Standpipe Pressure Research Articles

Research on Denoising Algorithm of Standpipe Pressure Signal in FCC Device Based on Tortuosity and Denoising Intensity.

The pressure signal of the standpipe in the regeneration device for the catalytic cracking reaction is commonly used as a basis for equipment fault diagnosis. The device operates in environments of high temperature, strong vibration, and high flow rate. The standpipe pressure signal is easily interfered with by noise, making it difficult to extract pressure characteristics. In this study, a signal denoising algorithm based on pressure curve tortuosity and denoising intensity is proposed. A power spectrum threshold vector is utilized to filter the pressure in the circulating standpipe, and the denoising intensity and tortuosity after denoising are calculated. Taking the ratio of noise reduction intensity and tortuosity as the objective function, the noise reduction scheme of the pressure signal at different heights of the standpipe is determined. The denoised pressure signals are reconstructed, and the characteristic pressure signals of the circulating standpipe are extracted. The results indicate that the method suppressed the noise, enhanced the significance of characteristic parameters, and improved the accuracy of the fault diagnosis in the FCC units.

Explainable machine-learning-based prediction of equivalent circulating density using surface-based drilling data

When drilling wells for energy explorations, it is important to regulate the formation pressures appropriately to prevent kicks, which can lead to unimaginable loss of lives and properties. This is usually done by controlling the equivalent circulating density (ECD), which responds to the dynamic conditions that occur during drilling. The conventional approach to determine ECD is via mathematical modeling or downhole measurements. However, the downhole measurement tools can be very expensive, and the mathematical models do not provide a high degree of accuracy. Some previous authors have proposed using machine learning (ML) techniques to improve the degree of accuracy of the ECD predictions. In this work, we employed an extreme gradient-boosting (XGBoost) methodology to predict ECD values. The model's accuracy was determined using correlation coefficients (R2) and root mean square errors (RMSE) as their performance metrics. The results showed a strong prediction capability with an R2 and RMSE of 1.00 and 0.0005 for the training data and an R2 and RMSE of 0.989 and 0.023 for the testing/blind data set, respectively. The developed model outperformed those obtained using other popular machine learning techniques. Lastly, an interpretation of the model results showed that mud weight, weight on hook, and standpipe pressure contributed the most to the ECD prediction values.

Prediction of penetration rate and optimization of weight on a bit using artificial neural networks

Relevance. Achieving the greatest rate of penetration is the aim of every drilling engineer because it is one of the most significant factors influencing drilling costs. However, a variety of drilling conditions could have an impact on rate of penetration, complicating its forecast. Aim. To suggest a novel strategy to accurately predict rate of penetration and optimize drilling parameters. Objects. Real-time drilling data of a few wells in the Ca Tam oil field, Vietnam, with more than 900 datasets including significant parameters like rotary speed, weight on bit, standpipe pressure, flow rate, weight of mud, torque. Methods. Various methods using Artificial Neural Network was proposed to estimate rate of penetration. Results. The number of neurons in a hidden layer was varied then the results of different Artificial Neural Network models were compared in order to obtain the optimal model. The final Artificial Neural Network model shows high exactness when contrasted with actual rate of penetration, in this manner, it tends to be suggested as a successful and reasonable approach to predict the rate of penetration of different wells in the Ca Tam oil field. Based on the proposed Artificial Neural Network model, the optimal weight on bit was determined for the drilling interval from 1800 to 2300 m of oil wells in the research region.

Intelligent Inversion Analysis of Drilling Gas Kick Characteristic Parameters and Advanced Predictions of Development Trends

