Prediction of Rate of Penetration for wells at Nam Con Son basin using Artificial Neural Networks models

The rate of penetration (ROP) is an important parameter that affects the success of a drilling operation. In this paper, the research approach is based on different artificial neural network (ANN) models to predict ROP for oil and gas wells in Nam Con Son basin. The first is the process of collecting and evaluating drilling parameters as input data of the model. Next is to find the network model capable of predicting ROP most accurately. After that, the study will evaluate the number of input parameters of the network model. The ROP prediction results obtained from different ANN models are also compared with traditional models such as the Bingham model, Bourgoyne & Young model. These results have shown the competitiveness of the ANN model and its high applicability to actual drilling operations.

Machine learning algorithm for prediction of stuck pipe incidents using statistical data: case study in middle east oil fields

One of the most troublesome issues in the drilling industry is stuck drill pipes. Drilling activities will be costly and time-consuming due to stuck pipe issues. As a result, predicting a stuck pipe can be more useful. This study aims to use an artificial intelligence technology called hybrid particle swarm optimization neural network (PSO-based ANN) to predict the probability of a stuck pipe in a Middle East oil field. In this field, a total of 85 wells were investigated. Therefore, to predict this problem, we must examine and determine the role of drilling parameters by creating an appropriate model. In this case, an artificial neural network is used to solve and model the problem. In this way, by processing the parameters of wells with and without being stuck in this field, the stuck or non-stuck of drilling pipes in future wells is predicted. To create a PSO-based ANN model database, mud characteristics, geometry, hydraulic, and drilling parameters were gathered from well daily drilling reports. In addition, two databases for directional and vertical wells were established. There are two types of datasets used for each database: stuck and non-stuck. It was discovered that the PSO-based ANN model could predict the incidence of a stuck pipe with an accuracy of over 80% for both directional and vertical wells. This study divided data from several cases into four sections: 17 ½″, 12 ¼″, 8 ½″, and 6 1/8″. The key reasons for sticking and the mechanics have been thoroughly investigated for each section. The methodology presented in this paper enables the Middle East drilling industry to estimate the risk of stuck pipe occurrence during the well planning procedure.

Experimental Investigations into Assessment of Thrust Force and Temperature in Bone Drilling

Drilling of bone is a challenging task for surgeons due to its effect on bone tissues. During drilling, it is noted that the temperature of bone increases. This increase in temperature if above 47°C causes thermal necrosis. Experiments were conducted to study the effect of input drilling parameters and drill bit parameters on bone health. To plan experiments a full factorial design method was used. An analysis is done on the effect of input parameters on thrust force and temperature of bone. The analysis of results shows an increase in thrust force and temperature when the feed rate increases and the spindle speed decreases. Further, the analysis of results shows an increase in thrust force and temperature when point angle increases and helix angle decreases. The increase in thrust force results in temperature rise. Scanning electron microscopy is done to analyze the surface topography of drilled hole. SEM image analysis shows an increase micro-crack in the drilled area when the thrust force and temperature increases.

Sensitivity Analysis of GFRP Composite Drilling Parameters and Genetic Algorithm-Based Optimisation

In this article, a genetic algorithm (GA) is used for optimizing a metamodel of surface roughness (R_a ) in drilling glass-fibre reinforced plastic (GFRP) composites. A response surface methodology (RSM) based three levels (-1, 0, 1) design of experiments is used for developing the metamodel. Analysis of variance (ANOVA) is undertaken to determine the importance of each process parameter in the developed metamodel. Subsequently, after detailed metamodel adequacy checks, the insignificant terms are dropped to make the established metamodel more rigorous and make accurate predictions. A sensitivity analysis of the independent variables on the output response helps in determining the most influential parameters. It is observed that f is the most crucial parameter, followed by the t and D. The optimization results depict that the R_a increases as the f increases and a minor value of drill diameter is the most appropriate to attain minimum surface roughness. Finally, a robustness test of the predicted GA solution is carried out.

