Method to tailor performance properties of cement barriers
A model-based approach using cement performance characteristics models optimizes cement mortar formulations to meet specific performance criteria, addressing the inconsistency in wellbore cementing by predicting and controlling properties like compressive strength and Young's modulus.
Patent Information
- Application Number
- SA1120262593
- Authority / Receiving Office
- SA · SA
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2026-03-30
- Publication Date
- 2026-10-05
AI Technical Summary
Existing methods for designing cementitious mortar formulations lack a systematic approach to ensure that the final cement meets specific performance characteristics such as compressive strength, tensile strength, and Young's modulus under varying wellbore conditions, leading to inconsistent wellbore cementing results.
A model-based approach using cement performance characteristics models that incorporate factors like cement composition, well conditions, and interactions between components to predict and control performance characteristics, employing equations and machine learning techniques to optimize cement mortar formulations.
Ensures that the cement mortar formulations meet predefined performance requirements, resulting in a reliable wellbore cement barrier with predictable and satisfactory properties under varying downhole conditions.
Smart Images

Figure 051F9354-C355-47B9-9012-924757E90C1C 
Figure 0530A168-3646-4FC3-B213-0A2BDA5F62C5 
Figure 0F1E3917-50E7-48A4-8259-A1786D3A8F9C
Abstract
Description
This discovery may relate generally to methods and systems for cementitious cementing, and more specifically, its manifestations may relate to the design of cement mortar formulations using a cement performance characteristics model, as well as the preparation of cement mortar based on a cement mortar formulation. In particular, a model-based approach to designing a target cement performance characteristic has been revealed. Existing methods and systems may employ a model-based approach that links performance characteristics to barrier composition, hardening, and test conditions. Cementitious mortars may contain cement, supplementary cementitious additives, inert materials, and chemical additives. Cementitious mortar formulations, sometimes referred to as cement designs or by other equivalent names, may be specific to each well to meet the different design requirements of each well. Cementitious mortar formulations can be developed so that the mortar prepared from the cementitious mortar formulation meets the design requirements related to cementitious mortar placement, for example, viscosity, density, and flow properties, and so that the cement resulting from the hardening of the cementitious mortar meets all the design requirements related to long-term safety, for example, compressive strength, tensile strength, and Young's modulus. When developing a cementitious mortar formulation, demonstration samples of the cementitious mortar can be prepared and tested in the laboratory to verify the availability of the required physical properties in the cementitious mortar and the cementitious setting. The performance characteristics of cement prepared from a cement mortar recipe can be measured in the laboratory using standard tests to ensure that the performance characteristics remain above the target performance characteristics.Once it is verified that a cement mortar mix meets the design requirements, the mix can be selected to begin preparation and the resulting cement mortar is introduced into the wellbore. The cement mortar used for wellbore cementing is typically mixed at the wellbore casing location using cement mixing equipment and pumped into the wellbore using cement pumps. Models of cement performance characteristics are presented here. One or more basic cement performance characteristics models may map the cement mortar composition and test conditions to predict API performance characteristics. These models can be used to design cement mortar formulations to achieve a target performance characteristic. One or more basic cement performance characteristics models may account for the influence on performance characteristics caused by mixed bulk components such as Portland cement, cementitious additives, crystalline silica, weighting agents such as beads, and the amount of water. One or more basic cement performance characteristics models may also capture the interactions between major components, as well as the effects of temperature on the contribution of each mortar component to the predicted performance characteristics. The performance characteristics of cement mortar are controlled by several factors, including the amount of water in the mortar (w), the type and amount of Portland cement (Pi), the type and amount of pozzolanic material (Poz) (e.g., fly ash, pumice, silica, cement kiln dust, etc.), the type and amount of cementitious complementary material (Sj), the type and amount of inert material (Ik) (e.g., lightweight beads, heavy metal oxides, rotation loss materials, rubber, fibers, etc.), the type and amount of chemical additives (e.g., accelerators, retarders, polymers, latex, polyvinyl alcohol, resin, crystalline silica, etc.) that may affect the performance characteristics (Cl), curing temperature (T), curing time (t), and ambient confining pressure (P). Furthermore, there may be other factors such as sand, salts, and any other type of additive.The general formula for performance characteristics may include Equation 1, where f can be a linear or nonlinear function, and the indices i, j, k, and l represent the possibility of multiple Portland cement types, various pozzolanic materials, complementary cementitious materials, inert materials, and chemical additives. In some embodiments, the function f can be described as a neural network, a decision tree, or an explicit mathematical formula. Equation 1 Illustrative performance characteristics include ultimate compressive strength, tensile strength, cohesion, angle of friction, or other elastic properties (EP). Elastic properties may include Young's modulus and Poisson's ratio. Without being bound by a specific testing technique, compressive strength is generally the ability of a material or a solid particle structure to withstand axial thrust forces. The compressive strength of a cementitious component can be measured at a specific time after the cementitious component has been mixed with water and the resulting cement mortar has been kept under specific temperature and pressure conditions. For example, compressive strength can be measured 24 to 48 hours (or longer) after the cement mortar has been mixed and kept at temperatures ranging from 40°F to approximately 500°F and at atmospheric or high pressure. Compressive strength can be measured using either a destructive or non-destructive method. The destructive method involves physically testing the strength of cement samples by crushing them in a compression testing machine. Compressive strength can be calculated by dividing the breaking load by the cross-sectional area resisting the load and is reported in pounds per square inch (psi).Non-destructive methods may typically utilize an Ultrasonic Cement Analyzer ("UCA"), available from Fann® Instrument Company, Houston, Texas. Compressive forces can be determined according to API RP 10B-2, Recommended Practice for Testing Well Cements, First Edition, July 2005. Without being bound by any specific testing technique, tensile strength is generally the ability of a molecular solid to withstand loads that tend to cause elongation, as opposed to compressive strength. For example, the tensile strength of a cementitious component can be measured at a specific time after the cementitious component has been mixed with water and the resulting cement mortar has been kept under specific temperature and pressure conditions. For instance, tensile strength can be measured 24 to 48 hours (or longer) after the sample has been mixed and kept at a temperature between 40°F and approximately 500°F and under atmospheric or high pressure. Tensile strength can be measured using any suitable method, including, but not limited to, the procedure described in ASTM C307. This means that samples may be prepared in briquette molds shaped like dog biscuits with a 1-inch square cross-sectional area in the center. Tensile pressure can then be applied to the swollen edges of the samples until the samples break in the central area. The tensile strength at which the model breaks, measured in pounds per square inch, represents the tensile strength of the molecular solid being tested. Young's modulus, also known as the elastic modulus, is a measure of the relationship between applied stress and resulting strain. Generally, a highly deformable (resilient) solid particulate material will exhibit a lower modulus of elasticity. Therefore, Young's modulus is an elastic constant that describes the ability of a tested particulate solid to deform under applied loads. Several different laboratory techniques can be used to measure the Young's modulus of a sample containing a pozzolanic component after allowing the sample to solidify for a period of time under specific temperature and pressure conditions. Without being bound by any specific testing technique, the angle of friction is the angle at which a granular material can settle or incline naturally without collapsing or experiencing shear failure; it measures the internal resistance to shear within the material. More specifically, the angle of friction is the angle of the inclined surface at which the material begins to slide or deform. Without being bound by any specific testing technique, cohesion is generally the ability of cement components to bond together. Cohesion may be a function of ultimate compressive strength and the angle of friction, as shown in Equation 2. Equation 2 is valid for isotropic brittle materials, such as cement. In Equation 2, UCS is the ultimate compressive strength, cohesion is the bond strength, and φ is the angle of friction. Equation 2 One of the model-based techniques for predicting performance properties involves Equation 3. In Equation 3, is a performance property of the cement (e.g., elastic property), w is the mass of water, Mass Poz is the mass of pozzolan, Vol Poz is the volume of pozzolan, Voli is the volume of the pozzolanic species, Mass Addj is the mass of the additive j, Mass blendk is the mass of the blend species k, Well Condition l is the well condition, e.g., temperature, pressure, etc., and a0, ai, aj, ak, and al are constants. Equation 3 Equation 3 takes a linear form. The model-based technique may, instead, or in addition, include Equation 4, which has a non-linear form. In Equation 4, is a performance property of the cement (e.g., elasticity), w is the mass of water, Mass Poz is the mass of the pozzolan, Vol Poz is the volume of the pozzolan, Voli is the volume of the pozzolan type i, Mass Addj is the mass of the additive j, Mass Blendk is the mass of the blend type k, Well Condition l is the well condition, e.g., temperature, pressure, etc., and a0, ai, aj, ak, and al are constants. Equation 4 The model-based techniques described here may, instead of or in addition to these, incorporate other forms of equations, such as polynomial equations, physics-based equations, linear regressions, etc., that relate one or more elastic properties to cement composition and well conditions. These equation-type models may allow personnel to interpret the behavior of different factors without requiring new predictions for a property each time a new cement composition is analyzed. For example, the sign and magnitude of the constant a0 will predict whether the elastic property will increase or decrease with the addition of more water, and will also determine the magnitude of the increase or decrease. The model-based techniques described here may, instead of or in addition to, incorporate one or more black-box models, including, for example, various neural networks, tree-based models (e.g., XG Boost), random forest models, and others. Cement performance characteristic models may be able to capture the complex relationship between different model parameters and may be particularly useful when deterministic models perform poorly. The outputs of random forest models are shown in subsequent figures relating ultimate compressive strength, cohesion, tensile strength, Young's modulus, and Poisson's ratio to the cement composition and well conditions (e.g., curing temperature, curing pressure, setting time, and ambient pressure). Other forms of black-box models may include, but are not limited to: various machine learning algorithms and / or artificial