Computer-implemented method for detecting and diagnosing failures in the fluid properties in a drilling system and computer-readable storage media

Abstract

The present disclosure relates to a computer-implemented method for detecting and diagnosing failures in the fluid properties in a drilling system, including: performing training using a first set of training data; sending the training data to the database; defining the limit local density values, coefficient of variation, and derivative window limits used to establish the steady state regime; receiving data from the drilling system with the point to be analyzed; calculating the local outlier factor; comparing the local outlier factor with the outlier factor of the training set; calculating the numerical derivative of the point to be analyzed; comparing the number of points in failure and the number of points in steady state with the tolerance of points in failure and the tolerance of points in steady state; calculating the dispersion coefficient of the new set of training data; isolating the failure and identifying with a label the variable that contributes most to the occurrence of a failure; isolating the failure and identifying with a label the variable that contributes most to the occurrence of a failure.

Description

CROSS-REFERENCE TO RELATED APPLICATION(S) AND CLAIM OF PRIORITY

[0001] This application claims the benefit of priority of application Ser. No. 1020250028034, filed in Brazil on Feb. 12, 2025, the complete disclosure of which is incorporated herein by reference.FIELD

[0002] Some embodiments of the present disclosure pertain to the technical field of well drilling and completion. Some embodiments of the present disclosure also relate to a computer-implemented method for detecting and diagnosing failures in the fluid properties in a drilling system and a computer-readable storage media.BACKGROUND

[0003] The process of drilling oil wells requires the use of fluids for certain applications, such as: loading well cuttings to the surface; cooling and lubricating the drill bit; maintaining the stability in uncased sections of the well; and maintaining hydrostatic pressure in the well annulus, among others. Consequently, it is desirable that the fluid properties be constantly monitored, so that any anomaly present is reported, and control and correction actions are triggered.

[0004] In the current context, with the increase in the amount of data collected relating to the properties of the drilling fluid, the evaluation of data by a human operator makes the monitoring process vulnerable to human error.

[0005] In this sense, there is a need for a method to assist the operator in making decisions to correct the properties of the drilling fluids, record operating data, and process and evaluate the current state of the drilling fluid through failure detection and diagnosis techniques. Thus, it is desirable to provide an answer to the operator indicating whether or not an anomaly has occurred in one or more monitored variables.

[0006] Currently, it is necessary to collect fluid samples and then send the samples to the analysis laboratory, where tests are performed, and the results are sent to the rig operator. Monitoring is carried out by following bench analyses, in which the operators analyze the results of a set of variables simultaneously over long periods. In this way, both the measurement process and the analysis of the results are time-consuming tasks, causing delays in the flow of information within the rig. As a consequence, the process causes adversities in the evaluation of the system, leading to risks for the entire operation.STATE OF THE ART

[0007] The document “Anomaly detection in oil-producing wells: a comparative study of one-class classifiers in a multivariate time series data set” (Journal of Petroleum Exploration and Production Technology 14:343-363, 2024) discloses a method for detecting anomalies in oil-producing wells using machine learning techniques, a publicly available 3W data set comprising multivariate time series data, presenting a comparison of different anomaly detection methods, specifically one-class classifiers, including Isolation Forest, One-class Support Vector Machine (OCSVM), Local Outlier Factor (LOF), Elliptical Envelope, among others. However, that paper does not provide details on the application of LOF to a method for anomaly detection.

[0008] Furthermore, document BR 102024003349-3 also deals with a diagnostic system for anomalies in the properties of drilling fluids. However, the techniques used in that document are statistical models for monitoring the process, namely, Principal Component Analysis (PCA) and Dynamic Principal Component Analysis (DPCA). After training the model, the algorithm calculates the process health statistics in order to quantify the fluid quality, identifying the state of operation.

[0009] Some embodiments of the present disclosure differ from document BR 102024003349-3 in at least two main points: (i) type of technique used and (ii) adaptability to new operational levels.

[0010] The type of technique used in some embodiments of the present disclosure, Local Outlier Factor (LOF), aims at detecting outliers by monitoring the local density of a point relative to its nearest k-neighbors. Therefore, points that have a substantially lower local density than their k-neighbors are classified as outlier candidates, and for this system, the outliers are considered anomalies.

[0011] The Principal Component Analysis (PCA), as used in document BR 102024003349-3, on the other hand, monitors the process based on the system variability. The process health statistics are trained under normal conditions, and during monitoring, if the system variability exceeds this detection limit, the process is considered to be in failure. Furthermore, through the contribution of each variable to the statistics, it is possible to infer in which component the failure originates.

[0012] In summary, LOF focuses on anomaly detection through the local density, while PCA reduces the dimensionality of the system and monitors the process through detection limits.

[0013] Another point that differentiates the present disclosure from document BR 102024003349-3 is the number of trainings performed. Once the BR 102024003349-3 algorithm trains the techniques and the detection limits are defined, the process is then monitored. If the process fails, the system will maintain the failure status until one of the previously defined steady states is reached. In this way, all possible steady states that the fluid may pass through during operation must be included in the training, which is sometimes a limitation of monitoring, since operating conditions change depending on the region and type of rock being drilled. If an operating state different from that performed in training is identified, the monitoring system will indicate that the fluid is in failure, even if this is a new desired operating level.

[0014] In turn, some embodiments of the present disclosure are able to adapt the normal operating state based on monitoring the derivative window and identify a new steady state. In this way, new operating levels can be entered throughout the process monitoring, provided there is consent from the human operator, since any operational changes in the fluid property conditions must be made by the person responsible for the operation.BRIEF DESCRIPTION OF THE DISCLOSURE

[0015] One object of the computer-implemented method for detecting and diagnosing failures in the fluid properties of a drilling system of the present disclosure is the monitoring, diagnosis of anomalies, and recovery of the process to normal or standard conditions. Specifically, the detection and identification of anomalies, more specifically, failures in the properties of drilling fluids such as density, apparent viscosity, electrical conductivity, and water-in-oil content, among others. In this sense, the result of the process health assists the operator in acting on the control system and restructuring the system to normal conditions.

[0016] The oil well drilling, especially in marine environments, is a very expensive operation where minimizing drilling time and damage to the producing reservoir is fundamental. Drilling normally occurs through the application of weight and rotation to the string, the end of which is coupled to a cutting bit. Simultaneously, drilling fluid is circulated through the inside of the well, according to the following path: the fluid is injected from inside the string, passes through holes in the drill bit and returns through the annular space formed by the well walls and the drill string. FIG. 1 illustrates the fluid circulation system, which is common n in oil and natural gas drilling activities.

[0017] The drilling fluids are specified in order to ensure fast and safe drilling. Thus, it is desirable that the fluid has the following features: be chemically stable, not cause damage to the producing formations, withstand chemical and physical treatment, prevent corrosion of the drill string and other equipment in the circulation system, be pumpable, have a cost compatible with the operation, mechanically and chemically stabilize the well walls, keep solids in suspension when at rest, facilitate the separation of solids generated by the drill bit (cuttings) on the surface, and facilitate the geological interpretations of the material removed from the well.

[0018] The drilling fluids, in addition to having the features described above, desirably also have some basic functions, which are highlighted below:

[0019] to exert hydrostatic pressure on the formations in order to prevent the influx of undesirable fluids, which can cause serious safety problems for the entire surface team;

[0020] to maintain the physical integrity of the well, preventing a possible collapse of the formation, especially the more friable ones;

[0021] to clean the bottom of the well, bringing back to the surface the solids generated by the drill bit (cuttings); and

[0022] to cool and lubricate the drill string and the drill bit.

