A method for processing data of multi-element signal sensing while drilling and a pre-warning system

Abstract

The application provides a drilling multi-element signal sensing data processing method and early warning system, and belongs to the technical field of data processing, and specifically comprises the following steps: determining a vibration monitoring signal type of interest of a vibration monitoring signal based on monitoring processing data of the vibration monitoring signal in a drilling signal; determining a monitoring analysis method of the vibration monitoring signal based on the vibration monitoring signal type of interest and an update condition of monitoring analysis data under different vibration monitoring signal types of interest; determining a usable analysis monitoring signal type of the vibration monitoring signal according to the monitoring analysis method; and determining a drilling signal sensing data processing method according to the usable analysis monitoring signal type and a distribution condition of monitoring analysis data under different usable analysis monitoring signal types, so that the accuracy of trajectory control processing is improved.

Description

Technical Field

[0001] This invention belongs to the field of data processing technology, and in particular relates to a method and early warning system for processing multi-signal sensing data while drilling. Background Technology

[0002] The current state of coal mining in China is characterized by "deepening depth, high stress in coal seams, high gas content and low permeability". This leads to problems such as short drilling distance, large workload, limited pressure relief range, low extraction efficiency and uncontrollable drilling trajectory when using traditional ordinary drilling equipment. In particular, problems such as blowouts and top drilling will occur frequently when operating in soft and low permeability coal seams.

[0003] To address the aforementioned technical problems, the invention patent application CN202511727882.6, "Image Processing-Based Rotary Steering Drilling Adaptive Trajectory Control Method," specifically addresses this by receiving raw logging-while-drilling image data frames from the downhole transmission channel, generating a high-resolution predicted geological texture map, simultaneously generating a feature uncertainty distribution map characterizing the pixel probability distribution features in the high-resolution predicted geological texture map, generating a geological illusion entropy, receiving the geological illusion entropy and sensor contamination marker signals, and driving the downhole steering tool according to the activated mode. This avoids drilling trajectory deviation caused by artificially constructed textures. However, the above technical solution has the following drawbacks: In the process of drilling downhole, existing technical solutions neglect to determine the vibration mode monitoring and analysis strategy based on the severity of the zero drift variation in the drilling signal. As a result, it is difficult to ensure the reliability of the assessment and analysis of the aging state of the drill bit and the assessment and analysis of the zero drift when the zero drift is relatively severe, and thus it is difficult to ensure the accuracy of trajectory control.

[0004] Specifically, this application provides a method and early warning system for processing multi-signal sensing data during drilling. Summary of the Invention

[0005] To achieve the objectives of this invention, the following technical solution is adopted: Specifically, this application provides a method for processing multi-element signal sensing data during drilling, which includes: S1 uses the sensing data of the drilling signal to construct the formation category based on the model. According to the trajectory deviation data of different formation categories in the target area, the correction processing strategy is determined. The offset data of the drilling signal is determined using the correction processing strategy. When it is determined that the vibration monitoring signal in the drilling signal needs to be monitored and analyzed based on the offset data, the next step is performed. S2 determines the type of vibration monitoring signal of interest based on the monitoring and processing data of the vibration monitoring signal in the drilling signal, and determines the monitoring and analysis method of the vibration monitoring signal based on the type of vibration monitoring signal of interest and the update status of the monitoring and analysis data under different types of vibration monitoring signal of interest. S3 determines the available analytical monitoring signal type of the vibration monitoring signal according to the monitoring and analysis method, and determines the sensing data processing method of the drilling signal according to the available analytical monitoring signal type and the distribution of monitoring and analysis data in different available analytical monitoring signal types.

[0006] The beneficial effects of this invention are as follows: Based on the types of vibration monitoring signals of interest and the update status of monitoring and analysis data under different types of vibration monitoring signals, the monitoring and analysis methods for vibration monitoring signals are determined. Specifically, based on the number of vibration monitoring signal types of interest and the distribution data of rock with these types of vibration monitoring signals during different time periods, the sufficiency and distribution density of vibration monitoring signal types currently available for drill bit life assessment are determined. Using the sufficiency and distribution density of vibration monitoring signal types currently available for drill bit life assessment, it is determined under which type of vibration monitoring signal of rock should the aging state assessment and analysis be performed. This ensures the reliability of the aging state assessment and analysis while also improving the efficiency of updating and processing the types of vibration monitoring signals of interest.

[0007] Based on the types of available analytical monitoring signals and the distribution of monitoring and analysis data under different types of available analytical monitoring signals, the sensing data processing method for drilling signals is determined. Specifically, based on the distribution data of the time periods used for assessing and analyzing the aging status of the drill bit and the number of assessment and analysis time periods under different types of available analytical monitoring signals, the timeliness and matching degree of the assessment and analysis of the aging status of the brick under the current state are determined. That is, the more time periods used for assessing and analyzing the aging status of the drill bit, and the more available analytical monitoring signal types with a large number of assessment and analysis time periods, the higher the reliability of the assessment and analysis of the aging status. The reliability of the assessment and analysis of the aging status is used to determine the sensing data processing method for drilling signals, namely the update strategy of the temperature calibration method. This also lays the foundation for further improving the efficiency and accuracy of correction processing when the reliability of the assessment and analysis of the aging status is poor.

[0008] Furthermore, the sensing data of the drilling signal is determined based on the monitoring data of different types of drilling monitoring devices.

[0009] Furthermore, the process of constructing stratigraphic categories based on the model includes: Based on the sensing data of the drilling signals, the formation categories are constructed using the output of the drilling lithology identification model.

[0010] Furthermore, the stratigraphic category identification data is determined based on the stratigraphic category identification results in different depth ranges within the target area.

[0011] Furthermore, the trajectory deviation data under the stratigraphic category is determined based on the number of times the positioning trajectory under the stratigraphic category has a deviation.

[0012] Furthermore, the method for determining the correction processing strategy is as follows: S11 determines the number of times the positioning trajectory deviates during drilling under the different formation categories based on the trajectory deviation data under those formation categories; S12 takes the number of times the positioning trajectory deviates during drilling under the formation category as the trajectory deviation number under the formation category; S13 determines the correction processing strategy based on the number of trajectory deviations under different stratigraphic categories in the target area.

[0013] Furthermore, it was determined that monitoring and analysis of vibration monitoring signals from the drilling signals were necessary, specifically including: S21 Based on the offset data, determine the residual between the inclinometer monitoring data and the model prediction data in the drilling signal; S22 takes the correction process where the residual exceeds the instrument’s nominal accuracy as the deviation identification process, and determines whether it is necessary to perform monitoring and analysis of the vibration monitoring signal in the drilling signal based on the deviation identification process data.

[0014] Secondly, the present invention provides an early warning system, employing the aforementioned method for processing multi-element signal sensing data during drilling, specifically including: Data sensing module, early warning module; The data sensing module is responsible for sensing and processing the drilling signal using the sensing data processing method for the drilling signal. The early warning module is responsible for assessing the degree of drill bit aging using drilling signals and neural network models corresponding to different rock types, and for issuing early warnings based on the assessment and analysis results of the degree of drill bit aging.

[0015] Other features and advantages will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0016] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0017] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0018] Figure 1 This is a flowchart of a method for processing multi-signal sensing data during drilling; Figure 2 This is a flowchart illustrating the method for determining the corrective action strategy; Figure 3 This is a flowchart illustrating the method for determining the need for monitoring, analyzing, and processing vibration monitoring signals from drilling signals. Detailed Implementation

[0019] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that the invention will be thorough and complete, and the concept of the exemplary embodiments will be fully conveyed to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and therefore their detailed description will be omitted.

[0020] The terms “a,” “one,” “the,” and “the” are used to indicate the existence of one or more elements / components / etc.; the terms “including” and “having” are used to indicate an open-ended meaning of inclusion and that other elements / components / etc. may exist in addition to the listed elements / components / etc.

[0021] Example 1 To solve the above problems, according to one aspect of the present invention, such as Figure 1 As shown, a method for processing multi-element signal sensing data while drilling is provided, specifically including: S1 uses the sensing data of the drilling signal to construct the formation category based on the model. According to the trajectory deviation data of different formation categories in the target area, the correction processing strategy is determined. The offset data of the drilling signal is determined using the correction processing strategy. When it is determined that the vibration monitoring signal in the drilling signal needs to be monitored and analyzed based on the offset data, the next step is performed. S2 determines the type of vibration monitoring signal of interest based on the monitoring and processing data of the vibration monitoring signal in the drilling signal, and determines the monitoring and analysis method of the vibration monitoring signal based on the type of vibration monitoring signal of interest and the update status of the monitoring and analysis data under different types of vibration monitoring signal of interest. S3 determines the available analytical monitoring signal type of the vibration monitoring signal according to the monitoring and analysis method, and determines the sensing data processing method of the drilling signal according to the available analytical monitoring signal type and the distribution of monitoring and analysis data in different available analytical monitoring signal types.