intelligence; a supervised, semi-supervised, unsupervised, and / or enhanced model; a binary classification model; a multi-class classification model; a regression model; decision trees; a random forest classifier; logistic regression; support vector machine algorithms (SVM); a naive Bayes classifier; k-nearest neighbors (K-NN) algorithms; clustering; k-means clustering; a dimensionality reduction algorithm; a gradient boosting algorithm; a probabilistic classifier; one or more physics-informed neural networks (PINNs); and any combination thereof. When cement performance modeling in this investigation incorporates machine learning, the method may rely on pre-trained models, which may have been trained using training data. For example, the training data might include measured or calculated performance characteristics of real cement compositions, which may include data from, or data derived from, any of the laboratory techniques mentioned here. The pre-trained models can be formed, for example, from machine learning algorithms, of which there are a wide variety of suitable classes and types. In general, machine learning algorithms that may be able to capture underlying relationships within a dataset can be divided into different categories. One such category might include whether the machine learning algorithm functions use supervised, unsupervised, semi-supervised, or reinforcement learning. The goal of a supervised learning algorithm might be to identify one or more dependent variables based on their relationship with one or more independent variables.Directed learning algorithms are called such because the dataset contains both independent and corresponding dependent values, with the dependent value being the "answer." The model then seeks to predict outcomes based on implicit relationships within the dataset. Thus, the goal of a model developed using a directed learning algorithm can be to predict the outcome of one or more scenarios for which no outcome is yet known. Directed learning algorithms can also be categorized according to their function as classification and regression algorithms. When the dependent variable is a labeled or categorical value, the algorithm can be referred to as a classification algorithm. When the dependent variable is an extended numerical value, the algorithm can be a regression algorithm. In an unrestricted example, algorithms used for directed learning might include neural networks, K-nearest neighbors, Naïve Bayes, decision trees, classification trees, regression trees, random forests, linear regression, support vector machines (SVMs), step-support regression, genetic algorithms, and perception backpropagation. The goal of unsupervised machine learning may be to identify similarities and / or differences between data points within a dataset, potentially allowing the dataset to be divided into groups or clusters without knowing which group or cluster the data belong to. Datasets used in supervised learning may not include a dependent variable, as a target function of this type of algorithm identifies one or more groups or categories within the dataset. In an unrestricted example, algorithms used for unsupervised machine learning could include K-means clustering, K-means classification, fuzzy C-means, Gaussian mixtures, the Hidden Markov model, neural networks, and hierarchical algorithms. Independent variables in a machine learning model, or any other type of machine learning-driven cement performance characteristic model, may include, for example, setting temperature, setting pressure, setting time, ambient pressure, and material composition. Setting temperature is the temperature at which the cement paste hardens. It may refer to a single temperature, a range of temperatures, or, in some cases, a function of time. In some instances, setting temperature may represent the heat of reaction during the setting process. Thus, the setting temperature used as an independent variable in a machine learning model may, in some cases, encompass a range of values over a period of time, or a single starting temperature. Another independent variable may be setting pressure. Setting pressure may include a single pressure value or multiple values. It may also be time-related and influenced by various factors, such as species concentration and cement reactivity, before being used as an independent variable in machine learning or prediction.In some cases, setting pressure may simply be the pressure exerted on the cement mortar when it is first placed in the location where it will permanently set. Setting time may refer to the amount of time the cementitious composition is expected to remain unaffected by well operations. In some cases, setting time is measured based on the time it takes for the mortar to develop more than 90% hydration. Ambient pressure, which differs from setting pressure, refers to the subsurface geological pressure exerted on the cement by the formation after the cement has been allowed to set. The dependent variables determined by the machine learning model may include any of the performance characteristics described here, including but not limited to, ultimate compressive strength, tensile strength, Young's modulus, Poisson ratio, cohesion, and angle of friction. Cement mortar can be prepared by mixing dry cement mix with water. Dry cement mixes can be prepared in bulk for cementing. A cement mortar recipe may include a list of solid and liquid cementitious components and their quantities to be included in the dry bulk mix. Cement mortar recipes can be designed to have specific performance characteristics, ensuring that the mortar prepared according to the recipe has predictable value. The components of a dry bulk cement mix are typically measured and then mixed together using appropriate mixing equipment. A cement mortar recipe may include a list of bulk materials and their quantities, such as cementitious materials that develop compressive strength when mixed with water, as well as supplementary cementitious materials that do not develop compressive strength when mixed with water but contribute to compressive strength when mixed with cement and water.A cement mortar recipe may also include a list of inert materials and chemical additives, along with their quantities. These include chemical additives that modify the physical properties of the cement mortar prepared using the recipe or the cement itself. Additives may include, but are not limited to, salts, retarders, accelerators, gel strength regulators, and suspending agents. A cement mortar recipe may also specify the amount of water to be added when preparing the mortar, based on the desired density. The dry bulk cement mix can be transported to a location, such as a wellbore, where it can be mixed with water, either in the quantity specified in the recipe or to obtain a cement mortar of the required density, which can then be injected into the wellbore. In some formulations, a cement mortar recipe may also include liquid additives that can be mixed with the cement mortar. The performance characteristics of cement slurry depend primarily on the selection of its main components. Therefore, the choice of raw materials can influence the performance characteristics observed under specific well conditions, such as pressure and temperature. The dependence of performance characteristics on temperature and pressure can be expressed in various ways, including, but not limited to, as an Arrhenius expression or a polynomial. The temperature and pressure dependence can also be expressed as an exponential, logarithmic, trigonometric, or any other transcendental or analytic function, or even described by neural networks, decision trees, or a combination thereof. Performance characteristics may also depend, for example, on particle shape, size, size and shape distribution, surface charges, soluble species, and so on. In some cases, interactions may exist between the components of the cement slurry formulation. These interactions may enhance the control effect on performance characteristics, resulting in a net decrease in performance characteristics when negatively reacting components are included together in the cement slurry.Alternatively, performance characteristics can be improved, resulting in greater net performance characteristics, by including positively reacting components in the cement mortar. Various factors can contribute to these component reactions. A model can be written to represent the performance characteristics of a specific additive for a particular cement formulation under defined test conditions as a function of the mix, water content, pozzolan content, pozzolan volume, specific type of additive, additive quantity, and well conditions. This function may be linear or nonlinear, depending on the type and quantity of the additive used to control the performance characteristics. The function may be polynomial, exponential, logarithmic, trigonometric, or any other transcendental or analytical expression, or it may even be described using neural networks or decision trees. Figure 1 is an example of System 5 for preparing cement slurry, including the delivery of the cement slurry to the drilling well. The cement slurry can be any cement slurry identified here. A cement slurry recipe can be developed, for example, using the cement performance characteristics models shown here, and the cement slurry can be prepared based on that recipe. As shown, the cement slurry can be mixed in mixing equipment 10, such as a jet mixer, recirculating mixer, or batch mixer, and then pumped through pumping equipment 15 to the wellbore. In some examples, mixing equipment 10 and pumping equipment 15 can be mounted on one or more cement trucks, as will become clear to those with average skill in the field. In some examples, a jet mixer can be used to continuously mix a dry mix containing cement slurry with water as it is pumped to the wellbore. An illustrative technique for placing cement mortar in an underground formation will now be described with reference to Figures 2 and 3. Figure 2 shows an example of surface equipment 20 that can be used for placing cement mortar. The cement mortar can be any cement mortar identified here. A cement mortar recipe can be developed, for example, using the cement performance characteristics models shown here, and the cement mortar can be prepared based on the cement mortar recipe. It should be noted that while Figure 2 generally illustrates a surface operation, those skilled in the field will recognize that the principles described here are equally applicable to offshore operations using floating or seabed-mounted drilling platforms and rigs, without going beyond the scope of the description. As illustrated by Figure 2, the surface equipment 20 can include cement 25 stabilization units and can include one or more cement trucks. A cement stabilization unit can include 25 mixing equipment, 10 pumping equipment, and 15 pumping equipment (for example, Figure 1), as will become clear to those with average skill in the field.The cementing unit 25 can pump cement slurry 30 through a feed pipe 35 and into a cementing head 36, which then transfers the cement slurry 30 down the well. As shown, one or more of the operations described in Figure 2 can be performed using (for example, with the assistance of) Information Processing System 114, which will be discussed in more detail. Now turning to Figure 3, cement mortar 30 can be placed in a subterranean formation 45. As illustrated, a well 50 can be drilled in one or more subterranean formations 45. While the extension of a well 50 is generally shown to be vertical in one or more subterranean formations 45, the principles described here are also applicable to wells extending at an angle through one or more subterranean formations 45, such as horizontal and inclined wells. As illustrated, wellbore 50 includes walls 55. In the example shown, surface casing 60 has been inserted into wellbore 50. Surface casing 60 can be cemented to the walls of wellbore 50 by cement casing 65. In the example shown, one or more additional gutters (e.g., intermediate casing, production casing, liners, etc.), as illustrated here in the image of casing 70, can also be placed in wellbore 50. As illustrated, there is an annular space 75 of a wellbore formed between the casing 70 and the walls of wellbore 50 and / or surface casing 60.One or more of the 80 centering devices can be attached to the 70 casing, for example, to center the 70 casing in the 50 wellbore before or during the cementing process. With continued reference to Figure 3, cement slurry 30 can be pumped down the inner surface of casing 70. Cement slurry 30 can be allowed to flow down the inner surface of casing 70 