Gel-Volume Optimization in Water Conformance Applying Risk and Uncertainty Analysis

Abstract

Gels have been used successfully in water conformance. One of the keyfactors in the application of this technology is the volume of gel, which hasbeen estimated by several techniques: a) using a relationship between thevolume and the water rate, b) calculating it given a fixed volume per feet, orc) invading a chosen penetration radius.The present work shows a newapproach to estimate the optimum volume of gel considering uncertainties andrisks. The proposed approach considers a model for the expected water and oilproductivity indexes integrated with nodal and economical analysis. Clearly allthese three major steps depend on the volume of gel, whose optimum value is thegoal of this research, then an iterative method is required. Besides theproposed method used should consider the lack of accuracy in the inputparameters, some of them highly sensitive.Three real cases are analyzedin order to show the potential of this approach. The results aid to establishpriorities to answer the following questions: Where to focus the efforts inField data and research?, What is the accuracy required in the inputparameters?, How much would an activity improve the profit of the treatment?Even though the methodology and the software prototype considered only threewater production problems, the general idea can be easily applied in othercases such as channeling in water injection wells. Introduction Water conformance refers to the application of appropriate processes inreservoirs and boreholes to reduce water production, enhance oil recoveryefficiency, or satisfy a broad range of reservoir management as well aseconomical and environmental objectives. An effective Conformance solution requires the following steps to handle theproblem:Determining the "source" of water;Selecting the treatment,Achiving a proper field operation and,Monitoring and evaluation [1]. This paper focuses in the optimization of the gel injection treatmentsdealing with watered-out layer without crossflow and fractures or faults from awater layer. In this work the optimum volume of gel was estimated by considering andintegrating models to calculate the following parameters:Radius of gel penetration and the corresponding expected future water andoil productivity indexes.Performance of the artificial lift system for the new water cut (expectedproduction by nodal analysis).Net present value (economical evaluation) of the treatment for a period oftime (treatment lifetime). Each parameter can take different input values, some of them are verysensitive and have associated a considerable uncertainty. For example; Fractureareas and the residual resistance factor affect the water and oil productivityindexes, both have considerable uncertainty; therefore, there is a possiblerange for the production expected in stead of a single value. The evaluation of the risk of a future gel injection project is not a simpletask because the events have not happened yet. To quantify the risk in thiscase, probabilistic and non-probabilistic models could be used. The probabilistic models set or establish the probability of any possibleevent based on statistical methods considering random variations. Two commonprobabilistic and economical indicators to evaluate a project are the expectednet present value and its corresponding variance easily obtained trough MonteCarlo simulation [2]. The non-probabilistic models are usually based on the perception of theexpert complemented with a methodology such as sensitivity analysis, tornadocharts and error propagation. The current software prototype only includes Monte Carlo to take intoaccount the uncertainties in the input parameters. Therefore, the main twosources of error in the current prototype are the physical models applied andthe uncertainty of the input parameters.