Fracture Pressure Prediction Research Articles

Fracture Pressure Prediction in Carbonate Reservoir Using Artificial Neural Networks

Fracture Pressure Prediction in Carbonate Reservoir Using Artificial Neural Networks

Prediction of formation fracture pressure based on reinforcement learning and XGBoost

Abstract Clearly determining the magnitude of fracture pressure is a crucial indicator for fracturing design. Traditional methods for predicting fracture pressure suffer from challenges such as difficulties in obtaining required data, low prediction accuracy, and local limitations in application. In light of these issues, the article proposes a fracture pressure prediction model based on reinforcement learning and XGBoost utilizing geophysical well logging data. Based on the relevance analysis, optimal input parameters, including DEPTH, DEN, AC, GR, CRL, and RT, are selected from geophysical well logging data. We have developed a framework for a fracture pressure prediction model based on XGBoost, wherein hyperparameters are fine-tuned using an improved Q-learning algorithm. The optimized XGBoost model for fracture pressure prediction attains outstanding performance metrics, including an R 2 value of 0.992, a root mean square error of 0.006%, and a mean absolute error of 0.539%. In direct comparison with grid search, Bayesian optimization, and ant colony optimization, the improved Q-learning algorithm emerges as the most effective optimization approach. The predictions generated by the proposed method exhibit remarkable consistency with fracture pressure data measured on-site. This approach successfully addresses the shortcomings encountered with traditional fracture pressure prediction methods, such as inadequate accuracy, demanding data prerequisites, and constrained applicability.

Intelligent prediction method for fracture pressure based on stacking ensemble algorithm

Fracture pressure is an important reference for wellbore stability analysis and hydraulic fracturing. Considering the low prediction accuracy, significant deviations, and limited applicability of traditional methods for predicting formation fracture pressure, this paper proposes an intelligent prediction method for fracture pressure using conventional well logging data based on the Stacking ensemble algorithm. The base learners of the model include RF, KNN, and LSTM algorithms with low correlation. The meta-learner adopts the XGBoost algorithm. The effectiveness of the model is validated using the fracture pressure data from Dagang Oilfield. The prediction results indicate that the stacking algorithm outperforms individual algorithms. After optimization with genetic algorithm, the R2 of the stacking model is 0.989, RMSE is 0.009%, and MAE is 0.32%. The global sensitivity analysis results show that AC and DEN in the well logging data have higher sensitivity to the fracture pressure. When using intelligent fracture pressure prediction methods, it is essential to ensure the accuracy of AC and DEN data. The work demonstrates the reliability and effectiveness of the method proposed for the intelligent prediction of fracturing pressure using conventional well logging data through Stacking ensemble algorithm to overcome the limitations of traditional methods.

New Models for Predicting Pore Pressure and Fracture Pressure while Drilling in Mixed Lithologies Using Artificial Neural Networks.

Precise prediction of pore pressure and fracture pressure is a crucial aspect of petroleum engineering. The awareness of both fracture pressure and pore pressure is essential to control the well. It helps in the elimination of the problems related to drilling, waterflooding project, and hydraulic fracturing job such as fluid loss, kick, differential sticking, and blowout. Avoiding these problems enhances the performance and reduces the cost of operation. Several researchers proposed many models for predicting pore and fracture pressures using well log information, rock strength properties, or drilling data. However, some of these models are limited to one type of lithology such as clean and compacted shale formation, applicable only for the pressure generated by under compaction, and some of them cannot be used in unloading formations. Recently, artificial intelligence techniques showed a great performance in petroleum engineering applications. Hence, in this paper, two artificial neural network models are developed to estimate both pore pressure and fracture pressure through the use of 2820 data sets obtained from drilling data in mixed lithologies of sandstone, carbonate, and shale. The proposed artificial neural network (ANN) models achieved accurate estimation of pore and fracture pressures, where the coefficients of determination (R2) for pore and fracture pressures are 0.974 and 0.998, respectively. Another data set from the Middle East was used to validate the developed models. The models estimated the pore and fracture pressures with high R2 values of 0.90 and 0.99, respectively. This work demonstrates the validity and reliability of the developed models to calculate pore and fracture pressures from real-time surface drilling parameters by considering the formation type to overcome the limitation of previous models.

