SUMMARY Researchers base their analysis on basic drilling parameters obtained during mud logging and demonstrate impressive results. However, due to limitations imposed by data quality often present during drilling, those solutions often tend to lose their stability and high levels of predictivity. In this work, the concept of hybrid modeling was introduced which allows to integrate the analytical correlations with algorithms of machine learning for obtaining stable solutions consistent from one data set to another.

This article is continuation of geomechanical estimations for wellbore stability and horizontal wells drilling optimization to M formation of the Boca de Jaruco oilfield, Republic of Cuba. Following the completion of the first part of the work presented in the article SPE-196897-RU, the first horizontal wells were successfully drilled. In the first part of the work, the stress-state analysis and horizontal in-situ stresses directions were analyzed using pressure behavior during steam injections, cross-dipole acoustics, as well as data from oriented caliper from vertical wells. As a result, according to the recommendations, an additional study in the deviated sections of the wells, including core sampling and micro-imager logging were conducted. Presence of wellbore failures at the inclined sections allowed to use the method of inverse in-situ stress modeling based on image logs interpretation. The main result of the work were the determination of the stress state and horizontal stresses direction, as well as classification of the failures detected using the image log into groups. Moreover, the genesis of breakouts which had deviation from each other different that 180° were evaluated.

No abstract

No abstract

Successful well planning and stimulation in complex geological settings (especially in the horizontal wells and wells with a high degree of deviation) is bound with conducting geomechanical estimations. Identification of the stress regime, which is an imperative basis for the geomechanical modeling, can significantly alter the reservoir production scheme. Moreover, knowledge of the stress regime directly impacts the efficiency of hydraulic fracturing procedures and wellbore stability. For example, in case of reverse stress regime, the hydraulic fracturing operations could be inefficient due to the problems with the fracture initiation, high injection pressures, and risks associated with the proppant fallout in the wellbore. Fields experiencing hydraulic fracturing problems should be assessed via the geomechanical frame of reference for the comprehensive understanding of the issue. Assessing the state of stress is challenged by the absence of direct measurement tools of maximum horizontal stress. Application of the stress estimation methods commonly used in the industry (including the breakout width, acoustic anisotropy inversion and poroelastic modeling with the assumption of tectonic coefficients) have certain limitations which often lead to a broad range of obtained values of maximum horizontal stress, thus adding uncertainty to the drilling and hydraulic fracturing recommendations. Thus, the main goal of this work is to develop and apply an instrument for qualitative assessment of stress regime and direction. The reliable mathematical model, built upon the minimal set of required data, which is able to forecast the rock behavior under far-field and near wellbore stresses can be an extremely useful instrument for effective operations of drilling, fracturing, well placement and reservoir development. The underlying method for the development of the stress inversion algorithm was based on limiting the range of possible values of horizontal stresses using Anderson's definition of stress regimes, the frictional theory of Mohr-Coulomb and Kirch equations. The subsequent analysis of the breakout azimuths at the wellbore walls of several inclined wells from the image log data results in a reliable prediction of reservoir stress regime and direction. The optimal usage of the method required knowledge of vertical stress and the borehole failures logged in the deviated wells with the inclination of at least 20° and varying azimuthal directions. The developed algorithm of the improved identification of stress regime was then applied for a real field case in order to understand the geomechanical roots of the problems experienced during hydraulic fracturing treatment. Learning the stress regime and the orientation of the horizontal stresses allowed building reliable geomechanical models, necessary for the optimization of the hydraulic fracturing program and improvement of well operating efficiency. The conclusion upon the conducted work was that the methods of horizontal stress detection should be further studied for the cases where the data is not enough (for example no deviated bores logged). Moreover since all methods of SHmax estimation are inverse, the most value can be brought only by creating a tool where all the techniques (existing and the ones under development) are integrated.