Integrating Deep Learning and Physics-Based Models for Improved Production Prediction in Unconventional Reservoirs

The physics of fluid flow and transport processes in hydraulically fractured unconventional reservoirs are not well understood. As a result, the predicted production behavior using conventional simulation often does not agree with the observed field performance data. The discrepancy is caused by potential errors in the simulation model and the physical processes that take place in complex fractured rocks subjected to hydraulic fracturing. Additionally, other field data such as well logs and drilling parameters containing important information about reservoir condition and reservoir characteristics are not conveniently integrated into existing simulation models. In this paper, we discuss the development of a deep learning model to learn the errors in simulation-based performance prediction in unconventional reservoirs. Once trained, the model is expected to forecast the performance response of a well by augmenting physics-based predictions with the learned prediction errors from the deep learning model. To learn the discrepancy between simulated and observed production data, a simulation dataset is generated by using formation, completion, and fluid properties as input to an imperfect physics-based simulation model. The difference between the resulting simulated responses and observed field data, together with collected field data (i.e. well logs, drilling parameters), is then used to train a deep learning model to learn the prediction errors of the imperfect physical model. Deep convolutional autoencoder architectures are used to map the simulated and observed production responses into a low-dimensional manifold, where a regression model is trained to learn the mapping between collected field data and the simulated data in the latent space. The proposed method leverages deep learning models to account for prediction errors originating from potentially missing physical phenomena, simulation inputs, and reservoir description. We illustrate our approach using a case study from the Bakken Play in North Dakota.

Predict Geomechanical Parameters with Machine Learning Combining Drilling Data and Gamma Ray

The aim of this work is the development of a fast and reliable method for geomechanical parameters evaluation while drilling using surface logging data. Geomechanical parameters are usually evaluated from cores or sonic logs, which are typically expensive and sometimes difficult to obtain. A novel approach is here proposed, where machine learning algorithms are used to calculate the Young's Modulus from drilling parameters and the gamma ray log. The proposed method combines typical mud logging drilling data (ROP, RPM, Torque, Flow measurements, WOB and SPP), XRF data and well log data (Sonic logs, Bulk Density, Gamma Ray) with several machine learning techniques. The models were trained and tested on data coming from three wells drilled in the same basin in Kuwait, in the same geological units but in different reservoirs. Sonic logs and bulk density are used to evaluate the geomechanical parameters (e.g. Young's Modulus) and to train the model. The training phase and the hyperparameter tuning were performed using data coming from a single well. The model was then tested against previously unseen data coming from the other two wells. The trained model is able to predict the Young's modulus in the test wells with a root mean squared error around 12 GPa. The example here provided demonstrates that a model trained with drilling parameters and gamma ray coming from one well is able to predict the Young Modulus of different wells in the same basin. These outcomes highlight the potentiality of this procedure and point out several implications for the reservoir characterization. Indeed, once the model has been trained, it is possible to predict the Young's Modulus in different wells of the same basin using only surface logging data.

Double Casing Exit for Side-Track to Commingle Two Borehole Sizes Based Sections in One Slim Shot to Well TD

This project began when a 9-5/8" in 43.5 ppf production casing became inaccessible due to the existing cemented pipe inside, preventing further reservoir section exposure and necessitating a mechanical side-track meanwhile introducing the challenge of loosing one section and imposimg slim hole challenges. The size and weight of the double-casing made for challenging drilling, as did the eight very different formations, which were drilled. The side-track was accomplished in two steps, an 8½ in hole followed by a single long 6⅛ in section, rather than the three steps (16 in, 12¼ in, 8½ in) that are typically required. The optimal kick off point carfully located across the dual casing by running electromagnetic diagnostics, the casing collar locator, and the cement bond log. The double casing mill was carefully tailored to successfully accomplish the exit in one run. Moreover, an extra 26 ft. MD rathole was drilled, which helped to eliminate the mud motor elongation run. A rotary steerable system was utilized directly in a directional BHA to drill an 8½ in open hole building section from vertical to a 30⁰ inclination. A 7.0 in liner was then set to isolate weak zones at the equivalent depth of the outer casing (13-3/8"). Subsequently, a single 6⅛ in section was drilled to the well TD through the lower eight formations. Drilling a 6⅛ in section through eight formations came with a variety of challenges. These formations have different challenging behaviors relative to the wellbore pressure that typically leads to the drilling being done in two sections. Modeling the geo-mechanical characteristics of each formation allowed the determination of a mud weight range and rheology that would stabilize the wellbore through all eight formations. The slim, 6⅛ in, hole was stabilized with higher equivalent circulating density (ECD) values than is typically used in larger boreholes. Optimizing mud weight and drilling parameters, while managing differential sticking with close monitoring of real-time ECD, helped to stabilize the high-pressurized zones to deliver the well to the desired TD with a single borehole. This project represents the first time in Kuwait that double casings in such large sizes have been cut and sidetracked. It is also the first time these eight formations have been cut across such a smaller hole size, slim hole (6⅛ in) in a single shot. Geo-mechanical modeling allowed us to stabilize the pressurized formations and to control the ECD. The well also deployed the longest production liner in the field commingling multiple reservoirs with differnt pore pressure ramps, with excellent cement quality providing optimal zonal isolation.

