the production system design and operation optimization as they enable sensitivity studies for large number of design variations and production condition sets – provided they present an industrial application in terms of short simulation runtime and low resource effort. Recently, GE Oil & Gas made an effort for examining the validity of existing erosion prediction correlations and modelling approaches in commercial Computational Fluid Dynamics (CFD) software. Sand particles transport and erosion laboratory experiments – on different materials as well as for OGPS-like flow loop configurations with test conditions close to the actual OGPS operation conditions – were performed to gather test data that can be used for the validation of the numerical simulations. The objective of this paper is to present part of the effort, focusing on the geometry and piping elements' arrangements (in particular bends arrangements) impact on the erosion rates and erosion hotspot locations, the validation results of 1D tools and 3D CFD erosion prediction models and the comparison to the test data. Furthermore, future development required to enhance the accuracy of the modelling results and the uncertainties are addressed. The major challenges to all methods are 1) erosion prediction for very small sand (<50 μm), 2) lack of quality field data for validation and improvement, 3) high level of uncertainty due to the complexity of the erosion process. Significant inconsistency between the erosion predicted by different models was the driver for the effort. Choosing one or the other in certain applications is not always obvious and can be difficult to justify, yet they can give very different results. The community needs to fully understand the situation and look for a solution.