A Novel Method for Fracture Pressure Prediction in Shallow Formation During Deep-Water Drilling

Abstract Large numbers of deep-water drilling practices have shown that more than 60% of deep-water wells have complex leak-off during the drilling process, which poses great difficulties and challenges for the safety and operation time of deep-water drilling. The purpose of this article is to establish a method for predicting the fracture pressure in shallow formations. In this study, the deep-water shallow formation was divided into the upper unconsolidated soil layer, and the lower diagenetic rock layer according to the geotechnical distribution characteristics of the deep-water shallow formation. The location of the transition soil/rock layer zone was determined using the upper soil layer density trend line, and the lower rock layer density log data regression trend line. The deep-water shallow fracture pressure prediction model was established based on the soil/rock transition zone. The shear failure criterion was used above the transition zone, and the tensile failure criterion is used below the transition zone. The shallow fracture pressure of six drilled exploratory wells in the X block from the South China Sea was calculated using this new method and the calculation errors were all less than 3.18%. Moreover, the shallow fracture pressure body in this block was established using the Kriging interpolation method based on six drilled exploratory wells data. This shallow fracture pressure body established here was used to predict nine development wells shallow fracture pressure with a predictive error of less than 1.7% and there were no drilling accidents. The case study demonstrates that the new model can significantly improve the prediction accuracy has good prospects for popularization and application.

Quantitative characterization of collapse and fracture pressure uncertainty based on Monte Carlo simulation

The complex geological conditions of drilling, the difficulty of formation collapse and fracture pressure prediction in South Sichuan work area lead to the complex drilling and frequent failure, which seriously restricts the safe and efficient development of shale gas. In view of this problem, this paper has carried out relevant research. First of all, the existing calculation model of formation collapse and fracture pressure is established and improved; on this basis, the sources of uncertainty in the calculation model of collapse and fracture pressure are analyzed, mainly the in-situ stress and rock mechanics parameters, which have a lot of uncertainties; then, the uncertainty of rock mechanics parameters and in-situ stress is analyzed, and its probability is determined. Finally, based on Monte Carlo simulation, the quantitative characterization method of formation collapse and fracture pressure uncertainty is established. The prediction result of collapse and fracture pressure is no longer a single curve or value, but an interval, which is more practical for drilling in complex geological environment. The results of this study are helpful to better describe the collapse and fracture pressure of complex formation and can provide more valuable reference data for drilling design.

Study on Method of Determining the Safe Operation Window of Drilling Fluid Density with Credibility in Deep Igneous Rock Strata

It is difficult to determine the safe operation window of drilling fluid density (SOWDFD) for deep igneous rock strata. Although the formation three-pressure (pore pressure, collapse pressure, and fracture pressure) prediction method with credibility improves the accuracy of formation three-pressure prediction, it still has a large error for deep igneous strata. To solve this problem, a modified method of the SOWDFD in deep igneous rock strata is proposed based on the leakage statistics of adjacent wells. This method is based on the establishment of the SOWDFD with credibility. Through statistical analysis of drilling fluid density of igneous rock leaky formation group in adjacent wells, the fracture leakage law of the formation is revealed and the upper limit of leak-off pressure containing probability information is obtained. Finally, the modified SOWDFD with credibility for deep igneous rock strata is formed. In this work, the proposed method was used to compute the SOWDFD with credibility of SHB well in Xinjiang, China. Results show that the modified density window is consistent with the field drilling conditions and can reflect the narrow density window in the Permian and lower igneous strata. Combined with the formation three-pressure prediction method with credibility and the actual leakage law of adjacent wells, it can effectively improve the prediction accuracy of the SOWDFD for deep igneous rock strata. The findings of the study can help in better understanding of the complex downhole geological environment in deep igneous rock strata and making reasonable drilling design scheme.

