A drilling early warning method, device and equipment

Abstract

This specification relates to the technical field of drilling status identification methods, and more particularly to a drilling early warning method, apparatus, and equipment. The drilling early warning method includes: calculating a trend value of first drilling data; the first drilling data includes real-time drilling key parameter values ​​at multiple consecutive time points; if the trend value of the first drilling data does not meet a preset confidence condition, obtaining a baseline value of second drilling data; the second drilling data includes normal historical drilling key parameter values ​​at multiple consecutive time points; and quantifying the degree of anomaly of the first drilling data based on the baseline value of the second drilling data to generate a corresponding early warning signal for the first drilling data. The embodiments of this specification can calculate the trend value of the first drilling data, and when the trend value of the first drilling data is not within a preset trend confidence interval, quickly and accurately determine the degree of anomaly of the first drilling data based on the baseline value of the second drilling data, thereby generating a timely early warning signal for the first drilling data.

Description

Technical Field

[0001] The embodiments in this specification relate to the field of oil and gas drilling, specifically to a drilling early warning method, device, and equipment. Background Technology

[0002] In oil and gas drilling operations, well control safety is always the core issue to ensure the smooth progress of the operation. Well control accidents, especially overflows and lost circulation, often lead to serious economic losses and safety hazards. In current well control operations, although conventional monitoring systems can collect data through multi-source sensors, their key shortcoming lies in the insufficient early warning and processing capabilities.

[0003] A Chinese patent with publication number CN111119864A discloses an overflow monitoring method and system based on gas intrusion pressure characteristics. This method compares raw pressure data and real-time pressure derivative data at the same location, dividing the current real-time pressure derivative data by the raw pressure data from the previous moment to obtain a quotient. Finally, it compares the quotient with a preset threshold. When the quotient is greater than the preset threshold, it records the current location and time as the overflow location and time.

[0004] The aforementioned early warning mechanisms / algorithms rely on fixed upper and lower threshold alarms, which are simplistic and crude. This method has significant shortcomings when dealing with rapidly changing and complex fluctuating operating conditions, especially when the parameter changes of the operating conditions are nonlinear. In such cases, the system response is slow and a large number of false alarms or missed alarms may occur. The current challenge facing the industry is no longer merely monitoring multi-source data, but rather how to quickly and accurately capture potential risk trends based on this multi-dimensional operating condition data through real-time processing and intelligent trend analysis, and issue effective early warning signals at an early stage.

[0005] Therefore, overcoming the problems of high false alarms / missed alarms and slow response speed in traditional early warning mechanisms / algorithms, and proposing a highly accurate and fast drilling early warning method to effectively capture early signs of accidents such as overflows and lost circulation in complex and ever-changing downhole environments, and thus issue early warning signals in a timely manner, is a key problem that urgently needs to be solved. Summary of the Invention

[0006] The purpose of the embodiments in this specification is to provide a drilling early warning method, device, and equipment to overcome the problems of high false alarms / missed alarms and slow response speed in traditional early warning mechanisms / algorithms. It proposes a drilling early warning method with high accuracy and speed to effectively capture early signs of accidents such as overflows and lost circulation in complex and ever-changing downhole environments, and then issue early warning signals in a timely manner.

[0007] On one hand, the embodiments of this specification propose a drilling early warning method, which includes: calculating a trend value of first drilling data; the first drilling data includes real-time drilling key parameter values ​​at multiple consecutive time points; if the trend value of the first drilling data does not meet a preset confidence condition, obtaining a benchmark value of second drilling data; the second drilling data includes normal historical drilling key parameter values ​​at multiple consecutive time points; and quantifying the degree of anomaly of the first drilling data based on the benchmark value of the second drilling data to generate a corresponding early warning signal for the first drilling data.

[0008] On another front, a drilling early warning device includes: a calculation module for calculating a trend value of first drilling data; the first drilling data includes real-time drilling key parameter values ​​at multiple consecutive time points; an acquisition module for acquiring a baseline value of second drilling data if the trend value of the first drilling data does not meet a preset confidence condition; the second drilling data includes normal historical drilling key parameter values ​​at multiple consecutive time points; and a quantization module for quantifying the degree of anomaly of the first drilling data based on the baseline value of the second drilling data to generate a corresponding early warning signal for the first drilling data.

[0009] In another aspect, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor performs the above-described drilling early warning method.

[0010] As can be seen from the technical solutions provided in the embodiments of this specification above, the drilling early warning method provided in the embodiments of this specification can calculate the trend value of the first drilling data; the first drilling data includes real-time drilling key parameter values ​​at multiple consecutive time points; if the trend value of the first drilling data does not meet the preset confidence condition, a benchmark value of the second drilling data is obtained; the second drilling data includes normal historical drilling key parameter values ​​at multiple consecutive time points; based on the benchmark value of the second drilling data, the degree of anomaly of the first drilling data is quantified to generate a corresponding early warning signal for the first drilling data. Compared with existing methods, when the trend value of the first drilling data is not within the preset trend confidence interval, the degree of anomaly of the first drilling data can be quickly and accurately determined based on the benchmark value of the second drilling data, thereby generating a corresponding early warning signal for the first drilling data in a timely manner, greatly reducing the false alarm / missed alarm rate. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the accompanying drawings used in the description of the embodiments or prior art will be briefly introduced below.

[0012] Figure 1 This is a flowchart of a drilling early warning method provided in the embodiments of this specification;

[0013] Figure 2 This is a schematic diagram illustrating the overall working principle of a drilling early warning method provided in the embodiments of this specification;

[0014] Figure 3 This is a schematic diagram illustrating a specific example of early warning based on a second drilling data benchmark value provided in the embodiments of this specification;

[0015] Figure 4 This is a schematic diagram of the outlet flow warning model constructed based on the proposed drilling early warning method, as provided in the embodiments of this specification.