through the casing shoe 85 at the bottom of casing 70 and upwards around casing 70 and into the annulus 75 of the wellbore. Cement slurry 30 can be allowed to penetrate the annulus 75 of the wellbore, for example, to form a cemented sleeve that holds casing 70 in place and anchors it in wellbore 50. Although not illustrated, other techniques can also be used to introduce cement slurry 30. For example, reverse circulation techniques can be used to introduce cement slurry 30 into the aquifer 45 via the annulus 75 of the wellbore instead of through casing 70. Upon insertion, cement slurry 30 can displace other fluids 90, such as drilling fluids and / or separating fluids that may be present in the inner surface of the casing 70 and / or the annulus 75 of the wellbore. At least some of the displaced fluids 90 can escape from the annulus of the wellbore 75 through the flow line 95 and settle, for example, in one or more retention holes 100 (e.g., the mud hole), as illustrated in Figure 2. Referring back to Figure 3, a bottom plug 105 can be inserted into the wellbore 50 prior to the cement slurry 30, for example, to separate the cement slurry 30 from other fluids 90 that may be present within the casing 70 prior to the cementing process. After the bottom plug 105 reaches the landing collar 110, the diaphragm or other suitable device should be torn to allow cement grout 30 through the bottom plug 105. In Figure 3, the bottom plug 105 is shown on the landing collar 110. In the example shown, the top plug 115 can be inserted into wellbore 50 behind the cement grout 30.Upper plug 115 can separate cement mortar 30 from displacement fluid 120 and push cement mortar 30 through lower plug 105. Figure 4 is a 400 workflow diagram for preparing a cement mortar recipe. Method 400 may begin with Unit 402, where the cement composition is entered into the 400 workflow. The inputs may include one or more initial estimates for the cement mortar recipe, or these inputs may be used to generate that estimate. The inputs and / or initial estimates may be based on, or selected from, one or more bulk materials, cementitious complements, or other cement mortar components available in a particular geographic area. For example, some bulk materials may be available in certain areas but not in others. The inputs may also include, in some examples, the condition of one or more wells, as noted above. Well conditions may include, but are not limited to, temperature, pressure, wettability, the amount and / or composition of the filter cake adhering to at least part of the wellbore annulus, acidity, combinations thereof, or similar factors.As explained, one or more operations, for example, some or all of the operations of any of the 400 workflow modules, can be performed by the 114 Information Processing System, which will be discussed in more detail. The performance characteristics of cement are predicted in Unit 406. The predictions for Unit 406 may include, or be derived from, one or more of the cement performance characteristics models in Unit 404. The cement performance characteristics model(s) for Unit 404 may include any of the equation-type and / or black-box-type models described here. Thus, the predictions for the cement performance characteristics model(s) for Unit 404 may include, for example, Young's modulus of elasticity, Poisson's ratio, tensile strength, ultimate compressive strength, cohesion, angle of friction, and combinations thereof. In examples where multiple cement performance characteristics are predicted by a combination of cement performance characteristics models, each predicted performance characteristic may correspond to a single cement performance characteristics model. Alternatively, in some examples, a single cement performance characteristics model (such as a neural network) may be used to predict multiple cement performance characteristics. In addition to the performance characteristics of cured cement mortar prepared according to a preliminary estimate of the cement mortar recipe, Unit 404 can also be used to calculate the properties of cement mortar before or during the hardening process, in some examples.These properties may include fluid properties such as flowability, density, viscosity, etc., as well as thickening time, heat generation, mixability, water requirements, fluid loss, transfer time, and other properties of cement mortar, for example. After predicting performance characteristics in Unit 406, workflow 400 can move to Unit 408. In Unit 408, an evaluation can be performed to determine whether the predictions meet one or more predefined requirements. Based on this evaluation, workflow 400 can move to Unit 410, where the cement mortar recipe is modified. The operations of Units 404, 406, and / or 408 can be repeated until, for example, it is determined in Unit 408 that the cement prepared from the cement mortar recipe meets one or more cement property requirements. This might involve, in some examples, comparing one or more (e.g., one, two, three, five, etc.) predictions generated in Unit 406 with one or more predefined requirements. For example, one method might involve predicting tensile strength, Young's modulus, ultimate compressive strength, cohesion, and angle of friction, and comparing these predictions with the limiting values or predefined requirements.Because the properties predicted by Unit 406 are for a hypothetical cement mortar, which, if prepared and allowed to harden, would result in hardened cement, Workflow 400 may include hypothetical cement mortar recipes as well as hypothetical cement formulations. When cement formulations are prepared according to Workflow 400 predictions or recommendations, one or more measurable properties of the resulting cement prepared according to the hypothetical cement mortar may, in some examples, closely match the final Unit 406 prediction for the recommended cement mortar recipe. Therefore, existing methods and systems may have the advantage of ensuring that the cement mortar formulation produces a cement well barrier with satisfactory performance characteristics for the downhole environment. Once workflow 400 produces a cement mortar formulation with satisfactory expected performance characteristics for one or more predefined limit values, workflow 400 can progress to module 412, which may involve performing a stabilization operation according to the cement mortar formulation used in the final projections for module 406. This stabilization can be performed using one or more of the techniques illustrated in Figures 2 and 3, for example. In one example, a proposed initial cement mortar recipe might be selected, which could include the cement components and their mass ratios, and the water and its mass ratio. In this illustrative method, any of the cement performance characteristic models developed above can be used to calculate the expected performance characteristics of the first proposed cement mortar recipe. The calculated performance characteristics can then be compared to one or more predefined requirements. If the expected performance characteristics meet (e.g., satisfy, exceed) one or more of the predefined requirements, the method might involve preparing an actual mortar based on the first proposed cement mortar recipe and testing it using laboratory techniques to measure performance characteristics and verify that the cement mortar meets the performance characteristic control requirements. If the expected performance characteristics do not meet the performance requirements, the method might involve modifying the cement mortar recipe and rerunning the predictions for the modified cement mortar recipe. Multiple prediction reruns can be performed as needed until the method converges to a satisfactory solution, for example, one that adequately meets the predefined requirements.Modifying the cement mortar mix may involve varying cementitious components and / or varying mass frictions, and / or chemical additives and their components. One or more of the expected performance characteristics of the second proposed cement mortar mix can be compared to one or more predefined thresholds. If the expected performance characteristics meet the required performance standard, the mortar can be prepared based on the second proposed cement mortar mix and tested using laboratory techniques to confirm that the performance characteristics do indeed meet the required specifications. Alternatively, in some cases, the well drilling process can proceed directly without using laboratory techniques. Otherwise, the process can be repeated until a cement mortar mix is found that meets, exceeds, or falls short of the required performance characteristics. The cement mortars described here may generally contain hydraulic cement and water. Various types of hydraulic cement may be used, according to this disclosure, including but not limited to: hydraulic cements containing calcium and / or aluminum and / or silicon and / or oxygen and / or iron and / or sulfur, which set and harden upon reaction with water. Suitable types of hydraulic cement may include, but are not limited to, Portland cements, pozzolana cements, gypsum cements, high alumina content cements, silica cements, and any combination thereof. In certain examples, the hydraulic cement is Portland cement. In some examples, Portland cement types can include Portland cements classified into classes A, C, H, and G according to the American Petroleum Institute, API Specification for Materials and Testing for Well Cements, API Specification 10, Fifth Ed., July 1, 1990.In addition, hydraulic cements may include cements classified by the American Society for Testing and Materials (ASTM) as C150 (Standard Specification for Portland Cement), C595 (Standard Specification for Blended Hydraulic Cement), or C1157 (Performance Specification for Hydraulic Cement) as Type I, II, or III cements according to ASTM. Hydraulic cement can be incorporated into cement mortar in any quantity suitable for a given formulation. For example, hydraulic cement can be incorporated into cement mortars in quantities ranging from 10% to approximately 80% by weight of the dry mix. For example, hydraulic cement can be present in an amount ranging from and / or including any of approximately 10%, approximately 15%, approximately 20%, approximately 25%, approximately 30%, approximately 35%, approximately 40%, approximately 45%, approximately 50%, approximately 55%, approximately 60%, approximately 65%, approximately 70%, approximately 75%, or approximately 80% by weight of the cement mortars. Water can be from any source, provided it does not contain excessive amounts of compounds that could adversely affect the other components in cement mortars. For example, cement mortar can contain either fresh or salt water. Salt water generally contains one or more dissolved salts and can be saturated or unsaturated depending on the specific application. Seawater or brine solutions may be suitable in some cases. Furthermore, sufficient water can be present to create a pumpable mortar. In certain applications, the water content in cement mortar can range from approximately 33% to approximately 200% by weight of the cementitious materials. For example, water may be present in quantities ranging from approximately 33%, 50%, 75%, 100%, 125%, 150%, 175%, or 200% by weight of the cementitious materials. The cementitious materials referred to may include all components that contribute to the compressive strength of the cement mortar, such as hydraulic cement and supplementary cementitious materials, for example. As mentioned above, cement mortar may contain supplementary cementitious materials. These supplementary materials may be any substance that contributes to the desired properties of the cement mortar. Some examples of supplementary cementitious materials include, but are not limited to, fly ash, blast furnace slag, silica fume, pozzolanic materials, furnace dust, and clays. Cementitious mortar may include kiln dust as a supplementary cementitious material. "Kiln dust," as used here, refers to a solid material produced as a byproduct of heating certain materials in kilns. The term "kiln dust," as used here, includes kiln dust prepared as described and its equivalent forms. Depending on its source, kiln dust can exhibit cementitious properties, as it can curdle and harden in the presence of water. Suitable examples of kiln dust include cement kiln dust, lime kiln dust, and combinations thereof. Cement kiln dust can be produced as a byproduct of cement production, removed from the gas stream, and collected, for example, in a dust collector. Large quantities of cement kiln dust are typically collected during cement production and are generally disposed of as