One-Dimensional Convolutional Neural Network with Adaptive Moment Estimation for Modelling of the Sand Retention Test

Stand-alone screens (SASs) are active sand control methods where compatible screens and slot sizes are selected through the sand retention test (SRT) to filter an unacceptable amount of sand produced from oil and gas wells. SRTs have been modelled in the laboratory using computer simulation to replicate experimental conditions and ensure that the selected screens are suitable for selected reservoirs. However, the SRT experimental setups and result analyses are not standardized. A few changes made to the experimental setup can cause a huge variation in results, leading to different plugging performance and sand retention analysis. Besides, conducting many laboratory experiments is expensive and time-consuming. Since the application of CNN in the petroleum industry attained promising results for both classification and regression problems, this method is proposed on SRT to reduce the time, cost, and effort to run the laboratory test by predicting the plugging performance and sand production. The application of deep learning has yet to be imposed in SRT. Therefore, in this study, a deep learning model using a one-dimensional convolutional neural network (1D-CNN) with adaptive moment estimation is developed to model the SRT with the aim of classifying plugging sign (screen plug, the screen does not plug) as well as to predict sand production and retained permeability using a varying sand distribution, SAS, screen slot size, and sand concentration as inputs. The performance of the proposed 1D-CNN model for the slurry test shows that the prediction of retained permeability and the classification of plugging sign achieved robust accuracy with more than a 90% value of R2, while the prediction of sand production achieved 77% accuracy. In addition, the model for the sand pack test achieved 84% accuracy in predicting sand production. For comparative model performance, gradient boosting (GB), K-nearest neighbor (KNN), random forest (RF), and support vector machine (SVM) were also modelled on the same datasets. The results showed that the proposed 1D-CNN model outperforms the other four machine learning models for both SRT tests in terms of prediction accuracy.

Understanding the Plugging Performance of HPAM-Cr (III) Polymer Gel for CO2 Conformance Control

Summary With the demand for conformance control in carbon dioxide (CO2) flooding fields, hydrolyzed polyacrylamide-chromium [HPAM-Cr (III)] polymer gel has been applied in fields for CO2 conformance control. However, the field application results are mixed with success and failure. This paper is intended to understand the HPAM-Cr (III) polymer gel plugging performance in CO2 flooding reservoirs through laboratory experiments and numerical analysis. We conducted core flooding tests to understand how the cycles of CO2 and water affect the HPAM-Cr (III) polymer gel plugging efficiency to CO2 and water during a water-alternating-gas (WAG) process. Berea Sandstone cores with the permeability range of 107 to 1225 md were used to evaluate the plugging performance in terms of residual resistance factor and breakthrough pressure, which is the minimum pressure required for CO2 to enter the gel-treated cores.We compared the pressure gradient from the near-wellbore to far-field with the gel breakthrough pressure, from which we analyzed under which conditions the gel treatment could be more successful. Results show that HPAM-Cr (III) polymer gel has higher breakthrough pressure in the low-permeability cores. The polymer gel can reduce the permeability to water much more than that to CO2. The disproportionate permeability reduction performance was more prominent in low-permeability cores than in high-permeability cores. The gel resistance to both CO2 and brine significantly decreased in later cycles. In high-permeability cores, the gel resistance to CO2 became negligible only after two cycles of water and CO2 injection. Because of the significant reduction of pressure gradient from near-wellbore to far-field in a radial flow condition and the dependence of breakthrough pressure on permeability and polymer concentration, we examined hypothetical reservoirs with no fractures, in which impermeable barriers separated high- and low-permeability zones and in which the gel was only placed in the high-permeability zone. We considered two scenarios: CO2 breaking through the gel and no CO2 breakthrough. No breakthrough represents the best condition in which the gel has no direct contact and can be stable in reservoirs for long. In contrast, the breakthrough scenario will result in the gel’s significant degradation and dehydration resulting from CO2 flowing through the gel, which will cause the gel treatment to fail.

