Search citation statements
Paper Sections
Citation Types
Year Published
Publication Types
Relationship
Authors
Journals
With the growing demands of challenging well construction operations in the oil and gas industry, cementing operations have become increasingly important. While oilwell cement properties in the short term are largely understood, longer term properties are largely ignored due to difficulties in measuring them. This is problematic because the lifetime of oilwells has grown as technology has improved, with some wells experiencing decades of life. Several of these physical and mechanical properties are dependent on the formulation of the cement – especially the composition, water content, curing conditions as well as conditions downhole in the wellbore. Using limited data available from experimental evaluations, it is possible to evaluate these properties longer term using machine learning approaches, as well as identify possible patterns in the dataset. This paper tests this by subjecting a dataset of representative cement properties which were collected from previous experimental evaluations to different machine learning algorithms such as K-Means and Support Vector Machines (SVM) to create a predictive model. Although there is a lot of work being done on machine learning and evaluating cement characteristics and properties, a lot of it is focused on the construction industry, with little work focusing on oilwell cement. Use of clustering and predictive algorithms can help solve and classify data in real-world oil and gas applications when a large amount of unlabeled field data pertaining to cements is available. The dataset used for the machine learning evaluations comprised of laboratory testing results of over 1100 distinct samples of Class G, H, and C cement, of different formulations and aged for periods ranging from a few days to several months and cured at 25 and 75 degrees Celsius. Among the mechanical and physical properties measures, of note were the densities, unconfined compressive strengths (UCS), pulse velocities (UPV) as well as physical dimensions of the samples. While generating the ML model, the dataset is split into two groups, with 30% of the datapoints used as a validation subset. Once the models are trained and tested, blind analysis is performed to determine possible trends in the cement types, as well as possibly predict the UCS using the available data. Given the availability of sufficient datapoints, machine learning techniques demonstrate promise in properly estimating cement's UCS as well as identifying broad trends in the formulation of the cement samples. When using the K-Means algorithm to identify trends in the cement dataset, the model correctly classified the available datapoints into five separate classes – each corresponding to the class of cement used, as well as the ageing period of the samples. The accuracy of the clustering was verified using blind data as well as by using a K-Nearest Neighbor algorithm to determine the accuracy metrics. UCS of samples was also reliably estimated using the SVM model, which showed excellent error margins and R2 values between actual and predicted datapoints. Optimal analysis of properties for any cement slurry will come from a combination of these approaches and computing the statistical confidence of all predicted datapoints.
With the growing demands of challenging well construction operations in the oil and gas industry, cementing operations have become increasingly important. While oilwell cement properties in the short term are largely understood, longer term properties are largely ignored due to difficulties in measuring them. This is problematic because the lifetime of oilwells has grown as technology has improved, with some wells experiencing decades of life. Several of these physical and mechanical properties are dependent on the formulation of the cement – especially the composition, water content, curing conditions as well as conditions downhole in the wellbore. Using limited data available from experimental evaluations, it is possible to evaluate these properties longer term using machine learning approaches, as well as identify possible patterns in the dataset. This paper tests this by subjecting a dataset of representative cement properties which were collected from previous experimental evaluations to different machine learning algorithms such as K-Means and Support Vector Machines (SVM) to create a predictive model. Although there is a lot of work being done on machine learning and evaluating cement characteristics and properties, a lot of it is focused on the construction industry, with little work focusing on oilwell cement. Use of clustering and predictive algorithms can help solve and classify data in real-world oil and gas applications when a large amount of unlabeled field data pertaining to cements is available. The dataset used for the machine learning evaluations comprised of laboratory testing results of over 1100 distinct samples of Class G, H, and C cement, of different formulations and aged for periods ranging from a few days to several months and cured at 25 and 75 degrees Celsius. Among the mechanical and physical properties measures, of note were the densities, unconfined compressive strengths (UCS), pulse velocities (UPV) as well as physical dimensions of the samples. While generating the ML model, the dataset is split into two groups, with 30% of the datapoints used as a validation subset. Once the models are trained and tested, blind analysis is performed to determine possible trends in the cement types, as well as possibly predict the UCS using the available data. Given the availability of sufficient datapoints, machine learning techniques demonstrate promise in properly estimating cement's UCS as well as identifying broad trends in the formulation of the cement samples. When using the K-Means algorithm to identify trends in the cement dataset, the model correctly classified the available datapoints into five separate classes – each corresponding to the class of cement used, as well as the ageing period of the samples. The accuracy of the clustering was verified using blind data as well as by using a K-Nearest Neighbor algorithm to determine the accuracy metrics. UCS of samples was also reliably estimated using the SVM model, which showed excellent error margins and R2 values between actual and predicted datapoints. Optimal analysis of properties for any cement slurry will come from a combination of these approaches and computing the statistical confidence of all predicted datapoints.