[0023] During a drilling operation, the properties of the fluid are constantly monitored in order to guarantee the fulfillment of its functions. The most important physical properties and those frequently measured in the drilling rigs are density, rheological parameters, filtration parameters, and solids content. In this sense, by quickly identifying anomalies, the operators can take corrective actions before small failures turn into breakdowns, minimizing risks and improving the safety of operations, thus reducing the frequency and severity of the corrective actions required. As a consequence, the method of some embodiments of the present disclosure allows the removal of personnel exposed to unhealthy and highly hazardous environments.

[0024] Furthermore, automatic monitoring of the properties significantly reduces the response time regarding the fluid state, which provides improvements in the ability to correct and control well drilling operations. This results in a more agile and effective response capacity, allowing the operators to make immediate adjustments to drilling operations. As a consequence, longer periods of stability can be achieved, avoiding interruptions and optimizing the workflow. Additionally, some embodiments of the disclosure have a direct impact on reducing the expenditure on inputs for correcting the properties of the drilling fluids.

[0025] In addition, some embodiments of the disclosure enable the reduction of operational costs associated with man-hours dedicated to data analysis and, by freeing operators from repetitive tasks, the disclosure allows them to focus on more strategic activities, increasing the productivity.

[0026] In summary, the disclosure of a failure automatic monitoring system based on the LOF technique effectively addresses the difficulties faced in well drilling. By ensuring the early detection of anomalies, improving safety, optimizing costs, and promoting sustainable practices in accordance with one or more embodiments disclosed herein, this technology transforms the way operations are conducted, resulting in significant benefits for both operators and the environment.

[0027] The present disclosure, in a preferred embodiment thereof, defines a computer-implemented method for detecting and diagnosing failures in the fluid properties in a drilling system, comprising the following steps:

[0028] performing training using a first set of training data;

[0029] sending the training data to the database;

[0030] defining the values of limit local density, coefficient of variation, and derivative window limits used to establish the steady state regime;

[0031] receiving data from the drilling system with the point to be analyzed;

[0032] calculating the local outlier factor, by using the equation:LOFMinPts(p)=∑o∈NMinPts(p)IrdMinpts(o)IrdMinpts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>,equation⁢ 1wherein:Nk-dist⁡(p)(p)={q∈D⁢{p}❘d⁡(p,q)≤k-dist⁡(p)},equation⁢ 2reach-distk(p,o)=max⁢{k-distance(o),d⁡(p,o)},equation⁢ 3IrdMinpts(p)=(∑o∈NMinPts(p)Rk(p,o)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>)-1,equation⁢ 4and wherein:

[0034] p and q are objects belonging to the data set D;

[0035] d(p,q) is the minimum distance between p and the object q in D;

[0036] k-dist(p) is the distance between p and an object o belonging to D, such that, for at least k objects o′∈D\{p}, it is valid that d(p,o′)≤d(p,o) and for at most k−1 objects o∈D\{p} it is valid that d(p,o′)<d(p,o); and

[0037] k is a real and positive value;

[0038] comparing the local outlier factor with the outlier factor of the training set;

[0039] calculating the numerical derivative of the point to be analyzed, by using the finite difference technique, and analyzing whether this point is contained in the interval defined by the equation below:OR=μtraining±x·σtraining,equation⁢ 5wherein:

[0041] μtraining represents the mean of the training series;

[0042] x represents the number of standard deviations used; and

[0043] σtraining represents the standard deviation of the training series;

[0044] comparing the number of points in failure and the number of points in steady state with the tolerance for points in failure and the tolerance for points in steady state; wherein, if the comparisons exceed the established limits, identifying a possible new steady state and collecting a new set of training data;

[0045] calculating the dispersion coefficient of the new set of training data using the equation below:cv=sx,equation⁢ 6wherein:

[0047] s represents the sample standard deviation;

[0048] x is the sample mean;

[0049] wherein, if the dispersion coefficient of the new set of training data is equal to or less than the first set of training data, the drilling system is established in a new steady state and a decision window is opened;

[0050] isolating the failure and identifying with a label the variable that most contributes to the occurrence of a failure in the drilling system, through the local outlier factor, by using the equations below:ContrjLOF=ηp⁢∑o∈NMinPts(p)[ξjT(xp-xo)]2,equation⁢ 7ηp=1<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>2⁢∑o∈NMinPts(p)lrdMinPts(o),equation⁢ 8whereξjTis the vector of the j-th column of the identity matrix; andJ is the number of variables being analyzed at each point;issuing an output message including at least one of: the variable identified with the label that most contributes to the occurrence of failure in the drilling system; the health of the operation, whether in training, normal or anomaly; and the components of the system that are in failure, if the operation is in anomaly.Furthermore, according to another embodiment of the disclosure, the first set of training data includes at least one variable, wherein the at least one variable includes at least one of: rheological properties of the fluids, fluid density, electrical conductivity, percentage of water in synthetic fluids, and apparent viscosity.

[0055] In addition, if the local outlier factor is greater than the outlier factor of the training set, the density of this point relative to its k-neighbors is lower compared to the training set, and the point is classified as a failure; and if the number of points classified as failures is greater than a predefined value, the number of points classified as failures is continuously summed; wherein a normal point zeroes the number of points in failure.

[0056] Additionally, if the derivative of the point is contained in the interval and the number of points in steady state is greater than a predefined value, then the number of points in steady state is summed; and wherein, if a point whose derivative is not contained in this interval, the number of points in steady state is zeroed.

[0057] In a complementary way, if the dispersion coefficient of the new set of training data is equal to or less than the first set of training data, the drilling system is established in a new steady state and a decision window is opened; wherein the operator decides whether or not to perform the training in this new steady state; if the operator decides to perform the training in this new steady state, the operator confirms that this new steady state is a desired state, and the training is performed.

[0058] In addition, the output message can be issued in an application to an operator.

[0059] Further, according to another preferred embodiment of the present disclosure, a computer-readable storage media is defined comprising, stored therein, a set of computer-readable instructions, which when executed by a computer, executes the computer-implemented method for detecting and diagnosing failures in the fluid properties in a drilling system, as described above.BRIEF DESCRIPTION OF THE FIGURES

[0060] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

[0061] In order to complement the present description and to obtain a better understanding of the features of the present disclosure, and in accordance with a preferred embodiment thereof, there is attached a set of figures that, by way of example, though in a non-limiting way, represent its preferred embodiment.

[0062] FIG. 1 illustrates the fluid circulation system, according to the state of the art.

[0063] FIG. 2 illustrates the flowchart of the computer-implemented method for detecting and diagnosing failures in the fluid properties in the drilling system of the present disclosure.

[0064] FIG. 3 presents an example of a diagnostic result, showing graphs of the LOP factor x time(s) and process health x time(s).

[0065] FIG. 4 represents an example of an output screen with a sampling graph with various fluid parameters (conductivity, flow rate, specific mass, temperature, apparent viscosity and ambient temperature).

[0066] FIG. 5 illustrates an example of the result for statistical monitoring of the system health.

[0067] FIG. 6 shows an example of the drilling system operating in anomaly.

[0068] FIG. 7 highlights an example of the drilling system operating without anomaly.

[0069] FIG. 8 presents an example of a screen from the real-time data receiving system.