Prediction of Fracture Pressure in Niger Delta Deep Offshore

During drilling operations, it is essential to keep the wellbore pressure within the maximum value of the fracture pressure and minimum value of the pore pressure of the formation. To handle this challenge, the fracture pressure of the formation must be known as it is significant to determining the mud window design. This study developed a correlation that could predict the formation fracture pressure in the Niger Delta deep offshore field. Two different fields were considered for this model named Field 1 and 2. From these fields, fracture pressure data were gotten from 21 wells during leak off test (LOT) at different casing shoe depths. While carrying-out the analysis of data, assumptions were made that the formations throughout the Niger Delta basin obeys the principle of horizontality. Also, that the fracture pressure at same depth is uniform with the pressure at other location in the Delta. Scatter plot was used as the tool for the data analysis. A line of best fit was drawn to arrive at the correlation. This correlation has an R2 coefficient values of 0.9969. In conclusion, the correlation gotten from this study for predicting fracture pressure has shown to align with some data sets from the Niger Delta fields with very little variation. This can be used for planning of further drilling operations in the Niger Delta to make it easier, faster and more economical.

Prediction of Fracture Pressure in Niger Delta Deep Offshore

During drilling operations, it is essential to keep the wellbore pressure within the maximum value of the fracture pressure and minimum value of the pore pressure of the formation. To handle this challenge, the fracture pressure of the formation must be known as it is significant to determining the mud window design. This study developed a correlation that could predict the formation fracture pressure in the Niger Delta deep offshore field. Two different fields were considered for this model named Field 1 and 2. From these fields, fracture pressure data were gotten from 21 wells during leak off test (LOT) at different casing shoe depths. While carrying-out the analysis of data, assumptions were made that the formations throughout the Niger Delta basin obeys the principle of horizontality. Also, that the fracture pressure at same depth is uniform with the pressure at other location in the Delta. Scatter plot was used as the tool for the data analysis. A line of best fit was drawn to arrive at the correlation. This correlation has an R2 coefficient values of 0.9969. In conclusion, the correlation gotten from this study for predicting fracture pressure has shown to align with some data sets from the Niger Delta fields with very little variation. This can be used for planning of further drilling operations in the Niger Delta to make it easier, faster and more economical.

The Niger Delta basin fracture pressure prediction

Accurate knowledge of formation fracture pressure is essential to optimize well design at all stages of the field development. However, erroneous prediction of formation fracture pressure can lead to process safety incidents such as surface and underground blowouts. While fracture pressure prediction models have been developed for some sedimentary basins, it is difficult to transfer these models to areas beyond the regions of study. In the Niger Delta basin, few fracture pressure prediction models have been developed. However, these models were developed primarily from leak-off test data acquired from the normally pressured intervals. Basically, the existing Niger Delta fracture pressure prediction models lack the leak-off test measurements in the overpressure intervals, because such data are not available. In this paper, a new fracture pressure prediction model that can be applied to normally pressured intervals and overpressure zones is being proposed. Model development is based on establishing a relationship between fracture pressure, true vertical depth and magnitude of overpressure using several leak-off test data acquired from over 100 wells in various fields scattered across the basin. Unlike the previous models, the newly developed model incorporates leak-off test measurements from the overpressure intervals in the basin. In general, the newly proposed model can be used with a high degree of confidence to predict the fracture pressure required for safe and economic well planning across the entire basin.