waste. The chemical analysis of cement kiln dust obtained from many cement manufacturers varies based on a range of factors, including the specific kiln feed, cement production process efficiency rates, and associated dust collection systems.Cement kiln dust generally contains a variety of oxides, such as SiO2, Al2O3, Fe2O3, CaO, MgO, SO3, Na2O, and K2O. The chemical analysis of lime kiln dust varies among lime manufacturers based on several factors, including the specific limestone or dolomitic limestone feedstock, kiln type, kiln operating mode, lime production process efficiency rates, and associated dust collection systems. Lime kiln dust can generally contain varying amounts of free lime and free magnesium, limestone, and / or dolomitic limestone, and a variety of oxides, such as SiO2, Al2O3, Fe2O3, CaO, MgO, SO3, Na2O, and K2O, as well as other components, such as chloride compounds. Cement kiln dust can be added to cement mortar before, during, or after activation. Cement kiln dust may include the calcined kiln feed stream, which is partially removed from the gas stream and collected in a dust collector during cement manufacturing. The chemical analysis of CKD varies between different cement plants depending on several factors, including the specific kiln feed, the efficiency of the cement production process, and the associated dust collection systems.CKD may generally include a variety of oxides, such as SiO2, Al2O3, Fe2O3, CaO, MgO, SO3, Na2O, and K2O. CKD and / or lime kiln dust may be included in cement mortar examples in quantities appropriate for a specific application. In some formulations, cement mortar contains one or more of the following as a supplementary cementitious material: slag, natural glass, shale, amorphous silica, or metakaolin. Slag is generally a granular byproduct of blast furnace production in the ironmaking process, containing oxidized impurities found in the iron ore. Cement may also include shale. A variety of shale types may be suitable, including those containing silicon, aluminum, calcium, and / or magnesium. Examples of suitable shale types include vitrified shale and / or calcined shale. In some cases, cement mortar may also include silica as a supplementary cementitious material. Amorphous silica is a powder that may be incorporated into formulations to increase the compressive strength of the cement. Amorphous silica is generally a by-product of the ferrosilicon production process, where amorphous silica can be formed by oxidation and condensation of gaseous silicon oxide, SiO, which is formed as an intermediate compound during the process. In some examples, cementitious mortar may also include a variety of fly ash as cementitious admixtures, which may include fly ash classified as Class C, Class F, or Class N according to the American Petroleum Institute, API Specification for Materials and Testing for Well Cements, API Specification 10, Fifth Ed., July 1, 1990. In some examples, cementitious mortar may also include zeolite compounds as cementitious admixtures. Zeolites are generally porous aluminosilicate materials that are either natural or synthetic. Synthetic zeolites are based on the same type of structural cell as natural zeolites and may include hydrated aluminosilicate compounds. As used here, the term "zeolite" refers to all natural and synthetic forms of zeolite. When used, one or more of the aforementioned cementitious additives may be present in the cement mortar. For example, but not limited to, one or more cementitious additives may be present in amounts ranging from approximately 0.1% to approximately 80% by weight of the cement mortar. For instance, cementitious additives may be present in amounts ranging from approximately 0.1%, approximately 10%, approximately 20%, approximately 30%, approximately 40%, approximately 50%, approximately 60%, approximately 70%, or approximately 80% by weight of the cement. In some examples, cement mortar may also include hydrated lime. As used here, it will be understood that the term "hydrated lime" refers to calcium hydroxide. In some formulations, hydrated lime may be supplied as quicklime (calcium oxide) that is hydrated when mixed with water to form hydrated lime. Hydrated lime may be included in cement mortars, for example, to create a hydraulic mix with complementary cementitious components. For example, hydrated lime may be included in a cementitious-to-hydrate ratio ranging from approximately 10:1 to approximately 1:1 or from 3:1 to approximately 5:1. Where present, hydrated lime may be included in the setting cement mortar in a quantity ranging from approximately 10% to approximately 100% by weight of the cement mortar, for example. In some examples, hydrated lime can be present in amounts ranging from and / or including any of approximately 10%, approximately 20%, approximately 40%, approximately 60%, approximately 80%, or approximately 100% by weight of the cement mortar.In some cases, the cementitious components in cement mortar may consist primarily of one or more complementary cementitious materials and hydrated lime. For example, the cementitious components may consist mainly of complementary cementitious materials and hydrated lime without any additional components (e.g., Portland cement, fly ash, slag cement) that are suspected of being hydraulically impregnable due to the presence of water. Lime can be present in cement mortar in various forms, including calcium oxide and / or calcium hydroxide, or as a reaction product such as the reaction of Portland cement with water. Alternatively, lime can be included in cement mortar through the amount of silica present. Cement mortar can be designed to have a target lime-to-silica ratio by weight. This target lime-to-silica ratio can be expressed as a molar ratio, molar ratio, or any other equivalent method of expressing the relative amount of silica to lime. Any suitable target lime-to-silica ratio can be chosen, ranging from approximately 10 / 90 lime-to-silica by weight to approximately 40 / 60 lime-to-silica by weight. Alternatively, ranging from about 10 / 90 lime to silica by weight to about 20 / 80 lime to silica by weight, about 20 / 80 lime to silica by weight to about 30 / 70 lime to silica by weight, or about 30 / 70 lime to silica by weight to about 40 / 63 lime to silica by weight. Other admixtures suitable for use in underground cementing processes may also be included in cement mortar formulations. These admixtures include, but are not limited to, weighting agents, lightweight admixtures, gas-producing admixtures, mechanical property-enhancing admixtures, loss circulation admixtures, filtration control admixtures, fluid loss control admixtures, defoaming agents, foaming agents, gelling admixtures, and combinations thereof. In formulations, one or more of these admixtures may be added to the cement mortar after storage and before the cement mortar is placed in the underground formation. In some cases, the cement mortar may include a dispersant. Examples of suitable dispersants include, but are not limited to, sulfonated formaldehyde-based dispersants (e.g., sulfonated acetone formaldehyde condensate) or polycarboxylate-treated ether dispersants. In some examples, the dispersing material may be included in cement mortar in an amount ranging from about 0.01% to about 5% by weight of the cementitious materials.In specific examples, the dispersed material can be present in amounts ranging from and / or include any of approximately 0.01%, approximately 0.1%, approximately 0.5%, approximately 1%, approximately 2%, approximately 3%, approximately 4%, or approximately 5% by weight of the cementitious materials. In some examples, cement mortar may also include a setting retarder. A wide variety of setting retarders can be suitable for use in cement mortars. For example, a setting retarder may include phosphonic acids, such as ethylenediaminetetraacetic acid (methylene phosphonic acid), diethylenediaminepentaacetic acid (methylene phosphonic acid), and so on; lignosulfonate compounds, such as sodium lignosulfonate, calcium lignosulfonate, and so on; salts such as tin sulfate, lead lactate, monobasic calcium phosphate; organic acids, such as citric acid, tartaric acid, and so on; cellulose derivatives such as hydroxyl ethyl cellulose (HEC) and carboxymethyl hydroxyethyl cellulose (CMHEC); Synthetic copolymers or tertiary polymers that include sulfonate and carboxylic acid groups, such as sulfonate-activated acrylamide-acrylic acid copolymers; borate compounds such as alkali borates, sodium metaborate, sodium tetraborate, potassium pentaborate; their derivatives, or mixtures thereof.Examples of suitable setting retarders include, among others, phosphonic acid derivatives. Generally, a setting retarder can be present in cement mortar in a quantity sufficient to delay setting for the required period. In some cases, the setting retarder may be present in cement mortar in a quantity ranging from approximately 0.01% to approximately 10% by weight of the cementitious material. In specific cases, the setting retarder may be present in a quantity ranging from and / or including approximately 0.01%, approximately 0.1%, approximately 1%, approximately 2%, approximately 4%, approximately 6%, approximately 8%, or approximately 10% by weight of the cementitious material. In some cases, cement mortar may also contain an accelerator. A wide variety of accelerators can be suitable for use in cement mortars. For example, accelerators may include, but are not limited to, aluminum sulfate, alum compounds, calcium chloride, calcium nitrate, calcium nitrite, calcium formate, calcium sulfoaluminate, calcium sulfate, gypsum hemihydrate, sodium aluminate, sodium carbonate, sodium chloride, sodium silicate, sodium sulfate, ferric chloride, or a combination thereof. In some cases, accelerators may be present in cement mortar in amounts ranging from approximately 0.01% to approximately 10% by weight of the cementitious materials. In specific examples, accelerators can be present in amounts ranging from and / or include any of approximately 0.01%, approximately 0.1%, approximately 1%, approximately 2%, approximately 4%, approximately 6%, approximately 8%, or approximately 10% by weight of the cementitious materials. In general, cement mortars should have a density appropriate for a specific application. For example, the density of cement mortar may range from about 8 pounds per gallon (ppg) (959 kg / m³) to about 20 pg (2397 kg / m³), or from about 8 pg to about 12 pg (1437 kg / m³), or from about 12 pg to about 16 pg (1917.22 kg / m³), or from about 16 pg to about 20 pg, or any range in between. Examples of cement mortars can be foamed or non-foamed, or they may incorporate other means of reducing their density values, such as hollow microspheres, low-density flexible beads, or other density-reducing additives known in the field. The cementitious mortars described here can be used in a range of subsurface applications, including primary and remedial cementing. The cementitious mortars can be injected into a subsurface formation and allowed to set. In primary cementitious applications, for example, the cementitious mortars can be injected into the annular space between a stream in a wellbore and the wellbore walls (and / or a longer stream in the wellbore), where the wellbore penetrates the subsurface formation. The cementitious mortar can then be allowed to set within the annular space, forming a hardened annular cement seal. The cementitious mortar can then act as a barrier, preventing fluid movement within the wellbore. It can also support, for example, the stream within the wellbore. In remedial cementitious applications, the cementitious mortars can be used, for example, in cement injection or cement plug placement.For example, cement mortar can be placed in a wellbore to seal an opening (e.g., a void or crack) in the formation, in a gravel fill, in the channel, in the cement casing, and / or between the cement casing and the channel (e.g., the fine annular space). In one or more examples, one or more workflow modules may include, or be preceded by, the measurement of the performance characteristics of one or more cement samples. This may be done to provide sufficient data for the training dataset used to train one or more cement performance characteristic models. However, in some examples, the training data may be retrieved from a database containing historical data from previous laboratory tests or installations. The measurement of the performance characteristics of each selected cement component may involve several laboratory techniques and procedures, including, but not limited to, various destructive and / or non-destructive tests. These tests may include API