[0022] Furthermore, the sensing data of the drilling signal is determined based on the monitoring data of different types of drilling monitoring devices.

[0023] Furthermore, the process of constructing stratigraphic categories based on the model includes: Based on the sensing data of the drilling signals, the formation categories are constructed using the output of the drilling lithology identification model.

[0024] Furthermore, the stratigraphic category identification data is determined based on the stratigraphic category identification results in different depth ranges within the target area.

[0025] Furthermore, the trajectory deviation data under the stratigraphic category is determined based on the number of times the positioning trajectory under the stratigraphic category has a deviation.

[0026] Specifically, such as Figure 2 As shown, the method for determining the correction processing strategy is as follows: In this embodiment, the deviation in the drilling process within the target area is determined based on the degree of trajectory deviation under different formation types. The deviation is then used to determine the sensor correction strategy. Specifically, the more formation types with trajectory deviations, or the higher the severity of trajectory deviation under any given formation type, the greater the need to determine the severity of zero drift affecting the current inclinometer through correction processing. This determination of the correction strategy based on the severity of zero drift affects the current inclinometer lays the foundation for further timely and effective monitoring and analysis of vibration data based on the correction processing data, thereby improving the reliability of trajectory control during drilling.

[0027] The sensing data of the drilling signal refers to the formation physical parameters and drilling engineering parameters collected in real time during the drilling process by drilling monitoring devices (such as drilling criterion meters, drilling seismometers, etc.); the formation category refers to the classification and identification of the formations in the target area based on factors such as lithology, geological age, and sedimentary environment; the trajectory deviation data refers to the quantitative record of the deviation between the actual drilling trajectory and the designed trajectory; the deviation correction strategy refers to the specific implementation plan for correcting the influence of zero drift of the criterion meter; and the offset data refers to the deviation between the criterion meter monitoring data and the actual data.

[0028] For example, suppose a drilling project's target area contains multiple formation categories. Let there be a preset threshold for the number of formation categories with deviation, a preset deviation rate threshold, a preset number of deviations threshold, a preset deviation process proportion threshold, and a preset demand threshold. Attitude data is collected during drilling using a logging-while-drilling (LMWD) instrument. Formation characteristics are analyzed using a WMWD lithology identification model to construct a formation category system for the target area. The number of trajectory deviations under each formation category is then counted to determine the corrective action strategy.

[0029] This step establishes a complete technical chain from drilling-while-feeding data to deviation correction strategies. Its significance lies in the timely detection of the zero drift effect of the inclinometer through systematic formation analysis and trajectory deviation assessment, thereby ensuring the accuracy and reliability of drilling trajectory control and laying the foundation for in-depth analysis of subsequent vibration monitoring signals.

[0030] S11 determines the number of times the positioning trajectory deviates during drilling under the different formation categories based on the trajectory deviation data for those formation categories.

[0031] The trajectory deviation data refers to the recorded information of deviations between the actual measured trajectory and the designed trajectory during drilling operations in a specific geological category; the number of times the positioning trajectory has deviations refers to the cumulative number of trajectory deviation events that occur during drilling in a certain geological category.

[0032] Assuming the target area is divided into multiple stratigraphic categories (such as mudstone, sandstone, limestone, etc.), when drilling is carried out in a certain stratigraphic category, record each instance of trajectory deviation and count the total number of times the positioning trajectory deviates under that stratigraphic category.

[0033] This step, by quantifying the frequency of trajectory deviations under each stratigraphic category, is significant in identifying the spatial distribution characteristics of trajectory control problems, thereby distinguishing between high-risk and stable strata, and providing data support for the differentiated formulation of subsequent correction strategies.

[0034] S12, the number of times the positioning trajectory deviates during drilling under the formation category is taken as the trajectory deviation number under the formation category.

[0035] The number of trajectory deviations refers to a quantitative indicator formed by aggregating the number of times the positioning trajectory has deviations, which is used to characterize the stability of trajectory control in a specific stratigraphic category.

[0036] If the number of times a certain mudstone layer deviates from its trajectory during drilling is a certain value, then this value is taken as the number of trajectory deviations for that mudstone layer, and used for subsequent formation deviation analysis and determination of correction strategies.

[0037] This step, by establishing a quantitative definition of the number of trajectory deviations, is significant in that it provides a unified metric for comparing the degree of deviation among different stratigraphic types, thereby making the trajectory control performance between different stratigraphic types comparable and facilitating the identification of stratigraphic types with deviations that require special attention.

[0038] S121 uses the number of trajectory deviations to determine the stratum category in which the positioning trajectory has deviated, and uses it as the deviation stratum category. It then determines whether the number of deviation stratum categories is greater than a preset threshold for the number of deviation stratum categories. If so, it determines that the correction processing strategy is to perform sensor correction processing whenever the deviation rate between the temperature in the sensor and the calibrated temperature is less than a preset deviation rate threshold. If not, it proceeds to step S122.

[0039] The deviation formation category refers to the formation category that has deviated from the trajectory during drilling; the preset deviation formation category number threshold refers to the critical value of the number of formation categories used to determine whether a high-frequency correction strategy is needed; the calibration temperature refers to the standard temperature reference value established when the inclinometer is calibrated before drilling; the deviation rate threshold refers to the critical value of the maximum allowable deviation ratio between the sensor temperature and the calibration temperature.

[0040] Assume the target area has multiple stratigraphic categories, and the number of stratigraphic categories causing trajectory deviation is a certain value. Let a preset threshold for the number of deviation stratigraphic categories be a certain value. If the number of deviation stratigraphic categories exceeds this threshold, it indicates a widespread trajectory control problem. In this case, the correction strategy is determined as follows: as long as the deviation rate between the sensor temperature and the calibration temperature is less than the preset deviation rate threshold, sensor correction processing is performed.

[0041] This step, which uses a threshold to determine the number of deviation formation categories, is significant in that it dynamically adjusts the triggering conditions for correction processing based on the spatial distribution of the trajectory control problem. When there are many deviation formation categories, a more sensitive correction strategy is adopted, thereby promptly eliminating the impact of inclinometer zero drift on trajectory control and improving the reliability of the drilling process.

[0042] S122, using the number of trajectory deviations, determine the number of trajectory deviations in different historical drilling processes under the formation category, and take the historical drilling processes with a trajectory deviation number above a preset number as trajectory deviation drilling processes. Determine whether there is a formation category where the proportion of trajectory deviation processes is greater than a preset deviation process proportion threshold. If so, determine that the correction processing strategy is that as long as the deviation rate between the temperature in the sensor and the calibration temperature is less than a preset deviation rate threshold, the sensor correction processing needs to be performed to determine whether a deviation has occurred. If not, proceed to step S13.

[0043] The historical drilling process refers to the record of multiple independent drilling operations carried out in the same formation category; the trajectory deviation drilling process refers to the drilling operation in the historical drilling process where the number of trajectory deviations reaches or exceeds a preset threshold; the preset deviation process proportion threshold is a critical proportion value used to determine whether the proportion of trajectory deviation drilling processes in a certain formation category is too high.

[0044] Assuming there are multiple historical drilling processes under a certain formation category, the number of trajectory deviations in each historical drilling process is counted. Historical drilling processes with a trajectory deviation count exceeding a preset number are identified as trajectory deviation drilling processes. If there exists a formation category where the proportion of trajectory deviation drilling processes to the total number of historical drilling processes in that formation category is greater than a preset deviation process proportion threshold, then a high-frequency deviation correction strategy is determined to be adopted.

[0045] This step, by analyzing the proportion of trajectory deviations during the drilling process, is significant in that it assesses the occurrence pattern of trajectory control problems from a time perspective, identifies the types of strata where trajectory deviations repeatedly occur, and thus strengthens the correction efforts for these strata in a targeted manner. This lays the foundation for improving the reliability of identifying the impact of zero drift and for determining further strategies for monitoring and analyzing vibration signals.

[0046] S13 determines the correction processing strategy based on the number of trajectory deviations under different stratigraphic categories in the target area.

[0047] The correction strategy refers to the inclinometer zero drift correction scheme determined based on the trajectory deviation analysis results, including two types: high-frequency correction strategy and conditional correction strategy. The high-frequency correction strategy refers to performing correction processing when the deviation rate between the sensor temperature and the calibration temperature is less than a threshold. The conditional correction strategy refers to performing correction processing only when the deviation rate between the sensor temperature and the calibration temperature is less than a threshold and no correction processing has been performed within the most recent preset time period.