[0016] Figure 5 This is a schematic diagram of the total well volume early warning model constructed based on the proposed drilling early warning method, provided in the embodiments of this specification.

[0017] Figure 6 This is a schematic diagram of the all-hydrocarbon early warning model constructed based on the proposed drilling early warning method, as provided in the embodiments of this specification;

[0018] Figure 7 This is a schematic diagram of the structural composition of a drilling early warning device provided in the embodiments of this specification;

[0019] Figure 8 This is a schematic diagram of the structural composition of the computer device provided in the embodiments of this specification. Detailed Implementation

[0020] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0021] Figure 1 This is a flowchart of a drilling early warning method provided as an embodiment of this specification. Figure 2 This is a schematic diagram illustrating the overall working principle of a drilling early warning method provided in the embodiments of this specification. In specific implementation, the method includes the following steps:

[0022] S101: Calculate the trend value of the first drilling data; the first drilling data includes real-time drilling key parameter values ​​at multiple consecutive time points.

[0023] In some embodiments, real-time drilling data can be collected during the drilling process.

[0024] By collecting real-time drilling data during the drilling process, a data foundation is laid for obtaining the first drilling data, namely the key real-time drilling parameter values, in the real-time drilling data.

[0025] During real-time drilling, sensors can collect relevant real-time drilling data. Each piece of real-time drilling data includes multiple fields: a timestamp field, an electronic fluid level data field, a logging tool data field, and a sensor data field. The timestamp field records the time point when the first drilling data was collected. It is used for time-series analysis, helping to determine the order of data acquisition and aligning timelines from different data sources. The timestamp field can use the ISO 8601 standard (e.g., YYYY-MM-DDTHH:MM:SS) or other suitable time representation formats, which will not be elaborated further here. The electronic fluid level data field records electronic measurement data of the fluid level height or changes during drilling. It reflects the dynamic changes in the fluid level, helping to monitor well pressure, fluid flow, etc. The electronic fluid level data field can include indicators such as fluid level height, fluid level fluctuation amplitude, and fluid level change trend. The logging tool data field contains various parameter data collected by the logging tool. The integrated logging tool data fields provide comprehensive monitoring information during the drilling process, including geological parameters, drilling parameters, and mud parameters. These fields can include mud outlet flow rate, hydrocarbon value, and gas composition information, as well as data such as drill pressure, rotational speed, torque, mud density, mud viscosity, and cuttings description. Sensor data fields are data collected by various sensors installed on the drilling rig. These fields provide real-time monitoring of various physical and chemical parameters during the drilling process. They can include measurements from multiple sensors, such as temperature, pressure, vibration, noise, and fluid flow rate.

[0026] In some embodiments, key parameters can be determined based on the real-time drilling data; and first drilling data can be determined based on the key parameters.

[0027] By determining the values ​​of key real-time drilling parameters in real-time drilling data, a data foundation is laid for trend analysis of these key parameters.

[0028] Real-time drilling data contains various drilling-related data. This data can be filtered to obtain parameters highly relevant to drilling accidents, i.e., key drilling parameters. Key parameters can be one or more of the following: mud inlet flow rate, mud outlet flow rate, hydrocarbon value, standpipe pressure (SPP), total pit volume (PITTOT), weight on bit (WOB), bit depth (BITDEP), and gas logging anomalies (TGAS). Mud inlet flow rate refers to the amount of mud entering the drilling system. It is crucial for maintaining wellbore pressure balance, cooling the drill bit, and carrying cuttings. Insufficient flow rate may lead to insufficient wellbore pressure, while excessive flow rate may increase system load. Mud outlet flow rate refers to the amount of mud returning to the surface from the bottom of the well. It is a key parameter for blow detection. Monitoring mud outlet flow rate helps determine if there are mud leaks or wellbore collapses. Furthermore, comparing inlet and outlet flow rates allows for the assessment of mud system efficiency and fluid balance. Hydrocarbon value indicates the content or concentration of hydrocarbon gases in drilling mud during real-time drilling. Monitoring hydrocarbon value is crucial for determining whether a formation contains oil and gas and for preventing safety accidents such as blowouts. Standby pressure indicates the pressure of drilling fluid in the standby tub. Standby pressure is an important basis for judging the bottom hole pressure status and helps prevent safety accidents such as blowouts. Total mud volume indicates the total volume of mud in the mud pit. Monitoring the total mud volume helps to detect mud leaks or wellbore collapses that cause changes in mud volume. Drilling pressure (DP) indicates the vertical force applied to the drill bit during real-time drilling. DP is a key factor affecting drilling speed and drill bit life. Reasonable DP can improve drilling efficiency and reduce drill bit wear. Drill bit depth indicates the position of the drill bit downhole during real-time drilling. Accurately recording drill bit depth helps in planning the drilling path, assessing drilling progress, and preventing downhole accidents. Gas logging anomalies indicate the content of abnormal gases detected in the drilling mud by gas detection instruments. Gas logging anomalies are an important means of determining the gas content of the formation and preventing safety accidents such as blowouts and fires. Timely detection and handling of gas logging anomalies helps ensure the safe conduct of drilling operations. After determining the key drilling parameters, the first drilling data, i.e., the real-time key drilling parameter values, can be determined based on real-time drilling data.

[0029] In some embodiments, the first drilling data can be smoothed based on Kalman filtering.