tests, as described in the API Recommended Practice for Well Cement Testing (published as ANSI / API Recommended Practice 10B-2). Other, but not limited to, examples of the various laboratory techniques used in some examples to determine performance characteristics may include: microscopy, spectroscopy, X-ray diffraction, X-ray fluorescence, particle size analysis, and requirements analysis.Water, scanning electron microscopy, energy-dispersive X-ray spectroscopy, surface area, specific gravity analysis, thermogravimetric analysis, shape analysis, infrared spectroscopy, ultraviolet-visible spectroscopy, mass spectroscopy, secondary ion mass spectroscopy, electron energy mass spectroscopy, X-ray dispersive spectroscopy, Auger electron spectroscopy, inductively coupled plasma analysis, thermal ionization mass spectroscopy, glow discharge mass spectroscopy, x-ray photoelectron spectroscopy, rheological properties, and combinations thereof. The cement components mentioned here can be analyzed in some examples to determine their water requirements. Water requirement is generally defined as the amount of water needed to mix a powdered solid material to form a suspension of a specific consistency. The amount of water required for a particular cement component can be determined through a process that includes: a) preparing a Waring mixer with a specific amount of water, b) stirring the water at a specific mixer speed, c) adding the powdered solid material being tested to the water until a specific consistency is achieved, and d) calculating the required amount of water based on the water-to-solids ratio needed to reach the desired consistency. In addition to measuring the performance characteristics of cement, laboratory tests can also be conducted to determine the behavior of cementitious components in cement mortar. For example, the cementitious components in cement mortar can be analyzed to determine the development of their compressive strength and mechanical properties. For instance, a predetermined quantity of cementitious components can be combined with water and lime (if necessary for certainty). The performance characteristics of the cement mortar, including compressive strength, tensile strength, and Young's modulus, can then be determined. Any of a variety of different testing conditions can be used, provided the conditions are consistent for the different cementitious components. Figure 16 is an example of an information handling system (ICS) that can be used to perform the various steps, methods, and techniques described here. Those with a basic understanding of the field will easily recognize that other examples of the system are possible. As shown, ICS includes a CPU (or processor) (CPU) (1602) and a system bus (1604) that pairs various system components, including system memory (SEM) (1606), such as read-only memory (ROM) (1608) and random-access memory (RAM) (1610), with CPU (1602). All the processors shown here can be images of the aforementioned CPU (1602). ICS can also include a cache (cache) (1612) of high-speed memory directly connected to, located near, or integrated with CPU (1602). The information processing system 114 can copy data from memory 1606 and / or storage device 1614 into buffer memory 1612, allowing processor 1602 to access it quickly. In this way, buffer memory 1612 enhances performance, preventing delays for processor 1602 while waiting for data.These and other modules can control, or be configured to control, the 1602 processor to perform various operations or procedures. Other system memory (1606) may also be available for use. 1606 memory can include several different types of memory with varying performance characteristics. This detection can be implemented on an information processing system (114) with more than one 1602 processor, or on a group or cluster of computers connected together via a network to provide greater rigidity. The 1602 processor includes any general-purpose processor and a hardware or software module, such as a first module (1616), a second module (1618), and a third module (1620) stored in storage device (1614), configured to control the 1602 processor, as well as a special-purpose processor where the code is embedded in the 1602 processor. A 1602 processor can be a standalone computer system, containing multiple cores or processors, a bus, a memory controller, a buffer, and so on. A multi-core processor can be symmetric or asymmetric. A 1602 processor can include many processors, such as a system with several physically separate processors in different sockets, or a system with many processor cores on a single physical chip. Similarly, a 1602 processor can include many distributed processors located in several separate computers, but working together, for example, over a network. Multiple processors or processor cores can share resources, such as 1606 memory or 1612 buffer, or they can operate using independent resources. A 1602 processor can include one or more state machines, an application-specific integrated circuit (ASIC), or a programmable gate array (PGA) that includes a field PGA (FPGA). Each of the independent components discussed above can be paired with the System Bus 1604, which can connect all the independent components to each other. The System Bus 1604 can be any of the various bus architectures that include a memory bus or memory controller, a peripheral bus, and a local bus using any combination of bus architectures. The Basic Input / Output (BIOS) stored in ROM 1608 or similar can provide the basic routine that helps transfer information between the elements within the Information Processing System 114, for example, during startup. The Information Processing System 114 also includes 1614 storage devices or computer-readable storage media, such as a hard disk drive, magnetic disk drive, optical disk drive, tape drive, solid-state drive, RAM drive, removable storage devices, a redundant array of inexpensive disks (RAID), a hybrid storage device, or similar.Storage device 1614 can include software modules 1616, 1618, and 1620 to control processor 1602. Information processing system 114 can include other hardware or software modules. Storage device 1614 communicates with system bus 1604 via an interface. Hard drives and computer-readable storage devices provide non-volatile storage for computer-readable instructions, data structures, software modules, and other data for information processing system 114. In one sense, a computer module that performs a specific function includes the software component stored in a physical computer-readable storage device along with the necessary hardware components, such as processor 1602, system bus 1604, and so on, to execute that specific function. In another sense, the system can use a processor and a computer-readable storage device to store instructions which, when executed by the processor, cause the processor to perform specific operations, methods, or procedures.The basic components and appropriate variations can be modified according to the type of device, for example, whether the 114 Information Processing System is a handheld microcomputer, a desktop computer, or a computer server. When the 1602 processor executes instructions to perform "operations," it can perform the operations directly and / or facilitate, direct, or collaborate with another device or component to perform the operations. As illustrated, the Information Processing System 114 utilizes a storage device 1614, which can be a hard disk or other types of computer-readable storage devices that can store data accessible by a computer, such as magnetic cassettes, flash memory cards, digital versatile disks (DVDs), cartridges, random access memory (RAM) units 1610, read-only memory (ROM) units 1608, cable containing bit current, and the like, in the demonstration operating environment. Computer-readable storage media, computer-readable storage devices, or computer-readable memory devices explicitly exclude media such as transitory waves, energy, carrier wave signals, electromagnetic waves, and signals themselves. To facilitate user interaction with the Information Processing System (IPS) 114, an input device (IP) 1622 represents any combination of input mechanisms, such as a microphone for speech input, a touchscreen for motion or graphics input, a keyboard, a mouse, motion input, speech input, and so on. An output device (IP) 1624 can also be one or more of a range of output mechanisms familiar to those skilled in the field. In some cases, multi-media systems enable the user to provide multiple types of input for communication with IPS 114. The Communications interface (I) 1626 generally controls and manages user input and system output. There is no restriction on operation with any specific computer component arrangement; therefore, the basic hardware component pictured can easily be used instead of enhanced hardware or firmware configurations during development. As illustrated, each individual component described above is depicted and revealed as individual functional units. The functions represented by these units can be provided by either a shared or dedicated device, including, but not limited to, a device capable of executing software, and a device, such as the 1602 processor, specifically designed to function as the software execution equivalent of a general-purpose processor. For example, the functions of one or more of the processors shown in Figure 16 can be provided by a single shared processor or multiple processors. (The use of the term "processor" should not be interpreted as referring exclusively to a computer component capable of executing a program.) Illustrative representations may include a microprocessor and / or a digital signal processor (DSP), 308 read-only memory (ROM) to store a program performing the operations described below, and 310 random access memory (RAM) to store the results. Hard-scale component representations of very large-scale integration (VLSI) circuits, as well as a specially designed VLSI circuit in combination with a general-purpose DSP circuit, may also be provided. The logical operations of the various methods described below are implemented as follows: (1) a series of steps, operations, or procedures performed by a computer operating on a programmable circuit within a general-purpose computer; (2) a series of steps, operations, or procedures performed by a computer operating on a specific-use programmable circuit; and / or (3) modules of a connected machine or program engines within programmable circuits. The Information Processing System 114 may exercise all or some of the aforementioned methods, may be part of the aforementioned systems, and / or may operate according to instructions in the aforementioned computer-readable physical storage devices. These logical operations may be implemented as modules configured to control the Processor 1602 to perform specific functions according to the programming of the Program Modules 1616, 1618, and 1620. In these examples, one or more parts of an analog information processing system (A114) can be virtualized, up to and including the entire A114 system. For example, a virtual processor can be a software component that executes according to a specific instruction set, even when no physical processor of the same type as the virtual processor is available. A virtualization layer, or virtual "host," can help virtual components of one or more different computing devices or hardware types by translating virtual operations into physical ones. However, ultimately, the virtual hardware components of each type are implemented or performed by specific underlying physical hardware components. Thus, a virtual computer layer can operate on top of a physical computer layer. A virtual computer layer can include one or more virtual machines, an overlay network, a virtual machine monitor, a virtual switch, and any other virtual application. Figure 17 shows a diagram of another Information Processing System 114, which contains a chipset architecture that can be used to implement a method for creating and displaying a graphical user interface (GUI). Information Processing System 114 is an example of the computer hardware, software, and firmware that can be used to implement the technology revealed. Information Processing System 114 can include a processor 1602, representing any combination of different physical and / or logical resources capable of executing programs, firmware, and computer components configured to perform specific calculations. The processor 1602 can communicate with a chipset 1700, which can control the inputs and outputs of the processor 1602. In this example, the chipset 1700 outputs information to an output device 1624, such as a