Paper presented at SPE Liquids-Rich Basins Conference – North America, Midland, Texas, USA, September 2017.

Abstract

Fracture pressure plays a key role in designing the mud weight and the cement slurry density in the drilling operation. Knowing the fracture pressure will eliminate many problems such as loss of circulation and hence reduce the time and the cost of the drilling operation. Many empirical models reported in the literature were used to calculate the fracture pressure based on different parameters. Most of these models used only formation and rock properties to estimate fracture pressure. Other models predicted the fracture pressure based on log data using a few real field data. Artificial intelligence techniques once optimized can be used to predict the fracture pressure with high accuracy. The objective of this research is to predict the fracture pressure based only on surface drilling parameters which are easy to get namely weight on bit (WOB), rotary speed (RPM), drilling torque (τ), rate of penetration (ROP), mud weight (MW) and formation pressure (Pf). More than 4700 real field data points are used to predict fracture pressure using Functional Networks (FN) which is a method of artificial intelligence (AI). Functional Networks (FN) tool was compared with different empirical models. The result showed that FN methods outperformed all the fracture pressure equations by high margin (very high correlation coefficient (R) of 0.986 and a very low average absolute percentage error (AAPE) of 0.201). the developed technique will help the drilling engineers to design the cement slurry and determine the casing setting depth. In addition, the drilling engineers will be able to eliminate the common drilling problems such as loss of circulation.

Presented at SPE Kingdom of Saudi Arabia Annual Technical Symposium and Exhibition, Dammam, Saudi Arabia, April 2018.

Abstract

Formation pressure is the main function that affects drilling operation economically and efficiently. Knowing the pore pressure and the parameters that affect it will help to reduce the cost of the drilling process. Many empirical models reported in the literature were used to calculate the formation pressure based on different parameters. Some of these models used only drilling parameters to estimate pore pressure. Other models predicted the formation pressure based on log data. All of these models required different trends such as normal or abnormal to predict the pore pressure. Few researchers applied artificial intelligence (AI) techniques to predict the formation pressure by only one method or a maximum of two methods of AI. The objective of this research is to predict the pore pressure based on both drilling parameters and log data namely; weight on bit (WOB), rotary speed (RPM), rate of penetration (ROP), mud weight (MW), bulk density (RHOB), porosity (ϕ) and compressional time (Δt). A real field data is used to predict the formation pressure using Fuzzy Logic (FL) which is one technique of AI. Fuzzy Logic (FL) tool was compared with different empirical models. FL method estimated the formation pressure with a high accuracy (high correlation coefficient (R) of 0.998 and low average absolute percentage error (AAPE) of 0.234%). FL outperformed all previously published models. The advantage of the new technique is its simplicity, which represented from its estimation of pore pressure without the need of different trends as compared to other models which require a two different trend (normal or abnormal pressure).

Presented at SPE Kingdom of Saudi Arabia Annual Technical Symposium and Exhibition, Dammam, Saudi Arabia, April 2018.

Abstract

Rate of penetration (ROP) is the main function that affects drilling operation economically and efficiently. Many theoretical models reported in the literature were produced to predict ROP based on different parameters. Most of these models used only drilling parameters to estimate ROP. Few models have considered the effects of drilling fluid on ROP using a simulated data or a few real field data. Some of the researchers used artificial intelligence to predict ROP by only one method. The objective of this research is to predict ROP based on both drilling parameters and mud properties such as weight on bit (WOB), rotary speed (RPM), pump flow rate (Q), standpipe pressure (SPP), drilling torque (τ), mud density (MW), plastic viscosity (PV), funnel viscosity (FV), yield point (YP) and solid (%). More than 400 real field data in shale formation are used to predict ROP using support vector machine (SVM) which is a method of artificial intelligence (AI) and compare it with different mathematical models. The result showed that support vector machine (SVM) technique outperformed all the theoretical equations of ROP by a high margin as shown by a very high correlation coefficient (CC) of 0.997 and a very low average absolute percentage error (AAPE) of 2.83%.

Salaheldin M. Elkatatny and Hisham A. Nasr-El-Din
Presented at International Journal of Oil, Gas and Coal Technology, 2014 Vol.7 No.2, pp.169 – 188

Abstract

Chelate solutions, GLDA (pH of 3.3 to 13) and HEDTA (pH of 4 and 7) were incompatible with α-amylase over a wide range of temperatures. GLDA (pH 3.3) and HEDTA (pH 4) can be used to remove the filter cake in one step. GLDA (20 wt% in a 200 g solution and pH of 3.3) and HEDTA (20 wt% in a 200 g solution and pH 4) had 100% removal efficiency of the filter cake. The retained permeability was 110% and 106% for Berea sandstone and Indiana limestone cores, respectively when using GLDA (13.3 wt% in 300 g solution and pH of 3.3). The retained permeability was 115% and 100% for Berea sandstone and Indiana limestone cores, respectively when using HEDTA (20 wt% in 200 g solution and pH of 3.3). No core damage was observed when using GLDA and HEDTA solutions as a breaker to remove water-based filter cake.