Summary Gas kicks are a common and complex occurrence in oil and gas exploration and development. Obtaining the characteristic parameters promptly and accurately after a gas kick occurs is of utmost importance. The kick characteristic parameters are essential foundational data for accurately understanding the multiphase flow state within the wellbore, analyzing kick development trends, and formulating well-killing design. Traditional methods for determining gas kick characteristic parameters often involve trial estimation, a process that can introduce significant arbitrariness. Alternatively, using specialized equipment for measurement and analysis is an option, but it can be time-consuming and limited by technology and cost constraints. This paper proposes an intelligent inversion method for kick characteristic parameters based on surface logging data, including standpipe pressure and mud pit volume. First, a gas kick simulation model is developed based on wellbore multiphase flow theory. This model can accurately simulate the changes in surface logging parameters that occur when a gas kick occurs. Next, the particle swarm optimization algorithm is used to optimize the kick parameters, such as the gas-influx index and pore pressure, in conjunction with the gas kick simulation model. The optimization evaluation criterion is the Fréchet distance, which is used to identify the calculated curve that is most similar to the actual logging parameter change curve. Subsequently, kick trends can be predicted using the simulation model based on these kick parameters. The case analysis demonstrates that the method proposed in this paper can rapidly acquire the gas kick characteristic parameters based on the early change characteristics of logging parameters. It effectively simulates and reproduces the true distribution of wellbore fluids, enabling advanced prediction of kick development. This approach helps in preparing preventive measures in advance, reducing the risk of accidents, and minimizing financial losses.

APPLICATION OF PSO-LSSVM IN PREDICTION AND ANALYSIS OF SLOW DRILLING (RATE OF PENETRATION)

The benefits of drilling include reducing the total time, maintaining the lowest possible risk, saving costs, and increasing efficiency, which occurs in (the planning and exploration stages). Slow drilling refers to a rate of penetration (ROP) that is not at the desired level. ROP characterizes the speed at which the drill bit penetrates the underlying rock to deepen the borehole, as it is directly related to controlling the speed and efficiency of drilling which ultimately impacts development costs. Predicting ROP is a very important step to optimize drilling with Machine Learning that can assist in solving complex problems with maximum possible efficiency. The model used is PSO-LSSVM treats the penetration drill bit as a continuous process. It takes drilling data sequentially, continuously predicts ROP, and achieves better ROP prediction results. In this case, Hole Depth, weight on bit (WOB), Bit Rotation per minute (RPM), Torque, Bit Depth, Time of Penetration, Hook Load, and Standpipe Pressure, demonstrated influence in keeping ROP at a high level. According to the results, the PSO-LSSVM algorithm can be used for the prediction of ROP in well X. thus providing a solution for prediction and control of operating effects which can result in a fast penetration rate and better efficiency in drilling.

Hybrid Approach Using Physical Insights and Data Science for Stuck-Pipe Prediction

Summary Early detection of stuck-pipe incidents is crucial because of the enormous costs of recovering from such incidents. Previous studies have leaned significantly toward a physics-based or data science approach. However, both approaches have challenges, such as the uncertainty of the physics-based model and the lack of data in the data science approach. In this study, we propose a hybrid approach using physical insights and data science and discuss the possibility of stuck-pipe prediction. The proposed method comprises two steps. In the first step, a data-driven model with physical insights is trained using the historical data of the in-situ well to estimate some of the drilling variables. In the second step, the risk of stuck-pipe occurrence (hereafter referred to as sticking risk) is calculated based on the historical and current measured data and the estimation of the trained model. This approach is expected to overcome the limitations of the previous methods as it allows the construction of a detection model tuned to the in-situ well. In the case studies, models for estimating the topdrive torque and standpipe pressure were constructed. The performance of the models is discussed using actual drilling data from drilling fields, including 21 stuck-pipe incidents during drilling operations. The proposed method was first examined using short-term output. The output confirmed that the sticking risk increased shortly (up to 20 hours) before the stuck-pipe incident occurred in 15 cases. This increase in sticking risk was consistent with physical considerations. Subsequently, this study examined the long-term output over several months; this was rarely done in previous studies. Even within this long-term output, some cases had good performance with only a few false alarms, while others had problems with many false alarms. For cases of low performance, several model improvements, such as adding mud information or making the learning and threshold-setting methods more robust to outliers, were found to have the potential to improve performance. The novelty of our research lies in creating a broad framework for the stuck-pipe prediction by using both physical insights and data science methods. The proposed hybrid approach demonstrated the potential to reduce false alarms and improve interpretability compared with previous methods. The framework is highly extensible, and further performance improvements can be expected in the future.

