Permeability Predictions for Tight Sandstone Reservoir Using Explainable Machine Learning and Particle Swarm Optimization

High-precision permeability prediction is of great significance to tight sandstone reservoirs. However, while considerable progress has recently been made in the machine learning based prediction of reservoir permeability, the generalization of this approach is limited by weak interpretability. Hence, an interpretable XGBoost model is proposed herein based on particle swarm optimization to predict the permeability of tight sandstone reservoirs with higher accuracy and robust interpretability. The porosity and permeability of 202 core plugs and 6 logging curves (namely, the gamma-ray (GR) curve, the acoustic curve (AC), the spontaneous potential (SP) curve, the caliper (CAL) curve, the deep lateral resistivity (RILD) curve, and eight lateral resistivity (RFOC) curve) are extracted along with three derived variables (i.e., the shale content, the AC slope, and the GR slope) as data sets. Based on the data preprocessing, global and local interpretations are performed according to the Shapley additive explanations (SHAP) analysis, and the redundant features in the data set are screened to identify the porosity, AC, CAL, and GR slope as the four most important features. The particle swarm optimization algorithm is then used to optimize the hyperparameters of the XGBoost model. The prediction results of the PSO-XGBoost model indicate a superior performance compared with that of the benchmark XGBoost model. In addition, the reliable application of the interpretable PSO-XGBoost model in the prediction of tight sandstone reservoir permeability is examined by comparing the results with those of two traditional mathematical regression models, five machine learning models, and three deep learning models. Thus, the interpretable PSO-XGBoost model is shown to have more advantages in permeability prediction along with the lowest root mean square error, thereby confirming the effectiveness and practicability of this method.

A review of the various treatments of oil-based drilling fluids filter cakes

Treatment of the filter cake layer after drilling is essential for better cement integrity and to retain the original reservoir permeability. Compared to water-based filter cake, oil-based mud filter cake removal is more sophisticated as oil encloses the filter cake’s particles. Therefore, oil-based mud clean-up requires wettability alteration additives (mutual solvents and/or surfactants) for permitting acid/filter cake reaction. With an appropriate acid, microemulsions were reported to be very efficient in cleaning oil-based filter cakes, due to their low interfacial tension and high acid solubility. The objective of this paper is to provide an overview of the different techniques and treatment solutions utilized in oil-based filter cake clean-up. Furthermore, a synopsis of the various treatments for drilling fluids densified with different weighting materials is presented. Subsequently, the research limitations and opportunities have been highlighted for future work. In the light of the review that has been presented in this paper, it's recommended to conduct further investigation on some areas related to filter cake removal. The removal of filter cake formed from weighting materials other than barite, calcium carbonate, ilmenite, and manganese tetroxide needs to be investigated thoroughly. Additionally, the overall efficiency of oil-based mud removal needs to be studied under wide ranges of temperature, salinity, and pH. The utilization of surfactant-free microemulsions in filter cake treatment could also be investigated.

Challenges of Gas Wells Killing Operation with Emphasis on Reservoir & Completion Integrity Issues

Well killing always remains a most radical part during the life cycle of gas production wells with reservoir and completion integrity issues. In moderate permeability gas reservoirs, it will be more challenging due to below issues; Low pressure gas reservoir with moderate reservoir permeability where hydrostatic head of water is almost double the formation pressure Well with the sustainable annulus pressures (Production & first cemented annulus) Well with complex layered scale / asphalting deposition Completion jewelry component integrity breached Recycle reservoir with pressure maintenance Noise logs / corrosion logs generally conducted in order to assess the downhole completion jewelry and potential leak source prior any attempt for killing the well. To achieve the desired accessibility extensive scale analysis for better designing of scale clean out operation carried out specially to access the SPM. Variation of reservoir permeability considered for designing of optimized kill fluid for Depleted horizontal gas reservoir to cater challenge of complete losses. Effective fluid loss solutions designed and implemented to avoid abnormal fluid losses. Further more Polymer based gels used to kill and prevent the gas peculation to surface. Wells having completion and reservoir integrity issue isolated by considering cement zonal isolation, salt plugs, thru-tubing bridge plugs and nipple less plugs. All these barriers having their advantages and disadvantages with reference to work over objectives and their application limitations with respect to well conditions and detail study conducted for each candidate prior execution. Depletion Gas well killing and securing operation considered to be complex in nature and may result serious concern of rig intervention or well future objectives in case of improper execution. Gas wells having reservoir integrity issues and in case of 1st cemented Annulus pressure can be isolated by using thru-tubing bridge plugs. For retrieval of dummy from SPM must be done after setting of downhole plug to avoid any heavy suction for wire line operation. Cement plug operation is not suitable for such wells due to severe losses and fluid circulation limitation. Adequate selection of kill gel fluid as per reservoir characteristics will improve the killing efficiency.

