Abstract

Scaling is a common and severe phenomenon of degradation for sub-sea wells of offshore oil and gas production facilities.Scaling itself,as well as intervention activities aiming to remove scaling may lead to significant production losses over their service lives and this calls for a careful consideration of how the management of scaling formation may best be optimized.However,since scale formation is associated with substantial uncertainty,its management comprises a rather non-trivial challenge.To achieve this challenge,it is of crucial importance to understand the processes leading to and governing scaling,and to represent the best available knowledge on these processes consistently.

Based on scaling models proposed by Zhen-Wu et al.[1]and Dawe and Zhang[2]together with ongoing research undertaken at the Danish Hydrocarbon Research and Technology Center,the present paper introduces novel probabilistic models for the representation of the available knowledge and dominating uncertainties associated with calcium carbonate(CaCO3)and barium sulphate(BaSO4)scaling processes[3].The proposed probabilistic models are formulated as functions of the main parameters governing scaling processes,including the temporal variability of the chemical composition of the production,temperature and pressure.The dependency of scale propagation,on temporal and spatial variations in production temperature and pressure is represented through a Poisson square wave process.On this basis,it is possible to model the scale growth probabilistically over the service lives of sub-sea wells-and this in turn forms the basis for optimal subsea well integrity management.The proposed probabilistic models are illustrated through a principal example,assessing the significance of model assumptions and the sensitivities of the probabilistic characteristics of scale formation with respect to the considered uncertainties.