[0048] Furthermore, by determining the number of trajectory deviations under different stratigraphic categories, the proportion of trajectory deviation processes under different stratigraphic categories is determined. Combined with the number of deviation stratigraphic categories, a deviation analysis requirement value is determined. It is then determined whether the deviation analysis requirement value is greater than a preset requirement threshold. If so, the correction processing strategy is determined to be that sensor correction processing is required whenever the deviation rate between the temperature in the sensor and the calibration temperature is less than a preset deviation rate threshold to determine whether a deviation has occurred. If not, the correction processing strategy is determined to be that sensor correction processing, i.e., inclinometer correction processing, is required whenever the deviation rate between the temperature in the sensor and the calibration temperature is less than a preset deviation rate threshold and the sensor has not undergone correction processing within the most recent preset time period to determine whether a deviation has occurred.

[0049] Assuming that the number of deviation strata categories is not greater than a preset threshold based on the judgments in S121 and S122, and there are no strata categories with a trajectory deviation process proportion greater than the preset threshold, then the deviation analysis requirement value is calculated. Let the preset requirement threshold be a certain value. If the deviation analysis requirement value is not greater than this threshold, then the correction strategy is determined to be a conditional correction strategy. That is, the inclinometer correction is only performed when the deviation rate between the sensor temperature and the calibration temperature is less than a preset deviation rate threshold, and the sensor has not undergone correction processing within the most recent preset time period.

[0050] This step involves a comprehensive evaluation of trajectory deviation data. Its significance lies in adopting a more economical conditional correction strategy when the trajectory control problem is not serious. This reduces unnecessary correction operations while ensuring trajectory control accuracy, thereby extending the sensor's lifespan and improving drilling efficiency.

[0051] This embodiment, through the execution of steps S11 to S13, achieves intelligent determination of the correction processing strategy based on formation type and trajectory deviation data. Its core value is reflected in three aspects: First, by making dual judgments on the number of deviation formation types and the proportion of trajectory deviation process, it realizes the differentiated formulation of correction strategies; second, by flexibly switching between high-frequency correction strategies and conditional correction strategies, it balances trajectory control accuracy and operational efficiency; and third, it provides a precondition judgment mechanism for the subsequent monitoring and analysis of vibration monitoring signals, ensuring that vibration monitoring and analysis are only triggered when the zero drift effect is severe, thereby improving the systematicness and reliability of the overall drilling signal sensing data processing.

[0052] In one possible specific embodiment: The target area of ​​a certain drilling project contains six formation categories: mudstone, sandstone, limestone, shale, conglomerate, and coal seam. Drilling data for each layer was collected using a logging-while-drilling (MLD) system, and the formation categories were constructed using a MLD lithology identification model. The statistical analysis showed that the trajectory deviation frequency for each formation category was as follows: mudstone 8 times, sandstone 5 times, limestone 2 times, shale 12 times, conglomerate 3 times, and coal seam 1 time.

[0053] During the historical drilling process in the mudstone layer, the percentage of drilling processes with more than 6 trajectory deviations was 0.40%, not exceeding 0.60%; the percentage in the sandstone layer was 0.33%, not exceeding 0.60%; and the percentage in the shale layer was 0.67%, exceeding 0.60%. Therefore, there is a formation type (shale layer) where the percentage of trajectory deviation processes exceeds the preset threshold. The deviation correction strategy is determined to be a high-frequency correction strategy, meaning that sensor deviation correction is performed whenever the deviation rate between the temperature in the sensor and the calibrated temperature is less than the preset deviation rate threshold of 0.05.

[0054] Using this correction processing strategy, the offset data of the drilling signal is calculated in real time during the drilling process. When the residual between the inclinometer monitoring data and the model prediction data exceeds the instrument's nominal accuracy (well deviation residual greater than 0.2°, azimuth residual greater than 2°), it is determined that the vibration monitoring signal in the drilling signal needs to be monitored and analyzed, and then proceed to S2.

[0055] Furthermore, the calibration temperature is determined based on the temperature calibration results before drilling.

[0056] Specifically, this includes: multi-point temperature calibration: calibration should not be performed only at room temperature or at a single point, but should be performed at least 5 to 8 temperature points within the simulated downhole temperature range (e.g., 20℃–150℃) to accurately establish a temperature polynomial model (usually second or third order) with zero bias and scaling factor.

[0057] Turntable tumbling test: Under the calibrated temperature, the inclinometer is tumbled on the turntable at different angles (such as 0°, 90°, 180°, 270° tool face angle) to verify the orthogonality and linearity of the gravity sensor and magnetic sensor and eliminate installation errors.

[0058] Furthermore, the offset data of the drilling signal is determined based on the deviation between the monitoring data of the inclinometer in the drilling signal and the actual data.

[0059] Specifically, such as Figure 3 As shown, it is determined that monitoring and analysis of vibration monitoring signals in the drilling signals are required, specifically including: In this embodiment, the severity of the zero drift effect of the inclinometer in its current state is determined based on the degree of deviation between the monitoring data of the inclinometer and the model prediction data. When the severity of the zero drift effect of the inclinometer in its current state is high, vibration signal monitoring and analysis are performed in a timely manner, thereby laying the foundation for determining the monitoring, analysis and processing strategy for the aging state of the drill bit.

[0060] S21 determines the residual between the inclinometer monitoring data and the model prediction data in the drilling signal based on the offset data.

[0061] The residual refers to the difference between the actual monitoring data of the inclinometer after compensation processing for temperature, vibration, etc., and the model prediction data, which is used to quantify the degree of unmodeled drift of the inclinometer.

[0062] Assuming the actual well inclination angle monitored by the inclinometer at a certain depth is a certain value, and the well inclination angle predicted by the model is also a certain value, then the residual is the difference between the actual value and the predicted value. Similarly, the residual of the azimuth angle can be calculated.

[0063] This step, by calculating the residuals, is significant in establishing a quantitative index of the inclinometer's performance deviation. A residual exceeding the instrument's nominal accuracy represents a true deviation from the instrument's own state, thus providing an objective basis for determining whether further vibration monitoring and analysis are needed.

[0064] S22 takes the correction process where the residual exceeds the instrument’s nominal accuracy as the deviation identification process, and determines whether it is necessary to perform monitoring and analysis of the vibration monitoring signal in the drilling signal based on the deviation identification process data.

[0065] The deviation identification process refers to the process of handling the inclinometer residual error exceeding the instrument's nominal accuracy during the deviation correction process; the instrument's nominal accuracy refers to the measurement accuracy index that the inclinometer can guarantee under standard conditions, such as the well inclination residual threshold and the azimuth residual threshold.

[0066] For example, assume the instrument's nominal accuracy is a deviation residual of no more than 0.2° and an azimuth residual of no more than 2°. During multiple correction processes, count the number of processes where the residual exceeds the above thresholds. If the proportion of deviation identification processes in the correction processes is higher than the preset proportion, then it is determined that vibration monitoring signal monitoring and analysis processing is required.

[0067] This step, through the analysis of the proportion of the deviation identification process, is significant in determining whether the inclinometer is susceptible to external interference. When the proportion of the deviation identification process is high, it indicates that the zero drift of the inclinometer is seriously affected, and further analysis of the vibration monitoring signal is needed to determine the aging status of the drill bit, so as to avoid the risk of displacement deviation caused by the overlap of inclinometer deviation and drill bit aging problems.

[0068] This embodiment, through the execution of steps S21 to S22, realizes the judgment of vibration monitoring and analysis requirements based on offset data. Its core value is reflected in two aspects: First, by calculating residuals and statistically analyzing the deviation identification process, a quantitative method for evaluating the performance of the inclinometer is established; Second, by judging the threshold of the proportion of the deviation identification process, a condition triggering mechanism for vibration monitoring and analysis is realized, ensuring that vibration monitoring and analysis will only be initiated when the influence of inclinometer zero drift is severe, thereby avoiding unnecessary consumption of computing resources and improving the overall efficiency of drilling signal sensing data processing.

[0069] During drilling, the inclination meter collects inclination angle and azimuth angle data in real time. Within a certain depth range, the actual monitored inclination angle was 85.3°, the model predicted 85.1°, and the inclination residual was 0.2°; the actual monitored azimuth angle was 127.5°, the model predicted 125.8°, and the azimuth residual was 1.7°. Assuming the instrument's nominal accuracy is no greater than 0.2° for the inclination residual and no greater than 2° for the azimuth residual, and setting the preset process percentage threshold to 0.30.

[0070] In the most recent 100 correction processes, the residuals of 35 of them exceeded the instrument’s nominal accuracy. The deviation identification process accounted for 0.35. Since 0.35 is greater than 0.30, it was determined that the vibration monitoring signal in the drilling signal needed to be monitored and analyzed.

[0071] Furthermore, the type of vibration monitoring signal of interest is determined based on the vibration monitoring signals of the rock during the drilling process. Specifically, the vibration monitoring signals of rocks whose similarity coefficients meet the requirements and whose frequency of occurrence exceeds a preset threshold are taken as the type of vibration monitoring signal of interest.