[0030] By using Kalman filtering to smooth the first drilling data, the statistical characteristics of the first drilling data measurement and acquisition process can be effectively modeled. Then, based on these statistical characteristics, noise in the first drilling data can be eliminated, reducing the impact of noise on the first drilling data.

[0031] Kalman filtering, based on linear dynamic equations and the Gaussian distribution assumption, recursively smooths initial drilling data through two steps: prediction and updating. Kalman filtering predicts the current real-time key drilling parameters (CTMPs) based on the previous time point's CTMP values. Specifically, it performs time-series prediction modeling on relevant key parameters, including the dynamic relationship between real-time CTMP values ​​and time points. Based on the established time-series prediction model, it predicts the current CTMP values ​​using the previous time point's CTMP values. The optimal CTMP values ​​can then be calculated based on the error between observed and predicted values, as well as their respective covariance matrices. By continuously predicting and fusing CTMP values, Kalman filtering can update state estimates in real time and smooth the observed data. This smoothing effectively reduces the impact of noise and interference on the observed data, improving data accuracy and reliability.

[0032] In some embodiments, M-level differential operations can be performed on the real-time drilling key parameter values ​​at the multiple consecutive time points to obtain the differential operation result of the first drilling data; the order of each differential operation in the M-level differential operation is different; the order is a preset positive integer less than the number of consecutive time points in the first drilling data; the trend value of the first drilling data can be calculated based on the preset M-level differential weight coefficient, the preset nonlinear amplification factor and the M-level differential operation result of the first drilling data.

[0033] Based on the real-time drilling key parameter values ​​at the multiple consecutive time points, a formula can be used. Calculate the trend value of the first drilling data; where S is the trend value of the first drilling data; K is the difference order, used to determine the time span for calculating the trend value; N is the number of consecutive time points included in the first drilling data; ω k X is a differential weighting coefficient used to adjust the influence of different time spans k on the trend value of the first drilling data; i+k X represents the real-time drilling key parameter value corresponding to the (i+k)th time point in the first drilling data; i The value of the key drilling parameter corresponding to the i-th time point in the first drilling data; Used to amplify X i+k and X i The differences between them.

[0034] By introducing M-level differential operations to calculate the trend value of the first drilling data, the limitations of a single time window are avoided, enhancing the comprehensiveness of trend analysis and improving the scientific validity and rationality of the trend value. Furthermore, the introduction of M-level differential weighting coefficients and nonlinear exponents in the calculation of the first drilling data trend value reduces the impact of noise in the first drilling data on the trend value, improving the adaptability and robustness of the calculation results in complex environments.

[0035] M-level differential operations can be performed on real-time drilling key parameter values ​​at multiple consecutive time points to obtain the differential operation result of the first drilling data. Based on preset M-level differential weighting coefficients, preset nonlinear amplification factors, and the M-level differential operation result of the first drilling data, the trend value of the first drilling data can be calculated. For example, for real-time drilling key parameter values ​​at multiple consecutive time points, the trend value of the first drilling data can be calculated using the following formula:

[0036]

[0037] In the formula, S is the trend value of the first drilling data; K is the difference order, used to determine the time span for calculating the trend value; N is the number of consecutive time points included in the first drilling data; ω k X is a differential weighting coefficient used to adjust the influence of different time spans k on the trend value of the first drilling data; i+k X represents the real-time drilling key parameter value corresponding to the (i+k)th time point in the first drilling data; i The value of the key drilling parameter corresponding to the i-th time point in the first drilling data; It is a preset nonlinear amplification factor used to amplify X. i+k and X i The difference between them. A difference weighting coefficient ω is introduced into the formula. k Adjusting the influence of different time spans k on the trend value of the first drilling data reduces the impact of noise in the first drilling data on the trend value calculation results. A nonlinear exponential δ is introduced into the formula to amplify X. i+k and X i The differences between these methods help to further reduce the impact of noise in the first drilling data on the trend value calculation results and improve the sensitivity to the magnitude of changes in the first drilling data. Introducing multiple time windows into the formula can avoid the limitation of using a single time window to calculate the trend value, which can only obtain a local specific trend of change in the first drilling data. Specifically, introducing multiple time windows to calculate the trend value of the first drilling data is more comprehensive, and can capture multiple hidden trends in the first drilling data, thus obtaining a more scientific and reasonable trend value.

[0038] Real-time drilling key parameters can be one or more of the following: mud inlet flow rate, mud outlet flow rate, hydrocarbon value, riser pressure, total tank volume, drilling pressure, drill bit depth, and gas logging anomalies. Specifically, for each key parameter, corresponding differential weights and nonlinear exponents can be set, and then the corresponding trend value can be calculated for each key parameter separately.

[0039] S102: If the trend value of the first drilling data does not meet the preset confidence condition, obtain the benchmark value of the second drilling data; the second drilling data includes normal historical drilling key parameter values ​​at multiple consecutive time points.

[0040] In some embodiments, normal historical drilling data can be obtained; based on the normal historical drilling data and key drilling-related parameters, second drilling data can be determined.

[0041] The second drilling data was determined by using normal historical drilling data and key drilling-related parameters, which laid the data foundation for obtaining the benchmark values ​​of key parameters in normal historical drilling.

[0042] Normal historical drilling data can be obtained from the drilling database. Each normal historical drilling data entry includes multiple fields, namely, timestamp, electronic fluid level, integrated logging tool data, and sensor data, which will not be elaborated here. Key parameters can be one or more of the following: mud inlet flow rate (FLOWIN), mud outlet flow rate (FLOWOUT), hydrocarbon value (HYDROCARBON VALUE), standpipe pressure (SPP), total pit volume (PITTOT), bit weight (WOB), bit depth (BITDEP), and gas logging anomaly (TGAS). Based on the drilling-related key parameters, second drilling data, namely the normal historical drilling key parameter values, can be determined based on the historical drilling data.