[0070] FIG. 9 presents an example of an anomaly reported in the rheology of the drilling fluid.DETAILED DESCRIPTION

[0071] The computer-implemented method for detecting and diagnosing failures in the fluid properties in a drilling system, as illustrated in the flowchart in FIG. 2, comprises the following steps:

[0072] performing training (1) using a first set of training data, wherein the first set of training data includes at least one variable, wherein the at least one variable includes at least one of: fluid rheological properties, fluid density, electrical conductivity, percentage of water in synthetic fluids, and apparent viscosity;

[0073] sending the training data to the database (2);

[0074] defining the local density limit values, coefficient of variation, derivative window limits used to establish the steady state regime;

[0075] receiving data from the drilling system with the point to be analyzed;

[0076] calculating the local outlier factor (3), by using the equation:LOFMinPts(p)=∑o∈NMinPts(p)IrdMinpts(o)IrdMinpts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>,equation⁢ 1wherein:Nk-dist⁡(p)(p)={q∈D⁢{p}❘d⁡(p,q)≤k-dist⁡(p)},equation⁢ 2reach-distk(p,o)=max⁢{k-distance(o),d⁡(p,o)},equation⁢ 3IrdMinpts(p)=(∑o∈NMinPts(p)Rk(p,o)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>)-1,equation⁢ 4and wherein:

[0078] p and q are objects belonging to the data set D;

[0079] d(p,q) is the minimum distance between p and the object q in D;

[0080] k-dist(p) is the distance between p and an object o belonging to D, such that, for at least k objects o′∈D\{p}, it is valid that d(p,o′)≤d(p,o) and for at most k−1 objects o∈D\{p} it is valid that d(p,o′)<d(p,o); and

[0081] k is a real and positive value;

[0082] comparing the local outlier factor with the outlier factor of the training set (4); wherein, if the local outlier factor is greater than the outlier factor of the training set, it means that the density of this point relative to its k-neighbors is substantially lower compared to the training set, and the point is classified as a failure (4.1);

[0083] if the number of points classified as failures is greater than a predefined value (5), the number of points classified as failures is continuously summed (5.1); wherein a normal point zeroes the number of points in failure;

[0084] calculating the numerical derivative of the point to be analyzed (6), by using the finite difference technique, and analyzing whether this point is contained in the interval defined by the equation below:OR=μtraining±x·σtraining,equation⁢ 5wherein:

[0086] μtraining represents the mean of the training series;

[0087] x represents the number of standard deviations used; and

[0088] σtraining represents the standard deviation of the training series;

[0089] wherein, if the derivative of the point is contained in the interval (7.1) and the number of points in steady state is greater than a predefined value (8), then the number of points in steady state is summed (8.1); and wherein, if it is a point whose derivative is not contained in this interval, the number of points in steady state is zeroed;

[0090] comparing the number of points in failure and the number of points in steady state with the tolerance for points in failure and the tolerance for points in steady state; wherein, if the comparisons exceed the established limits, identifying a possible new steady state and collecting a new set of training data (9);

[0091] calculating the dispersion coefficient of the new set of training data using the equation below:cv=sx,equation⁢ 6wherein:

[0093] s represents the sample standard deviation;

[0094] x is the sample mean;

[0095] wherein, if the dispersion coefficient of the new set of training data is equal to or less than the first set of training data (10), the drilling system is established in a new steady state and a decision window is opened; wherein the operator decides whether or not to perform the training in this new steady state; if the operator decides to perform the training in this new steady state, the operator confirms that this new steady state is a desired state, and the training is performed;

[0096] isolating the failure and identifying with a label the variable that most contributes to the occurrence of a failure in the drilling system, through the local outlier factor, by using the equations below:ContrjLOF=ηp⁢∑o∈NMinPts(p)[ξjT(xp-xo)]2,equation⁢ 7ηp=1<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>2⁢∑o∈NMinPts(p)lrdMinPts(o),equation⁢ 8whereξjTis the vector of the j-th column of the identity matrix; andJ is the number of variables being analyzed at each point;issuing an output message including at least one of: the variable identified with the label that most contributes to the occurrence of failure in the drilling system; the health of the operation, whether in training, normal or anomaly; and the components of the system that are in failure, if the operation is in anomaly; wherein the output message can be issued in an application to an operator.TechniquesLocal Outlier Factor (LOF)The Local Outlier Factor (LOF) technique is an unsupervised anomaly detection method that calculates the deviation of the local density of a given point relative to its neighbors in a data set. Unlike traditional techniques, in which the outlier is considered in a Boolean way, in LOF each point is assigned a factor that measures the distance of the point from its neighbors. In the LOF technique, two levels are produced from the training data: the first level, called the anomaly-free state; and the second, called the state with anomalies. It is necessary that the selected training data satisfactorily describe the steady state of the drilling fluid, so that the classifier is able to generate the anomaly-free region. Once trained, the classifier will compare the new monitored samples and label the points as being in an anomalous state or anomaly-free state.Signal Variability Monitoring and Confidence Window

[0101] The signal variability monitoring technique consists of identifying in which level of variability are the properties of the drilling fluid. To define the variability of the drilling fluid, the derivatives of the monitored signals are calculated from the training data. Specifically, a statistical confidence region is drawn by summing the mean of the absolute value of the derivatives with the standard deviation of the training data; it is then defined that any monitored variation within this confidence region is a normal variability of a steady state, and any variation outside the confidence region is a variation belonging to a transient state of the drilling fluid properties, that is, it indicates that the drilling fluid properties are varying, and it is necessary to wait for stabilization before performing a new training.Operator Support Application

[0102] The results of the method of the present disclosure can be presented in an interface to present the data to the human operator in understandable formats, so as to allow him / her to analyze the data and make informed decisions.

[0103] Furthermore, the method of the present disclosure can also issue the information in the form of monitoring graphs, which represent the temporal evolution of the outlier factor over time, making it possible to observe the moment when the failure occurs. In addition, the process label and the variable in failure are presented to the operator through the diagnostic message.

[0104] In a complementary way, the present disclosure relates to a computer-readable storage media comprising, stored therein, a set of computer-readable instructions, wherein when the computer-readable instruction set is executed by one or more processors, the one or more processors implement the method of the present disclosure as described above.

[0105] In particular, the computer-readable storage media may be a memory, wherein the memory may be of a non-volatile type, such as a hard disk drive (HDD) or a solid-state drive (SSD), or it may be a volatile memory, such as random-access memory (RAM). Furthermore, the readable storage media may be any other medium or media that can carry or store or record the expected program code in the form of an instruction or a data structure or a set of instructions, and can be accessed by one or more computers or one or more processors, but is not limited to the same. The readable storage media can alternatively be a circuit or any other device or means that can implement a storage, carry, or recording function, such as a signal or carrier.

[0106] Specifically, the set of computer-readable instructions represents the algorithm or computer program code or a data structure that performs the method of the present disclosure described above.

[0107] The processor can be a general-purpose processor, which can be a microprocessor or any conventional processor or similar.EXAMPLES AND RESULTS OF THE DISCLOSURE

[0108] The validation of f the method of the present disclosure required a high volume of data relating to the properties of drilling fluids, varying set point values and generating noise in some cases. Such data were obtained by means of a pilot plant for the automated preparation of drilling fluids. The data were collected in closed-loop circulation operation equipped with sensors to measure apparent viscosity (rotational viscometer—TT-100), density (process densimeter, RHM 20), water-in-oil ratio (for oil-based fluids) and electrical conductivity (process conductivity meter—Strato PRO).

[0109] The method of the present disclosure records, monitors, detects, identifies failures, and saves the fluid property data in real time. System noises are smoothed by using the normalization based on the sets of training data. The possibility of set point changes was handled through trend monitoring of variables. If the points in failure exceed a certain tolerance and are contained within a confidence region, the algorithm performing the method of the disclosure retrains the algorithm. With the method of the present disclosure, it was possible to identify different types of disturbances in the characteristics of the drilling fluid. The algorithm was tested through numerical simulations and validated in the pilot unit.