Experimental Research on the Optimization and Evaluation of the Polymer/Chromium-Ion Deep Profile Control System for the Fractured Low-Permeability Reservoirs

The matrix/fracture conductivity of a fractured low-permeability reservoir is variable, and its heterogeneity is serious. When carrying out deep profile control measures, it is difficult to inject under the premise of ensuring the plugging effect. According to the characteristics of the fractured low-permeability reservoir in Chaoyanggou Oilfield, the polymer/chromium ion deep profile control system was optimized via a viscosity evaluation experiment, liquidity experiment and oil displacement experiment. The experimental results show that the high molecular weight main agent/low concentration system and low molecular weight main agent/high concentration system can meet the gel strength requirement. The evaluation results of the injection ability and plugging performance of the fractured low-permeability core show that a high molecular weight profile control system is difficult to inject, while a low molecular weight profile control system has a poor plugging performance and high cost after simulated shear. Therefore, the formulation of the profile control system was determined to be a polymer with a molecular weight of 16 million g·mol−1 as the main agent with a concentration of 1000–1500 mg·L−1. As assisting agents, the concentrations of thiourea, NaCl and NaHCO3 were 900 mg·L−1, 800 mg·L−1 and 700 mg·L−1, respectively. The plugging rate of the system was 87.6%, and the resistance coefficient was 19.2. Finally, a fractured low-permeability core model with parallel long cores was designed, and the optimal profile control system was used for the core oil-displacement experiment. Compared with the water-flooding experiment, the plugging rate can be increased by 6.9–8.0%.

Pore-Scale Investigation on the Plugging Behavior of Submicron-Sized Microspheres for Heterogeneous Porous Media with Higher Permeability

Reservoir heterogeneity is regarded as one of the main reasons leading to low oil recovery for both conventional and unconventional reservoirs. High-permeability layers or fractures could result in ineffective water or gas injection and generate nonuniform profile. Polymer microspheres have been widely applied for the conformance control to overcome the bypass of injected fluids and improve the sweep efficiency. For the purpose of examining the plugging performance of submicron-sized microspheres in high-permeability porous media, systematic investigations were implemented incorporating macroscale blocking rate tests using core samples and pore-scale water migration analysis via nuclear magnetic resonance (NMR). Experimental results indicate that microsphere particle size dominates the plugging performance among three studied factors and core permeability has the least influence on the plugging performance. Subsequently, microsphere flooding was conducted to investigate its oil recovery capability. Different oil recovery behaviors were observed for cores with different permeability. For cores with lower permeability, oil recovery increased stepwise with microsphere injection whereas for higher permeability cores oil recovery rapidly increased and reached a plateau. This experimental work provides a better understanding on the plugging behavior of microspheres and could be employed as a reference for screening and optimizing the microsphere flooding process for profile control in heterogeneous reservoirs.

Performance Evaluation of Degradable Temporary Plugging Agent in Laboratory Experiment

Temporary plugging fracturing is an effective way to enhance the fracture complexity and increase the stimulated reservoir volume (SRV) of unconventional reservoirs. The performance of temporary plugging agents (TPA) directly affects the success rate of temporary plugging. Currently, laboratory evaluation of the plugging effects of the TPA is rarely reported, and there are no industrial standards on laboratory evaluation of TPA plugging. In this study, two new experimental methods were used to evaluate a novel particulate TPA. The plugging performance of the TPA to the core end face and the propped fractures was measured through displacement experiments of cores, and the applicability of its basic performance to the temporary plugging fracturing was verified. Furthermore, the large-scale true triaxial simulation experiment of temporary plugging fracturing was carried out to confirm the influence mechanism of different factors on fracture propagation during temporary plugging. Finally, the influence rule of different types of combinations of TPA and placement patterns on the plugging was obtained based on laboratory evaluation of the conductivity. The results show that the novel TPA causes effective temporary plugging on the core end face and the propped fractures and has the strong plugging performance, and the TPA solubility in the carrying fluids decreases with the increase in the TPA concentration. The basic performance of the TPA meets the requirements of temporary plugging fracturing. If the proppants and 20% fibers are placed within the fracture in the mixed pattern, the fracture is initiated along the direction of the horizontal maximum principal stress. The preset fracture reduces the fracture initiation pressure. The fracture complexity is closely related to the placement pattern of TPA and proppants. If the preset fractures are filled by the uniform mixture or the plug of the 20/40 mesh or 20/80 mesh particulate TPA (4%), fibers (1%), and proppants, the fracture initiation pressure significantly increases, and the complex fractures are formed after fracturing. Effective plugging cannot be formed only by mixing the fibers with the proppants, and the uniform mixture of the proppants and 4% particulate TPA and the 6% particulate TPA at the front end of the fracture form a temporary plugging belt, achieving effective plugging. The fibers improve the conductivity under the low closure stress, and it has a certain effect of temporary plugging under the closure stress above 30 MPa. The research results provide the design consideration for creating the complex fracture by temporary plugging.