[0043] In some embodiments, the corrected variance of the first drilling data can be calculated; the trend confidence value of the first drilling data can be determined based on the trend value and the corrected variance value of the first drilling data; if the trend confidence value of the first drilling data is not within a preset trend confidence interval and the slope value of the first drilling data is greater than or equal to a preset slope threshold, the baseline value of the second drilling data can be obtained.

[0044] By obtaining the baseline value of the second drilling data when the trend confidence value of the first drilling data is not within the preset trend confidence interval and the slope value of the first drilling data is greater than or equal to the preset slope threshold, the normal trend change can be characterized based on the baseline value of the second drilling data. This lays the data foundation for judging whether the trend change of the first drilling data belongs to the normal trend change based on the trend confidence value and slope value of the first drilling data.

[0045] All stalemate groups related to the first drilling data can be identified. Then, stalemate correction can be used to eliminate the influence of these stalemate groups and calculate the corrected variance of the first drilling data. Based on the corrected variance of the first drilling data, the trend confidence value (Z) corresponding to the first drilling data can be calculated according to different trend values. A trend confidence interval can be pre-defined. For example, the trend confidence interval corresponding to the trend confidence value (Z) can be set to the absolute value of Z being less than 1.645. A trend confidence level (trend significance level) α = 0.1 can be set. If |Z| ≥ Z1 - α / 2, then at the confidence level α (significance level), the first drilling data exhibits a significant upward or downward trend. A positive Z value indicates an upward trend, and a negative Z value indicates a downward trend. Furthermore, according to the standard normal distribution table, when α = 0.1, Z1 - α / 2 = Z0.95, Z0.95 = 1.645. At this time, the trend confidence value Z corresponding to the first drilling data is not within the preset trend confidence interval, and Z > 0, meaning that the first drilling data has an upward trend. When the trend confidence value of the first drilling data is not within the preset trend confidence interval and the slope value of the first drilling data is greater than or equal to the preset slope threshold, the baseline value of the second drilling data can be obtained.

[0046] In some embodiments, a benchmark value for the second drilling data can be calculated based on a median filtering algorithm.

[0047] By using a median filtering algorithm to calculate the baseline value of the second drilling data, the median value of the normal historical drilling key parameter values ​​at multiple time points can be sorted and taken. Furthermore, the interference of abnormal / noise points in the normal historical drilling key parameter values ​​at multiple time points on the baseline data can be eliminated, thereby obtaining a more reasonable baseline value that can better represent the normal drilling trend.

[0048] Median filtering can be used to sort the second drilling data within a fixed-size sliding window and select the median to replace the original data points, thereby smoothing the second drilling data and filtering out random noise. Specifically, for a set of second drilling data {a1, a2, a3, a4, ..., an} containing normal historical drilling key parameter values ​​at multiple time points, calculate ΔA = (a1 + a2 + ... + an) / n (ΔA: represents the average of n numbers, n: represents the number of numbers in the set, ai (1 <= i <= n): represents each element in the set). If a1 to an are not within the range of ΔA ± f, then that value is not taken and a new array {b1, b2, b3, b4, ..., bn} is formed. Calculate ΔB = (b1 + b2 + ... + bn) / n (ΔB: represents the average of n numbers, n: represents the number of numbers in this set, bi (1 <= i <= n): represents each element in this set of numbers), and so on to form ΔC, which can be the benchmark value P of the second drilling data after the median filtering algorithm.

[0049] For example, see Figure 3 As shown, the normal historical key parameter values ​​from 10-20 minutes prior to the current time can be taken as the second drilling data, based on the timeline and following the first-in-first-out principle. The second drilling data is processed using a median filtering algorithm to generate a baseline value P. During the generation of the regression baseline value P, multi-level processing is performed on the second drilling data to ensure the accuracy and stability of the baseline value. Outlier removal technology is used to eliminate noisy data that does not conform to the set threshold or trend, reducing interference with the baseline value calculation. After median filtering, a smoothing method is further used to reduce the impact of high-frequency noise, thereby improving the stability of the baseline value. This baseline value is used to represent the historical data trend over a period of time, serving as a benchmark for comparison with the first drilling data. For example, if the first drilling data continuously deviates from the regression baseline value (greater than or less than 0) for 30-60 seconds, a blowout or lost-hole warning is triggered. The values ​​greater than or less than 0 indicate whether the difference between the first drilling data and the baseline value P exceeds a preset deviation threshold. When the difference exceeds the positive threshold, it indicates a significant increase in the first drilling data, which may trigger a overflow warning. When the difference is below the negative threshold, an abnormal decrease may occur, prompting a leak warning. A deviation must persist for 30-60 seconds to be considered a significant anomaly and trigger a warning. The baseline value is updated every 2-5 minutes; if the same warning occurs within 5 minutes, the system will not issue a repeat warning.

[0050] In some embodiments, multiple stalemate groups can be identified in the first drilling data; one stalemate group includes multiple identical drilling data; each drilling data in one stalemate group uniquely corresponds to a real-time drilling key parameter value at a point in time in the first drilling data; local variance values ​​of the multiple stalemate groups in the first drilling data can be calculated; and a corrected variance value of the first drilling data can be calculated based on the overall variance value of the first drilling data and the local variance values ​​of the multiple stalemate groups in the first drilling data.