[0110] Specifically, the validation test occurred with the disturbance in the properties of industrial water by means of the addition of a viscosifier, in this case carboxymethylcellulose (CMC), and a thickener, such as barite, for example. The test began with the circulation of industrial water through the system, which contains approximately 100 liters, for about 15 minutes to ensure that the fluid properties were in a steady state. After establishing a steady state for all monitored variables, the perturbation test was initiated by the gradual addition of carboxymethylcellulose (CMC) in sufficient quantities, ranging from 1 lb / bbl (2.853 kg / m3) to 2 lb / bbl (5.706 kg / m3), so that the apparent viscosity of the fluid, measured at 511 s−1, reached a value of 2.20 cP. Thirty minutes were allowed for the CMC particle swelling process to complete. Next, a second perturbation was performed on the fluid by adding barite in sufficient quantities, ranging from 30% w / w to 50% w / w, so that the fluid density reached 1.04 g / cm3. In addition to the disturbed properties, the diagnostic system observed the behavior of the other properties monitored in the plant, such as conductivity, water content, and flow rate.

[0111] The addition of viscosifier caused a continuous increase in the apparent viscosity of the fluid and, as this property was not being controlled, the system stabilized at a new level after the absorption of all the viscosifier. The sudden addition of a thickener caused a similar disturbance in the density and apparent viscosity properties, stabilized at another steady state level after the fluid circulated in a closed loop through the system.

[0112] The diagnostic system received, processed, and reported the integrity of the operation to the human operator in real time. At the end of the test, the algorithm provided a text file containing all the monitored variables and the process label. The process monitoring system was able to satisfactorily detect and identify the instant and indicate in which property the failure occurred. After the initial training, the algorithm maintained the information that the operation was the expected normality until the disturbance in apparent viscosity was sensed by the system. As soon as this property left the training interval, the process monitoring system almost instantly identified the disturbance, changing the process label to failure. FIG. 3 describes the entire process explained above. Due to the slow dilution of the CMC, the diagnostic system attempts to perform the training in the middle of the apparent viscosity increase curve; however, this region shows a greater dispersion compared to the original training, so this set is excluded, and when the apparent viscosity reaches 7.8 cP and the other properties remain constant, the system opens a window on the work screen so that the operator can indicate whether they want to perform a new training in this new steady state or maintain the previous training. With the operator's consent, the label is changed to normal and remains so until a new disturbance is detected by the system.

[0113] Next, the thickening agent was added to the system and, as soon as it was incorporated, the process monitoring changed the label from normal to failure and remained in this status until a new steady state was identified and the algorithm performed a new training and the label remained normal until the end of the test.

[0114] In addition to detecting the occurrence of the failure, the developed monitoring system identifies the cause of the failures. The algorithm identifies the predominance of the apparent viscosity as the root cause of the first failure identified by the algorithm, while for the second failure the main cause is shown as a variation in the fluid density followed to a lesser degree by apparent viscosity, since these two properties exhibit a certain synergy. Such diagnoses corroborate the test performed, which demonstrates the algorithm's ability to efficiently detect and diagnose disturbances in the properties of the drilling fluids.

[0115] FIGS. 4, 5, 6, and 7 illustrate the interface of an operator support application that receives the results of the method of the present disclosure. FIG. 4 shows an output screen with the result of a contribution of the monitored variables in relation to the disturbance of the system as a whole. In FIG. 5, the screen with the results of the computer-implemented method can be understood in 3 parts: a field for statistical monitoring of the system's health; the upper right part displays the communication status of the system with the test or analysis plant, and in the lower right part there is a field for detecting failures / anomalies and process state (green indicates and red indicates any anomaly detected). FIG. 6 shows the system operating under anomaly conditions. Once a failure is detected and alarmed, a secondary interface presents the operator with the result of the interpretation of the contribution maps of the process health statistics for the diagnosis of the detected failure. In turn, FIG. 7 illustrates an interface of the computer-implemented method in which the system is operating without anomalies.

[0116] The method of the present disclosure can be applied by receiving data from the monitoring systems of the properties of the drilling fluid. The opinions issued by the method of the present disclosure, integrated into supervisory systems, can speed up the operator's response to the process disturbances. This integration requires uploading diagnostic scripts and algorithms, training on the healthy state of the fluid, and output of results. Once loaded into the code repository, the outputs of the present method must be reconciled with the other results from the supervisory systems.

[0117] As a way to validate these methodologies with field data, the algorithm was set to monitor the drilling of wells A and B operated by PETROBRAS through the RTO live real-time data reception platform. During the drilling of well A, no anomalies were identified by the algorithm, as indeed occurred throughout the operation. FIG. 8 shows a screen of the real-time data reception system and in the last track on the right of the figure, in the “ANNOTATIONS” field, the algorithm's response after analyzing the data of the fluid properties. In turn, in the analyses performed by the algorithm for Well B, an anomaly was reported in the fluid rheology, as can be identified in FIG. 9. In this case, it was a sending failure at the root of the data, which was replicating the rheological reading value θ300 in the rheological reading value θ3. The operational team was contacted and the problem was resolved.

[0118] The rules implemented in the algorithm were able to identify this anomaly, showing that if any significant change or a change outside the expected range for the properties of drilling fluids occurs, alerts will be triggered so that the operational team can take corrective actions.

[0119] Those skilled in the art will appreciate the knowledge presented herein and may reproduce the disclosure in the presented embodiments and other variants, encompassed within the scope of the attached claims.

Examples

Embodiment Construction

[0071]The computer-implemented method for detecting and diagnosing failures in the fluid properties in a drilling system, as illustrated in the flowchart in FIG. 2, comprises the following steps:[0072]performing training (1) using a first set of training data, wherein the first set of training data includes at least one variable, wherein the at least one variable includes at least one of: fluid rheological properties, fluid density, electrical conductivity, percentage of water in synthetic fluids, and apparent viscosity;[0073]sending the training data to the database (2);[0074]defining the local density limit values, coefficient of variation, derivative window limits used to establish the steady state regime;[0075]receiving data from the drilling system with the point to be analyzed;[0076]calculating the local outlier factor (3), by using the equation:

LOFMinPts(p)=∑o∈NMinPts(p)IrdMinpts(o)IrdMinpts(p)❘"\[LeftBracketingBar]"NMinPts(p)❘"\[RightBracketingBar]",equation⁢ 1wherein:Nk-d...

CROSS-REFERENCE TO RELATED APPLICATION(S) AND CLAIM OF PRIORITY

[0001] This application claims the benefit of priority of application Ser. No. 1020250028034, filed in Brazil on Feb. 12, 2025, the complete disclosure of which is incorporated herein by reference.FIELD

[0002] Some embodiments of the present disclosure pertain to the technical field of well drilling and completion. Some embodiments of the present disclosure also relate to a computer-implemented method for detecting and diagnosing failures in the fluid properties in a drilling system and a computer-readable storage media.BACKGROUND

[0003] The process of drilling oil wells requires the use of fluids for certain applications, such as: loading well cuttings to the surface; cooling and lubricating the drill bit; maintaining the stability in uncased sections of the well; and maintaining hydrostatic pressure in the well annulus, among others. Consequently, it is desirable that the fluid properties be constantly monitored, so that any anomaly present is reported, and control and correction actions are triggered.

[0004] In the current context, with the increase in the amount of data collected relating to the properties of the drilling fluid, the evaluation of data by a human operator makes the monitoring process vulnerable to human error.

[0005] In this sense, there is a need for a method to assist the operator in making decisions to correct the properties of the drilling fluids, record operating data, and process and evaluate the current state of the drilling fluid through failure detection and diagnosis techniques. Thus, it is desirable to provide an answer to the operator indicating whether or not an anomaly has occurred in one or more monitored variables.