Salaheldin Elkatatny, Zeeshan Tariq, Mohamed Mahmoud, Abdulazeez Abdulraheem & Ibrahim Mohamed
Paper presented at Neural Computing and Applications volume 31, pages4123–4135 (2019)

Abstract

Elastic parameters play a key role in managing the drilling and production operations. Determination of the elastic parameters is very important to avoid the hazards associated with the drilling operations, well placement, wellbore instability, completion design and also to maximize the reservoir productivity. A continuous core sample is required to be able to obtain a complete profile of the elastic parameters through the required formation. This operation is time-consuming and extremely expensive. The scope of this paper is to build an advanced and accurate model to predict the static Young’s modulus using artificial intelligence techniques based on the wireline logs (bulk density, compressional time, and shear time). More than 600 measured core data points from different fields were used to build the AI models. The obtained results showed that ANN is the best AI technique for estimating the static Young’s modulus with high accuracy [R2 was 0.92 and the average absolute percentage error (AAPE) was 5.3%] as compared with ANFIS and SVM. For the first time, an empirical correlation based on the weights and biases of the optimized ANN model was developed to determine the static Young’s modulus. The developed correlation outperformed the published correlations for static Young’s modulus prediction. The developed correlation enhanced the accuracy of predicting the static Young’s modulus. (R2 was 0.96 and AAPE was 6.2%.) The developed empirical correlation can help geomechanical engineers determine the static Young’s modulus where laboratory core samples are not available.

Elkatatny, Salaheldin; Tariq, Zeeshan; Mahmoud, Mohamed; Mohamed, Ibrahim; Abdulraheem, Abdulazeez
Paper presented at Arabian Journal for Science & Engineering (Springer Science & Business Media B.V. ) . Nov2018, Vol. 43 Issue 11, p6375-6389. 15p.

Abstract

Compressional (P-wave) and shear (S-wave) velocities are used to estimate the dynamic geomechanical properties including: Poisson’s ratio, Young’s modulus, and Lamé parameters. These parameters are mainly used in estimating the static properties of the formation rocks as well as the in situ stresses. The sonic logs are not always available, epically for old wellbores. Also, in several occasions when the sonic logs are available, missing sections found in the well logs might affect the analysis results. To the authors’ knowledge, there is no single straightforward correlation that can be used to accurately estimate both P- and S-wave travel times directly from the well log data. Most of the existing correlations use the P-wave velocity to measure the S-wave velocity. The main purpose of this study is to develop accurate and simple empirical models using wireline log data (bulk density, gamma ray, and neutron porosity) to predict the sonic travel times (P-wave and S-wave). These wireline logs are slandered wireline log data that are commonly recorded in most of the wells. Three robust artificial intelligence techniques, namely: support vector machine (SVM), artificial neural network (ANN), and adaptive neurofuzzy interference systems (ANFIS), were employed and compared based on their prediction performance. Ultimately, using the weights and biases of optimized ANN model, a simple generalized empirical correlation is derived that can be used without the need of costly commercial software’s to run the AI models. The obtained results showed that ANN, ANFIS, and SVM can be used to estimate P-wave and S-wave travel times. ANN outperformed the ANFIS and SVM by yielding the lowest average absolute percentage error (AAPE) and the highest coefficient of determination R2<inline-graphic></inline-graphic> for predicting P-wave and S-wave travel times. ANN model could predict the P-wave and S-wave travel times from wireline log data with high accuracy giving R2<inline-graphic></inline-graphic> of 0.98 when compared to actual field data. In addition, the developed empirical correlations prediction completely matched the ANNs prediction. The AAPE of the predicted P and S-waves travel times was less than 5%. The developed correlations are very accurate and can help geomechanical engineers to determine the dynamic geomechanical properties (Poisson’s ratio and Young’s modulus) and propose any operation in case where sonic logs are missing.