[0051] By calculating the local variance values ​​of multiple stale groups in the first drilling data and introducing stale correction, the influence of the same real-time drilling key parameter values ​​at different time points in the first drilling data on the overall variance is eliminated. This helps to obtain a more reasonable trend confidence value of the first drilling data based on the corrected variance value and trend value.

[0052] Each stalemate group contains multiple data points from different time points but with the same data value. Stalemate groups can be identified among the real-time drilling key parameter values ​​at multiple time points in the first drilling data. Each stalemate group includes multiple identical drilling parameter values, and each drilling parameter value in a stalemate group uniquely corresponds to a real-time drilling key parameter value at a specific time point in the first drilling data. After identifying all stalemate groups related to the first drilling data, the local variance values ​​of multiple stalemate groups in the first drilling data can be calculated. Based on the overall variance value of the first drilling data and the local variance values ​​of the multiple stalemate groups in the first drilling data, the corrected variance value of the first drilling data can be calculated. For example, the corrected variance value of the first drilling data can be calculated using the following formula:

[0053]

[0054] In the formula, V is the corrected variance of the first drilling data; Q is the number of average groups included in the first drilling data; N is the number of consecutive time points included in the first drilling data; t q Let q be the number of drilling data points included in the q-th tie group. Introducing a tie correction into the above formula can eliminate the influence of identical real-time drilling key parameter values ​​at different time points in the first drilling data on the overall variance. This allows for a more reasonable trend confidence value for the first drilling data to be obtained based on the corrected variance and trend values.

[0055] In some embodiments, the trend value of the first drilling data can be standardized and normalized; based on the standardized and normalized trend value and the corrected variance of the first drilling data, a preset test statistical model can be used to determine the trend confidence value of the first drilling data.

[0056] Standardizing the trend value based on different values ​​of the first drilling data and performing normalization processing, and using a preset test statistical model to determine the trend confidence value of the first drilling data, helps to more comprehensively and reasonably characterize the changing trend of the first drilling data under different values.

[0057] The trend value of the first drilling data can be standardized and normalized. A pre-defined statistical model can then be used to calculate the trend confidence value of the first drilling data by comparing the standardized trend value with the corrected variance. For example, based on the trend value and corrected variance of the first drilling data, the trend confidence value can be determined using the following formula:

[0058]

[0059] In the formula, Z represents the trend confidence value of the first drilling data; S represents the trend value of the first drilling data; and V represents the corrected variance value of the first drilling data. When the trend value of the first drilling data is greater than 0, the trend confidence value can be expressed as the ratio of the trend value minus one to the square root of the corrected variance value. When the trend value of the first drilling data is equal to 0, the trend confidence value can be represented as 0. When the trend value of the first drilling data is less than 0, the trend confidence value can be expressed as the ratio of the trend value plus one to the square root of the corrected variance value. Considering different values ​​of the trend value of the first drilling data and using different calculation methods to calculate the trend confidence value allows for a more comprehensive and reasonable characterization of the changing trend of the first drilling data under different value conditions.

[0060] In some embodiments, based on the first drilling data, trend curves of real-time drilling key parameter values ​​at multiple time points can be constructed; based on the trend curves of real-time drilling key parameter values ​​at the multiple time points, the slope value of the first drilling data can be calculated.

[0061] By constructing trend curves for real-time drilling key parameter values ​​at multiple time points, the slope value of the first drilling data can be calculated quickly and accurately. On the one hand, the calculated slope value can effectively characterize the direction and rate of change of the first drilling data. On the other hand, the calculated slope value can be used together with the trend value to identify potential overflow or well control risks in the first drilling data, helping to improve the accuracy and real-time performance of drilling early warning.

[0062] Based on the real-time drilling key parameter values ​​at multiple time points included in the first drilling data, trend curves for these values ​​can be plotted. These trend curves provide a visual representation of how the real-time drilling key parameter values ​​change over time. The slope of the trend curve represents the rate of change of the parameter value over time. By calculating the slope of the trend curve, the direction and rate of change of the first drilling data in different time periods can be obtained.

[0063] S103: Based on the baseline value of the second drilling data, quantify the degree of anomaly of the first drilling data to generate a corresponding early warning signal for the first drilling data.

[0064] In some embodiments, based on the baseline value of the second drilling data, the anomaly weight of the first drilling data can be calculated; based on the anomaly weight of the first drilling data, the information entropy of the first drilling data can be calculated; based on the information entropy of the first drilling data, the degree of anomaly of the first drilling data can be quantified.

[0065] Information entropy-based methods can comprehensively quantify the overall deviation of the distribution of first-well drilling data, rather than being limited to the detection of anomalies in individual data points. This global perspective can comprehensively evaluate the distribution pattern of first-well drilling data, avoiding misjudgments caused by isolated data anomalies. Information entropy-based methods can also sensitively reflect the uncertainty and volatility of the distribution of first-well drilling data. The introduction of information entropy dilutes the weight of short-term anomalies, avoiding false alarms. For persistent deviations or significant anomalies, information entropy can amplify their impact, improving the reliability of early warnings.

[0066] After determining the baseline values ​​for the second drilling data, the anomaly weights of the real-time drilling key parameter values ​​at multiple time points included in the first drilling data can be calculated. Then, the degree of anomaly in the first drilling data can be quantified based on its information entropy. For example, based on the anomaly weights of the real-time drilling key parameter values ​​at multiple time points included in the first drilling data, the degree of anomaly in the first drilling data can be quantified using the following formula:

[0067]

[0068] In the formula, A represents the anomaly level (information entropy) of the first drilling data; N represents the number of consecutive time points included in the first drilling data; p i The anomaly weight is the real-time drilling key parameter value at the i-th time point in the first drilling data. Introducing information entropy into the above formula to quantify the degree of anomaly in the first drilling data allows for a comprehensive assessment of the data's distribution pattern, avoiding misjudgments caused by isolated data anomalies. Especially for persistent deviations or significantly abnormal distributions, information entropy can amplify their impact and improve the reliability of early warnings.