[0072] Vibration monitoring signals during drilling were collected using vibration sensors, with a preset threshold of 20 occurrences and a preset similarity coefficient requirement of 0.85. Analysis identified four rock types whose vibration monitoring signals met the similarity coefficient requirement: mudstone (occurring 28 times), sandstone (occurring 35 times), limestone (occurring 15 times), and shale (occurring 42 times). Since mudstone, sandstone, and shale all occurred more than 20 times, these three vibration monitoring signal types were selected as the focus of the analysis and proceeded to step S3 for determining the monitoring and analysis method.

[0073] Furthermore, the method for determining the monitoring and analysis method of the vibration monitoring signal is as follows: In this embodiment, based on the number of vibration monitoring signal types of interest and the distribution data of rock periods containing these types of vibration monitoring signals, the sufficiency and distribution density of vibration monitoring signal types currently available for drill bit life assessment are determined. By utilizing the sufficiency and distribution density of these vibration monitoring signal types, it is determined under which type of rock vibration monitoring signal to perform aging state assessment and analysis. This ensures the reliability of aging state assessment and analysis while also improving the efficiency of updating the types of vibration monitoring signals of interest.

[0074] The available analytical monitoring signal types are all monitoring signal types within a group whose aging degree assessment and analysis results are consistent within the same unit time period, including vibration monitoring signal types that meet the monitoring and analysis methods for vibration monitoring signals and monitoring signal types of interest. The aging degree assessment and analysis process refers to the process of using the characteristic parameters of vibration monitoring signals and the drill bit aging state assessment model to determine the current aging degree level of the drill bit. In addition to the monitoring signal types of interest, there are other vibration monitoring signal types that have also undergone aging state assessment and analysis during the drilling process, and the assessment results of these types also participate in the group division and the selection of available signal types.

[0075] Assuming that multiple types of vibration monitoring signals of interest are identified through S2, and the monitoring and analysis methods for these signals are determined through S31 to S33, and the unit time is set to a certain value, based on the determined monitoring and analysis methods, the aging degree of the vibration monitoring signals of interest and other signal types that have undergone monitoring and analysis is assessed using a drill bit aging condition assessment model within the same unit time. Signal types with consistent assessment results are grouped together, and all signal types in the group with the largest number of signal types are selected as the available monitoring signal types for analysis.

[0076] This step establishes a technical path for assessing the reliability of aging evaluation from the perspective of focusing on characteristic signals and screening usable signal features. Its significance lies in identifying signal type groups with high signal stability and consistent evaluation results by judging the consistency of aging evaluation results of different signal types within the same unit of time. This ensures that the determination of subsequent drilling signal sensing data processing methods is based on a reliable signal foundation.

[0077] The available analytical monitoring signal types are all monitoring signal types within a group whose aging degree assessment analysis results are consistent within the same unit time period, including vibration monitoring signal types that meet the monitoring analysis method for vibration monitoring signals and monitoring signal types of interest; the aging degree assessment analysis process refers to the process of using the characteristic parameters of vibration monitoring signals and the drill bit aging state assessment model to determine the current aging degree level of the drill bit; the monitoring analysis method refers to the analysis strategy determined based on the reliable time period distribution characteristics of the identified vibration monitoring signal types of interest, used for aging assessment in subsequent drilling periods.

[0078] For example, assuming that multiple types of vibration monitoring signals of interest are identified through S2, and the monitoring and analysis methods for vibration monitoring signals are determined through S31 to S33, in the subsequent drilling interval after the completion of the reliable period for identifying the types of vibration monitoring signals of interest, the aging degree of vibration monitoring signal types that meet the conditions is assessed based on the determined monitoring and analysis methods. Signal types with consistent assessment results are grouped into a group, and all signal types in the group with the largest number of signal types are selected as available analytical monitoring signal types.

[0079] This step establishes a technical path for assessing the reliability of aging evaluation from the perspective of focusing on characteristic signals and screening usable signal features. Its significance lies in identifying signal type groups with high signal stability and consistent evaluation results by judging the consistency of aging evaluation results of different signal types within the same unit of time. This ensures that the determination of subsequent drilling signal sensing data processing methods is based on a reliable signal foundation.

[0080] S31 determines the number of vibration monitoring signal types based on the vibration monitoring signal types of interest.

[0081] The number of vibration monitoring signal types of interest refers to the total number of rock vibration monitoring signal types that meet the similarity coefficient requirements and whose occurrence frequency reaches a preset threshold; the difference in the occurrence frequency indicates a difference in signal stability, and the higher the signal stability, the higher the reliability of the aging degree assessment.

[0082] Assuming that mudstone vibration signal, sandstone vibration signal, and shale vibration signal are selected as the vibration monitoring signal types of interest through S2, then the number of vibration monitoring signal types of interest is 3.

[0083] This step involves statistically analyzing the number of vibration monitoring signal types. Its significance lies in assessing the richness of signal types that can be used for drill bit aging assessment, providing a basis for selecting subsequent monitoring and analysis methods.

[0084] S32, based on the update status of monitoring data under different types of vibration monitoring signals of interest, determine the time period in which the type of vibration monitoring signal of interest exists, and use it as the reliable time period for identification.

[0085] The reliable time period is determined by analyzing the monitoring data of vibration monitoring signals of the same type of vibration monitoring signal to identify the period during which the drill bit is aging. The update status of the monitoring data refers to the real-time acquisition and accumulation of each type of vibration monitoring signal during the drilling process. The reliable time period is used to analyze the distribution characteristics of the signal types to determine the monitoring and analysis method for aging assessment in subsequent drilling.

[0086] Assuming that mudstone vibration signals are detected at depths of 1000-1050 meters and 1100-1150 meters during drilling, these depth ranges are considered reliable periods for identifying mudstone vibration signals, which can be used to analyze the distribution density and continuity of this signal type in subsequent analyses.

[0087] This step, by identifying reliable time periods, is significant in establishing a spatiotemporal correspondence between vibration monitoring signals and drilling locations. It provides a data foundation for analyzing the distribution density and continuity of signal types, thereby supporting the rational selection of monitoring and analysis methods and determining the drilling time period range for subsequent aging assessment.

[0088] S33, using the number of vibration monitoring signal types of interest and the distribution data of the reliable time periods for identification, determine the monitoring and analysis method for the vibration monitoring signals.

[0089] The monitoring and analysis method refers to a strategy determined based on the number of vibration monitoring signal types of concern and the distribution characteristics of reliable identification time periods. This strategy is used to assess and analyze the aging status of drill bits in subsequent drilling periods after the reliable identification time period of the vibration monitoring signal types of concern ends. It includes multiple real-time analysis methods based on preset occurrence thresholds, second preset occurrence thresholds, and third preset occurrence thresholds.

[0090] Assuming the number of vibration monitoring signal types of interest is a certain value, and a preset threshold value is also set, the vibration monitoring signal analysis method to be used in subsequent drilling periods is determined based on the comparison between the number of interest vibration monitoring signal types and the threshold, as well as the distribution data of reliable time periods. Then, based on this method, aging assessments are performed on signal types that meet the conditions, and usable analytical monitoring signal types are selected.

[0091] This step, by comprehensively evaluating the number of signal types of concern and the distribution of reliable time periods, is significant in that it dynamically selects the most suitable monitoring and analysis method based on the sufficiency of data conditions. This ensures that in subsequent drilling after the reliable time period for identifying the vibration monitoring signal types of concern ends, reliable aging assessment processing can be performed on the vibration monitoring signal types that meet the conditions, thereby supporting the subsequent screening of available signal types.

[0092] It should be noted that if the number of vibration monitoring signal types of concern is less than a preset number threshold, then as long as the number of times the rock of the vibration monitoring signal type has appeared in history is greater than a preset number of occurrences threshold, then as long as a vibration monitoring signal of the rock of the vibration monitoring signal type is detected at any time, monitoring and analysis processing will be performed to determine the aging state of the drill bit.

[0093] The preset occurrence threshold refers to the lowest critical value of historical occurrences used to trigger real-time analysis; the drill bit aging status assessment model refers to an assessment model that uses characteristic parameters such as amplitude, frequency, and energy of vibration monitoring signals to determine the current aging level of the drill bit (normal, mild aging, moderate aging, severe aging). The higher the occurrence frequency, the higher the signal stability and the more reliable the assessment result; the monitoring analysis and processing are carried out in the subsequent drilling period after the reliable period for identifying the vibration monitoring signal type ends.

[0094] Assuming the number of vibration monitoring signal types under observation is less than a preset threshold, and setting the preset occurrence threshold to a certain value, in the subsequent drilling interval after the reliable identification period for all vibration monitoring signal types under observation has ended, for vibration monitoring signal types with a historical occurrence count exceeding the preset occurrence threshold, whenever a vibration monitoring signal of this type of rock is detected at any time, the drill bit aging condition assessment model is used for monitoring and analysis to determine the aging condition of the drill bit.