display, and can read and write information to a storage device 1614, which can include, for example, magnetic media and solid-state media. The 1700 chipset can also read data from and write data to RAM 1610.The 1702 bridge can be provided to connect to a variety of 1704 user interface components for connection to the 1700 chipset. These 1704 user interface components can include a keyboard, microphone, touch detection and hardening circuit, signaling device such as a mouse, and so on. In general, input to the 114 information processing system can come from any variety of sources, both machine-generated and / or human-generated. The 1700 chipset can also connect to one or more of the 1626 interfaces, which can house various physical interfaces. These interfaces can include wired and wireless local area networks (LANs), wireless broadband networks (WANs), and personal network interfaces (PNAs). Some of the GUI production, display, and usage methods described here involve receiving sorted datasets via the physical interface or generating them directly from the machine itself using the 1602 processor, which parses data stored in storage device 1614 or RAM 1610. Furthermore, the 114 information processing system receives user input via the 1704 user interface components and executes appropriate functions, such as navigation, by interpreting this input using the 1602 processor. In examples, an information processing system (IPS) can also include computer-readable physical and / or non-temporary storage devices to hold or retrieve computer-executable instructions or data structures. Computer-readable physical storage devices can be any available device accessible by a general-purpose or purpose-built computer, including the functional design of any purpose-built processor as described above. For example, but not limited to, computer-readable physical storage devices can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage devices, magnetic disk storage, other magnetic storage devices, or any other device that can be used to hold or store preferred program code as computer-executable instructions, data structures, or the design of a processor chip. When information or instructions are provided to a computer via a network or other communication link (either wired, wireless, or a combination thereof), the computer appropriately presents the link as a computer-readable medium.Thus, any such connection is appropriately called a computer-readable medium. Combinations of the above should also be included within the scope of computer-readable storage devices. In examples, an information processing system (114) can also include physical and / or computer-readable non-temporary storage devices to hold or retrieve computer-executable instructions or data structures stored on them. Computer-readable non-temporary media (148) can store programs or method instructions described here. Computer-readable non-temporary media (148) can include any tools or toolkits that may retain data and / or instructions for a period of time.Non-temporary computer-readable media 148 may include, for example, storage media such as a direct-access storage device (e.g., a hard disk drive or floppy disk drive), a sequential-access storage device (e.g., a tape drive), a compact disc, CD-ROM, DVD, RAM, ROM, electrically erasable programmable read-only memory (EEPROM), and / or flash memory; communication media such as wires, fiber optics, microwaves, radio waves, and other electromagnetic and / or optical carriers; and / or any combination thereof. Computer-readable physical storage devices can be any available device accessible by a general-purpose or purpose-built computer, including the functional design of any purpose-built processor as described above. For example, but not limited to, computer-readable physical storage devices may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage devices, magnetic disk storage or other magnetic storage devices, or any other device that can be used to carry or store preferred program code in the form of computer-executable instructions, data structures, or processor chip designs. When information or instructions are provided to a computer via a network or other communication link (either wired, wireless, or a combination thereof), the computer appropriately presents the link as a computer-readable medium. Thus, any such link is appropriately called a computer-readable medium. Combinations of the above should also be included within the scope of computer-readable storage devices. Computer-executable instructions include, for example, instructions and data that cause a general-purpose computer, a special-purpose computer, or a special-purpose hardener to perform a specific function or set of functions. Computer-executable instructions also include program modules executed by computers in standalone or networked environments. In general, program modules can include routines, programs, components, data structures, objectives, and functions inherent in the design of special-purpose processors, and so on, that perform specific tasks or execute specific abstract data types. Computer-executable instructions, their associated data structures, and program modules represent examples of how program code implements the method steps revealed here. The specific sequence of those executable instructions or their associated data structures represents examples of the procedures corresponding to the execution of the functions described in these steps. In further examples, the methods can be practiced in networked computing environments with many types of computer system architectures, including personal computers, handheld devices, multiprocessor systems, programmable or microprocessor-based consumer electronics, network computers, microcomputers, mainframe computers, and the like. Examples can also be implemented in distributed computing environments where tasks are performed by local and remotely connected hard drives that are linked (either by hardwired links, wireless links, or a combination thereof) via a communications network. In a distributed computing environment, software modules can be placed on both local and remotely connected hard drives. Figure 18 shows an example of a resource setup in an 1800 computing network that might use the operations and techniques described here, although many other techniques are, of course, possible. As mentioned above, an Information Processing System (IPS) 114 can, as part of its functionality, use data that includes files, directories, metadata (e.g., creation / editing dates of the access control list (ACLS) associated with the data, etc.), and other data objects. Typically, the data in an IPS 114 is a raw copy (e.g., a production copy). During copying, backup, archiving, or other storage operations, an IPS 114 can send a copy of some data objects (or some of their components) to a secondary storage computer 1804 by using one or more data agents 1802. The 1802 data agent can be a desktop application, a website application, or any software-based application running on the 114 Information Processing System. As illustrated, the 114 Information Processing System can be located at any drilling rig site (e.g., refer to Figure 1) or repair and manufacturing center. The 1802 data agent can communicate with a secondary storage computer (1804) using the 1808 communication protocol located in a wired or wireless system. The 1808 communication protocol can operate or perform its function as input to a website application. In the website application, relevant field data for previous and subsequent operations, generated DTCs, notes, and the like can be loaded. In addition, the 114 Information Processing System can use the 1808 communication protocol to access stiffness measurements, operations with similar DTCs, troubleshooting results, historical operating data, and / or the like. This information is accessed from the secondary storage computing device 1804 via the data agent 1802, which is loaded onto the information processing system 114. The 1804 secondary storage computing device may create secondary copies of underlying data objects (or some of their components) in various 1806A-N cloud storage locations. Additionally, the 1804 secondary storage computing device may run deterministic algorithms on data loaded from one or more 138 information processing systems, which will be explained in detail later. Communication between 1804 secondary storage computing devices and 1806A-N cloud storage locations may use REST (Descriptive State Transfer Interfaces) protocols that satisfy the underlying C / R / U / D semantics (create / read / update / delete semantics), or other protocols based on the Hypertext Transfer Protocol ("HTTP") or File Transfer Protocol ("FTP") (such as Simple Object Access Protocol). In addition to creating secondary copies in the 1806A-N cloud storage sites, the 1804 secondary storage computing device may also index local content and / or deduplication at the object, subobject, or module level when performing storage operations involving different 1806A-N cloud storage sites. The 1806A-N cloud storage sites may also log DTC code records for each well operation or run, assign DTC codes, store repair and maintenance data, store operating data, and / or provide output from critical algorithms run in the 1806A-N cloud storage sites. As previously mentioned, the 114 information processing system may be telemetry-enabled to receive well measurements on surface 108. The following data may describe certain incarnations of the revelation, but should not be understood as being limited to any specific incarnation. Item 1: A method for designing cement mortar that includes the following: (a) Providing the cement mortar design requirements, whereby the cement design requirements include at least one performance property from among the following: compressive strength, tensile strength, cohesion, angle of friction, Young's modulus, Poisson ratio, or any combination thereof; (b) Providing a hypothetical cement mortar recipe representing water and its concentration, and one or more cementitious materials and their concentrations; (c) Inputting the well condition and the hypothetical cement mortar into a cement performance characteristics model; (d) Predicting the performance characteristics of the hypothetical cement mortar recipe using a cement performance characteristics model, whereby these characteristics include at least one property from among the following: compressive strength, tensile strength, cohesion, angle of friction, Young's modulus, and Poisson ratio; (e) Comparing the predicted performance characteristics with the cement design requirements.(f) Prepare cement mortar according to the default cement mortar recipe if the expected performance characteristics of the default cement mortar recipe meet the cement design requirements or repeat steps (b) - (f) if the default cement mortar recipe does not meet the cement design requirements, where the step to provide the default cement mortar recipe involves providing a default cement mortar recipe with a different concentration of water, a different concentration of one or more cementitious materials, and / or a different chemical type of one or more cementitious materials. Item 2: Method according to Item 1, where performance characteristics also include fluid loss and consistency thickening time. Item 3: The method according to items 1 or 2, where the hypothetical cement mortar recipe also represents at least one cement additive selected from the group consisting of an accelerator, retarder, gum, polyvinyl alcohol, crystalline silica, and any combination thereof. Item 4: The method according to any of items 1-3, where the cement performance characteristics model includes a neural network, decision tree or random forest model. Item 5: The method according to any of items 1-4, where the method also includes training one or more machine learning algorithms using the training data to form a model of cement performance characteristics, where the training data includes a set of cement composition data and cement properties that include ultimate compressive strength, tensile strength, angle of friction, cohesion, Young's modulus, and Poisson ratio for each of the multiple cement composition data. Item 6: The method in accordance with any of items 1-5, where one or more cementitious materials include at least one pozzolanic material selected from the group consisting of fly ash, pumice, silicate, cement kiln dust, and any combination thereof. Item 7: The method according to any of items 1-6, whereby the cement mortar consists of at least one density modifier chosen from the group consisting of a light weight bead, a heavy weight