Enhanced Geothermal Systems – Promises and Challenges

Geothermal energy plays a very important role in the energy basket of the world. However, understanding the geothermal hotspots and exploiting the same from deep reservoirs, by using advanced drilling technologies, is a key challenge. This study focuses on reservoirs at a depth greater than 3 km and temperatures more than 150°C. These resources are qualified as Enhanced Geothermal System (EGS). Artificially induced technologies are employed to enhance the reservoir permeability and fluid saturation. The present study concentrates on EGS resources, their types, technologies employed to extract energy and their applications in improving power generation. Studies on fracture stimulation using hydraulic fracturing and hydro shearing are also evaluated. The associated micro-seismic events and control measures for the same are discussed in this study. Various simulators for reservoir characterization and description are also analyzed and presented. Controlled fluid injection and super critical CO2 as heat transmission fluid are described for the benefit of the readers. The advantages of using CO2 over water and its role in reducing the carbon footprint are brought out in this paper for further studies.

The problem of fluid filtration through cement stone on gas fields with low reservoir permeability

Partial destruction of cement stone or an increase in its permeability during the development of oil and gas fields is a frequent occurrence. There are a huge number of natural and man-made factors that lead to an increase in the gas permeability of the cement ring. This article attempts to investigate the permeability of cement stone in active and abandoned wells. It presents an overview of experimental works devoted to determining the real permeability of cement stone on samples and according to the results of studying the state of wells in various conditions.

Low-Permeable Reservoirs as High Potential Assets for EGR

Natural gas fields that are being developed in Ukraine, mainly relate to the high and medium permeability reservoirs, most of which are at the final stage of field life. In this situation one of the main sources of additional gas production is unconventional fields. This paper presents the analysis of challenges concerning development of low-permeable reservoirs and experimental results of conducted research, which provide the opportunity to establish technologies for enhance gas recovery factor. For this purpose, a series of laboratory experiments were carried out on the sand packed models of gas field with different permeability (from 9.7 to 93 mD) using natural gas. The pressure in the experiments varied from 1 to 10 MPa, temperature – 22-60 °C. According to the features of low-permeable gas fields development the research of displacement desorption with carbon dioxide and inert gas stripping by nitrogen was conducted. These studies also revealed the influence of pressure, temperature, reservoir permeability and non-hydrocarbon gases injection rate on the course of adsorption-desorption processes and their impact on the gas recovery factor. According to the experimental results of relative adsorption capacity determination it can be concluded that the carbon dioxide usage as the displacement agent can lead to producing adsorbed gas by more than 30% than by using nitrogen. To remove the adsorbed gas just reservoir pressure lowering is not enough due to the nature of adsorption isotherms. Particularly at pressure decreasing by 8-10 times compared to initial reservoir pressure only about 30-40% of the total amount of initially adsorbed gas is desorbed. And at considerable reservoir pressure reduction the further deposit development is not economically profitable. According to the results it was found that in the case of nitrogen usage the most effective method is full voidage replacement at injection pressure of 0.8 of the initial reservoir pressure, and in case of carbon dioxide usage - full voidage replacement method at pressure of 0.6 of the initial reservoir pressure. Taking into account availability of N2 and CO2, N2injection is recommended for further implementation. The influence of displacement agent injection pressure on gas recovery was experimentally proved. The physical sense of the processes taking place during natural gas desorption stimulation by non-hydrocarbon gases was justified. The effect of temperature, pressure and reservoir permeability on methane adsorption capacity were determined. The mathematical model for estimating adsorbed gas amount depending on the reservoir parameters was developed. Obtained results were summarized and recommendations for practical implementation of elaborated technological solutions were suggested.