Experimental investigation on the effect of confining pressure on the tensile strength of sandstone using hollow cylinder tensile test method

Rock tensile strength is a factor that affects formation fracturing. Most of the formation fracture pressure prediction models treated it as a constant or ignored it due to its small magnitude. However, the rock tensile strength was observed increasing with elevated crustal stress, so it might be influential at a large depth. In this work, hollow cylinder tensile test method was adopted to measure the tensile strength of two types of sandstones under different confining pressures. Sample in this method has the similar stress state compared with the borehole wall rock, which is important because tensile strength varies with the stress state. The result revealed that there was a significant linear enhancement in rock tensile strength with the increase of confining pressure for both types of sandstones. The effect degree of confining pressure on the tensile strength of two different sandstones was nearly the same.

The Effect of Thermochemical Factors on Fracturing Pressure in Shale Rock Characterized by Tensisle Strength Anisotropy

Drilling mud characteristics affect the stress distribution around the borehole, while anisotropic tensile strength determines the fracture behavior of the formation, but the combined effect of these factors is rarely considered in the prediction of fracture pressure. In this work, tensile strength anisotropy of shale rock was analyzed based on the Brazilian disc test (BDT), and the corresponding anisotropic tensile criteria were reviewed and contrasted with experimental results. The N-Z (Nova – Zaninetti) criterion is adopted to describe tensile strength anisotropy of shale rock. Based on the stress distribution model and the N-Z criterion, a model of shale rock fracture under the combined action of thermal and chemical factors was constructed. The solution of the model shows that chemical and thermal factors have a different effect on pore pressure distribution around the borehole. The effect of sedimentary layers, tensile strength anisotropy, in-situ stress and pore pressure on equivalent density of fracture pressure (EDFP) was also investigated. It is shown that the EDFP decreases with increasing dip angle at a given strike of the bedding plane and reaches a minimum value when the strike of the bedding plane is along the direction of the minimum horizontal stress. The decrease in EDFP caused by tensile strength anisotropy reaches or exceeds 10% of the value calculated for isotropic conditions. The more pronounced the anisotropy of strength, the smaller the possible value of EDFP. The higher the ratio of horizontal stresses and pore pressure, the more significant the effect of anisotropy on EDFP. It is also notable that increasing the temperature of the wellbore can improve EDFP parameters and enlarge the SMDW.

Fracture pressure prediction for layered formations with anisotropic rock strengths

Strength anisotropy is one of the most distinct features of layered formations, but anisotropic tensile strength has received scarce attention in the prediction of fracture pressure. Therefore, the characteristics of anisotropic tensile strength of layered rocks were investigated, main anisotropic tensile failure criteria were reviewed and compared, and the fracture pressure models were deduced for three different anisotropic failure criteria. The results demonstrate that the tensile strength of layered rocks is strongly anisotropic. The Nova–Zaninetti criterion is recommended for the characterization of the anisotropic tensile strength. The variation of the fracture pressure caused by anisotropic tensile strength is close to 10% (it even exceeds 10%), and the anisotropy has a greater influence on the fracture pressure with increasing anisotropic degree. The higher the degree of anisotropy is, the lower the fracture pressure is; the higher the horizontal stress ratio and pore pressure are, the higher the impact of anisotropy on the fracture pressure is. A low dip angle of the bedding plane is beneficial for the fracture of the vertical wellbore, and it is the worst when the minimum horizontal stress is perpendicular to the bedding planes.

Fracture pressure model for inclined wells in layered formations with anisotropic rock strengths