metal oxide, an elastomer, fibers, and any combination thereof. Item 8: The method according to any of items 1-7, where the well condition includes solidification temperature, solidification pressure, solidification time, and ambient pressure. Item 9: The method according to any of items 1-8, where the expected performance characteristics include the angle of friction, and where the angle of friction is predicted using measurements of ultimate compressive strength and cohesion. Item 10: The method according to any of items 1-9, where the expected performance characteristics include ultimate compressive strength and cohesion, where the method also includes determining the angle of friction from the expected ultimate compressive strength and cohesion. Item 11: The method according to Item 10, where determining the angle of friction involves using an equation as follows: Where UCS is the expected maximum compressive strength, Cohesion is the expected cohesion, and φ is the angle of friction. Item 12: The method is in accordance with Item 10, whereby the cement mortar is prepared in accordance with the new cement mortar recipe if the expected angle of friction is greater than or equal to the requirements for the angle of friction. Item 13: The method according to any of items 1-12, where the cement performance characteristics model includes a linear model as follows: Where it is a performance property of cement, w is the mass of water, Mass Poz is the mass of pozzolan, Vol Poz is the volume of pozzolan, Voli is the volume of pozzolan type i, Mass Addj is the mass of the added material j, Mass blendk is the mass of the blend type k, Well Condl is the condition of the well l, and a0, ai, aj, ak, and al are constants. Item 14: The method according to any of items 1-13, where the cement performance characteristics model includes a non-linear model as follows: Where it is a performance property of cement, w is the mass of water, Mass Poz is the mass of pozzolan, Vol Poz is the volume of pozzolan, Voli is the volume of pozzolan type i, Mass Addj is the mass of the additive j, Mass blendk is the mass of the blend type k, Well Condl is the well condition l, and a0, ai, aj, ak, and al are constants. Item 15: The method according to any of items 1-14, which also includes the introduction of cement mortar into a subterranean composition. Item 16: A method for designing cement mortar that includes the following: (a) Providing the cement mortar design requirements, whereby the cement design requirements include at least one of the following performance properties: compressive strength, tensile strength, cohesion, angle of friction, Young's modulus, Poisson ratio, or any combination thereof; (b) Providing a hypothetical cement mortar recipe that represents at least the water and its concentration, and one or more cementitious materials and their concentrations; (c) Introducing the setting temperature, setting pressure, setting time, and ambient pressure into the cement performance properties model, as well as the hypothetical cement mortar into the cement performance properties model; (d) Predicting the performance properties of the hypothetical cement mortar using the cement performance properties model, whereby these properties include at least one of the following: compressive strength, tensile strength, cohesion, angle of friction, Young's modulus, and Poisson ratio; (e) Comparing the predicted performance properties with the cement design requirements; (f) Modifying the default cement mortar formula if the expected performance characteristics do not conform to the cement design requirements, where the modification involves formulating a new default cement mortar.With a different concentration of at least one of the cementitious materials or the preparation of cement mortar according to the default cement mortar recipe if the expected performance characteristics meet or exceed the cement design requirements. Item 17: The method in accordance with Item 16, where the cement performance characteristics model includes a neural network, tree-based model, or random forest model. Item 18: The method according to Item 17, where the cement performance characteristics model includes a linear model as follows: Where it is a performance property of cement, w is the mass of water, Mass Poz is the mass of pozzolan, Vol Poz is the volume of pozzolan, Voli is the volume of pozzolan type i, Mass Addj is the mass of the additive j, Mass blendk is the mass of the blend type k, Well Condl is the well condition l, and a0, ai, aj, ak, and al are constants. Item 19: The method according to any of items 16-18, where the cement performance characteristics model includes a non-linear model as follows: Where it is a performance property of cement, w is the mass of water, Mass Poz is the mass of pozzolan, Vol Poz is the volume of pozzolan, Voli is the volume of pozzolan type i, Mass Addj is the mass of the additive j, Mass blendk is the mass of the blend type k, Well Condl is the well condition l, and a0, ai, aj, ak, and al are constants. Item 20: The method in accordance with Item 13 also includes the introduction of cement mortar into a subterranean composition. To facilitate a better understanding of the present invention, the following examples are given of some aspects of certain embodiments. These examples should in no way be interpreted as limiting or restricting the scope of the disclosure as a whole. Example In this example, a random forest model was trained using historical cement data that included the cement mortar composition, mortar physical properties, hardened cement physical properties, and wellbore placement conditions. The mortar compositions were of varying designs with different cement types, cementitious additives, and densities. Figures 5–15 show neutral graphs of the properties measured from the historical cement data versus the properties predicted by the random forest model. Figure 5 is a neutral graph of the training data, which includes the predicted versus measured ultimate compressive strength for a range of cement barrier designs. This training was used to develop a cement performance characteristics model using a range of cement barrier designs, based on one or more of these findings. In the example shown, the cement performance characteristics model was constructed using random forests to correlate the ultimate compressive strength with the composition, curing, and test conditions. Figure 6 is a neutral graph of the test data used to test the predictions of the cement performance characteristics model trained in Figure 5. This data also includes the predicted versus measured ultimate compressive strength. Each data point in Figure 6 represents a prediction made by the cement performance characteristics model trained in Figure 5 for a unique cement barrier design. Figure 7 is a neutral graph of the training data, including the expected versus measured cohesion, for training a cement performance characteristics model using a range of cement barrier designs, according to one or more embodiments of this disclosure. In the example shown, the cement performance characteristics model was constructed using random forests to correlate cohesion with composition, hardening, and test conditions. Figure 8 is a neutral graph of the test data showing the predictions of the cement performance characteristics model trained in Figure 7, which also includes the expected versus measured cohesion for a range of cement barrier designs, according to one or more embodiments of this disclosure. Each data point in Figure 8 represents a prediction by the cement performance characteristics model trained in Figure 7 for a unique cement barrier design. Figure 9 is a neutral graph of the training data, which includes the predicted tensile strength versus the measured tensile strength, for training a cement performance characteristics model using a range of cement barrier designs, according to one or more embodiments of this disclosure. In the example shown, the cement performance characteristics model was constructed using random forests to correlate tensile strength with composition, curing, and test conditions. Figure 10 is a neutral graph of the test data to show the predictions of the cement performance characteristics model trained in Figure 9, which also includes the predicted tensile strength versus the measured tensile strength for a range of cement barrier designs, according to one or more embodiments of this disclosure. Each data point in Figure 10 represents a prediction by the cement performance characteristics model trained in Figure 9 for a unique cement barrier design. Figure 11 is a graph showing the predicted and measured angle of friction using the measured cohesion and ultimate compressive strength, according to one or more of the embodiments of this detection. In the example shown, the angle of friction (φ) was calculated using Equation 2. One or both of the ultimate compressive strength and cohesion used in calculating the angle of friction can be measured or predicted, for example, using the trained cement performance characteristic model(s) in Figures 5 and 7, in some examples. In the graph shown in Figure 11, the cohesion is calculated and compared to its measured values for a range of cement barrier designs. As shown, with the exception of some data points, the calculated cohesion compares well to its measured value, demonstrating the good accuracy of the trained cement performance characteristic models in Figures 5 and 7 for use in the methods and systems of this detection, for example, the 400 workflow in Figure 4. Figure 12 is an equivalence plot of the training data, including the predicted versus measured Young's modulus, for training the cement performance characteristics model using a range of cement barrier designs, according to one or more embodiments of this disclosure. Figure 13 is an equivalence plot of the test data for testing the cement performance characteristics model predictions for the cement performance characteristics model trained in Figure 12, which also includes the predicted versus measured Young's modulus for a range of cement barrier designs, according to one or more embodiments of this disclosure. As shown, the equivalence plot and R² values show excellent agreement. Without adhering to any theory, the measurement differences shown in Figures 12 and 13 and other figures are believed to be attributable to measurement repeatability, reproducibility, and other factors. Figure 14 is a neutral graph of the training data, including the predicted versus measured Poisson ratio, for training a cement performance characteristics model using a range of cement barrier designs, based on one or more embodiments of this finding. In the example shown, the cement performance characteristics model was constructed using a random forest to correlate the Poisson ratio with the composition, hardening, and test conditions. Figure 15 is a neutral graph of the test data for evaluating the accuracy of the cement performance characteristics model predictions for the trained cement performance characteristics model in Figure 14, which also includes the predicted versus measured Poisson ratio for a range of cement barrier designs, based on one or more embodiments of this finding. Each data point in Figure 15 represents a prediction by the cement performance characteristics model trained in Figure 14 for a unique cement barrier design. It is possible to improve the R2 of the cement performance characteristics model for the Poisson ratio shown in Figure 14 with further data cleaning and fine-tuning.As with the other trained cement performance characteristic models shown in the present figures, the trained cement performance characteristic models for Young’s modulus and Poisson ratio in Figures 12 to 15 show good accuracy and are usable with the methods and systems mentioned in this disclosure, for example, the 400 workflow in Figure 4. The methods and systems described above represent an improvement over current technology in that they provide techniques for predicting cement failure properties that are both rapid and reliable. Predictions of these and other cement performance characteristics can help engineers make more informed decisions regarding cement designs. In some cases, these predictions may be readily available, such as when digital technology (e.g., Information Processing System 114 in Figure 4) is used to display predictions in real time, allowing them to be shown to the engineer or other personnel in real time, such as during the installation process (e.g., while cement is actively being pumped into a borehole). Other specific improvements associated with one or more implementations of this detection