Uncertainty Driven Formation Damage Control Using Analytical Technique

Abstract

The aim of this work is to develop an analytical technique for characterizing formation damage. The oil reservoir of the East Siberian Yaraktinskoe field suffers from salt and organic scales precipitation leading to skin damage. Besides, injection water has sulfates, which precipitate as gypsum in the near wellbore area of production wells and at bottomhole. Historically pressure build-ups (PBU) were used to characterize the evolution and extent of the damage. The use of PBUs leads to the shut in of production. Additionally PBUs in the reservoir provide conclusive results in no more than 22% cases. Based on inconsistent results from PBUs and their cost in production losses, it was of interest to find a better and preferable technique for formation damage control using existing data. The result of that initiative is analytical technique that provides dimensionless productivity index (Jd) range monitoring over time, Jd range comparison to the technical potential and identification of the performance gap range. By identifying the performance gap range, stimulation actions are ordered reestablishing oil production, productivity index (PI) and Jd. The technique is based on transmissibility (kh/µB or T) model derived from Kamal and Pan study (2010) and reservoir pressure (Pres or P) model. Stochastic part of the technique is provided by T and Pres error functions. The functions are probability distribution functions (PDF) derived from comparison of the modeled T and Pres with well test measured historical values. Using this T and Pres models and historical data of liquid rates and bottomhole pressures (BHP), we can calculate current and historical Jd, Jd drop relative to historical performance or potential and oil rate potential increment with uncertainty margins (10th, 50th and 90th percentile or P10-50-90). The margins are calculated from 10000 stochastic iterations of T and Pres within the PDFs of their error. The technique has enabled to find 14 stimulation candidates during 6 month of use. Overall, 15 stimulations were implemented since one well was stimulated twice. Ten of 14 stimulations increased oil production rate by 4161 bbl/day. Five stimulations were economically unsuccessful due to inappropriate stimulation technology implementation. The technique shows acceptable uncertainty level to make efficient and appropriate decisions for the appropriately chosen stimulation technology. Modeled P50 PIs have good match with more than 85% correlation with well test measured PIs after economically successful stimulation. New analytical technique is presented here, which can be utilized as an automatic process without repeating well tests for routine generation of accurate stimulation plan with numerical assessment of success probability and anticipated oil rate increment uncertainty range. Realization of stimulation potential is simplified to the task of appropriate treatment technology selection and implementation for the candidates from the rating.