Exploration
Mustafa Yasser Elgindy; Ahmed Zakaria Nooh; Ali Mostafa Wahba
Abstract
Kick monitoring, detection, and control are key elements to ensure safe drilling operations and avoid catastrophic blow-out incidents that can cause loss of life, equipment, and environmental damage. Conventional kick detection systems such as the pit volume totalizer and the flow in/out sensors identify ...
Read More
Kick monitoring, detection, and control are key elements to ensure safe drilling operations and avoid catastrophic blow-out incidents that can cause loss of life, equipment, and environmental damage. Conventional kick detection systems such as the pit volume totalizer and the flow in/out sensors identify the kick after a large amount of influx has been recorded on the surface. So, we aim to recognize the kick before it enters the wellbore by detecting the abnormal formation pressure once the bit penetrates the rock. This paper proposes a new machine learning model as an alternative solution using field drilling parameters such as true vertical depth, porosity, bulk density, resistivity, rate of penetration, weight on bit, rotation per minute, torque, standpipe pressure, flow rate, flowline temperature, and gas level. The data-driven models were developed using three separate algorithms: K-Nearest Neighbor, Random Forest, and XG Boost. 6022 field data points were split for training, testing, and validation processes. On average, the model using the random forest algorithm showed the highest accuracy in forecasting the formation pressure, with root mean squared error values and coefficient of determination values of 12.868 and 0.9638, respectively. Streamlit Deployment tool was used to create a user interface program to facilitate the prediction process. The program was tested using 196 field data points and recorded a high accuracy of 95%.
Exploitation
Rym Khettabi; Issam Touil; Mohamed Kezzar; Mohamed R. Eid; Fatima.Z Derdour; Kamel Khounfais; Lakhdar Khochmane
Abstract
It is well-established that the response surface methodology (RSM) is commonly employed to establish the differences between the predicted values and those observed experimentally. This study mainly goals on the impact of four drilling factors including weight on the bit (WOB), the rotating rapidity ...
Read More
It is well-established that the response surface methodology (RSM) is commonly employed to establish the differences between the predicted values and those observed experimentally. This study mainly goals on the impact of four drilling factors including weight on the bit (WOB), the rotating rapidity of the bit, RPM, cutting angle , and rock resistance on the penetration rate of the drilling tool. In this examination, three kinds of limestone rocks were considered. The planned assessments were carried out at three stages of the considered four input variables. The statistical analysis was realized using both RSM approach and analysis of variance (ANOVA). This analysis allowed us to develop the appropriate penetration model with a higher determination coefficient of 96.19%, which demonstrates the high correlation between the predicted and experimental data, and consequently, it can be concluded that the obtained model is highly suitable for the prediction of the penetration rate. Also from variance analysis, the results obtained show that rotational speed, RPM, and weight on the bit (WOB) parameters, as well as the nature of the rock, which is determined by the rock compressive resistance, having a significant effect on the penetration rate; however, the rake angle has little effect. Finally, the optimal parameters were determined to find the best possible penetration rate of the drilling tool.
M. Anemangely; A. Ramezanzadeh; B. Tokhmechi
Abstract
Achieving minimum cost and time in reservoir drilling requires evaluating the effects of the drilling parameters on the penetration rate and constructing a drilling rate estimator model. Several drilling rate models have been presented using the drilling parameters. Among these, the Bourgoyne and Young ...
Read More
Achieving minimum cost and time in reservoir drilling requires evaluating the effects of the drilling parameters on the penetration rate and constructing a drilling rate estimator model. Several drilling rate models have been presented using the drilling parameters. Among these, the Bourgoyne and Young (BY) model is widely utilized in order to estimate the penetration rate. This model relates several drilling parameters to the penetration rate. It possesses eight unknown constants. Bourgoyne and Young have suggested the multiple regression analysis method in order to define these constants. Using multiple regressions leads to physically meaningless and out of range constants. In this work, the Cuckoo Optimization Algorithm (COA) is utilized to determine the BY model coefficients. To achieve this goal, the corresponding data for two wells are collected from one of the oilfields located in SW of Iran. The BY model constants are determined individually for two formations in one of the wells. Then the determined constants are used to estimate the drilling rate of penetration in the other well having the same formations. To compare the results obtained for COA, first, the two mathematical methods including progressive stochastic and multiple regressions were implemented. Comparison between these methods indicated that COA yields more accurate and reliable results with respect to the others. In the following, Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) as meta-heuristic algorithms were applied on the field data in order to determine BY model’s coefficients. Comparison between these methods showed that the COA has fast convergence rate and estimation error less than others.