[0095] This approach, which focuses on the threshold of the number of vibration monitoring signal types, is significant because, when the number of available signal types is limited, a simplified real-time analysis method can be used to reduce the requirements for signal stability in subsequent drilling after the reliable identification period ends, thereby improving the reliability of the selection of available signal types.

[0096] Additionally, it should be noted that if the number of vibration monitoring signal types of interest is not less than a preset threshold, the following content is also included: Case 1: If the number of vibration monitoring signal types of concern is not less than a preset threshold, the interval between adjacent reliable time periods is determined based on the distribution data of the reliable time periods. If the average interval between adjacent reliable time periods is less than a preset interval threshold, then as long as the number of times the rock of the vibration monitoring signal type appears in history is greater than a second preset occurrence threshold, then as long as a vibration monitoring signal of the rock of the vibration monitoring signal type is detected at any time, monitoring and analysis processing is performed to determine the aging state of the drill bit.

[0097] The interval between adjacent reliable identification time periods refers to the time or depth interval between two adjacent reliable identification time periods after all the reliable identification time periods of the vibration monitoring signal types of interest are sorted by depth location (regardless of signal type, adjacent time periods across signal types are also included); the second preset occurrence threshold is greater than the preset occurrence threshold, which refers to the historical occurrence threshold value used to trigger real-time analysis in subsequent drilling periods under the condition that the number of vibration monitoring signal types of interest is sufficient and the time period distribution is dense.

[0098] Assuming the number of vibration monitoring signal types of interest is not less than a preset threshold, the reliable identification time periods of all vibration monitoring signal types of interest are mixed and sorted, and the interval between adjacent reliable identification time periods is calculated (regardless of signal type). If the average interval length is less than the preset interval length threshold, it indicates that the signal types are generally densely distributed. In this case, during subsequent drilling after the reliable identification time period ends, a second preset occurrence threshold is used for real-time analysis and processing.

[0099] This approach, which uses the average interval of all reliable identification periods to assess the overall distribution density of all signal types of interest, allows for the use of a higher historical occurrence threshold in subsequent drilling periods when the overall signal distribution is dense. This improves the reliability of the analysis and reduces unnecessary analysis frequency.

[0100] Also includes the following: Case 2: If the average interval between adjacent reliable identification time periods is not less than the preset interval length threshold, the historical drilling time period is divided into multiple sub-time periods using unit time. It is determined whether the proportion of sub-time periods with reliable identification time periods is less than the preset reliable time period number threshold. If so, as long as the rock of the vibration monitoring signal type appears more than the preset occurrence number threshold in history, as long as a vibration monitoring signal of the rock of the vibration monitoring signal type is detected at any time, monitoring and analysis processing is performed to determine the aging state of the drill bit. If not, proceed to Case 3.

[0101] The unit duration refers to the standard time or depth interval used to divide the drilling period into sub-periods; the sub-period with a reliable identification period refers to a sub-period unit containing at least one reliable identification period (regardless of signal type); the preset reliable period quantity threshold refers to the critical proportion value used to determine the sparseness of the distribution of reliable identification periods.

[0102] Assuming the average interval between adjacent reliable identification periods is not less than a preset interval threshold, it indicates that the overall distribution of signal types is relatively dispersed. In this case, the historical drilling period is divided into multiple sub-periods according to unit duration, and the proportion of sub-periods containing at least one reliable identification period is counted. If this proportion is less than a preset reliable period number threshold, then a preset occurrence frequency threshold is used for real-time analysis and processing in subsequent drilling after the reliable identification period ends.

[0103] The significance of identifying the proportion of reliable sub-periods in this situation lies in further assessing the signal coverage density when the overall distribution of signal types is dispersed. When the coverage density is low, a simplified real-time analysis method can be used in subsequent drilling to improve the reliability of acquiring available vibration monitoring signals.

[0104] Also includes the following: Scenario 3: If there is no reliable identification period within the most recent preset time period, then as long as the number of occurrences of the rock of the vibration monitoring signal type in history is above a preset occurrence threshold, then as long as a vibration monitoring signal of the rock of the vibration monitoring signal type is detected at any time, monitoring and analysis processing will be performed to determine the aging state of the drill bit. If there is a reliable identification period within the most recent preset time period, then as long as the number of occurrences of the rock of the vibration monitoring signal type in history is above a third preset occurrence threshold, then as long as a vibration monitoring signal of the rock of the vibration monitoring signal type is detected at any time, monitoring and analysis processing will be performed to determine the aging state of the drill bit.

[0105] The most recent preset time period refers to the specific depth interval closest to the current position; the third preset occurrence threshold is greater than the second preset occurrence threshold, and refers to the highest historical occurrence threshold value used to trigger real-time analysis in subsequent drilling under the condition that the coverage density of the identified reliable time period is high and reliable data exists in the most recent time period.

[0106] Assuming the proportion of sub-periods in the identified reliable time period is not less than a preset reliable time period number threshold, it indicates a high signal type coverage density. In this case, it is determined whether a reliable time period exists within the most recent preset time period (regardless of signal type). If it does, a third preset occurrence frequency threshold is used for real-time analysis during subsequent drilling after the reliable time period ends; otherwise, a preset occurrence frequency threshold is used for real-time analysis.

[0107] This situation, by identifying the existence of reliable time periods within the most recent preset time period, is significant because it further considers the timeliness of data when the signal type coverage density is high. When reliable data exists in the most recent time period, a higher threshold is used for real-time analysis in subsequent drilling to ensure that the available vibration signal types for analysis and processing of aging conditions can be obtained more comprehensively.

[0108] It should be noted that the drill bit aging condition assessment model is based on the vibration signal variation state of different vibration monitoring signal characteristic types, specifically including: The drill bit aging condition assessment model uses a neural network model based on a specific vibration monitoring signal type for evaluation, specifically including: Neural Network Model Architecture: The neural network model adopts a hybrid architecture combining convolutional neural networks (CNNs) and long short-term memory networks (LSTMs). The CNNs are used to extract the time-frequency domain features of the vibration monitoring signals, including the signal amplitude distribution, frequency components, and energy concentration; the LSTMs are used to capture the dynamic trends of the vibration signals over time, including the signal fluctuation patterns, periodic characteristics, and anomalous abrupt changes.

[0109] Neural network models for specific signal types: For mudstone vibration signals, a dedicated neural network model for mudstone signals is constructed. The model's input layer receives the raw time-domain waveform data of the mudstone vibration signal. The first hidden layer is a convolutional layer using 32 convolutional kernels to extract local features of the signal. The second hidden layer is a pooling layer to reduce feature dimensionality. The third hidden layer is an LSTM layer containing 64 memory units to capture the signal's temporal dependence. The output layer outputs a four-class classification of drill bit aging (normal, mildly aging, moderately aging, and severely aging). The model uses root mean square amplitude, low-frequency energy proportion, and temporal waveform stability as core feature parameters, and learns the characteristic patterns of drill bit aging during mudstone drilling through training.

[0110] For sandstone vibration signals, a dedicated neural network model for sandstone signals is constructed. The model's input layer receives the raw time-domain waveform data of the sandstone vibration signal. The first hidden layer is a convolutional layer using 64 convolutional kernels to extract high-frequency impact features. The second hidden layer is a pooling layer. The third hidden layer is an LSTM layer containing 128 memory units to capture the time-series features of the impact signal. The output layer outputs a four-class classification result indicating the degree of drill bit aging. The model uses peak acceleration, impact frequency, and spectral complexity as core feature parameters and learns the characteristic patterns of drill bit cutting tooth breakage during sandstone drilling through training.

[0111] For shale vibration signal types, a dedicated neural network model for shale signals is constructed. The model's input layer receives the raw time-domain waveform data of the shale vibration signal. The first hidden layer is a convolutional layer using 48 convolutional kernels to extract the bedding response features of the signal. The second hidden layer is a pooling layer. The third hidden layer is an LSTM layer containing 96 memory units to capture the anisotropic variation features of the signal. The output layer outputs a four-class classification result of the drill bit aging degree. The model uses root mean square amplitude, dominant frequency component, and energy spectral density as core feature parameters, and learns the characteristic patterns of drill bit cutting efficiency decline during shale drilling through training.

[0112] For limestone vibration signals, a dedicated neural network model for limestone signals is constructed. The model's input layer receives the raw time-domain waveform data of the limestone vibration signal. The first hidden layer is a convolutional layer using 64 convolutional kernels to extract high-frequency wear features. The second hidden layer is a pooling layer. The third hidden layer is an LSTM layer containing 128 memory units to capture the cumulative wear features of the signal. The output layer outputs a four-class classification result of the drill bit aging degree. The model uses the proportion of high-frequency energy, amplitude variation coefficient, and harmonic distortion as core feature parameters, and learns the characteristic patterns of drill bit wear during limestone drilling through training.