[0006] Currently, it is necessary to collect fluid samples and then send the samples to the analysis laboratory, where tests are performed, and the results are sent to the rig operator. Monitoring is carried out by following bench analyses, in which the operators analyze the results of a set of variables simultaneously over long periods. In this way, both the measurement process and the analysis of the results are time-consuming tasks, causing delays in the flow of information within the rig. As a consequence, the process causes adversities in the evaluation of the system, leading to risks for the entire operation.STATE OF THE ART

[0007] The document “Anomaly detection in oil-producing wells: a comparative study of one-class classifiers in a multivariate time series data set” (Journal of Petroleum Exploration and Production Technology 14:343-363, 2024) discloses a method for detecting anomalies in oil-producing wells using machine learning techniques, a publicly available 3W data set comprising multivariate time series data, presenting a comparison of different anomaly detection methods, specifically one-class classifiers, including Isolation Forest, One-class Support Vector Machine (OCSVM), Local Outlier Factor (LOF), Elliptical Envelope, among others. However, that paper does not provide details on the application of LOF to a method for anomaly detection.

[0008] Furthermore, document BR 102024003349-3 also deals with a diagnostic system for anomalies in the properties of drilling fluids. However, the techniques used in that document are statistical models for monitoring the process, namely, Principal Component Analysis (PCA) and Dynamic Principal Component Analysis (DPCA). After training the model, the algorithm calculates the process health statistics in order to quantify the fluid quality, identifying the state of operation.

[0009] Some embodiments of the present disclosure differ from document BR 102024003349-3 in at least two main points: (i) type of technique used and (ii) adaptability to new operational levels.

[0010] The type of technique used in some embodiments of the present disclosure, Local Outlier Factor (LOF), aims at detecting outliers by monitoring the local density of a point relative to its nearest k-neighbors. Therefore, points that have a substantially lower local density than their k-neighbors are classified as outlier candidates, and for this system, the outliers are considered anomalies.

[0011] The Principal Component Analysis (PCA), as used in document BR 102024003349-3, on the other hand, monitors the process based on the system variability. The process health statistics are trained under normal conditions, and during monitoring, if the system variability exceeds this detection limit, the process is considered to be in failure. Furthermore, through the contribution of each variable to the statistics, it is possible to infer in which component the failure originates.

[0012] In summary, LOF focuses on anomaly detection through the local density, while PCA reduces the dimensionality of the system and monitors the process through detection limits.

[0013] Another point that differentiates the present disclosure from document BR 102024003349-3 is the number of trainings performed. Once the BR 102024003349-3 algorithm trains the techniques and the detection limits are defined, the process is then monitored. If the process fails, the system will maintain the failure status until one of the previously defined steady states is reached. In this way, all possible steady states that the fluid may pass through during operation must be included in the training, which is sometimes a limitation of monitoring, since operating conditions change depending on the region and type of rock being drilled. If an operating state different from that performed in training is identified, the monitoring system will indicate that the fluid is in failure, even if this is a new desired operating level.

[0014] In turn, some embodiments of the present disclosure are able to adapt the normal operating state based on monitoring the derivative window and identify a new steady state. In this way, new operating levels can be entered throughout the process monitoring, provided there is consent from the human operator, since any operational changes in the fluid property conditions must be made by the person responsible for the operation.BRIEF DESCRIPTION OF THE DISCLOSURE

[0015] One object of the computer-implemented method for detecting and diagnosing failures in the fluid properties of a drilling system of the present disclosure is the monitoring, diagnosis of anomalies, and recovery of the process to normal or standard conditions. Specifically, the detection and identification of anomalies, more specifically, failures in the properties of drilling fluids such as density, apparent viscosity, electrical conductivity, and water-in-oil content, among others. In this sense, the result of the process health assists the operator in acting on the control system and restructuring the system to normal conditions.

[0016] The oil well drilling, especially in marine environments, is a very expensive operation where minimizing drilling time and damage to the producing reservoir is fundamental. Drilling normally occurs through the application of weight and rotation to the string, the end of which is coupled to a cutting bit. Simultaneously, drilling fluid is circulated through the inside of the well, according to the following path: the fluid is injected from inside the string, passes through holes in the drill bit and returns through the annular space formed by the well walls and the drill string. FIG. 1 illustrates the fluid circulation system, which is common n in oil and natural gas drilling activities.

[0017] The drilling fluids are specified in order to ensure fast and safe drilling. Thus, it is desirable that the fluid has the following features: be chemically stable, not cause damage to the producing formations, withstand chemical and physical treatment, prevent corrosion of the drill string and other equipment in the circulation system, be pumpable, have a cost compatible with the operation, mechanically and chemically stabilize the well walls, keep solids in suspension when at rest, facilitate the separation of solids generated by the drill bit (cuttings) on the surface, and facilitate the geological interpretations of the material removed from the well.

[0018] The drilling fluids, in addition to having the features described above, desirably also have some basic functions, which are highlighted below:

[0019] to exert hydrostatic pressure on the formations in order to prevent the influx of undesirable fluids, which can cause serious safety problems for the entire surface team;

[0020] to maintain the physical integrity of the well, preventing a possible collapse of the formation, especially the more friable ones;

[0021] to clean the bottom of the well, bringing back to the surface the solids generated by the drill bit (cuttings); and

[0022] to cool and lubricate the drill string and the drill bit.

[0023] During a drilling operation, the properties of the fluid are constantly monitored in order to guarantee the fulfillment of its functions. The most important physical properties and those frequently measured in the drilling rigs are density, rheological parameters, filtration parameters, and solids content. In this sense, by quickly identifying anomalies, the operators can take corrective actions before small failures turn into breakdowns, minimizing risks and improving the safety of operations, thus reducing the frequency and severity of the corrective actions required. As a consequence, the method of some embodiments of the present disclosure allows the removal of personnel exposed to unhealthy and highly hazardous environments.

[0024] Furthermore, automatic monitoring of the properties significantly reduces the response time regarding the fluid state, which provides improvements in the ability to correct and control well drilling operations. This results in a more agile and effective response capacity, allowing the operators to make immediate adjustments to drilling operations. As a consequence, longer periods of stability can be achieved, avoiding interruptions and optimizing the workflow. Additionally, some embodiments of the disclosure have a direct impact on reducing the expenditure on inputs for correcting the properties of the drilling fluids.

[0025] In addition, some embodiments of the disclosure enable the reduction of operational costs associated with man-hours dedicated to data analysis and, by freeing operators from repetitive tasks, the disclosure allows them to focus on more strategic activities, increasing the productivity.

[0026] In summary, the disclosure of a failure automatic monitoring system based on the LOF technique effectively addresses the difficulties faced in well drilling. By ensuring the early detection of anomalies, improving safety, optimizing costs, and promoting sustainable practices in accordance with one or more embodiments disclosed herein, this technology transforms the way operations are conducted, resulting in significant benefits for both operators and the environment.

[0027] The present disclosure, in a preferred embodiment thereof, defines a computer-implemented method for detecting and diagnosing failures in the fluid properties in a drilling system, comprising the following steps:

[0028] performing training using a first set of training data;

[0029] sending the training data to the database;

[0030] defining the values of limit local density, coefficient of variation, and derivative window limits used to establish the steady state regime;

[0031] receiving data from the drilling system with the point to be analyzed;

[0032] calculating the local outlier factor, by using the equation:LOFMinPts(p)=∑o∈NMinPts(p)IrdMinpts(o)IrdMinpts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>,equation⁢ 1wherein:Nk-dist⁡(p)(p)={q∈D⁢{p}❘d⁡(p,q)≤k-dist⁡(p)},equation⁢ 2reach-distk(p,o)=max⁢{k-distance(o),d⁡(p,o)},equation⁢ 3IrdMinpts(p)=(∑o∈NMinPts(p)Rk(p,o)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>)-1,equation⁢ 4and wherein:

[0034] p and q are objects belonging to the data set D;

[0035] d(p,q) is the minimum distance between p and the object q in D;