Panchal Y., Kholy SM., Loloi M., Mohamed IM., and Abou-Sayed, O.
Presented at Recent Adv Petrochem Sci. 2017; 3(5): 555621. DOI:10.19080/RAPSCI.2017.03.555621

Abstract

Subsurface fractured injection (sometimes called cuttings re-injection, drill cuttings injection, or slurry injection) has been proven over the past decades to be the safest, most efficient, and the lowest-cost technology for disposal of certain kinds of oil and gas waste. This technology involves creating a hydraulic fracture in a subsurface injection formation followed by an intermittent process of pumping the slurrified waste into the fracture. The objective of this study is to investigate the impact of changing the rheological properties of the slurrified waste on the hydraulic fracture geometry. The investigation was conducted in two main steps: first, using the geophysical information a geotechnical earth model was built to estimate the mechanical properties of different subsurface formations. This allowed the selection of a porous/permeable injection formation which is over-laid and under-laid by proper stress barriers. Second, a commercial 3-D fracture simulator (@Frac 3D) was used to study the impact of changing the rheological properties of the injection fluid such as viscosity, solids concentration, and injection rate on the geometry of the hydraulic fracture and net pressure. The results show that solids concentration, injection rate and fluid viscosity are proportional to the fracture width and net pressure.

Mohamed, I.M., Block, G.I., Abou-Sayed O.A., and Abou-Sayed A.S. 2016
Presented at IPTC-18885-MS accepted for publication at the International Petroleum Technology Conference held in Bangkok, Thailand, 14–16 November 2016.

Abstract

Class I and Class II waste re-injection are the most important methods for disposing of fluid in North Dakota: in 2007, more than 96% of produced water were disposed of using underground injection, and by 2012 all produced water was being managed by underground injection. While Class II injection covers waste produced from most Exploration & Production (E&P) activities, Class I injection wells are used for disposing of a special class of industrial wastes, including waste generated by petroleum refining, metal production, chemical production, pharmaceutical production, commercial disposal, and food production. Non-hazardous industrial waste and Naturally Occurring Radioactive Materials (NORM) not associated with E&P can also be injected using Class I wells. In all cases, the primary concern for permitting and safe operations is to (1) predict the movement of the injected waste to ensure that it stays within pre-defined formations, and (2) ensure that pore-pressure increases caused by injection do not impact neighboring offset wells. Results from a geochemical study of the feasibility of disposal into the Dakota Sands (Inyan Kara formation) in North Dakota is being presented. Analyses were made using a compositional Reservoir Optimization (REVEAL) to predict the pore-pressure distribution, direction and movement of the injected fluid, as well as chemical reactions between formation brine/waste/formation rocks and the effect of these chemical reactions on formation injectivity and cap rock integrity. Forecasts indicate that for over 50 years of injection, the injected wastes will be completely trapped within the Dakota Sands (no fluid flow is expected to penetrate through the cap rock) and injection pressures are expected to remain well below the estimated fracture pressure. While the Inyan Kara formation is therefore a reasonable storage trap for industrial wastes, carbonate and sulfate scales may cause near wellbore formation damage and rising wellhead pressures which operators will need to address.

S. M. Kholy; I. M. Mohamed; M.. Loloi; O.. Abou-Sayed; A.. Abou-Sayed
Paper presented at SPE Liquids-Rich Basins Conference – North America, Midland, Texas, USA, September 2017.

Abstract

During hydraulic fracturing operations, conventional pressure fall-off analyses (G-Function, Square Root of Time, and Diagnostic Plots) are the main methods for predicting fracture closure pressure. However, there are situations when it is not practical to determine the fracture closure pressure using these analyses. These conditions occur when closure time is long, such as in mini-frac tests in very tight formations, or waste fluid injection in reservoirs where there is low native permeability or where there is significant near wellbore damage. In these situations, it can take several days for the shut-in pressure to stabilize enough for conventional pressure fall-off tests analyses to be used. Thus, the objective of the present study is to attempt to correlate the fracture closure pressure to the early time fall off data using the field-measured Initial Shut-in Pressure (ISIP), rock properties and pumped / injection volumes. A study of the injection pressure history of many injection wells with multiple hydraulic fractures in a variety of rock lithologies shows an interesting relationship between the fracture closure pressure and the initial shut-in pressure. An empirical equation has been created to calculate the fracture closure pressure as a function of the instantaneous shut-in pressure (ISIP) and the injection formation rock properties. Such rock properties include formation permeability, formation porosity, reservoir pressure, overburden pressure, Poisson’s ratio, and Young’s modulus. An empirical equation was developed using the injected volumes combined with data obtained from geomechancial models and core analysis of a wide range of injection horizons in terms of lithology type: Sandstone, Carbonate, and Shale. The empirical equation was validated using different case studies by comparing the predicted fracture closure pressure calculated using the developed empirical equation to the measured fracture closure pressure value. The reported correlation predicted the fracture closure pressure with a relative error of less than 6%. Also, the empirical equation was used to predict the fracture closure pressure in a shale formation with less than 3% error. The new empirical equation predicts the fracture closure pressure using a single point of falloff pressure data, the ISIP, without the need to conduct a conventional fracture closure analysis. This allows the operator to avoid having to collect pressure data between shut-in and until the actual fracture closure point which can take several days in highly damaged, very tight, and/or shale formations. Moreover, in operations with multiple batch injection events into the same interval / perforations, as is often found cuttings / slurry injection operations, the trends in closure pressure evolution can be tracked even if the fracture is never allowed to close.