[0113] Training and updating of neural network models: Neural network models for each specific signal type are trained using vibration monitoring signal samples from historical drilling data. Training samples include labeled data with known drill bit aging conditions. During model training, backpropagation is used to optimize network weights, cross-entropy loss is used to evaluate model performance, and Dropout is employed to prevent overfitting. As new drilling data accumulates, incremental training of each neural network model is performed periodically to update model parameters and adapt to changes in drill bit type and formation conditions.

[0114] Aging Degree Assessment Process: During actual drilling, when a vibration monitoring signal for a specific rock type is detected, the signal type is first identified, and then the corresponding dedicated neural network model is invoked. The original time-domain waveform data of the signal is input into the model, which automatically extracts features, performs classification, and outputs the assessment result of the drill bit aging degree for that period. The assessment result is used for subsequent signal type grouping and selection of available analytical monitoring signal types.

[0115] This embodiment, through S31 to S33 and a comprehensive judgment of three scenarios, achieves intelligent determination of vibration monitoring signal analysis methods and selection of available analytical monitoring signal types. Its core value is reflected in three aspects: First, by focusing on the number of vibration monitoring signal types and the overall distribution of reliable identification periods through multi-layered judgment, it achieves a fine match between monitoring analysis methods and data conditions; second, by conducting aging assessments on vibration monitoring signal types that meet the conditions of the monitoring analysis methods in subsequent drilling after the reliable identification period ends, and by grouping based on the consistency of assessment results, it achieves scientific selection of available signal characteristics; third, by selecting the group with the largest number of signal types as the available analytical monitoring signal types, it ensures the breadth of coverage and reliability of aging state assessment results.

[0116] S2 identifies three types of vibration monitoring signals to be of interest: mudstone vibration, sandstone vibration, and shale vibration. A preset threshold of three types is set, and since 3 is not less than 3, the process proceeds to condition judgment.

[0117] The reliable identification periods for each type of vibration monitoring signal of interest were statistically analyzed: mudstone vibration signals were detected at depths of 1000-1050 meters, 1100-1150 meters, and 1300-1350 meters in three periods; sandstone vibration signals were detected at depths of 1050-1100 meters, 1200-1250 meters, and 1400-1450 meters in three periods; and shale vibration signals were detected at depths of 1150-1200 meters, 1350-1400 meters, and 1600-1650 meters in three periods. The reliable identification time periods for all vibration monitoring signal types of interest were sorted by depth as follows: 1000-1050 meters (mudstone), 1050-1100 meters (sandstone), 1100-1150 meters (mudstone), 1150-1200 meters (shale), 1200-1250 meters (sandstone), 1300-1350 meters (mudstone), 1350-1400 meters (shale), 1400-1450 meters (sandstone), and 1600-1650 meters (shale), totaling nine time periods. The last reliable identification time period ended at a depth of 1650 meters.

[0118] Calculate the interval between adjacent reliable identification time periods (regardless of signal type, including cross-signal types): 0 meters (continuous between 1000-1050 meters and 1050-1100 meters), 0 meters (continuous between 1050-1100 meters and 1100-1150 meters), 0 meters (continuous between 1100-1150 meters and 1150-1200 meters), 0 meters (continuous between 1150-1200 meters and 1200-1250 meters), 50 meters (between 1200-1250 meters and 1300-1350 meters), 0 meters (continuous between 1300-1350 meters and 1350-1400 meters), 0 meters (continuous between 1350-1400 meters and 1400-1450 meters), and 150 meters (between 1400-1450 meters and 1600-1650 meters), for a total of 8 intervals. The average interval between adjacent reliable identification time periods is (0+0+0+0+50+0+0+150)÷8=25 meters.

[0119] If the preset interval time threshold is 30 meters, and 25 meters is less than 30 meters, then the average interval time between adjacent reliable identification time periods is less than the preset interval time threshold, and the process enters case 1.

[0120] The second preset occurrence threshold is set at 25 occurrences. After the reliable identification period for vibration monitoring signal types (1000-1650 meters) ends, during subsequent drilling at depths of 1650-1900 meters, vibration monitoring signal types that meet the occurrence threshold are monitored, analyzed, and processed. Among all vibration monitoring signal types, mudstone vibration signals have a historical occurrence count of 28, greater than 25; sandstone vibration signals have a historical occurrence count of 35, greater than 25; shale vibration signals have a historical occurrence count of 42, greater than 25; limestone vibration signals have a historical occurrence count of 30, greater than 25; and conglomerate vibration signals have a historical occurrence count of 12, less than 25, and therefore do not meet the threshold.

[0121] Based on the monitoring and analysis method determined in Scenario 1, in the subsequent drilling section (1650-1900 meters), the drill bit aging status assessment model was used to monitor and analyze the vibration signals of mudstone (1650-1700 meters), sandstone (1700-1750 meters), shale (1750-1800 meters), and limestone (1800-1850 meters). The aging degree assessment results of the drill bit were determined within the same unit time of 50 meters: the assessment result of the mudstone vibration signal within the unit time of 1650-1700 meters was moderate aging; the assessment result of the sandstone vibration signal within the unit time of 1700-1750 meters was moderate aging; the assessment result of the shale vibration signal within the unit time of 1750-1800 meters was moderate aging; and the assessment result of the limestone vibration signal within the unit time of 1800-1850 meters was slight aging.

[0122] Signal types with consistent assessment results were grouped as follows: Group 1 includes mudstone vibration signals, sandstone vibration signals, and shale vibration signals. The assessment results of all signal types in this group are moderately aged, totaling 3 signal types; Group 2 includes limestone vibration signals. The assessment results of all signal types in this group are mildly aged, totaling 1 signal type.

[0123] Comparing the number of signal types in each group, group 1 contains 3 signal types, while group 2 contains 1 signal type, indicating that group 1 has the most signal types. Therefore, the mudstone vibration signal, sandstone vibration signal, and shale vibration signal within group 1 are selected as the three available analytical monitoring signal types, and proceed to step S4.

[0124] Furthermore, the available analytical monitoring signal types are all vibration monitoring signal types within the monitoring signal type group that have the largest number of vibration monitoring signal types with consistent aging degree assessment analysis results within the same unit time period, including vibration monitoring signal types that meet the monitoring and analysis methods for vibration monitoring signals and monitoring signal types of interest.

[0125] Furthermore, the method for determining the sensing data processing method for the drilling signal is as follows: In this embodiment, based on the distribution data of the time periods used for evaluating and analyzing the aging state of the drill bit and the number of evaluation and analysis time periods under different available analytical monitoring signal types, the timeliness and matching degree of the evaluation and analysis of the aging state of the brick in the current state are determined. That is, the more time periods used for evaluating and analyzing the aging state of the drill bit, and the more available analytical monitoring signal types with a large number of evaluation and analysis time periods, the higher the reliability of the evaluation and analysis of the aging state. The reliability of the evaluation and analysis of the aging state is used to determine the sensing data processing method for the drilling signal, which also lays the foundation for further improving the efficiency and accuracy of the correction processing when the reliability of the evaluation and analysis of the aging state is poor.

[0126] The method for processing the sensing data of the drilling signal refers to the inclinometer correction strategy determined based on the monitoring and analysis matching coefficient of the available analytical monitoring signal types and the reliability of the aging state assessment. This includes a basic calibration method (calibration is performed as long as the number of time periods within any temperature range meets the requirements) and a conditional calibration method (calibration is performed only when the number of time periods within the temperature range meets the requirements and no correction has been performed within the most recent preset time period). The timeliness and matching degree of the aging state assessment and processing refer to the fact that the more time periods used for aging state assessment and analysis processing of the drill bit, and the more available analytical monitoring signal types with a large number of assessment and analysis time periods, the higher the reliability of the aging state assessment and analysis processing.

[0127] For example, assuming there are a certain number of available analytical monitoring signal types, and a preset threshold for the number of available signal types, the reliability of drill bit aging status assessment and processing under the current condition is determined based on the number of available analytical monitoring signal types and the monitoring and analysis matching coefficients for each type, thereby determining the sensing data processing method for drilling signals.

[0128] This step establishes a technical path from the types of available analytical monitoring signals to the determination of sensing data processing methods. Its significance lies in comprehensively evaluating the reliability of drill bit aging condition assessment and adjusting the deviation correction strategy of the inclinometer accordingly. In cases where the reliability of aging condition assessment is poor, it further improves the efficiency and accuracy of deviation correction, thereby ensuring the stability and reliability of drilling trajectory control.

[0129] S41, Based on the available analytical monitoring signal type data, determine the time period for the aging state assessment and analysis of the drill bit, and use it as the assessment and analysis time period.