[0036] k-dist(p) is the distance between p and an object o belonging to D, such that, for at least k objects o′∈D\{p}, it is valid that d(p,o′)≤d(p,o) and for at most k−1 objects o∈D\{p} it is valid that d(p,o′)<d(p,o); and

[0037] k is a real and positive value;

[0038] comparing the local outlier factor with the outlier factor of the training set;

[0039] calculating the numerical derivative of the point to be analyzed, by using the finite difference technique, and analyzing whether this point is contained in the interval defined by the equation below:OR=μtraining±x·σtraining,equation⁢ 5wherein:

[0041] μtraining represents the mean of the training series;

[0042] x represents the number of standard deviations used; and

[0043] σtraining represents the standard deviation of the training series;

[0044] comparing the number of points in failure and the number of points in steady state with the tolerance for points in failure and the tolerance for points in steady state; wherein, if the comparisons exceed the established limits, identifying a possible new steady state and collecting a new set of training data;

[0045] calculating the dispersion coefficient of the new set of training data using the equation below:cv=sx,equation⁢ 6wherein:

[0047] s represents the sample standard deviation;

[0048] x is the sample mean;

[0049] wherein, if the dispersion coefficient of the new set of training data is equal to or less than the first set of training data, the drilling system is established in a new steady state and a decision window is opened;

[0050] isolating the failure and identifying with a label the variable that most contributes to the occurrence of a failure in the drilling system, through the local outlier factor, by using the equations below:ContrjLOF=ηp⁢∑o∈NMinPts(p)[ξjT(xp-xo)]2,equation⁢ 7ηp=1<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>2⁢∑o∈NMinPts(p)lrdMinPts(o),equation⁢ 8whereξjTis the vector of the j-th column of the identity matrix; andJ is the number of variables being analyzed at each point;issuing an output message including at least one of: the variable identified with the label that most contributes to the occurrence of failure in the drilling system; the health of the operation, whether in training, normal or anomaly; and the components of the system that are in failure, if the operation is in anomaly.Furthermore, according to another embodiment of the disclosure, the first set of training data includes at least one variable, wherein the at least one variable includes at least one of: rheological properties of the fluids, fluid density, electrical conductivity, percentage of water in synthetic fluids, and apparent viscosity.

[0055] In addition, if the local outlier factor is greater than the outlier factor of the training set, the density of this point relative to its k-neighbors is lower compared to the training set, and the point is classified as a failure; and if the number of points classified as failures is greater than a predefined value, the number of points classified as failures is continuously summed; wherein a normal point zeroes the number of points in failure.

[0056] Additionally, if the derivative of the point is contained in the interval and the number of points in steady state is greater than a predefined value, then the number of points in steady state is summed; and wherein, if a point whose derivative is not contained in this interval, the number of points in steady state is zeroed.

[0057] In a complementary way, if the dispersion coefficient of the new set of training data is equal to or less than the first set of training data, the drilling system is established in a new steady state and a decision window is opened; wherein the operator decides whether or not to perform the training in this new steady state; if the operator decides to perform the training in this new steady state, the operator confirms that this new steady state is a desired state, and the training is performed.

[0058] In addition, the output message can be issued in an application to an operator.

[0059] Further, according to another preferred embodiment of the present disclosure, a computer-readable storage media is defined comprising, stored therein, a set of computer-readable instructions, which when executed by a computer, executes the computer-implemented method for detecting and diagnosing failures in the fluid properties in a drilling system, as described above.BRIEF DESCRIPTION OF THE FIGURES

[0060] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

[0061] In order to complement the present description and to obtain a better understanding of the features of the present disclosure, and in accordance with a preferred embodiment thereof, there is attached a set of figures that, by way of example, though in a non-limiting way, represent its preferred embodiment.

[0062] FIG. 1 illustrates the fluid circulation system, according to the state of the art.

[0063] FIG. 2 illustrates the flowchart of the computer-implemented method for detecting and diagnosing failures in the fluid properties in the drilling system of the present disclosure.

[0064] FIG. 3 presents an example of a diagnostic result, showing graphs of the LOP factor x time(s) and process health x time(s).

[0065] FIG. 4 represents an example of an output screen with a sampling graph with various fluid parameters (conductivity, flow rate, specific mass, temperature, apparent viscosity and ambient temperature).

[0066] FIG. 5 illustrates an example of the result for statistical monitoring of the system health.

[0067] FIG. 6 shows an example of the drilling system operating in anomaly.

[0068] FIG. 7 highlights an example of the drilling system operating without anomaly.

[0069] FIG. 8 presents an example of a screen from the real-time data receiving system.

[0070] FIG. 9 presents an example of an anomaly reported in the rheology of the drilling fluid.DETAILED DESCRIPTION

[0071] The computer-implemented method for detecting and diagnosing failures in the fluid properties in a drilling system, as illustrated in the flowchart in FIG. 2, comprises the following steps:

[0072] performing training (1) using a first set of training data, wherein the first set of training data includes at least one variable, wherein the at least one variable includes at least one of: fluid rheological properties, fluid density, electrical conductivity, percentage of water in synthetic fluids, and apparent viscosity;

[0073] sending the training data to the database (2);

[0074] defining the local density limit values, coefficient of variation, derivative window limits used to establish the steady state regime;

[0075] receiving data from the drilling system with the point to be analyzed;

[0076] calculating the local outlier factor (3), by using the equation:LOFMinPts(p)=∑o∈NMinPts(p)IrdMinpts(o)IrdMinpts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>,equation⁢ 1wherein:Nk-dist⁡(p)(p)={q∈D⁢{p}❘d⁡(p,q)≤k-dist⁡(p)},equation⁢ 2reach-distk(p,o)=max⁢{k-distance(o),d⁡(p,o)},equation⁢ 3IrdMinpts(p)=(∑o∈NMinPts(p)Rk(p,o)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>)-1,equation⁢ 4and wherein:

[0078] p and q are objects belonging to the data set D;

[0079] d(p,q) is the minimum distance between p and the object q in D;

[0080] k-dist(p) is the distance between p and an object o belonging to D, such that, for at least k objects o′∈D\{p}, it is valid that d(p,o′)≤d(p,o) and for at most k−1 objects o∈D\{p} it is valid that d(p,o′)<d(p,o); and

[0081] k is a real and positive value;

[0082] comparing the local outlier factor with the outlier factor of the training set (4); wherein, if the local outlier factor is greater than the outlier factor of the training set, it means that the density of this point relative to its k-neighbors is substantially lower compared to the training set, and the point is classified as a failure (4.1);

[0083] if the number of points classified as failures is greater than a predefined value (5), the number of points classified as failures is continuously summed (5.1); wherein a normal point zeroes the number of points in failure;

[0084] calculating the numerical derivative of the point to be analyzed (6), by using the finite difference technique, and analyzing whether this point is contained in the interval defined by the equation below:OR=μtraining±x·σtraining,equation⁢ 5wherein:

[0086] μtraining represents the mean of the training series;

[0087] x represents the number of standard deviations used; and

[0088] σtraining represents the standard deviation of the training series;

[0089] wherein, if the derivative of the point is contained in the interval (7.1) and the number of points in steady state is greater than a predefined value (8), then the number of points in steady state is summed (8.1); and wherein, if it is a point whose derivative is not contained in this interval, the number of points in steady state is zeroed;

[0090] comparing the number of points in failure and the number of points in steady state with the tolerance for points in failure and the tolerance for points in steady state; wherein, if the comparisons exceed the established limits, identifying a possible new steady state and collecting a new set of training data (9);

[0091] calculating the dispersion coefficient of the new set of training data using the equation below:cv=sx,equation⁢ 6wherein:

[0093] s represents the sample standard deviation;

[0094] x is the sample mean;