Real-Time Gamma Ray Log Generation from Drilling Parameters of Offset Wells Using Physics-Informed Machine Learning

Summary By 2026, USD 5.05 billion will be spent per year on logging while drilling (LWD) according to the market report from Fortune Business Insights (2020). Logging tools and wireline tools are costly services for operators to pay for, and the companies providing the services also have a high cost of service delivery. They are, however, an essential service for drilling wells efficiently. The ability to computationally generate logs in real time using known relationships between the rock formations and drilling parameters, namely, rate of penetration (ROP), rotations per minute (RPM), surface weight on bit (SWOB), surface torque (TQX), standpipe pressure (SPPA), and hookload (HKLD), provides an alternative method to generate formation evaluation information (analysis of the subsurface formation characteristics such as lithology, porosity, permeability, and saturation). This paper describes an approach to creating a digital formation evaluation log generator using a novel physics-informed machine learning (PIML) approach that combines physics-based approaches with machine learning (ML) algorithms. The designed approach consists of blocks that calculate mechanical specific energy (MSE), physical estimates of gamma ray (GR) using physical and empirical models, and formation information. All this information and the drilling parameters are used to build a classification model to predict the formations, followed by formation-based regression models to get the final estimate of GR log. The designed PIML approach learns the relationships between drilling parameters and the GR logs using the data from the offset wells. The decomposition of the model into multiple stages enables the model to learn the relationship between drilling parameters data and formation evaluation data. It makes it easier for the model to generate GR measurements consistent with the rock formations of the subject well being drilled. Because the computationally generated GR by the model is not just dependent on the relationships between drilling parameters and GR logs, this model is also generalizable and capable of being deployed into the application with only retraining on the offset wells and no change in the model structure or complexity. For this paper, the drilling of the horizontal section will not be discussed, as this was done as a separate body of work. Historically collected data from the US Land Permian Basin wells are used as the primary data set for this work. Results from the experiments based on the data collected from five different wells have been presented. Leave-one-out validation for each of the wells was performed. In the leave-one-out validation process, four of the wells represent the set of offset wells and the remaining one becomes the subject well. The same process is repeated for each of the wells as they are in turn defined as a subject well. Results show that the framework can infer and generate logs such as GR logs in real time. The average root-mean-squared error (RMSE) observed from the experiments is 27.25 api, representing about 10% error. This error value is calculated based on the mean estimate and does not consider the predicted confidence interval. Considering the confidence interval helps further reduce the error margin.

Applications of Different Classification Machine Learning Techniques to Predict Formation Tops and Lithology While Drilling.

Accurate prediction of formation tops and lithology plays a critical role in optimizing drilling processes, cost reduction, and risk mitigation in hydrocarbon operations. Although several techniques like well logging, core sampling, cuttings analysis, seismic surveys, and mud logging are available for identifying formation tops, they have limitations such as high costs, lower accuracy, manpower-intensive processes, and time or depth lags that impede real-time estimation. Consequently, this study aims to leverage machine learning models based on easily accessible drilling parameters to predict formation tops and lithologies, overcoming the limitations associated with traditional methods. Data from two wells (A and B) in the Middle East, encompassing drilling mechanical parameters such as rate of penetration (ROP), drill string rotation (DSR), pumping rate (Q), standpipe pressure (SPP), weight on bit (WOB), and torque, were collected for real-field analysis. Machine learning models including Gaussian naive Bayes (GNB), logistic regression (LR), and linear discriminant analysis (LDA) were trained and tested on the data set from well A, while the data set from well B was utilized for model validation as unseen data. The formations of wells A and B consist of four lithologies, namely, sandstone, anhydrite, carbonate/shale, and carbonates, necessitating the development of multiclass classification models. The drilling parameters, specifically the WOB and ROP, exhibited a strong influence on lithology identification. Among the models, GNB demonstrated exceptional performance in predicting formation lithology from the drilling parameters, achieving accuracy and nearly perfect precision, recall, and F1 score for the different classes. LDA and LR models accurately predicted sandstone and carbonate lithologies, although some misclassifications occurred in approximately 5% of points for anhydrite and around 20% in carbonate/shale formations. During validation, the models demonstrated accuracies of around 0.96, 0.95, and 0.92 for the GNB, LR, and LDA, respectively. The study highlights the efficacy of the developed machine learning models in accurately predicting the formation lithology and tops in real time. This is achieved by utilizing readily available drilling parameters, making the approach highly accurate and cost effective by leveraging existing real-time drilling data.