may include an overall improvement in the ability to design cement for cementing, a reduction in the time and expertise required for cement design, an improved ability to quickly determine long-term cement behavior in advance, and enhanced ability to make immediate decisions, among other benefits.In addition, some models may provide a digital system that recommends low-cost cement designs guaranteed not only to meet the requirements of the ultrasonic compressive strength analyzer, condensation, and fluid loss, but also to meet the requirements of long-term mechanical properties. The cement mentioned may directly or indirectly affect various equipment and tools located at the bottom of the well that may come into contact with well-hardening fluids during operations. This equipment and tools may include, but are not limited to, wellbore casing, wellbore liner, completion pipe string, inlet pipe string, drill pipe string, coiled pipe, slip cable, drill cable, drill pipe, drill collars, drilling motors, downhole motors and / or pumps, surface-mounted motors and / or pumps, centering rings, bladed centering rings, abrasives, floats (e.g., soles, collars, valves, etc.), logging instruments and related telemetry equipment, actuators (e.g., electromechanical devices, hydromechanical devices, etc.), slip sleeves, production sleeves, plugs, screens, filters, flow controllers (e.g., inlet flow controllers, standalone inlet flow controllers, outlet flow controllers, etc.), and couplings (e.g., wet electro-hydraulic couplings and dry couplings).Inductive couplers, etc.), control lines (e.g., electrical, fiber optic, hydraulic, etc.), monitoring lines, drill bits and reamers, sensors or distributed sensors, downhole heat exchangers, corresponding valves and actuators, tool seals, gaskets, cement plugs, bridge plugs, and other wellbore isolation devices or components, and the like. Any of these components can be included in the main systems and devices generally described above. It should be noted that constructs and methods are described as "incorporating," "containing," or "including" multiple components or steps, and may also be described as "consisting primarily of" or "comprising" multiple components and steps. Furthermore, the "indefinite singular" form, as used in protective elements, is defined here as meaning one or more of the elements it refers to. For the sake of brevity, only certain ranges are explicitly disclosed. However, ranges from any minimum to any maximum can be combined to indicate a range not explicitly stated, and similarly, ranges from any minimum to any other minimum can be combined to indicate a range not explicitly stated. Furthermore, when disclosing a numerical range with a minimum and a maximum, any number and any included range within that range are specifically disclosed. In particular, each of the value ranges (in the form of, "about A to about B," or equivalently, "about A to B," or equivalently, "about A-B") disclosed here is intended to indicate every number and range included within the broader range of values, even if not explicitly stated. Thus, each individual point or value, in the form of its minimum or maximum, can function as a combination with any other point or individual value, or any other minimum or maximum, to indicate a range not explicitly stated. Therefore, the present disclosure is well-suited to achieving the stated aims and features, as well as those inherent to it. The specific examples disclosed above are for illustrative purposes only, as the present disclosure can be modified and implemented in various but equally valid ways, readily apparent to those skilled in the field upon utilizing the information provided herein. Although individual examples are discussed, the disclosure covers all combinations of those examples in their entirety. Furthermore, there are no restrictions on the construction or design details presented herein, other than those described in the following elements of protection. Also, the terms used in the elements of protection have their ordinary, unambiguous meanings unless the patent holder explicitly and unambiguously defines otherwise. Thus, it is clear that the specific illustrative examples disclosed can be altered or modified, and all such variations should be considered within the scope and content of the present disclosure. In the event of a conflict between the use of a word or term in this specification and one or more patents or other documents that may be included herein for reference, the definitions consistent with this specification shall be used.
Claims
1. The cement slurry design method includes: (a) providing cement slurry design requirements, where cement design requirements include at least one cement performance property chosen from the group consisting of compressive strength, tensile strength, cohesion, friction angle, Young's modulus, Poisson's ratio, and any combination thereof; (b) providing a virtual cement slurry recipe representing at least water and its concentration and one or more cementitious materials and their concentration; (c) introducing at least one well condition and the virtual cement slurry into the cement performance property model; (d) Predicting the performance characteristics of a virtual cement slurry recipe using at least one cement performance property model, where the performance characteristics include at least one of the compressive strength and tensile strength.(e) Comparison of predicted performance properties with cement design requirements; and (f) Preparation of cement slurry according to the virtual cement slurry recipe if the predicted performance properties of the virtual cement slurry recipe meet the cement design requirements, or repeating steps (b) - (g) if the virtual cement slurry recipe does not meet the cement design requirements, where the step of providing the virtual cement slurry recipe includes providing a virtual cement slurry recipe with a different concentration of water, a different concentration of one or more cementitious materials, and / or a different chemical identity of one or more cementitious materials.
2. The method according to safeguard element 1, where the performance properties also include fluid loss and thickening time.
3. The method according to safeguard element 1, where the virtual cement slurry recipe also includes at least one cement additive selected from a set consisting of an accelerator, a retarder, latex, polyvinyl alcohol, crystalline silica, and any combination thereof.
4. The method according to safeguard element 1, where the cement performance property model includes a neural network, a decision tree, or a random forest model.
5. The method according to safeguard element 1, where the method also includes training one or more machine learning algorithms using training data to construct the cement performance property model, where the training data includes a set of cement composition data.Cement composition data and cement properties, including ultimate compressive strength, tensile strength, friction angle, cohesion, Young's modulus, and Poisson's ratio, are used for each of the multiple cement composition data.
6. The method according to safeguard element 1, where one or more cementitious materials include at least one pozzolanic material selected from the group consisting of fly ash, pumice, silicate, cement kiln dust, and any combination thereof.
7. The method according to safeguard element 1, where the cement slurry consists of at least one density modifier selected from the group consisting of a lightweight bead, a heavy-weight metal oxide, an elastomer, fibers, and any combination thereof.
8. The method according to protection element 1, where the well condition includes curing temperature, curing pressure, and curing time.
9. Method according to Protection Element 1, where the predicted performance properties include the friction angle, and where the friction angle is predicted using measurements of the ultimate compressive strength and cohesion.
10. Method according to Protection Element 1, where the predicted performance properties include the ultimate compressive strength and cohesion, where the method also involves determining the friction angle from the predictions of the ultimate compressive strength and cohesion.
11. Method according to Protection Element 10, where the process of determining the friction angle involves using an equation of the following form: where UCS is the predicted ultimate compressive strength, Cohesion is the predicted cohesion, and φ is the friction angle.
12. Method according to Protection Element 10, where the cement slurry is prepared according to the new cement slurry recipe if it isThe expected friction angle is greater than or equal to the friction angle requirements.
13. The method according to protection element 1, where the cement performance property model includes a linear model as follows: where is the performance property of the cement, w is the mass of water, Mass Poz is the mass of pozzolan, Vol Poz is the volume of pozzolan, Voli is the volume of pozzolan type i, Mass Addj is the mass of the additive j, Mass Blendk is the mass of the blended type k, Well Conditionl is the well conditionl, and a0, ai, aj, ak, and al are constants.
14. The method according to protection element 1, where the cement performance property model includes a non-linear model with the following form: where is the performance property of the cement, w is the mass of water, Mass Poz is the mass of pozzolan, Vol Poz is the volume of pozzolan, Voli is the volume of the pozzolanic species, Mass Addj is the mass of the additive j, Mass Blendk is the mass of the blended species k, Well Conditionl is the well conditionl, and a0, ai, aj, ak, and al are constants.
15. The method according toProtection element 1 also includes the introduction of cement slurry into a subterranean formation.
16. The cement slurry design method includes: providing cement slurry design requirements, where cement design requirements include at least one cement performance property chosen from the group consisting of compressive strength, tensile strength, cohesion, friction angle, Young's modulus, Poisson's ratio, and any combination thereof; providing a virtual cement slurry recipe representing at least water and its concentration, and one or more cementitious materials and their concentration; Input the minimum curing temperature, curing pressure, curing time, and ambient pressure into the cement performance property model, and the default cement slurry into the cement performance property model.Performance property model; predicting the performance properties of a virtual cement slurry using a cement performance property model, where the performance properties include at least one of the following: compressive strength, tensile strength, cohesion, friction angle, Young's modulus, and Poisson's ratio; comparing the predicted performance properties with the cement design requirements; and modifying the virtual cement slurry recipe if the predicted performance properties do not meet the cement design requirements. This modification may involve formulating a new virtual cement slurry with a different concentration of at least one or more cementitious materials, or preparing the cement slurry according to the virtual cement slurry recipe if the performance properties...The predicted performance properties meet or exceed the cement design requirements.
17. The method according to safeguard element 16, where the cement performance property model includes a neural network, a tree-based model, or a random forest model.
18. The method according to safeguard element 16, where the cement performance property model includes a linear model as follows: where is the performance property of the cement, w is the mass of water, Mass Poz is the mass of pozzolan, Vol Poz is the volume of pozzolan, Voli is the volume of the pozzolanic species, Mass Addj is the mass of the additive j, Mass Blendk is the mass of the blended species k, Well Conditionl is the well conditionl, and a0, ai, aj, ak, and al are constants.
19. The method according to protection element 16, where the cement performance property model includes a non-linear model with the following form: where is the performance property of the cement, w is the mass of water, and Mass Poz is the massPozzolan, where Vol Poz is the volume of the pozzolan, Voli is the volume of the pozzolanic species, Mass Addj is the mass of the added material, Mass Blendk is the mass of the blended species, Well Conditionl is the well condition, and a0, ai, aj, ak, and al are constants.
20. The method according to protection element 13 also includes the introduction of cement slurry into a subterranean formation.