Fracture pressure is one of the most important foundations of drilling engineering, and the anisotropy of tensile strength is also a distinct feature of layered formations, but the anisotropy of tensile strength has seldom received attention in the analysis of fracture pressure. In the present paper, a novel fracture pressure model was proposed for inclined wells in layered formations with anisotropic rock strengths. Firstly, the characteristics of anisotropic tensile strength were investigated. Secondly, four typical anisotropic tensile criteria were collected and assessed using the maximum absolute relative error (MARE), the average absolute relative error (AARE), and the standard error (SE). Thirdly, the fracture pressure model was proposed based on the stress distribution model and anisotropic tensile criterion. Finally, the influences of anisotropy ratio, bedding planes' occurrence, well path, in-situ stress and pore pressure on fracture pressure were investigated. The results indicated that the layered rocks display the distinct anisotropy in tensile strength, and the strength increases with the loading angle. The MARE, AARE and SE of the Nova-Zaninetti criterion is the lowest for all six kinds of typical layered rocks, followed by the Lee-Pietruszczak, Hobbs-Barron, and Single Plane of Weakness criteria, and therefore the Nova-Zaninetti criterion is recommended. The higher the anisotropy ratio in tensile strength, the higher the influence on fracture pressure involved. The influence of bedding planes’ occurrence mainly depends on the initiation position of the fractures, the influence increases with increasing the horizontal stress ratio and pore pressure. Thus, the influence of anisotropy cannot be ignored in the prediction of fracture pressure.

Estimating pore fluid pressure-stress coupling

Fracture Pressure (FP) prediction is a vital component in well planning for safe drilling in high pressure areas, in the estimation of seal capacity and seal breach risk analysis, for fluid injection and in fracture stimulation for oil/gas recovery from tight reservoirs and mature source rocks.Coupling of pore fluid pressure (PP) and FP at basin-scale has been recognised for several decades, with coupling ratio values mostly between 0.46 and 0.87, i.e. for every 10 MPa of PP increase at constant depth, FP increases by 4.6–8.7 MPa. These values are similar to estimates for reduction of FP with reservoir depletion of an oil/gas field during production. Poroelasticity has been suggested as the principal reason for coupling. Most previous coupling values were generated by examination of pressure gradients. A new method to estimate pore fluid pressure-stress coupling is proposed which plots the measured value of FP from a Leak Off Test (LOT) minus the expected FP value on the trend for normal pressures (termed the Fracture Pressure Residual, FPr) against overpressure (OP), which is the magnitude of PP above normal at the same depth. The coupling value is now the slope of FPr against OP. FPr:OP coupling values from 11 global basin datasets range from 0.24 to 0.43, approximately half the previous quoted basin-scale coupling values.A pressure-depth model for rapid deposition on a continental margin, with OP exclusively generated by disequilibrium compaction and no change in effective stress in the overpressured section, represents a base case for compaction OP without poroelasticity. This model helps to reveal (1) that coupling values vary with the ratio of FP to lithostatic pressure (Sv); (2) since the 11 case study coupling values exceed the base case, one or more other (minor) coupling mechanisms are involved, potentially including poroelasticity; (3) values may be independent of tectonic environment; (4) one case study involving carbonate (chalk) is within the same range as those from siliciclastic sediments, and (5) the high coupling pore pressure-stress coupling values in the literature result from using fluid and fracture gradients including the shallow data where PP is normal, which distort the analysis.

Prediction of pore pressure and fracture pressure in Cauvery and Krishna-Godavari basins, India

Geoscientific data from several wells drilled in onshore and offshore parts of the Cauvery and Krishna-Godavari basins, two main hydrocarbon producing basins located in the east coast of India, have been used to determine pore pressures and fracture pressures in the subsurface formations. We have estimated pore pressure based on Zhang's porosity model. Variations of normal compaction curves across the basins are demonstrated here. This study also proposes simple relationships among the parameters used in Eaton's equation for estimating the fracture pressure. Relations established based on the available data in this current study are compressional and shear sonic velocities against bulk density, Poisson's ratio against depth, and the overburden stress against depth. These empirical relationships can be used to predict fracture gradient for the future drilling locations in these basins. The pore pressure in Cauvery basin is shown to be almost hydrostatic in nature, which is due to normal sedimentation rate. High sedimentation rate in the Miocene section of the KG basin is found to be the main reason for overpressure development.