Carbon Capture, Utilization and Storage (CCUS) processes are increasingly being utilized as a viable solution for carbon removal and meet the goal of net-zero carbon emissions by 2050. Captured carbon dioxide (CO2) is stored deep underground – typically in depleted oil or gas (O&G) wells - utilizing technologies and methods currently employed by the energy industry. However, there are certain ongoing well integrity challenges that would need to be addressed – especially those relating to the cement layer. Cement present in wells used for CCUS applications – including old or abandoned wells - need to ensure zonal isolation, be resistant to deterioration, corrosion, or gas migration, as well as be suited for adverse downhole conditions. Oilwell cement present in existing or abandoned O&G assets have been exposed to a wide range of downhole conditions throughout their lifecycle. It is generally very difficult to determine the mechanical properties and physical condition of the cement downhole and a decline in these properties is expected over time. Experimental evaluations have shown that temperature plays a role in the setting and maturity of the cement, and in CCUS wells, corrosive factors are a major concern due to the acidic environment produced at the CO2 injection zone. These can significantly affect cement mechanical properties such as the Uniaxial Compressive Strength (UCS). Evaluations have shown Temperature or Acoustic Logs can be used to determine downhole properties which can then be correlated to the behavior of cements and the change in their mechanical properties over time using machine learning algorithms. Laboratory evaluations showed varying mechanical properties for oilwell cement at different temperatures and degradation over time. Overall, Class G cements developed the highest stress failure resistance, followed by Class H cements. Higher temperatures accelerated the setting time of all cement samples due to rapid dehydration. However, this in turn reduced the peak UCS developed, indicating a lower stress failure criterion. UCS also showed a direct relationship to acoustic data which can be utilized to evaluate mature and abandoned wells for their integrity. When modeled using supervised machine learning algorithms, field temperature data and acoustic data can reliably predict the mechanical properties of cements over time. An artificial neural network model, and two tree based models were developed, which showed good correlation in predicting compressive strength of downhole cements. Properly understanding the behavior of oilwell cement and the evolution of their mechanical properties is critical to ensure safe storage. Data driven algorithms which can correlate the dynamic mechanical properties of cement to the temperature gradient and acoustic logs can help reliability predict the integrity of the cement layer over time especially for CCUS applications.
Complex operations such as fracturing, and stimulations have become a mainstay in most drilling and completion operations around the world. Safe technologies have been adopted by the industry to mitigate issues in complex wells, HPHT conditions and difficult formations. However, well integrity problems - especially in the cement layer - are still a major concern in a lot of cases when performing workover, fracturing or re-completion operations in existing or older abandoned wells. Oilwell cement used in drilling and completion comes in several different classes and grades. Geopolymer based cements are also increasingly being considered for cementing operations, owing to their green credentials. Commonly used API Class C, Class H and Class G cements as wells as Geopolymers all have mechanical properties which vary widely, and a decline in these properties are expected after exposure to different downhole conditions over time. Experimental evaluations were performed to measure mechanical properties such as the Uniaxial Compressive Strength (UCS) and acoustic velocities and determine how they vary over time and under different physical environments. Finite element stress modeling was then performed to determine failure mechanisms in downhole conditions. Degradation of the cement layer due to ageing, as well as exposure to different downhole temperatures especially in the cement-casing interface are of particular interest. Each of the classes of oilwell cements perform differently and thereby have a different impact on the overall integrity of the well. Results from laboratory testing of samples showed significantly different mechanical properties during the mixing, setting and ageing periods for different oilwell cement classes and at different temperatures. Among the different formulations tested, Class G cement showed the highest failure stress with almost all samples showing a consistent peak UCS growth, before stabilizing. Class C cements and Geopolymers had the lowest stress failure resistance, indicating their unsuitability for HPHT operations. Higher temperatures accelerated the setting time, though reduced the UCS for all classes of cement. When stresses experienced during typical fracturing operations were modeled in a downhole scenario with these cements, propagating failure points were observed. Stresses can migrate and concentrate at different points - which in some cases can exceed the failure criteria of these cements leading to the formation of cracks. These can in turn cause integrity issues in the cement sheath and possibly a critical well integrity situation. Robust testing of oilwell cements and geopolymers is needed to properly understand their properties, as well as the development of stress failure points around the wellbore. Identifying potential well integrity issues for various cement formulations can in turn help in improving the quality and reliability of cementing operations, reduce the risks associated and ensure safe operations over the lifespan of a well.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.