Real-time prediction of logging parameters during the drilling process using an attention-based Seq2Seq model

In recent years, there has been a notable upsurge within the drilling industry regarding the construction of machine learning models that leverage logging parameters to augment decision-making processes. When building these models, the logging parameters are usually assumed to be fully accessible. However, it may not always hold true because certain logging parameters will be inaccessible or unavailable in real-time drilling operations, which greatly hinders the practical implementation of these models. Aiming at addressing the logging parameters' accessibility issue, this paper proposes an innovative attention-based sequence-to-sequence (Seq2Seq) model to enable real-time predictions. The proposed model integrates the Seq2Seq model, bidirectional Gated Recurrent Unit (GRU) model, and attention mechanism to enable multi-step ahead prediction. Furthermore, dynamic updates using the sliding window method with intermittent frequency are employed to enhance the model's performance. Meanwhile, hyperparameter tuning is automatically performed using an open-source toolbox to boost the model training process. Finally, the efficacy of the model is evaluated by predicting four commonly utilized logging parameters (surface torque, standpipe pressure, mud pit volume, and Gamma ray) based on two field drilling datasets. The comprehensive analysis demonstrates promising results, with mean absolute percentage errors below 5% across the datasets. Moreover, the attention-based Seq2Seq model surpasses base models (the vanilla GRU model and the Seq2Seq model) due to the inherent Seq2Seq structure, bidirectional architecture, and attention layer. This paper offers a practical approach to bridge the gap between the development of theoretical machine learning models and their implementation in real-world drilling processes.

The Impact of Drilling Operation Parameters on the Lost Circulation Problems for Oil and Gas Wells Based on Field Data

Lost circulation, the loss of drilling fluid into holes or fractures throughout the geological formations being drilled, is the most expensive problem encountered in drilling operations for oil and gas wells. It is known that when there is a loss of circulation during the drilling process, the operation activities are delayed or stopped for curing or treatment, resulting in extra time known as non-productive time (NPT). The NPT issues during the drilling operations are the most important priorities for planning any drilling program. Therefore, the lost circulation is one of the main problems which can control the NPT. To mitigate the lost circulation problems, an integrated investigation is necessarily needed for the main parameters which enhance/downgrade the rate of lost circulation.
 Drilling through the Dammam and Hartha formation is typically accompanied with substantial nonproductive time due to severe and often entire drilling fluid losses. The majority of the curing this problem require time for preparation, placement or setting. In this research, a field study was carried out to investigate the effect of 6 drilling parameters on the rate of the lost circulation. The investigated parameters are the rate of penetration, weight on bit, stand pipe pressure, mud weight, flow rate, and the equivalent circulation density. Furthermore, the effect of lithology type on the lost circulation rate was clearly determined and correlated. In addition, a graphical representation of bit depth vs the real drilling time/date was constructed to compare the lossy wells vs non-lossy wells named frithogram. This study was based on field data for 350 wells drilled in Rumaila Oilfield, southern Iraq. The results of this study showed that if these six operating parameters are standard and controlled, the significance of lost circulation problems could be reduced. However, the main factors which are responsible for the lost circulation problems are lithology types, facies and/or diagenesis process, which mainly distinguish the carbonate formations.

A Developed Robust Model and Artificial Intelligence Techniques to Predict Drilling Fluid Density and Equivalent Circulation Density in Real Time.