[0069] After determining the degree of anomaly in the first drilling data, a corresponding early warning signal can be generated. Different levels of early warning conditions can be preset: Minor Increase Warning: When the anomaly level of the first drilling data meets the criteria for a minor anomaly, the system issues a low-level warning. Moderate Increase Warning: When the anomaly level of the first drilling data reaches a moderate anomaly level, the system issues a medium-level warning. Rapid Increase Warning: When the anomaly level of the first drilling data reaches a severe anomaly level, the system issues a high-level warning.

[0070] In some embodiments, based on the baseline value of the second drilling data, the Q-order absolute difference result corresponding to the real-time drilling key parameter value at each time point in the first drilling data can be calculated; where Q is a preset positive integer; based on the Q-order absolute difference result corresponding to the real-time drilling key parameter value at each time point in the first drilling data, the cumulative absolute difference result of the first drilling data can be calculated; based on the Q-order absolute difference result corresponding to the real-time drilling key parameter value at each time point in the first drilling data and the cumulative absolute difference result of the first drilling data, the abnormal weight of the real-time drilling key parameter value at each time point in the first drilling data can be calculated.

[0071] By calculating the Q-order absolute difference between real-time drilling key parameter values ​​and a second drilling data benchmark, the deviation of real-time drilling key parameter values ​​is measured, thus quantifying the anomaly levels of real-time drilling key parameter values ​​at multiple time points. A relative weight is calculated using each value representing the deviation of each real-time drilling key parameter value; larger deviations receive greater weights, which helps to amplify outlier data.

[0072] Based on the baseline value of the second drilling data, the Q-order absolute difference result corresponding to the real-time drilling key parameter value at each time point in the first drilling data can be calculated. Based on the Q-order absolute difference result corresponding to the real-time drilling key parameter value at all time points in the first drilling data, the cumulative absolute difference result of the first drilling data can be calculated. Based on the Q-order absolute difference result corresponding to the real-time drilling key parameter value at each time point in the first drilling data and the cumulative absolute difference result of the first drilling data, the anomaly weights of the real-time drilling key parameter values ​​at multiple time points in the first drilling data can be calculated. For example, after determining the baseline value of the second drilling data, the anomaly weights of the real-time drilling key parameter values ​​at multiple time points in the first drilling data can be calculated using the following formula:

[0073]

[0074] In the formula, p i X represents the anomaly weight of the real-time drilling key parameter value at the i-th time point in the first drilling data; i P represents the real-time drilling key parameter value corresponding to the i-th time point in the first drilling data; i X is the reference value of the second drilling data corresponding to the real-time drilling key parameter value at the i-th time point in the first drilling data; N is the number of consecutive time points included in the first drilling data; X j P represents the real-time drilling key parameter value corresponding to the j-th time point in the first drilling data; jThe reference value of the second drilling data corresponding to the real-time drilling key parameter value at the j-th time point in the first drilling data; Q is the preset order; |X i -P i | Q This is the Q-order absolute difference result corresponding to the real-time drilling key parameter value at the i-th time point in the first drilling data; This represents the cumulative absolute difference result of the first drilling data. The absolute difference between the real-time drilling key parameter values ​​and the corresponding second drilling data benchmark values ​​is calculated to measure the degree of deviation of the real-time drilling key parameter values, enabling the quantification of the anomaly level of real-time drilling key parameter values ​​at multiple time points. Using each value to represent the degree of deviation of each real-time drilling key parameter value, a relative weight is calculated, which can amplify abnormal real-time drilling key parameter values.

[0075] Reference Figure 4 The diagram shows an outlet flow warning model constructed based on the proposed drilling early warning method. In actual operation, due to differences in measuring equipment or processes, the outlet flow measured by the parameters may be a relative or absolute value. Only when it is a relative value can a trend warning model be used to warn of the outlet flow trend.

[0076] Reference Figure 5 The diagram shows a total well volume early warning model constructed based on the proposed drilling early warning method. The total well volume early warning process uses two models: one for each stage of drilling, from pump start-up to stabilization, from stabilization to stabilization, and from stabilization to pump shutdown; and the other for each stage after pump shutdown, which is then monitored using its own model.

[0077] Reference Figure 6 The model shown is a total hydrocarbon early warning system constructed based on the proposed drilling early warning method. During the total hydrocarbon early warning process, two algorithms are used for early warning: one is based on trend changes, and the other is based on inlet / outlet interpolation combined with total hydrocarbon change analysis.

[0078] The following is a specific embodiment of this specification:

[0079] As shown in Table 1, the proposed drilling early warning method was used for overflow warnings in wells A, B, C, D, and E, with well A having accumulated over 285 days of operation. The proposed drilling early warning method possesses multi-parameter comprehensive analysis capabilities, accurately assessing complex downhole conditions and effectively identifying multiple minor anomalies and issuing warnings. This ensures the timeliness and accuracy of overflow identification, improving the safety and efficiency of drilling operations. By comparing the false alarm rates of overflow identification warnings, the false alarm rate using the proposed drilling early warning method was reduced by an average of 83%, significantly improving well control safety during drilling operations.

[0080] Table 1

[0081]

[0082] Based on the above-described drilling early warning method, this specification also provides embodiments of drilling early warning devices. For example... Figure 7 As shown, the drilling early warning device 700 may specifically include the following modules:

[0083] The calculation module 701 can be used to calculate the trend value of the first drilling data; the first drilling data includes real-time drilling key parameter values ​​at multiple consecutive time points.