[0130] The evaluation and analysis period refers to the effective period for evaluating and analyzing the aging status of the drill bit using vibration monitoring data of available analytical monitoring signal types. These periods have sufficient signal quality and data volume to support reliable evaluation and analysis. The evaluation and analysis period is located in the subsequent drilling interval after the reliable period for identifying the vibration monitoring signal type of interest ends.

[0131] This step, by determining the evaluation and analysis period, is significant in that it clarifies the effective data coverage that can be used for drill bit aging assessment, providing a statistical basis for the calculation of matching coefficients and the judgment of the reliability of aging assessment in subsequent monitoring and analysis.

[0132] S42, using the distribution data of evaluation analysis periods under different available analysis monitoring signal types, determine the number of evaluation analysis periods under different available analysis monitoring signal types, and determine the monitoring analysis matching coefficient of the available analysis monitoring signal type based on the number of evaluation analysis periods under the available analysis monitoring signal type.

[0133] The monitoring and analysis matching coefficient is determined by multiplying the number of evaluation and analysis periods within the most recent drilling time under the available analysis and monitoring signal type by a preset proportional factor; the more available analysis and monitoring signal types with a larger number of evaluation and analysis periods, the higher the reliability of the aging state evaluation and analysis processing.

[0134] Assuming that the number of evaluation and analysis periods of mudstone vibration signals within the most recent drilling time is a certain value, and the preset scaling factor is a certain value, then the monitoring and analysis matching coefficient of mudstone vibration signals is the product of the number of evaluation and analysis periods and the scaling factor.

[0135] This step, through monitoring and analyzing the calculation of the matching coefficient, is significant in that it quantifies the contribution of each available analytical monitoring signal type to the timeliness of drill bit aging assessment, providing a numerical basis for the selection of sensing data processing methods.

[0136] S43, using the available analytical monitoring signal types and the monitoring analysis matching coefficients for different available analytical monitoring signal types, determine the sensing data processing method for the drilling signal.

[0137] The determination of the sensing data processing method comprehensively considers the number of available analytical monitoring signal types during the evaluation and analysis period, the monitoring and analysis matching coefficient of each type, and the sum of the number of reliable monitoring signal types and the number of available analytical monitoring signal types during the evaluation and analysis period. The method is hierarchically determined to adopt either the basic calibration method or the conditional calibration method.

[0138] Assume the number of available monitoring signal types for the evaluation and analysis period is a certain value, and set a preset threshold for the number of available signal types. If this number is less than the preset threshold, the basic calibration method is adopted; if the number is not less than the preset threshold, the basic calibration method or the conditional calibration method is further determined based on the distribution of the monitoring and analysis matching coefficients.

[0139] This step, through a comprehensive evaluation of the number of available signal types and the monitoring and analysis matching coefficient, is significant in that it dynamically adjusts the triggering conditions for the inclinometer correction process based on the reliability of the drill bit aging condition assessment. When the assessment reliability is high, a conditional calibration method is used to reduce the frequency of corrections, while when the assessment reliability is low, a basic calibration method is used to ensure the timeliness of the correction process.

[0140] It should be noted that if the number of available analytical monitoring signal types during the evaluation and analysis period is less than the preset threshold for the number of available signal types, then the method for processing the sensing data of the drilling signal is determined to be that as long as the number of time periods within any temperature range meets the requirements, calibration processing is performed within the temperature range, thereby enabling more definitive correction processing of the inclinometer.

[0141] The preset threshold for the number of available signal types refers to a critical value used to determine whether the number of available analytical monitoring signal types is sufficient; the temperature range refers to multiple temperature segments divided according to the downhole temperature range, used for multi-point temperature calibration processing.

[0142] If the number of available analytical monitoring signal types for an evaluation and analysis period is less than a preset threshold, it indicates that there are insufficient signal types available for drill bit aging assessment. In this case, the basic calibration method is adopted, which means that as long as the number of time periods in any temperature range meets the requirements, calibration processing is performed within that temperature range.

[0143] This situation is determined by a threshold of the number of available signal types. Its significance lies in the fact that when the data basis for assessing the aging status of the drill bit is insufficient, a more frequent basic calibration method is adopted. By increasing the frequency of correction processing, the lack of assessment reliability is compensated for, thereby ensuring the measurement accuracy of the inclinometer and the reliability of trajectory control.

[0144] Additionally, it should be noted that if the number of available monitoring signal types during the evaluation and analysis period is not less than the preset threshold for the number of available signal types, the following content is also included: Case 1: If, based on the monitoring and analysis matching coefficients of different available monitoring and analysis signal types, it is determined that there is no available monitoring and analysis signal type with a monitoring and analysis matching coefficient greater than the preset monitoring and analysis matching coefficient threshold, then the processing method for the sensing data of the drilling signal is determined to be that as long as the number of time periods within any temperature range meets the requirements, calibration processing is performed within the temperature range, thereby more definitively performing the correction processing of the inclinometer.

[0145] The preset analysis matching coefficient threshold refers to the critical value used to determine the sufficiency of monitoring and analysis of a single available analysis monitoring signal type.

[0146] Assume that the number of available analytical monitoring signal types during the evaluation and analysis period is a certain value, not less than a preset threshold for the number of available signal types. If the monitoring and analysis matching coefficients of all available analytical monitoring signal types are not greater than the preset analysis matching coefficient threshold, it indicates that the monitoring and analysis sufficiency of each signal type is insufficient, and in this case, the basic calibration method is adopted.

[0147] The significance of judging the matching coefficient through single signal type monitoring and analysis in this case is that even if there are enough available signal types, if the monitoring and analysis of each signal type is insufficient, the basic calibration method is still used to ensure the timeliness of the correction process and avoid the decrease in evaluation reliability due to insufficient analysis of a single signal type.

[0148] Also includes the following: Scenario 2: If there are available analytical monitoring signal types with a monitoring and analysis matching coefficient greater than the preset analytical matching coefficient threshold, these types are considered reliable monitoring signal types. The monitoring and analysis matching coefficient is determined based on the sum of the reliable monitoring signal types and the number of available analytical monitoring signal types with evaluation and analysis periods. It is then determined whether the monitoring and analysis matching coefficient is greater than the preset matching coefficient threshold. If so, the method for processing the sensing data of the drilling signal is determined to be that calibration processing is performed within the temperature range provided the number of time periods within the target temperature range meets the requirements and no correction processing of the drilling signal has been performed within the most recent preset duration. This allows for more definitive correction processing of the inclinometer. If not, the method for processing the sensing data of the drilling signal is determined to be that calibration processing is performed within the temperature range provided the number of time periods within the target temperature range meets the requirements. This allows for more definitive correction processing of the inclinometer.

[0149] The reliable monitoring signal type refers to the available analytical monitoring signal type whose monitoring and analysis matching coefficient is greater than the preset analysis matching coefficient threshold; the preset matching coefficient threshold refers to the critical value used to comprehensively judge the overall sufficiency of monitoring and analysis; the number of target temperature intervals refers to the number of temperature intervals that need to meet the time period requirement when performing condition calibration processing.

[0150] If a monitoring and analysis matching coefficient for a certain available analytical monitoring signal type is greater than a preset matching coefficient threshold, then that signal type is considered a reliable monitoring signal type. The sum of the number of reliable monitoring signal types and the number of available analytical monitoring signal types within the evaluation and analysis period is calculated, and this sum is used as the comprehensive monitoring and analysis matching coefficient. If the preset matching coefficient threshold is a certain value, and the comprehensive monitoring and analysis matching coefficient is greater than this threshold, then the conditional calibration method is used; otherwise, the basic calibration method is used.

[0151] This situation is determined by a combination of reliable monitoring signal type and comprehensive monitoring analysis matching coefficient. Its significance lies in the fact that when the data basis for assessing the aging state of the drill bit is sufficient and the analysis quality is high, the condition calibration method is adopted to add the condition that no correction processing has been performed within the most recent preset time period, on the basis of meeting the requirements for the number of time periods in the temperature range. This reduces unnecessary correction operations, extends the service life of the sensor, and improves drilling efficiency.

[0152] This embodiment achieves intelligent determination of the drilling signal sensing data processing method through S41 to S43 and the comprehensive judgment of three situations. Its core value is reflected in three aspects: First, by evaluating the statistical analysis of the analysis period and calculating the matching coefficient, a quantitative evaluation system for the reliability of drill bit aging status assessment is established; second, by hierarchical judgment of the number of available signal types and the matching coefficient of a single signal type, the flexible switching between the basic calibration method and the conditional calibration method is realized; third, by comprehensively monitoring and analyzing the threshold judgment of the matching coefficient, the conditional calibration method is used to reduce the frequency of correction when the evaluation reliability is high, and the basic calibration method is used to ensure the timeliness of correction when the evaluation reliability is low, thereby achieving the best balance between trajectory control accuracy and operation efficiency.

