Stuck pipe incidents are one of the contributors to non-productive time (NPT), where they can result in a higher well cost. This research investigates the feasibility of applying machine learning to predict events of stuck pipes during drilling operations in petroleum fields. The predictive model aims to predict the occurrence of stuck pipes so that relevant drilling operation personnel are warned to enact a mitigation plan to prevent stuck pipes. Two machine learning methodologies were studied in this research, namely, the artificial neural network (ANN) and support vector machine (SVM). A total of 268 data sets were successfully collected through data extraction for the well drilling operation. The data also consist of the parameters with which the stuck pipes occurred during the drilling operations. These drilling parameters include information such as the properties of the drilling fluid, bottom-hole assembly (BHA) specification, state of the bore-hole and operating conditions. The R programming software was used to construct both the ANN and SVM machine learning models. The prediction performance of the machine learning models was evaluated in terms of accuracy, sensitivity and specificity. Sensitivity analysis was conducted on these two machine learning models. For the ANN, two activation functions—namely, the logistic activation function and hyperbolic tangent activation function—were tested. Additionally, all the possible combinations of network structures, from [19, 1, 1, 1, 1] to [19, 10, 10, 10, 1], were tested for each activation function. For the SVM, three kernel functions—namely, linear, Radial Basis Function (RBF) and polynomial—were tested. Apart from that, SVM hyper-parameters such as the regularization factor (C), sigma (σ) and degree (D) were used in sensitivity analysis as well. The results from the sensitivity analysis demonstrate that the best ANN model managed to achieve an 88.89% accuracy, 91.89% sensitivity and 86.36% specificity, whereas the best SVM model managed to achieve an 83.95% accuracy, 86.49% sensitivity and 81.82% specificity. Upon comparison, the ANN model is the better machine learning model in this study because its accuracy, sensitivity and specificity are consistently higher than those of the best SVM model. In conclusion, judging from the promising prediction accurateness as demonstrated in the results of this study, it is suggested that stuck pipe prediction using machine learning is indeed practical.

The present study investigates the effect of TiO₂ nanoparticles on the stability of Enhanced Oil Recovery (EOR)-produced stable emulsion. The chemical precipitation method is used to synthesize TiO₂ nanoparticles, and their properties were determined using various analytical characterization techniques such as X-ray Diffraction (XRD), High-Resolution Transmission Electron Microscopy (HRTEM), and Field Emission Scanning Electron Microscopy (FESEM). The effect of TiO₂ nanoparticles is evaluated by measuring oil/water (o/w) separation, rag layer formation, oil droplet size, and zeta potential of the residual EOR produced emulsion. The laser scattering technique is used to determine the o/w separation. The results showed that spherical-shaped anatase phase TiO₂ nanoparticles were produced with an average particle size of 122 nm. The TiO₂ nanoparticles had a positive effect on o/w separation and the clarity of the separated water. The separated aqueous phases’ clarity is 75% and 45% with and without TiO₂ nanoparticles, respectively. Laser scattering analysis revealed enhanced light transmission in the presence of TiO₂ nanoparticles, suggesting higher o/w separation of the ASP-produced emulsion. The overall increase in the o/w separation was recorded to be 19% in the presence of TiO₂ nanoparticles, indicating a decrease in the stability of ASP-produced emulsion. This decrease in the stability can be attributed to the improved coalescence’ action between the adjacent oil droplets and improved behavior of o/w interfacial film. An observable difference was found between the oil droplet size before and after the addition of TiO₂ nanoparticles, where the oil droplet size increased from 3 µm to 35 µm. A similar trend of zeta potential is also noticed in the presence of TiO₂ nanoparticles. Zeta potential was −13 mV to −7 mV, which is in the unstable emulsion range. Overall, the o/w separation is enhanced by introducing TiO₂ nanoparticles into ASP-produced stable emulsion.

Data-driven models with some evolutionary optimization algorithms, such as particle swarm optimization (PSO) and ant colony optimization (ACO) for hydraulic fracturing of shale reservoirs, have in recent times been validated as one of the best-performing machine learning algorithms. Log data from well-logging tools and physics-driven models is difficult to collate and model to enhance decision-making processes. The study sought to train, test, and validate synthetic data emanating from CMG’s numerically propped fracture morphology modeling to support and enhance productive hydrocarbon production and recovery. This data-driven numerical model was investigated for efficient hydraulic-induced fracturing by using machine learning, gradient descent, and adaptive optimizers. While satiating research curiosities, the online predictive analysis was conducted using the Google TensorFlow tool with the Tensor Processing Unit (TPU), focusing on linear and non-linear neural network regressions. A multi-structured dense layer with 1000, 100, and 1 neurons was compiled with mean absolute error (MAE) as loss functions and evaluation metrics concentrating on stochastic gradient descent (SGD), Adam, and RMSprop optimizers at a learning rate of 0.01. However, the emerging algorithm with the best overall optimization process was found to be Adam, whose error margin was 101.22 and whose accuracy was 80.24% for the entire set of 2000 synthetic data it trained and tested. Based on fracture conductivity, the data indicates that there was a higher chance of hydrocarbon production recovery using this method.

Several technical factors contribute to the flow of cuttings from the wellbore to the surface of the well, some of which are fundamentally due to the speed and inclination of the drill pipe at different positions (concentric and eccentric), the efficacy of the drilling mud considers plastic viscosity (PV) and yield point (YP), the weight of the cuttings, and the deviation of the well. Moreover, these overlaying cutting beds breed destruction in the drilling operation, some of which cause stuck pipes, reducing the rate of rotation and penetration. This current study, while it addresses the apropos of artificial intelligence (AI) with symmetry, employs a three-dimensional computational fluid dynamic (CFD) simulation model to validate an effective synthetic-based mud-drilling and to investigate the potency of the muds’ flow behaviours for transporting cuttings. Furthermore, the study examines the ratio effects of YP/PV to attain the safe transport of cuttings based on the turbulence of solid-particle suspension from the drilling fluid and the cuttings, and its velocity–pressure influence in a vertical well under a concentric and eccentric position of the drilling pipe. The resulting CFD analysis explains that the YP/PV of SBM and OBM, which generated the required capacity to suspend the cuttings to the surface, are symmetric to the experimental results and hence, the position of the drill pipe at the concentric position in vertical wells required a lower rotational speed. A computational study of the synthetic-based mud and its potency of not damaging the wellbore under an eccentric drill pipe position can be further examined.

In the common practice of geotechnical design, the state of soil is considered fully saturated, while recent examples show the importance of the consideration of unsaturated conditions as well due to the location of the groundwater table at deeper levels. However, it is problematic to design a foundation and calculate bearing capacity as the internal stress in unsaturated conditions needs to be quantified. The mechanics of unsaturated soils are critical in geotechnical engineering, especially when the case of the foundation design is required. This study aims to investigate the influence of suction on the shaft capacity of pile foundations based on the data obtained from in situ and laboratory testing. The three modified methods (i.e., modified α, β, and λ) were implemented in this study in order to calculate the shaft capacity of the pile foundation incorporating unsaturated soil mechanics principles. In addition, the effect of varying matric suction on shaft capacity was discussed. The results of obtained changes show the importance of rain infiltration in the design of piles in unsaturated conditions. It can be concluded that the most conservative method is found to be the modified β method in designing pile foundations incorporating the suction changes due to rainwater infiltration. However, cost optimization shall be considered while choosing the method for the design. As for the increase in GWT, the modified $\lambda$ method depicted the most conventional behavior. The possible rain infiltration decreases matric suction, mostly in the upper layers.