When drilling deep wells, it is important to regulate the formation pressure and prevent kicks. This is achieved by controlling the equivalent circulation density (ECD), which becomes crucial in high-pressure and high-temperature wells. ECD is particularly important in formations where the pore pressure and fracture pressure are close to each other (narrow windows). However, the current methods for measuring ECD using downhole sensors can be expensive and limited by operational constraints such as high pressure and temperature. Therefore, to overcome this challenge, two novel models named ECDeffc.m and MWeffc.m were developed to predict ECD and mud weight (MW) from surface-drilling parameters, including standpipe pressure, rate of penetration, drill string rotation, and mud properties. In addition, by utilizing an artificial neural network (ANN) and a support vector machine (SVM), ECD was estimated with a correlation coefficient of 0.9947 and an average absolute percentage error of 0.23%. Meanwhile, a decision tree (DT) was employed to estimate MW with a correlation coefficient of 0.9353 and an average absolute percentage error of 1.66%. The two novel models were compared with artificial intelligence (AI) techniques to evaluate the developed models. The results proved that the two novel models were more accurate with the value obtained from pressure-while-drilling (PWD) tools. These models can be utilized during well design and while drilling operations are in progress to evaluate and monitor the appropriate mud weight and equivalent circulation density to save time and money, by eliminating the need for expensive downhole equipment and commercial software.

Drilling operation optimization using machine learning framework

Optimizing the surface parameters to maximize the rate of penetration (ROP) could be achieved through a machine learning (ML) framework. Hence, predicting ROP is a crucial factor in any optimization problem. A defined range of surface parameters will provide the maximum ROP for a given bottom hole assembly (BHA) while drilling a particular formation lithology. This maximization of ROP value is valid for a specific range of certain combinations of surface parameters and subsurface equipment. However, tuning these surface parameters is an art rather than a science. Despite the abundance of offset wells data, local experience and trial-and-error methods are the current keys for this tuning. This work used a large drilling dataset (460,076 records) to develop two artificial neural network (ANN) models. The first ANN model is used to predict standpipe pressure (SPP) and the apparent surface torque for the drilling system. In addition to other inputs, these two parameters are used in the second ANN model to predict the ROP. The first ANN achieved 92.39% and 93.71% accuracy in predicting the apparent torque and standpipe pressure values respectively; while the second ANN achieved 87.63% accuracy when tested against 39,294 records that were not used in training. To apply the system in real applications, the ROP is predicted, and a genetic algorithm (GA) is used to optimize both the surface parameters and the downhole equipment while drilling a section of the hole. This system or framework (ANN models coupled with the GA algorithm) was used in a field trial in one of the Egyptian Western Desert routine wells. The framework was used in two sections within the 8 ½” hole and enhanced these sections' ROP by 15.05% and 14.41% compared to similar sections in different offset wells. It resulted in an overall best-in-class (BIC) well performance compared to similar wells in the field.

Lost circulation monitoring using bi-directional LSTM and data augmentation

In order to improve the lost circulation risk recognition accuracy of artificial intelligence (AI) based models using limited number of data samples, a lost circulation monitoring method involving data augmentation and Bi-directional Long Short-Term Memory (Bi-LSTM) network is proposed. Firstly, the collected lost circulation data samples including pit volume (PIT), flow-out rate (FOR), pump speed (PS) and standpipe pressure (SPP) data sequences as elements are augmented using percentage scaling and random dithering to produce a dataset with increased number of samples. Then, a Bi-LSTM-based lost circulation monitoring model, which can explore both the past and future information of the input data, is established and trained with the augmented dataset. Finally, the obtained lost circulation monitoring model is applied to the PIT, FOR, PS, and SPP field data sequences for risk monitoring. A collected field lost circulation dataset with 2000 sample points was used to train and test the recognition performance of the proposed method, the test results demonstrate that the recognition accuracies of the LSTM model and Bi-LSTM model without data augmentation are 84% and 89%, respectively. After data augmentation is applied, the recognition accuracy of the Bi-LSTM model is improved to 93%.