[0153] S3 determines that the available analysis and monitoring signal types are mudstone vibration signals, sandstone vibration signals, and shale vibration signals, a total of three types. The preset threshold for the number of available signal types is set to 2, and 3 is not less than 2; therefore, the process proceeds to situation judgment.

[0154] Determine the assessment and analysis period: Within the new drilling section (1900-2200 meters) following the completion of the subsequent drilling section (1650-1900 meters) of S3, conduct an aging assessment of each available analytical monitoring signal type based on monitoring and analysis methods. Assuming a unit duration of 50 meters, the new drilling section (1900-2200 meters) comprises six unit durations: 1900-1950 meters, 1950-2000 meters, 2000-2050 meters, 2050-2100 meters, 2100-2150 meters, and 2150-2200 meters.

[0155] Within the new drilling section, based on the monitoring and analysis method conditions (historical occurrence count exceeding the second preset occurrence threshold of 25 times), the mudstone vibration signal was evaluated and analyzed twice, once each within the unit time period of 1900-1950 meters and 2050-2100 meters; the sandstone vibration signal was evaluated and analyzed twice, once each within the unit time period of 1950-2000 meters and 2100-2150 meters; and the shale vibration signal was evaluated and analyzed once within the unit time period of 2000-2050 meters.

[0156] Assuming the most recent drilling time is 300 meters (1900-2200 meters), this interval is the new drilling interval. The number of evaluation and analysis periods for each type of available analytical monitoring signal within this drilling time is counted: mudstone vibration signal: 2 times (1900-1950 meters, 2050-2100 meters); sandstone vibration signal: 2 times (1950-2000 meters, 2100-2150 meters); shale vibration signal: 1 time (2000-2050 meters).

[0157] Assuming the preset scaling factor is 0.35, the monitoring and analysis matching coefficient for mudstone vibration signals is 2 × 0.35 = 0.70, the monitoring and analysis matching coefficient for sandstone vibration signals is 2 × 0.35 = 0.70, and the monitoring and analysis matching coefficient for shale vibration signals is 1 × 0.35 = 0.35.

[0158] Assuming a preset analysis matching coefficient threshold of 0.60, the monitoring and analysis matching coefficient for mudstone vibration signals is 0.70, which is greater than 0.60; the monitoring and analysis matching coefficient for sandstone vibration signals is 0.70, which is greater than 0.60; and the monitoring and analysis matching coefficient for shale vibration signals is 0.35, which is not greater than 0.60. Therefore, there are available analytical monitoring signal types (mudstone vibration signals and sandstone vibration signals) with monitoring and analysis matching coefficients greater than the preset analysis matching coefficient threshold. Mudstone vibration signals and sandstone vibration signals are then considered reliable monitoring signal types, proceeding to Case 2.

[0159] There are 2 reliable monitoring signal types, 3 available analytical monitoring signal types within the evaluation and analysis period, and a monitoring and analysis matching coefficient of 2+3=5. Assuming a preset matching coefficient threshold of 4, and since 5 is greater than 4, the conditional calibration method is determined as the data processing method for the drilling signals.

[0160] The target temperature range is set to two, with a preset duration of four hours. If the number of time periods within the two temperature ranges meets the requirements, and no correction processing of the drilling signal has been performed within the most recent four hours, then calibration processing will be performed within these two temperature ranges, thereby enabling more definitive correction processing of the inclinometer.

[0161] Example 2 Secondly, the present invention provides an early warning system, employing the aforementioned method for processing multi-element signal sensing data during drilling, specifically including: Data sensing module, early warning module; The data sensing module is responsible for sensing and processing the drilling signal using the sensing data processing method for the drilling signal. The early warning module is responsible for assessing the degree of drill bit aging using drilling signals and neural network models corresponding to different rock types, and for issuing early warnings based on the assessment and analysis results of the degree of drill bit aging.

[0162] Specifically, early warning procedures are implemented based on the assessment and analysis results of the drill bit aging degree, including: Based on the evaluation and analysis results of the drill bit aging degree of the signals of the available analysis and monitoring signal types, it is determined whether all of them belong to severe aging. When the evaluation and analysis results of the drill bit aging degree of the signals of all available analysis and monitoring signal types belong to severe aging, an early warning signal is output.

[0163] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0164] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0165] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A method for processing multi-element signal sensing data while drilling, characterized in that, Specifically, it includes: Using the sensing data of the drilling signal, the formation category is constructed based on the model. According to the trajectory deviation data of different formation categories in the target area, the deviation correction strategy is determined. The deviation correction strategy is used to determine the offset data of the drilling signal. According to the offset data, when it is determined that the vibration monitoring signal in the drilling signal needs to be monitored and analyzed, the next step is performed. Based on the monitoring and processing data of the vibration monitoring signals in the drilling signals, the type of vibration monitoring signal of interest is determined. Based on the type of vibration monitoring signal of interest and the update status of the monitoring and analysis data under different types of vibration monitoring signal of interest, the monitoring and analysis method of the vibration monitoring signal is determined. The available analytical monitoring signal type of the vibration monitoring signal is determined according to the monitoring and analysis method. Based on the available analytical monitoring signal type and the distribution of monitoring and analysis data of different available analytical monitoring signal types, the sensing data processing method of the drilling signal is determined.

2. The drilling multi-element signal sensing data processing method as described in claim 1, characterized in that, The sensing data of the drilling signal is determined based on the monitoring data of different types of drilling monitoring devices.

3. The drilling multi-element signal sensing data processing method as described in claim 1, characterized in that, The process of constructing stratigraphic categories based on the model includes: Based on the sensing data of the drilling signals, the formation categories are constructed using the output of the drilling lithology identification model.

4. The drilling multi-element signal sensing data processing method as described in claim 1, characterized in that, The stratigraphic category identification data is determined based on the stratigraphic category identification results in different depth ranges within the target area.

5. The drilling multi-element signal sensing data processing method as described in claim 1, characterized in that, The method for determining the correction processing strategy is as follows: Based on trajectory deviation data under different formation categories, determine the number of times the positioning trajectory deviates during drilling under the given formation category; The number of times the positioning trajectory deviates during drilling under the aforementioned formation category is taken as the trajectory deviation number under the aforementioned formation category; The trajectory deviation frequency under different stratigraphic categories in the target area is used to determine the correction processing strategy.

6. The drilling multi-element signal sensing data processing method as described in claim 5, characterized in that, The number of trajectory deviations under different stratigraphic categories is used to determine the proportion of trajectory deviation processes under different stratigraphic categories. Combined with the number of deviation stratigraphic categories, the deviation analysis requirement value is determined. It is then determined whether the deviation analysis requirement value is greater than a preset requirement threshold. If so, the correction processing strategy is determined to be that sensor correction processing is required whenever the deviation rate between the temperature in the sensor and the calibration temperature is less than a preset deviation rate threshold to determine whether a deviation has occurred. If not, the correction processing strategy is determined to be that sensor correction processing, i.e., inclinometer correction processing, is required whenever the deviation rate between the temperature in the sensor and the calibration temperature is less than a preset deviation rate threshold and the sensor has not undergone correction processing within the most recent preset time period to determine whether a deviation has occurred.

7. The drilling multi-element signal sensing data processing method as described in claim 6, characterized in that, The calibration temperature is determined based on the temperature calibration results before drilling.

8. The drilling multi-element signal sensing data processing method as described in claim 1, characterized in that, The offset data of the drilling signal is determined based on the deviation between the monitoring data of the inclinometer in the drilling signal and the actual data.

9. The drilling multi-element signal sensing data processing method as described in claim 1, characterized in that, The method for determining the sensing data processing method for the drilling signal is as follows: Based on the available analytical monitoring signal type data, the time period for evaluating and analyzing the aging status of the drill bit is determined and used as the evaluation and analysis time period; Using the distribution data of evaluation and analysis periods under different types of available analytical monitoring signals, the number of evaluation and analysis periods under different types of available analytical monitoring signals is determined, and the monitoring and analysis matching coefficient of the available analytical monitoring signal type is determined by the number of evaluation and analysis periods under the available analytical monitoring signal type. The sensing data processing method for the drilling signal is determined by using the available analytical monitoring signal types and the monitoring analysis matching coefficients for different available analytical monitoring signal types.

10. An early warning system, employing the drilling multi-element signal sensing data processing method according to any one of claims 1-9, characterized in that, Specifically, it includes: Data sensing module, early warning module; The data sensing module is responsible for sensing and processing the drilling signal using the sensing data processing method for the drilling signal. The early warning module is responsible for assessing the degree of drill bit aging using drilling signals and neural network models corresponding to different rock types, and for issuing early warnings based on the assessment and analysis results of the degree of drill bit aging.

Images