[0084] The acquisition module 702 can be used to acquire the baseline value of the second drilling data if the trend value of the first drilling data does not meet the preset confidence condition; the second drilling data includes normal historical drilling key parameter values ​​at multiple consecutive time points.

[0085] The quantization module 703 can be used to quantify the degree of anomaly of the first drilling data based on the baseline value of the second drilling data in order to generate a corresponding early warning signal for the first drilling data.

[0086] In some embodiments, the calculation module 701 can be specifically used to perform M-level differential operations on the real-time drilling key parameter values ​​of the multiple consecutive time points to obtain the differential operation result of the first drilling data; the order of each differential operation in the M-level differential operation is different; the order is a preset positive integer less than the number of consecutive time points in the first drilling data; the trend value of the first drilling data is calculated according to the preset M-level differential weight coefficient, the preset nonlinear amplification factor and the M-level differential operation result of the first drilling data.

[0087] In some embodiments, the acquisition module 702 may be specifically used to calculate the corrected variance value of the first drilling data; determine the trend confidence value of the first drilling data based on the trend value and the corrected variance value of the first drilling data; and if the trend confidence value of the first drilling data is not within a preset trend confidence interval and the slope value of the first drilling data is greater than or equal to a preset slope threshold, obtain the benchmark value of the second drilling data.

[0088] In some embodiments, the acquisition module 702 may further be used to determine multiple stalemate groups in the first drilling data; a stalemate group includes multiple identical drilling data; each drilling data in a stalemate group uniquely corresponds to a real-time drilling key parameter value at a point in time in the first drilling data; calculate the local variance values ​​of the multiple stalemate groups in the first drilling data; and calculate the corrected variance value of the first drilling data based on the overall variance value of the first drilling data and the local variance values ​​of the multiple stalemate groups in the first drilling data.

[0089] In some embodiments, the acquisition module 702 may further be used to perform standardized normalization processing on the trend value of the first drilling data; and to determine the trend confidence value of the first drilling data using a preset test statistical model based on the standardized normalized trend value and the corrected variance value of the first drilling data.

[0090] In some embodiments, the acquisition module 702 may further be used to construct trend curves of real-time drilling key parameter values ​​at multiple time points based on the first drilling data; and to calculate the slope value of the first drilling data based on the trend curves of the real-time drilling key parameter values ​​at multiple time points.

[0091] In some embodiments, the quantization module 703 can be specifically used to calculate the anomaly weight of the first drilling data based on the baseline value of the second drilling data; calculate the information entropy of the first drilling data based on the anomaly weight of the first drilling data; and quantify the degree of anomaly of the first drilling data based on the information entropy of the first drilling data.

[0092] In some embodiments, the quantization module 703 can further be used to calculate the Q-order absolute difference result corresponding to the real-time drilling key parameter value at each time point in the first drilling data based on the baseline value of the second drilling data; where Q is a preset positive integer; calculate the cumulative absolute difference result of the first drilling data based on the Q-order absolute difference result corresponding to the real-time drilling key parameter value at each time point in the first drilling data; and calculate the abnormal weight of the real-time drilling key parameter value at each time point in the first drilling data based on the Q-order absolute difference result corresponding to the real-time drilling key parameter value at each time point in the first drilling data and the cumulative absolute difference result of the first drilling data.

[0093] It should be noted that the units, devices, or modules described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. For ease of description, the above devices are described by dividing them into various modules according to their functions. Of course, in implementing this specification, the functions of each module can be implemented in one or more software and / or hardware, or the module that implements the same function can be implemented by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection between the devices or units shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0094] As can be seen from the above, the drilling early warning device provided in the embodiments of this specification can calculate the trend value of the first drilling data; the first drilling data includes real-time drilling key parameter values ​​at multiple consecutive time points; if the trend value of the first drilling data does not meet the preset confidence condition, the benchmark value of the second drilling data is obtained; the second drilling data includes normal historical drilling key parameter values ​​at multiple consecutive time points; based on the benchmark value of the second drilling data, the degree of anomaly of the first drilling data is quantified to generate a corresponding early warning signal for the first drilling data. Compared with existing methods, when the trend value of the first drilling data is not within the preset trend confidence interval, the degree of anomaly of the first drilling data can be quickly and accurately determined based on the benchmark value of the second drilling data, thereby generating a corresponding early warning signal for the first drilling data in a timely manner, greatly reducing the false alarm / missed alarm rate.

[0095] This specification also provides a computer device for a drilling early warning method, including a processor and a memory for storing processor-executable instructions. Specifically, the processor can execute the following steps according to the instructions: calculating a trend value of first drilling data; the first drilling data includes real-time drilling key parameter values ​​at multiple consecutive time points; if the trend value of the first drilling data does not meet a preset confidence condition, obtaining a baseline value of second drilling data; the second drilling data includes normal historical drilling key parameter values ​​at multiple consecutive time points; and quantifying the degree of anomaly of the first drilling data based on the baseline value of the second drilling data to generate a corresponding early warning signal for the first drilling data.

[0096] To execute the above instructions more accurately, please refer to... Figure 8 As shown in the embodiments of this specification, another specific computer device is also provided, wherein the computer device includes a network communication port 801, a processor 802, and a memory 803. The above structures are connected by internal cables so that the various structures can perform specific data interaction.