Integrating drilling parameters and machine learning tools to improve real-time porosity prediction of multi-zone reservoirs. Case study: Rhourd Chegga oilfield, Algeria

Porosity is a key variable for hydrocarbon reservoirs evaluation. It can be directly determined in laboratory tests using core samples or calculated indirectly from well logs. However, these methods are expensive and time consuming affecting the cost of supply of oil produced. Monitoring drilling variables in real time as a borehole is drilled can provide a cheap, initial insight to various rock formation properties. Investigating the relationships between rock porosity and drilling variables is therefore important. This study applies machines learning (ML) relate rock porosity to a range of measured drilling variables in several types of reservoir. Four ML models are evaluated: K-nearest neighbor (KNN), random forest (RF), support vector regression (SVR), and eXtreme Gradient Boosting (XGBoost). A measured drilling and porosity dataset from the Rhourd Chegga oilfield (SE Algeria) is compiled from its multi-zone reservoir system. The drilling variables from nine vertical wellbores considered are: weight on bit, torque, standpipe pressure, drill string rotation speed, rate of penetration and its inverse, pump rate and bit depth. The recorded data is derived from three distinct reservoir zones: Trias (T1), Rhourd Chegga sandstone (RDC), and the Quartzite Hamra tight reservoir (QH). Porosity predictions were derived in two ways: (1) for each reservoir separately; (2) for the three reservoirs combined. Prediction error analysis was conducted in terms of the coefficient of determination (R2), root mean squared error (RMSE) and mean absolute error (MAE). The XGBoost model yielded the best porosity predictions for the individual and combined reservoirs. Prediction errors generated for reservoir T1 were (R2 ≈ 0.865, RMSE≈0.866, MAE≈0.692); for reservoir RDC were R2 ≈ 0.935, RMSE≈0.848, and MAE≈0.772; and, for reservoir QH were R2 ≈ 0.743, RMSE≈0.794, and MAE≈0.573. For the combined (multi-zone) reservoir (T1+RDC + QH), porosity predictions were less accurate, and the XGBoost model generated prediction errors of R2 ≈ 0.76, RMSE≈1.259, and MAE≈1.027.

Derivation of rock's geomechanical parameters while drilling by combining surface drilling data, gamma ray data, and machine learning techniques in carbonate formations

This work presents a field application of an effective and reliable methodology for geomechanical parameters evaluation, using surface logging drilling data (rate of penetration, torque, weight on bit, stand pipe pressure, rotation per minute, flow rates) and well log data (sonic log, bulk density log, gamma ray log) to feed a model characterised by the combination of different machine learning algorithms (multiple linear regression, support vector regression, random forest, artificial neural network, and XGBoost). To expand the range of application to those cases where the downhole tools are not technically feasible or economically viable, a flexible workflow has been developed to derive a Synthetic Gamma Ray (SGR) from X-ray Fluorescence (XRF) analysis performed on cuttings. The methodology, applied to a dataset of 11 wells drilled in the same geological units but in different fields in Kuwait, proves its successful use to derive several geomechanical parameters, among which the results associated with Young’s Modulus, density, and Poisson’s Ratios are here presented.

Well head multi parameter data monitoring system based on LoRa

Oilfield wellhead data is an important indicator to reflect the normal operation of oil well equipment. This paper monitors the data of standpipe pressure, hanging weight, torque and other parameters, which is helpful to improve the safety and production efficiency of the entire oilfield. Therefore, this paper proposes a set of multi parameter monitoring system based on LoRa to further explore and study in order to reduce oilfield accidents.

How Underflow Standpipes Adjust to Maintain Pressure Balance Around a Circulation Loop

AbstractThe standpipe has become a standard element of most circulation loops. In the past, standpipes have been thought of as a device to only allow solids to transfer from a low to a higher‐pressure region. However, it also has another function in a solids circulation loop. It is also the device that adjusts its operation to balance the pressure around the solids circulation loop. An overflow standpipe adjusts the pressure drop across it by raising or lowering the solids level in the standpipe. However, the operation of the underflow standpipe is more complex. This self‐adjusting function can cause underflow standpipe pressure drops to be negative, which seems counter‐intuitive. How an underflow standpipe operating with a negative pressure build can occur in a circulation loop is described below using two commercial unit examples as illustrations of this phenomenon.

