سيدفرزين ميره بيگي جاماسبي

Matrix acidizing operations in carbonate reservoirs face fundamental challenges in design an‎d implementation, mainly due to reliance on inaccurate laboratory data an‎d the complexity of transferring laboratory data to the actual reservoir scale. This mismatch can cause deviations from optimal conditions, increase acid consumption costs, an‎d reduce operation efficiency. Therefore, in this study, an attempt has been made to develop an algorithm for online optimization of acidizing operations using well injection data, including coil injection rate an‎d well pressure during injection, an‎d using various models to simulate the behavior of the reservoir (areas around the well), well string, an‎d skin factor prediction models. The conceptual an‎d mathematical framework of the algorithm is based on the mechanistic model of Ali an‎d Ziauddin (2020) an‎d using the inverse solution approach, an‎d its efficiency has been validated through a synthetic case study. Skin factor prediction models usually include parameters that are a function of reservoir rock conditions, injection rate, rock reactivity coefficient, acid type, reservoir temperature an‎d pressure, porosity an‎d permeability, an‎d other petrophysical properties. The developed algorithm allows these parameters to be calculated online based on well data (which are always available) an‎d then the optimal injection rate is determined continuously. The necessary comman‎ds are then given to the coil to adjust the injection rate an‎d the injection process will be in optimal conditions. In this way, using this system, the acid treatment operation is optimized online; in such a way that the acid consumption volume is optimized, the appropriate wormhole structure is formed, the damages around the well are bypassed, an‎d ultimately, the well production an‎d productivity will be increased. The results show that the algorithm is able to scale an‎d calibrate the uncertain model parameters with high accuracy using only 3.5% of the initial operation data an‎d even in the presence of 20% noise. Comparison of these values with laboratory data showed that online calibration is not only useful but also essential for achieving optimal design. It was also found that the optimal injection rate is not a fixed value but changes dynamically with the growth of the wormhole. The proposed algorithm was able to achieve rapid learning of the actual behavior of the reservoir in the first phase of operation an‎d then, during the optimization phase, provide the optimal dynamic injection path. This approach led to rapid removal of the initial damage, efficient management of the excitation phase an‎d achievement of the target wormhole length (0.6 m) with minimal acid consumption. In addition, the developed system was able to identify the exact moment of achieving the technical goal an‎d stop the operation without wasting resources. The results also indicated the modelʹs ability to reproduce key effects, including the role of temperature on acidification efficiency, an‎d its complete agreement with independent laboratory data.

كرم‌چاله , ضريب پوسته , فشار , نرخ تزريق , اسيد

wormhole , skin factor , pressure , Injection Rate , acid

seyed farzin mirehbeygi jamasbi

Dr. miri