[0095] wherein, if the dispersion coefficient of the new set of training data is equal to or less than the first set of training data (10), the drilling system is established in a new steady state and a decision window is opened; wherein the operator decides whether or not to perform the training in this new steady state; if the operator decides to perform the training in this new steady state, the operator confirms that this new steady state is a desired state, and the training is performed;

[0096] isolating the failure and identifying with a label the variable that most contributes to the occurrence of a failure in the drilling system, through the local outlier factor, by using the equations below:ContrjLOF=ηp⁢∑o∈NMinPts(p)[ξjT(xp-xo)]2,equation⁢ 7ηp=1<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>2⁢∑o∈NMinPts(p)lrdMinPts(o),equation⁢ 8whereξjTis the vector of the j-th column of the identity matrix; andJ is the number of variables being analyzed at each point;issuing an output message including at least one of: the variable identified with the label that most contributes to the occurrence of failure in the drilling system; the health of the operation, whether in training, normal or anomaly; and the components of the system that are in failure, if the operation is in anomaly; wherein the output message can be issued in an application to an operator.TechniquesLocal Outlier Factor (LOF)The Local Outlier Factor (LOF) technique is an unsupervised anomaly detection method that calculates the deviation of the local density of a given point relative to its neighbors in a data set. Unlike traditional techniques, in which the outlier is considered in a Boolean way, in LOF each point is assigned a factor that measures the distance of the point from its neighbors. In the LOF technique, two levels are produced from the training data: the first level, called the anomaly-free state; and the second, called the state with anomalies. It is necessary that the selected training data satisfactorily describe the steady state of the drilling fluid, so that the classifier is able to generate the anomaly-free region. Once trained, the classifier will compare the new monitored samples and label the points as being in an anomalous state or anomaly-free state.Signal Variability Monitoring and Confidence Window

[0101] The signal variability monitoring technique consists of identifying in which level of variability are the properties of the drilling fluid. To define the variability of the drilling fluid, the derivatives of the monitored signals are calculated from the training data. Specifically, a statistical confidence region is drawn by summing the mean of the absolute value of the derivatives with the standard deviation of the training data; it is then defined that any monitored variation within this confidence region is a normal variability of a steady state, and any variation outside the confidence region is a variation belonging to a transient state of the drilling fluid properties, that is, it indicates that the drilling fluid properties are varying, and it is necessary to wait for stabilization before performing a new training.Operator Support Application

[0102] The results of the method of the present disclosure can be presented in an interface to present the data to the human operator in understandable formats, so as to allow him / her to analyze the data and make informed decisions.

[0103] Furthermore, the method of the present disclosure can also issue the information in the form of monitoring graphs, which represent the temporal evolution of the outlier factor over time, making it possible to observe the moment when the failure occurs. In addition, the process label and the variable in failure are presented to the operator through the diagnostic message.

[0104] In a complementary way, the present disclosure relates to a computer-readable storage media comprising, stored therein, a set of computer-readable instructions, wherein when the computer-readable instruction set is executed by one or more processors, the one or more processors implement the method of the present disclosure as described above.

[0105] In particular, the computer-readable storage media may be a memory, wherein the memory may be of a non-volatile type, such as a hard disk drive (HDD) or a solid-state drive (SSD), or it may be a volatile memory, such as random-access memory (RAM). Furthermore, the readable storage media may be any other medium or media that can carry or store or record the expected program code in the form of an instruction or a data structure or a set of instructions, and can be accessed by one or more computers or one or more processors, but is not limited to the same. The readable storage media can alternatively be a circuit or any other device or means that can implement a storage, carry, or recording function, such as a signal or carrier.

[0106] Specifically, the set of computer-readable instructions represents the algorithm or computer program code or a data structure that performs the method of the present disclosure described above.

[0107] The processor can be a general-purpose processor, which can be a microprocessor or any conventional processor or similar.EXAMPLES AND RESULTS OF THE DISCLOSURE

[0108] The validation of f the method of the present disclosure required a high volume of data relating to the properties of drilling fluids, varying set point values and generating noise in some cases. Such data were obtained by means of a pilot plant for the automated preparation of drilling fluids. The data were collected in closed-loop circulation operation equipped with sensors to measure apparent viscosity (rotational viscometer—TT-100), density (process densimeter, RHM 20), water-in-oil ratio (for oil-based fluids) and electrical conductivity (process conductivity meter—Strato PRO).

[0109] The method of the present disclosure records, monitors, detects, identifies failures, and saves the fluid property data in real time. System noises are smoothed by using the normalization based on the sets of training data. The possibility of set point changes was handled through trend monitoring of variables. If the points in failure exceed a certain tolerance and are contained within a confidence region, the algorithm performing the method of the disclosure retrains the algorithm. With the method of the present disclosure, it was possible to identify different types of disturbances in the characteristics of the drilling fluid. The algorithm was tested through numerical simulations and validated in the pilot unit.

[0110] Specifically, the validation test occurred with the disturbance in the properties of industrial water by means of the addition of a viscosifier, in this case carboxymethylcellulose (CMC), and a thickener, such as barite, for example. The test began with the circulation of industrial water through the system, which contains approximately 100 liters, for about 15 minutes to ensure that the fluid properties were in a steady state. After establishing a steady state for all monitored variables, the perturbation test was initiated by the gradual addition of carboxymethylcellulose (CMC) in sufficient quantities, ranging from 1 lb / bbl (2.853 kg / m3) to 2 lb / bbl (5.706 kg / m3), so that the apparent viscosity of the fluid, measured at 511 s−1, reached a value of 2.20 cP. Thirty minutes were allowed for the CMC particle swelling process to complete. Next, a second perturbation was performed on the fluid by adding barite in sufficient quantities, ranging from 30% w / w to 50% w / w, so that the fluid density reached 1.04 g / cm3. In addition to the disturbed properties, the diagnostic system observed the behavior of the other properties monitored in the plant, such as conductivity, water content, and flow rate.

[0111] The addition of viscosifier caused a continuous increase in the apparent viscosity of the fluid and, as this property was not being controlled, the system stabilized at a new level after the absorption of all the viscosifier. The sudden addition of a thickener caused a similar disturbance in the density and apparent viscosity properties, stabilized at another steady state level after the fluid circulated in a closed loop through the system.

[0112] The diagnostic system received, processed, and reported the integrity of the operation to the human operator in real time. At the end of the test, the algorithm provided a text file containing all the monitored variables and the process label. The process monitoring system was able to satisfactorily detect and identify the instant and indicate in which property the failure occurred. After the initial training, the algorithm maintained the information that the operation was the expected normality until the disturbance in apparent viscosity was sensed by the system. As soon as this property left the training interval, the process monitoring system almost instantly identified the disturbance, changing the process label to failure. FIG. 3 describes the entire process explained above. Due to the slow dilution of the CMC, the diagnostic system attempts to perform the training in the middle of the apparent viscosity increase curve; however, this region shows a greater dispersion compared to the original training, so this set is excluded, and when the apparent viscosity reaches 7.8 cP and the other properties remain constant, the system opens a window on the work screen so that the operator can indicate whether they want to perform a new training in this new steady state or maintain the previous training. With the operator's consent, the label is changed to normal and remains so until a new disturbance is detected by the system.

[0113] Next, the thickening agent was added to the system and, as soon as it was incorporated, the process monitoring changed the label from normal to failure and remained in this status until a new steady state was identified and the algorithm performed a new training and the label remained normal until the end of the test.

[0114] In addition to detecting the occurrence of the failure, the developed monitoring system identifies the cause of the failures. The algorithm identifies the predominance of the apparent viscosity as the root cause of the first failure identified by the algorithm, while for the second failure the main cause is shown as a variation in the fluid density followed to a lesser degree by apparent viscosity, since these two properties exhibit a certain synergy. Such diagnoses corroborate the test performed, which demonstrates the algorithm's ability to efficiently detect and diagnose disturbances in the properties of the drilling fluids.