Presented at 51st U.S. Rock Mechanics/Geomechanics Symposium, San Francisco, California, USA, June 2017.

Abstract

Unconfined compressive strength (UCS) is the key parameter to; estimate the in situ stresses of the rock, alleviate drilling problems, design optimal fracture geometry and to predict optimum mud weight. Retrieving reservoir rock samples throughout the depth of the reservoir section and performing laboratory tests on them are extremely expensive as well as time consuming. Therefore, mostly UCS predicted from empirical correlations. Most of the empirical correlations for UCS prediction are based on elastic parameters or on compressional wave velocity. These correlations were developed using linear or non-linear regression techniques. This paper presents a rigorous empirical correlation based on the weights and biases of Artificial Neural Network to predict UCS. The testing of new correlation on real field data gave a less error between actual and predicted data, suggesting that the proposed correlation is very robust and accurate. Therefore, the developed correlation can serve as handy tool to help geo-mechanical engineers in order to determine the UCS.

Presented at 51st U.S. Rock Mechanics/Geomechanics Symposium, San Francisco, California, USA, June 2017.

Abstract

Static Poisson’s ratio plays a vital role in calculating the minimum and maximum horizontal stresses which are required to alleviate the risks associated with the drilling and production operations. Incorrect estimation of Static Poisson’s ratio may wrongly lead to inappropriate field development plans which consequently result in heavy investment decisions. Static Poisson’s ratio can be determined by retrieving cores throughout the depth of the reservoir section and performing laboratory tests, which are extremely expensive as well as time consuming. The objective of this paper is to develop a robust and an accurate model for estimating static Poisson’s ratio based on 610 core sample measurements and their corresponding wireline logs data using artificial neural network. The obtained results showed that the developed ANN model was able to predict the static Poisson’s ratio based on log data; bulk density, compressional time, and shear time. The developed ANN model can be used to estimate static Poisson’s ratio with high accuracy; the correlation coefficient was 0.98 and the average absolute error was 1.3%. In the absence of core data, the developed technique will help engineers to estimate a continuous profile of the static Poisson’s ratio and hence reduce the overall cost of the well.

Presented at SPE J. 22 (05): 1530–1541.

Abstract

Carbon dioxide (CO2) sequestration in coal seams combines CO2 storage with enhancing methane (CH4) recovery. The efficiency of CO2 sequestration depends on the coal-formation properties and the operating conditions. This study investigated the effects of the sodium chloride (NaCl) salinity of coal-seam water, injection flow rate, injected-gas composition, and CO2 state (formation pressure) on CO2 sequestration in coal formations. Coreflood tests were conducted on nine coal cores to simulate the injection of CO2 into coal formations. The change in the effective water/coal permeability after CO2 injection was measured. A commercial simulator was used to match the pressure drop across the core from the experimental study by adjusting the relative permeability curves. Moreover, permeability dynamic measurements were conducted to estimate the absolute permeability reduction caused by coal swelling. The effective water permeability in the tested coal decreased during CO2 injection because of its adsorption onto the coal surface, coupled with a reduction in the relative water permeability. As salt concentration increased, the change in the pressure drop across the core increased, but this effect decreased as the formation pressure increased. Higher formation pressure and lower nitrogen (N2) concentrations led to further permeability reduction as a result of the higher CO2 adsorption onto the coal surface. Furthermore, as the injection flow rate increased, the contact time of CO2 at the coal surface decreased. Hence, the CO2 adsorption to the coal matrix decreased, and thus the difference in the effective water permeability slightly decreased. CO2 injectivity in fully water-saturated formations increased initially as the gas relative permeability increased, then the injectivity decreased as a result of matrix swelling and absolute permeability reduction. Moreover, the water salinity in coal formations decreased the overall gas relative permeability and increased the water relative permeability. Similar behavior occurred in the presence of N2. It is derived from these observations that the injection of CO2 into highly volatile bituminous coal seams for CO2 sequestration purpose is more efficient as the salt concentration increases, especially at high injection pressures.