Integrated Approach to Pore Pressure and Fracture Pressure Prediction Using Well Logs: Case Study of Onshore Niger-Delta Sedimentary Basin

This study investigated the cause of identified zones of overpressure in some selected wells in a field in the Niger Delta sedimentary basin. Two models were used each for predicting pore pressure and the corresponding fracture pressure using well log and drilling data. Shale lithology in Niger Delta is massive and characterized by high pore pressure; hence shale compaction theory is utilized in this study. The petrophysical data were evaluated using Ikon’s Science Rokdoc software. The two major pore pressure prediction techniques employed are the Eaton’s and Bowers’ models while the Eaton’s fracture pressure model and the Hubbert and Willis fracture pressure prediction models were utilized for fracture prediction. The density and sonic logs were used respectively to generate the shale trend and the shale normal compaction trend used for the prediction. The wells studied showed disequilibrium compaction of sediment to be the major mechanism that gave rise to overpressure in the Niger Delta. Clay diagenesis and fluid expansion were also observed as the secondary overpressure generation mechanism in well X-1. This secondary overpressure mechanism was observed to start approximately at depths of 10,000 ft (TVD). The top of overpressure and the pressure range in the wells studied varied from 6000 to 11,017 ft (TVD) and 1796.70 to 5297.00 psi respectively. The Eaton’s model under-predicts pore pressure at the depth interval where unloading mechanism is witnessed. Since the study revealed presence of secondary overpressure generation mechanism, Bowers model was observed to be the most reliable pore pressure prediction model in the area.

Study on Influence of Mud Pollution on Formation Fracture Pressure

The mud pollution may change the mechanical properties of rock during oil and gas drilling process, which affects the prediction of fracture pressure, leads to the failure of hydraulic fracturing treatment. Therefore, it is necessary to study influence of mud pollution on formation fracture pressure to improve the forecasting accuracy. The mud pollution has influences on the modulus of elasticity and the Poisson’s ratio of rock by the mud pollution experiment, the core microstructure is observed around the mud pollution. Based on the experiment and research, the effects of mud pollution on the fracturing pressure are studied by finite element software system ANSYS, the factors such as pollution depth, perforation length and Poisson’s ratio of polluted area are taken into account. The result of the experiment indicated that the modulus of elasticity of rock is reduced and the Poisson’s ratio of rock is increased by the mud pollution. Through computing and analyzing, it can be concluded that increases in pollution depth and Poisson’s ratio can lead to a vast increase in formation fracturing pressure. A calculation example is presented and the results show that the results of this research can provide valuable guidance to the designers of hydraulic fracturing treatment.

An automated method for azimuthal anisotropy analysis in marine seismic data

Interest in azimuthal anisotropy analysis on marine seismic data has grown significantly in recent years. One of the reasons for this growing interest is that insights can be gained from these analyses on fracture detection and pressure prediction. Another reason is that a number of 3-D processing problems, such as footprint and loss of frequency content in stacking, that are caused by azimuthal anisotropy can be solved if the effects of azimuthal anisotropy are taken into account. It is, therefore, important to quantify the azimuthal properties of 3-D seismic data. In this paper, we present an automated and efficient method for estimating the azimuthal anisotropy parameters for large volumes of data. The method is automated because it is data driven and it does not involve any picking of velocity or semblance. It involves mainly two steps: first, flattening the CMP gathers that have been NMO corrected to generate time-variant timeshifts; second, fitting these timeshifts to the azimuthal anisotropic traveltime equation in a least squares sense to generate the velocity ellipticity and its direction. Using a recent survey that was acquired off Western Australia, we demonstrate that this method is able to obtain dense anisotropies and directions in large-scale areas in an automated way.

831136 Prediction of fracture pressure for wildcat wells : J Pet Technol, V34, N4, April 1982, P863–872

831136 Prediction of fracture pressure for wildcat wells : J Pet Technol, V34, N4, April 1982, P863–872