Wellbore Stability Beyond Mud Weight

Drilling oil and gas wells with stable and good quality wellbores is essential to minimize drilling difficulties, acquire reliable openhole logs data, run completions and ensure well integrity during stimulation. Stress-induced compressive rock failure leading to enlarged wellbore is a common form of wellbore instability especially in tectonic stress regime. For a particular well trajectory, wellbore stability is generally considered a result of an interplay between drilling mud density (i.e., mud weight) and subsurface geomechanical parameters including in-situ earth stresses, formation pore pressure and rock strength properties. While role of mud system and chemistry can also be important for water sensitive formations, mud weight is always a fundamental component of wellbore stability analysis. Hence, when a wellbore is unstable (over-gauge), it is believed that effective mud support was insufficient to counter stress concentration around wellbore wall. Therefore, increasing mud weight based on model validation and calibration using offset wells data is a common approach to keep wellbore stable. However, a limited number of research articles show that wellbore stability is a more complex phenomenon affected not only by geomechanics but also strongly influenced by downhole forces exerted by drillstring vibrations and high mud flow rates. Authors of this paper also observed that some wells drilled with higher mud weight exhibit more unstable wellbore in comparison with offset wells which contradicts the conventional approach of linking wellbore stability to stresses and rock strength properties alone. Therefore, the objective of this paper is to analyze wellbore stability considering both geomechanical and drilling parameters to explain observed anomalous wellbore enlargements in two vertical wells drilled in the same field and reservoir. The analysis showed that the well drilled with 18% higher mud weight compared with its offset well and yet showing more unstable wellbore was, in fact, drilled with more aggressive drilling parameters. The aggressive drilling parameters induce additional mechanical disturbance to the wellbore wall causing more severe wellbore enlargements. We devised a new approach of wellbore stability management using two-pronged strategy. It focuses on designing an optimum weight design using geomechanics to address stress-induced wellbore failure together with specifying safe limits of drilling parameters to minimize wellbore damage due to excessive downhole drillstring vibrations. The findings helped achieve more stable wellbore in subsequent wells with hole condition meeting logging and completion requirements as well as avoiding drilling problems.

Application of a Drilling Simulator for Real-Time Drilling Hydraulics Training and Research

Challenging wells require an accurate hydraulic model to achieve maximum performance for drilling applications. This work was conducted with a simulator capable of recreating the actual drilling process, including on-the-fly adjustments of the drilling parameters. The paper focuses on the predictions of the drilling simulator's pressure losses inside the drill string and across the open-hole and casing annuli applying the most common rheological models. Comparison is then made with pressure losses from field data. Drilling data of vertical and deviated wells were acquired to recreate the actual drilling environment and wellbore design. Several sections with a variety of wellbore sizes were simulated in order to observe the response of the various rheological models. The simulator allows the input of wellbore and bottom-hole assembly (BHA) sizes, formation properties, drilling parameters, and drilling fluid properties. To assess the hydraulic model's performance during drilling, the user is required to input the drilling parameters based on field data and match the penetration rate. The resulting simulator hydraulic outputs are the equivalent circulation density (ECD) and standpipe pressure (SPP). The simulator's performance was assessed using separate simulations with different rheological models and compared with actual field data. Similarities, differences, and potential improvements were then reported. During the simulation, the most critical drilling parameters are displayed, emulating real-time measured values, combined with the pore pressure, wellbore pressure, and fracture pressure graphs. The simulation results show promise for application of real-time hydraulic operations. The simulated output parameters, ECD and SPP, have similar trends and values with the values from actual field data. The simulator's performance shows excellent matching for a simple BHA, with decreasing system's accuracy as the BHA design becomes more complex, an area of future improvement. The overall approach is valid for non-Newtonian drilling fluid pressure losses. The user can observe the output parameters, and by adding a benchmark safety value, the simulator gives a warning of a potential fracture of the formation or maximum pressure at the mud pumps. Thus, by simulating the drilling process, the user can be trained for the upcoming drilling campaign and reach the target depth safely and cost-effectively during actual drilling. The simulator allows emulation of real-time hydraulic operations when drilling vertical and directional wells, albeit with a simple BHA for the latter. The user can instantly observe the output results, which allows proper action to be taken if necessary. This is a step towards real-time hydraulic operations. The results also indicate that the simulator can be used as an excellent training tool for professionals and students by creating wellbore exercises that can cover different operating scenarios.