[0097] The processor 802 can specifically be used to calculate the trend value of the first drilling data; the first drilling data includes real-time drilling key parameter values ​​at multiple consecutive time points; if the trend value of the first drilling data does not meet the preset confidence condition, the benchmark value of the second drilling data is obtained; the second drilling data includes normal historical drilling key parameter values ​​at multiple consecutive time points; based on the benchmark value of the second drilling data, the degree of abnormality of the first drilling data is quantified to generate a corresponding early warning signal for the first drilling data.

[0098] The memory 803 can be used to store the corresponding instruction program.

[0099] In this embodiment, the network communication port 801 can be a virtual port bound to different communication protocols, thereby enabling the sending or receiving of different data. For example, the network communication port can be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for email data communication. Furthermore, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM or CDMA; it can also be a Wi-Fi chip; or it can be a Bluetooth chip.

[0100] In this embodiment, the processor 802 can be implemented in any suitable manner. For example, the processor can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers, etc. This specification is not limiting.

[0101] In this embodiment, the memory 803 includes volatile memory and non-volatile memory. The memory 803 can include multiple layers. In digital systems, anything that can store binary data can be a memory; in integrated circuits, a circuit with storage function but no physical form is also called a memory, such as RAM, FIFO, etc.; in a system, a storage device with a physical form is also called a memory, such as a memory stick, TF card, etc.

[0102] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0103] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0104] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0105] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0106] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A drilling early warning method, characterized in that, include: Calculate the trend value of the first drilling data; The first drilling data includes real-time drilling key parameter values ​​at multiple consecutive time points; If the trend value of the first drilling data does not meet the preset confidence condition, the baseline value of the second drilling data is obtained; the second drilling data includes normal historical drilling key parameter values ​​at multiple consecutive time points; Based on the baseline value of the second drilling data, the degree of anomaly of the first drilling data is quantified to generate a corresponding early warning signal for the first drilling data.

2. The method according to claim 1, characterized in that, The calculation of the trend value of the first drilling data includes: M-level differential operations are performed on the real-time drilling key parameter values ​​at the multiple consecutive time points to obtain the differential operation results of the first drilling data; the order of each differential operation in the M-level differential operation is different; the order is a preset positive integer less than the number of consecutive time points in the first drilling data; The trend value of the first drilling data is calculated based on the preset M-level differential weighting coefficient, the preset nonlinear amplification factor, and the M-level differential operation result of the first drilling data.

3. The method according to claim 1, characterized in that, If the trend value of the first drilling data does not meet the preset confidence condition, the step of obtaining the benchmark value of the second drilling data includes: Calculate the corrected variance of the first drilling data; Based on the trend value and corrected variance value of the first drilling data, determine the trend confidence value of the first drilling data; If the trend confidence value of the first drilling data is not within the preset trend confidence interval and the slope value of the first drilling data is greater than or equal to the preset slope threshold, the baseline value of the second drilling data is obtained.

4. The method according to claim 3, characterized in that, The calculation of the corrected variance value of the first drilling data includes: Identify multiple tie groups in the first drilling data; each tie group includes multiple identical drilling data; each drilling data in a tie group uniquely corresponds to a real-time drilling key parameter value at a point in time in the first drilling data; Calculate the local variance values ​​of multiple average groups in the first drilling data; The corrected variance of the first drilling data is calculated based on the overall variance of the first drilling data and the local variances of multiple average groups in the first drilling data.

5. The method according to claim 3, characterized in that, The step of determining the trend confidence value of the first drilling data based on the trend value and the corrected variance value of the first drilling data includes: The trend values ​​of the first drilling data are standardized and normalized. Based on the trend value after standardized normalization and the corrected variance value of the first drilling data, a preset test statistical model is used to determine the trend confidence value of the first drilling data.

6. The method according to claim 3, characterized in that, The method further includes: Based on the first drilling data, trend curves of real-time drilling key parameter values ​​at multiple time points are constructed. The slope value of the first drilling data is calculated based on the trend curves of the real-time drilling key parameter values ​​at the multiple time points.

7. The method according to claim 1, characterized in that, The step of quantifying the degree of anomaly in the first drilling data based on the baseline value of the second drilling data to generate a corresponding early warning signal for the first drilling data includes: Calculate the anomaly weight of the first drilling data based on the baseline value of the second drilling data; Calculate the information entropy of the first drilling data based on the anomaly weights of the first drilling data; The degree of anomaly in the first drilling data is quantified based on the information entropy of the first drilling data.

8. The method according to claim 7, characterized in that, The step of calculating the anomaly weight of the first drilling data based on the baseline value of the second drilling data includes: Based on the baseline value of the second drilling data, calculate the Q-order absolute difference result corresponding to the real-time drilling key parameter value at each time point in the first drilling data; where Q is a preset positive integer. Based on the Q-order absolute difference results corresponding to the real-time drilling key parameter values ​​at each time point in the first drilling data, the cumulative absolute difference results of the first drilling data are calculated. Based on the Q-order absolute difference results corresponding to the real-time drilling key parameter values ​​at each time point in the first drilling data and the cumulative absolute difference results of the first drilling data, the abnormal weights of the real-time drilling key parameter values ​​at each time point in the first drilling data are calculated.

9. A drilling early warning device, characterized in that, The device includes: The calculation module is used to calculate the trend value of the first drilling data; the first drilling data includes real-time drilling key parameter values ​​at multiple consecutive time points; The acquisition module is used to acquire the baseline value of the second drilling data if the trend value of the first drilling data does not meet the preset confidence condition; the second drilling data includes normal historical drilling key parameter values ​​at multiple consecutive time points; The quantization module is used to quantify the degree of anomaly of the first drilling data based on the baseline value of the second drilling data in order to generate a corresponding early warning signal for the first drilling data.

10. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1-8.

Images