The intersection of the pre-existing natural fractures (NF) with the hydraulic fractures (HF) forms a complex fracture network, which allows for increased well productivity. However, the complex fracture geometry may adversely affect the proppant placement and its transportation inside the fractures. The study reports the results of the Lattice numerical simulations of proppant placement at the intersection of fractures. Sensitivity analysis was performed to investigate the effect of pre-existing fracture friction angle, angle of approach as well as differential horizontal stress on HF-NF interaction mechanisms and proppant transport and placement. The results present the effect of these parameters on proppant transport and placement in NF.

In this study, a special focus was dedicated to the effect of elastic anisotropy of shales on the in-situ stress contrast between different layers and its implications on the vertical containment of hydraulic fractures (HF) and how they relate to the widely observed fracture driven interaction (FDI) phenomena and undesirable HF height growth. The reported elastic and mechanical properties of the main members of the Bakken petroleum system in the Williston Basin (i.e. Upper and Lower Bakken Shale, Middle Bakken, and Three Forks) (Ellafi et al., 2019) was used to estimate the in-situ stresses based on anisotropic rock properties and use the minimum horizontal stress profile for HF modeling. The estimated stress profile appeared to be very different from the one calculated based on the isotropic formation assumption. The anisotropic stress model, as reported by other researchers, is more realistic in transversely isotropic rocks and rock with a high volume of clay and TOC and generated more reliable results that conform better with other indicators and observations from other types of data associated with HF geometry.

One of the significant unconventional oil reserves in the USA is the Bakken Petroleum System located in the Williston Basin. It is known for its complex lithology, composed of three prominent members, Upper and Lower Bakken, with similar properties of organic-rich shale relatively uniform compared to the middle member with five distinct lithofacies, formed mainly from calcite, dolomite, or silica. The higher properties variability makes the reservoir characterization more challenging with low permeability and porosity. Understanding lithology by quantifying mineralogy is crucial for accurate geological modeling and reservoir simulation. Besides that, the reservoir’s capacity and the oil production are affected by the type and the mineral volume fractions, which impact the reservoir properties. Conventionally, to identify the mineralogy of the reservoir, the laboratory analysis (X-Ray Diffraction, XRD) using core samples combined with the well logs interpretation is widely used. The unavailability of the core data due to the high cost, as well as the discontinuities of the core section of the reservoir due to the coring failures and the destructive operations, are one of the challenges for an accurate mineralogy quantification. The XRD cores analysis is usually used to calibrate the petrophysical evaluation using well logs data because they are economically efficient. To remedy to these limitations, artificial intelligence and data-driven based models have been widely deployed in the oil and gas industry, particularly for petrophysical evaluation. This study aims to develop machine learning models to identify mineralogy by applying six different machine learning methods and using real field data from the upper, middle, and lower members of the Bakken Formation. Efficient pre-processing tools are applied before training the models to eliminate the XRD data outliers due to the formation complexity. The algorithms are based on well logs as inputs such as Gamma Ray, bulk density, neutron porosity, resistivity, and photoelectric factor for seven (07) wells. XRD mineral components for 117 samples are considered outputs (Clays, Dolomite, Calcite, Quartz, and other minerals). The results' validation is based on comparing the XRD Data prediction from the developed models and the petrophysical interpretation. The applied approach and the developed models have proved their effectiveness in predicting the XRD from the Bakken Petroleum system. The Random Forest Regressor delivered the best performance with a correlation coefficient of 78 percent. The rest of the algorithms had R-scores between 36 and 72 percent, with the linear regression having the lowest coefficient. The reason is the non-linearity between the inputs and outputs.

Sand production is one of the major problems in many oil and gas assets around the world. Uncontrollable sand production can affect hydrocarbon recovery and increase operational costs. This paper aims to develop a classification approach to suggest the optimal sand control method using machine learning algorithms. Four different models have been used, namely K-Nearest Neighbors, Support Vector Machine, Random Forest and Decision Tree. After extensive exploratory data analysis, nine parameters were included in the model: Sorting coefficient D10/D95, Mass fraction smaller than 44 micron, Well deviation angle through the pay zone, Pay-zone true vertical thickness, Bottom hole pressure, Maximum oil rate, Maximum gas rate, Permeability, Well type. By comparing the different models in this study Random Forest classifier achieved the highest evaluation metrics: f1-score of 0.9568, precision of 0.9580, and recall of 0.9568 respectively. These results combined with the confusion matrix to assess the model performance have shown that up to a certain level machine learning methods can ensure the adequate completion for the candidate well. The proposed work turns out to be a potential approach that rises to the level of a decision support tool and thus can help engineers set the right completion option. Introduction Sand production has always been one of the major concerns for the oil and gas industry. It is the result of the migration of failed sand grains because of the drag forces caused by fluids flow from an unconsolidated reservoir. Generally, this phenomenon leads to technical problems such as erosion of downhole equipment and surface facilities, production loss, well access obstruction, and economic effects represented by the additional cost of sand disposal. Sand control is, therefore, necessary to mitigate the effect of sand production on hydrocarbon wells. Judhan (2016) defined sand control as all the techniques that allow hydrocarbons production without sand grains production in the wellbore. Sand control consist of two methods: passive and active. Passive sand control methods are generally used to reduce sand during the first stage of production when the sanding rate is low, as this latter gradually increase, active sand control will be required.

Gas lift is one of the most commonly used artificial lift method in oil-producing wells. However, the technique requires constant optimization of gas allocation to maximize profit. The Gas Lift Performance Curves (GLPC) are the main design element that is used for optimized injection. Several authors have proposed models to fit the GLPC. These curves are generated by modeling wells in a multiphase steady-state simulator. Once the model is built, a sensitivity analysis is run, and the curves are generated. In this work, The common workflow to generate GLPC was followed. Then, a new correlation for GLPC was suggested. The correlation outperforms all the models in the literature in terms of R-score and root mean square error. The correlation was then used to formulate a case study for four wells located in North Africa. First, the wells and PVT models were used to create a simulation. Once the simulation was calibrated, a sensitivity analysis of the gas lift injection rate was run. The new correlation was used to fit the GLPC. The optimization problem was mathematically formulated, and stochastic optimization techniques were used, noting Grey Wolf Optimization (GWO) Algorithm and Genetic Algorithm (GA) to obtain the global optimum of the distribution of a limited gas lift quantity. Both algorithms’ results were compared. GWO slightly outperformed GA. The advantages of GWO over GA were discussed, and the optimum gas allocation was obtained.