Production Optimization: System Identification and Uncertainty Estimation

Abstract

1 Abstract Real-time optimization of oil and gas production requires a production model, which must be fitted to data for accuracy. A certain amount of uncertainty must typically be expected in production models fitted to data due to the limited information content in data. It is usually not acceptable to introduce additional excitation at will to reduce this uncertainty due to the costs and risks involved. The contribution of this paper is twofold. Firstly, this paper discusses estimation of uncertainty in production optimization resulting from fitting models to production data with low information content, a concept that has previously mainly been applied in reservoir management. Secondly, this paper illustrates how system identification can be used to find production models which can be solved with little computational effort and which are designed to be easily fitted to production data. The method is demonstrated on a synthetic example before being applied to a case study of a North Sea oil and gas field. In offshore oil and gas production, the suggested method is expected to have applications in the development of structured approaches to uncertainty handling, for instance excitation planning and real-time optimization under uncertainty. 2 Introduction Production in the context of offshore oil and gas fields, can be considered the total output of production wells, a mass flow of components including hydrocarbons, in addition to water, CO2, H2S, sand and possibly other components. Hydrocarbon production is for simplicity often lumped into oil and gas. Production travels as multiphase flow from wells through flow lines to a processing facility for separation, illustrated in Figure 1. Water and gas injection is used for optimizing hydrocarbon recovery of reservoirs. Gas lift can increase production to a certain extent by increasing the pressure difference between reservoir and well inlet. Multiphase flow rates are hard to measure. Measurements of total produced single phase oil and gas rates are usually available, and estimates of total water rates can often by found by adding different measured water rates after separation. To determine the rates of oil, gas and water produced from individual wells, the production of a single well is usually routed to a dedicated test separator where the rate of each separated component is measured. In single-rate well tests rates are only measured for the current setpoint, while rates are measured for several setpoints in multi-rate well tests. The total amount of oil, gas and water which can be separated and processed is constrained by the capacity of facilities, these capacities are themselves uncertain. Normally production is at setpoints where some of these capacities are at their perceived constraints, therefore a multi-rate well test cannot be performed without simultaneously reducing production at some other well, which may cause lost production and a cost. There is also a risk that changes in setpoints during testing may cause some part of the facilities to exceed the limits of safe operation, which may force an expensive shutdown and re-start of production. Well tests are only performed when a need for tests has been identified due to the costs and risks involved. Well tests are a form of planned excitation, some planned variation in one or more decision variables designed to reveal information on production through measurements. Production is constrained by several factors including, on the field level, the capacity of the facilities to separate components of production and the capacity of facilities to compress lift gas. The production of groups of wells may travel through shared flow lines or inlet separators which have a limited liquid handling capacity. The production of individual wells may be constrained due to slugging, other flow assurance issues or due to reservoir management constraints. In the context of oil and gas producing systems, real-time optimization has been defined as a process of measure-calculate-control cycles at a frequency which maintains the system's optimal operating conditions within the time-constant constraints of the system [1]. It has been suggested that real-time optimization could be divided into subproblems on different time scales to limit complexity, and to consider separately reservoir management, optimization of injection and reservoir drainage on the time scales of months and years, and production optimization, maximization of value from the daily production of reservoirs [1]. Reservoir management typically specifies constraints on production optimization to link these problems. The aim of production optimization is to determine setpoints for a set of chosen decision variables which are optimal by some criterion. These setpoints are implemented by altering the settings of production equipment. Decision variables may be any measured or computed variables associated with production which are influenced by changes in settings, but the number of decision variables is limited by the number of settings. We may for instance determine the settings of a gas lift chke by deciding a target lift gas rate, a target annulus pressure or a target gas lift choke settings and possibly by routing settings.

Oil and Gas Production Optimization; Lost Potential due to Uncertainty

“…In Bieker et al (2007) it has been proposed that uncertainty in production optimization should be related to production profit to allow structured business cases for uncertainty mitigation to be created. Elgsaeter et al (2008) suggested how uncertainty due to low information content in production data can be quantified using bootstrapping. No references have been found which discuss the value of information in the context of daily oil and gas production optimization.…”

“…where w[t] is a user-specified weighting function. When Z N has low information content, it may not be possible to determineθ uniquely from (12), resulting in an uncertainty inθ that can be quantified in terms of a probability density function f θ (θ)u s i n gbootstrapping,s e eE f r o na n d Tibshirani (1993) or Elgsaeter et al (2008). Bootstrapping is a computational approach which generates an estimate of uncertainty by re-solving (12) a number of times to fit the model to number of synthetically generated, or "re-sampled", data sets similar to (10).…”

“…A decline inq tot o was visible in the tuning set, and we chose to detrendq tot o and chose w[t] to weigh older measurements less than newer measurements when solving (12). For further details on the modeling, refer to Elgsaeter et al (2008). The set of indices of wells with flow assurance issues was W c = {1, 4, 10, 13, 14, 16}.…”

“…for all wells i and for oil, gas and water p ∈{ o, g, w}, based on Elgsaeter et al (2008). q i gl is gas lift rate and z i ∈ [0, 1] is the relative production valve opening of…”

Dynamic Real-time Optimization of Reservoir Production

“…In 2006, the method combining empirical formula with optimization theory is proposed by Saputelli [3,4], Elgsaeter [5], and Awasthi [6] and etc. Their goal focused on the short-term optimization of production and gained the optimized development program in time with convenient empirical formula.…”