Dynamic Well Modeling, Where are We?: Mahakam Operation Experience for Well Diagnostics & Optimization

Abstract

Since first time introduced in oil & gas well by Gilbert (1954) then popularized by Mach (1979) and Brown (1984) as Nodal Analysis™, system analysis is commonly used to analyze the performance of system composed of multiple interacting components. It is generally used for well diagnostic, performance matching, and prediction for optimization. In this concept, multiphase flow modeling in steady-state condition also introduced and widely used in industry to analyze flow performance inside wellbore with empirical or mechanistic approach. However by its nature, flow performance always as transient flow phenomenon and several cases could not be captured by steady-state model. Dynamic well modeling is needed to verify some of the conclusions from steady-state simulation therefore uncertainty could be reduced. Dynamic model features mechanistic & empirical approach with basic conservation equations (mass, energy, momentum), constitutive correlations and flow pattern transition modeling (Benidiksen et al, 1991; Staff et al., 2015). It is a powerful approach to generate model where dynamic-transient phenomena is become concern. Dynamic well modeling has been demonstrated in latest 5 years of Mahakam operation for specific purpose where steady-state well model which regularly used could not represent dynamic condition. Several studies are presented in this paper with certain objectives for well diagnostic, optimization, risk analysis or in general for petroleum engineering purpose: 1) Liquid loading prediction; 2) Well start-up prediction & analysis; 3) Static condition requirement analysis for well revival campaign; 3) Velocity String performance prediction; 4) Alternative water production estimation where direct measurement has limitations; 5) Wellhead shut-in pressure prediction with buffer zone effect to support unlocking high pressure reservoirs; 6) Blow-out modeling and dynamic well killing for blow-out contingency plan; 7) Unstable gas lift injection phenomena that impact to liquid production; 8) Liquid accumulation inside pipeline which impact to well back pressure, etc. The advantage and drawbacks from model implementation also presented as a practical reference to consider this approach. In summary, dynamic well modeling is beneficial in Mahakam operation, moreover in today's mature condition (operated more than 40 years) where the problem becomes more complex and better decision making is needed to consider efficient method with effective cost.