Early monitoring of gas kick in deepwater drilling based on ensemble learning method: A case study at South China Sea

Gas kick monitoring is of great significance for prevention of blow-out accidents, especially in deep drilling and deep-water drilling. In this study, a machine learning (ML) model for early-monitoring of gas kick is developed using the ensemble learning algorithms based on 7363 lines of drilling logging data at South China Sea. The selected input parameters based on mechanism analysis of gas kick are six fast engineering parameters, including hook load (WHO), weight on bit (WOB), torque (TOR), flow rate (FLW), rate of penetration (ROP) and stand-pipe pressure (SPP), and two slow mud property parameters, i.e. electrical conductivity (CON) and mud outlet density (DEN). The model is constructed using RUSboosted, Subspace-KNN and Bagged Trees algorithms, and is compared with the neural network algorithm. We propose a comprehensive error to quantitatively evaluate the performance of the gas kick monitoring models. The models for early-monitoring of gas kick are applied for a single well and multiple wells, respectively. The results indicate that: (i) The optimal combination of input parameters is made up of six fast engineering parameters and two slow mud parameters. When there is a higher requirement on timeliness, only use of the six fast engineering parameters is also acceptable. (ii) The ensemble learning models work well when the input data expand from single well to multiple wells in the same block. For most cases, the prediction error of the optimal model is below 10%. The RUSboosted algorithm performed best in most data sets. (iii) Gas kick identification from lots of drilling logging records is mathematically a small-sample problem. The output labelling of a potential gas kick should be based on the field practical requirement. The recommended positive length of continuous-point labelling method is 5 m for the studied area, which can effectively reduce the average error from 8.02% to 5.48%.

Wellbore multiphase flow behaviors of gas kick in deep water horizontal drilling

During the deepwater drilling, the complicated gas-liquid-solid multiphase flow will occur if the formation gas enters and migrates in the wellbore. Through understanding of the wellbore flow behaviors is of great importance for the blowout prevention and well control. Considering the dynamic mass and heat transfer process in wellbore caused by alternating ambient temperature field, a multiphase flow model of multicomponent fluid in wellbore is deduced and developed, including the continuity equation, momentum conservation equation and energy conservation equation. Furthermore, the corresponding initial and boundary conditions are proposed for different working conditions in deepwater drilling, and an efficient numerical solution method is established, including dynamic mesh generation method and discrete solution method of partial differential equations. Applied in a deep-water kicking well, the proposed model is used to analyze the multiphase flow rules in the wellbore. The results show that in the process of annular fluid returning from the bottomhole, the pressure generally decreases linearly, while the temperature change is nonlinear. The temperature first rises and then falls at the formation section, and first falls and then rises at the seawater section. Furthermore, the pit gain increases approximately in a quadratic polynomial relationship, caused by the rise and expansion of gas in the wellbore, and the pressure drop and gas influx rate increase at the bottomhole. In the process of kick evolution, the standpipe pressure and bottomhole pressure gradually decrease, which can be an important sign for kick detection.

Predicting Rate of Penetration and optimization Weight on bit using Artificial Neural Networks

Obtaining the maximum Rate of Penetration (ROP) by optimization drilling parameters is the aim of every drilling engineer. This is because it could save time, reduce cost and minimize drilling problems. However, ROP depends on a lot of parameters which lead to difficulties in its prediction. Therefore, it is necessary and important to investigate a solution predicting ROP with high accuracy to determine the suitable drilling parameters. In this study, a new approach using Artificial Neural Network (ANN) has been proposed to predict ROP from real – time drilling data of several wells in Nam Rong - Doi Moi field with more than 900 datasets included important parameters such as the weight on bit (WOB), weight of mud (MW), rotary speed (RPM), standpipe pressure (SPP), flow rate (FR), torque (TQ). In the process of training the network, algorithms and the number of neurons in the hidden layer were varied to find the optimal model. The ANN model shows high accuracy when compared to actual ROP, therefore it can be recommended as an effective and suitable method to predict the ROP of other wells in the research area. Besides, base on the proposed ANN model, authors carried out experiments and determind the optimal weight on bit value for the drilling interval from 1800 to 2300 m of wells in Nam Rong Doi Moi field