[0115] FIGS. 4, 5, 6, and 7 illustrate the interface of an operator support application that receives the results of the method of the present disclosure. FIG. 4 shows an output screen with the result of a contribution of the monitored variables in relation to the disturbance of the system as a whole. In FIG. 5, the screen with the results of the computer-implemented method can be understood in 3 parts: a field for statistical monitoring of the system's health; the upper right part displays the communication status of the system with the test or analysis plant, and in the lower right part there is a field for detecting failures / anomalies and process state (green indicates and red indicates any anomaly detected). FIG. 6 shows the system operating under anomaly conditions. Once a failure is detected and alarmed, a secondary interface presents the operator with the result of the interpretation of the contribution maps of the process health statistics for the diagnosis of the detected failure. In turn, FIG. 7 illustrates an interface of the computer-implemented method in which the system is operating without anomalies.

[0116] The method of the present disclosure can be applied by receiving data from the monitoring systems of the properties of the drilling fluid. The opinions issued by the method of the present disclosure, integrated into supervisory systems, can speed up the operator's response to the process disturbances. This integration requires uploading diagnostic scripts and algorithms, training on the healthy state of the fluid, and output of results. Once loaded into the code repository, the outputs of the present method must be reconciled with the other results from the supervisory systems.

[0117] As a way to validate these methodologies with field data, the algorithm was set to monitor the drilling of wells A and B operated by PETROBRAS through the RTO live real-time data reception platform. During the drilling of well A, no anomalies were identified by the algorithm, as indeed occurred throughout the operation. FIG. 8 shows a screen of the real-time data reception system and in the last track on the right of the figure, in the “ANNOTATIONS” field, the algorithm's response after analyzing the data of the fluid properties. In turn, in the analyses performed by the algorithm for Well B, an anomaly was reported in the fluid rheology, as can be identified in FIG. 9. In this case, it was a sending failure at the root of the data, which was replicating the rheological reading value θ300 in the rheological reading value θ3. The operational team was contacted and the problem was resolved.

[0118] The rules implemented in the algorithm were able to identify this anomaly, showing that if any significant change or a change outside the expected range for the properties of drilling fluids occurs, alerts will be triggered so that the operational team can take corrective actions.

[0119] Those skilled in the art will appreciate the knowledge presented herein and may reproduce the disclosure in the presented embodiments and other variants, encompassed within the scope of the attached claims.

Examples

Embodiment Construction

[0071]The computer-implemented method for detecting and diagnosing failures in the fluid properties in a drilling system, as illustrated in the flowchart in FIG. 2, comprises the following steps:[0072]performing training (1) using a first set of training data, wherein the first set of training data includes at least one variable, wherein the at least one variable includes at least one of: fluid rheological properties, fluid density, electrical conductivity, percentage of water in synthetic fluids, and apparent viscosity;[0073]sending the training data to the database (2);[0074]defining the local density limit values, coefficient of variation, derivative window limits used to establish the steady state regime;[0075]receiving data from the drilling system with the point to be analyzed;[0076]calculating the local outlier factor (3), by using the equation:

LOFMinPts(p)=∑o∈NMinPts(p)IrdMinpts(o)IrdMinpts(p)❘"\[LeftBracketingBar]"NMinPts(p)❘"\[RightBracketingBar]",equation⁢ 1wherein:Nk-d...

Claims

1. A computer-implemented method for detecting and diagnosing failures in the fluid properties in a drilling system, the method comprising:performing training using a first set of training data;sending the first set of training data to the database;defining values of limit local density, coefficient of variation, and derivative window limits used to establish a steady state regime;receiving data from the drilling system with a point to be analyzed;calculating a local outlier factor, by using the equation:LOFMinPts(p)=∑o∈NMinPts(p)IrdMinpts(o)IrdMinpts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>,equation⁢ 1wherein:Nk-dist⁡(p)(p)={q∈D⁢{p}❘d⁡(p,q)≤k-dist⁡(p)},equation⁢ 2reach-distk(p,o)=max⁢{k-distance(o),d⁡(p,o)},equation⁢ 3IrdMinpts(p)=(∑o∈NMinPts(p)Rk(p,o)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>)-1,equation⁢ 4and wherein:p and q are objects belonging to a data set D;d(p,q) is the minimum distance between p and the object q in D;k-dist(p) is a distance between p and an object o belonging to D, such that, for at least k objects o′∈D\{p}, it is valid that d(p,o′)≤d(p,o) and for at most k−1 objects o∈D\{p} it is valid that d(p,o′)<d(p,o); andk is a real and positive value;comparing the local outlier factor with an outlier factor of the training set;calculating a numerical derivative of the point to be analyzed, by using the finite difference technique, and analyzing whether this point is contained in an interval defined by the equation below:OR=μtraining±x·σtraining,equation⁢ 5wherein:μtraining represents the mean of the training series;x represents the number of standard deviations used; andσtraining represents the standard deviation of the training series;comparing the number of points in failure and the number of points in steady state with the tolerance for points in failure and the tolerance for points in steady state; wherein, if the comparisons exceed the established limits, identifying a possible new steady state and collecting a new set of training data;calculating the dispersion coefficient of the new set of training data using the equation below:cv=sx,equation⁢ 6wherein:s represents the sample standard deviation;x is the sample mean;wherein, if the dispersion coefficient of the new set of training data is equal to or less than the first set of training data, the drilling system is established in a new steady state and a decision window is opened;isolating the failure and identifying with a label the variable that most contributes to the occurrence of a failure in the drilling system, through the local outlier factor, by using the equations below:ContrjLOF=ηp⁢∑o∈NMinPts(p)[ξjT(xp-xo)]2,equation⁢ 7ηp=1<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[LeftBracketingBar]"< / annotation>< / semantics>NMinPts(p)<semantics definitionURL="">❘<annotation encoding="Mathematica">"\[RightBracketingBar]"< / annotation>< / semantics>2⁢∑o∈NMinPts(p)lrdMinPts(o),equation⁢ 8whereζjTis the vector of the j-th column of the identity matrix; andJ is the number of variables being analyzed at each point; andissuing an output message including at least one of: the variable identified with the label that most contributes to the occurrence of failure in the drilling system; the health of the operation, whether in training, normal or anomaly; and the components of the system that are in failure, if the operation is in anomaly.

2. The method according to claim 1, wherein the first set of training data includes at least one variable, wherein the at least one variable includes at: least one of: rheological properties of the fluids, fluid density, electrical conductivity, percentage of water in synthetic fluids, and apparent viscosity.

3. The method according to claim 1, wherein if the local outlier factor is greater than the outlier factor of the training set, the density of this point relative to its k-neighbors is lower compared to the training set, and the point is classified as a failure; wherein if the number of points classified as failure is greater than a predefined value, the number of points classified as failure is continuously summed; and wherein a normal point zeroes the number of points in failure.

4. The method according to claim 1, wherein if the derivative of the point is contained in the interval and the number of points in steady state is greater than a predefined value, then the number of points in steady state is summed; and wherein, if a point whose derivative is not contained in this interval, the number of points in steady state is zeroed.

5. The method according to claim 1, wherein if the dispersion coefficient of a new set of training data is equal to or less than the first set of training data, the drilling system is established in a new steady state and a decision window is opened; wherein the operator decides whether or not to perform the training in this new steady state; wherein if the operator decides to perform the training in this new steady state, the operator confirms that this new steady state is a desired state, and the training is performed.

6. The method according to claim 1, wherein the output message can be issued in an application to an operator.

7. A computer-readable storage media, for a process comprising storing, a set of computer-readable instructions, when executed by a computer, executes the method according to claim 1.

Images