Dolomite sanding degree evaluation model construction method based on digital drilling and dolomite sanding degree evaluation model product

Abstract

The invention provides a dolomite sanding degree evaluation model construction method based on digital drilling and a dolomite sanding degree evaluation model product, and relates to the technical field of hydraulic engineering drilling. The dolomite sanding degree evaluation model construction method comprises the following steps: determining drilling parameter data corresponding to different drilling depths in a digital drilling test; determining the quantized values of the sanding degrees of the dolomite samples taken by the drill bit at different drilling depths in the digital drilling test; mapping correspondence is carried out on the drilling parameter data and the quantized value of the sanding degree corresponding to the same drilling depth; and on the basis of a sample set formed by the drilling parameter data and the corresponding sanding degree quantized values, numerical fitting is conducted on the function relation between the drilling parameters and the sanding degree, and a sanding degree evaluation model is obtained. As the drilling parameters are objective, the evaluation result based on the drilling parameters is objective, meanwhile, the sanding degree can be calculated only by substituting the drilling parameter data into the function relation model, the real-time performance is high, and the effect of interpreting the sanding degree while drilling can be achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of water conservancy engineering drilling, and particularly to a method and product for constructing a dolomite sanding degree evaluation model based on digital drilling. Background Art

[0002] Rock sanding refers to the process in which rocks, under specific geological and climatic conditions, gradually become sandy or powdery after long-term physical, chemical, and biological weathering, resulting in the destruction of their original structure and a reduction in strength. The classification of sanding grades needs to consider multiple factors such as the physical properties, chemical composition, structural characteristics, and engineering geological behavior of the rocks. Existing evaluation methods for the degree of rock sanding generally fall into two categories: on-site worker judgment and laboratory judgment.

[0003] (1) On-site worker judgment Visual observation: Observe the color, luster, fissure development of the rock, and whether there are obvious signs of pulverization or sanding.

[0004] Hammer test: Tap the rock gently with a hammer and judge its strength based on the sound and the reaction of the rock (such as whether it is easy to break).

[0005] However, visual observation and hammer test largely rely on the experience and subjective judgment of workers, and different people may draw different conclusions. Moreover, environmental conditions, personal fatigue level, technical level, etc. may all affect the accuracy of the judgment.

[0006] (2) Laboratory judgment Scanning electron microscope: Observe the microstructure of the rock sample and analyze the particle morphology and connection situation.

[0007] Acoustic velocity test: Use an ultrasonic detector to measure the propagation velocity of sound waves in the rock. The acoustic velocity of sanded rocks is relatively low.

[0008] However, laboratory judgment often takes time, and data processing is not fast enough, which is not suitable for rapidly changing construction sites. Laboratory equipment such as scanning electron microscopes and acoustic velocity testers is expensive, and the testing cost is high. Moreover, laboratory tests require professional technical personnel to operate and analyze, and the data analysis stage may also be affected by the subjective judgment of technical personnel.

[0009] Therefore, there is an urgent need to propose an evaluation method for the degree of dolomite sanding that has strong real-time performance and is not affected by human subjective factors. Summary of the Invention

[0010] The present invention provides a method and product for constructing a dolomite sanding degree evaluation model based on digital drilling to solve the defects of strong subjectivity and poor real-time performance in the prior art and achieve real-time and objective evaluation of the degree of dolomite sanding.

[0011] The present invention provides a method for constructing a dolomite sanding degree evaluation model based on digital drilling, comprising the following steps.

[0012] Determine the drilling parameter data corresponding to different drilling depths in the digital drilling test; Determine the quantification values of the sanding degree of the dolomite rock samples taken by the drill bit at different drilling depths in the digital drilling test; Map and correspond the drilling parameter data and the sanding degree quantification values corresponding to the same drilling depth; Based on the sample set composed of the drilling parameter data and the corresponding sanding degree quantification values, numerically fit the functional relationship between the drilling parameters and the sanding degree to obtain a sanding degree evaluation model.

[0013] According to the method for constructing a dolomite sanding degree evaluation model based on digital drilling provided by the present invention, the drilling parameters include at least one of the drilling speed, drilling torque, and drilling pressure.

[0014] According to the method for constructing a dolomite sanding degree evaluation model based on digital drilling provided by the present invention, the drilling parameters include at least two types; Based on the sample set composed of the drilling parameter data and the corresponding sanding degree quantification values, numerically fitting the functional relationship between the drilling parameters and the sanding degree includes: Respectively take the drilling parameter data of different types as independent variables and the corresponding sanding degree quantification values as dependent variables, and numerically fit the relationship between the drilling parameters and the sanding degree to obtain multiple sanding degree evaluation sub-models; Perform weighted averaging on each of the sanding degree evaluation sub-models to obtain the sanding degree evaluation model.

[0015] According to the method for constructing a dolomite sanding degree evaluation model based on digital drilling provided by the present invention, based on the sample set composed of the drilling parameter data and the corresponding sanding degree quantification values, numerically fitting the functional relationship between the drilling parameters and the sanding degree includes: Take the drilling parameter data as the independent variable and the corresponding sanding degree quantification value as the dependent variable, and numerically fit the relationship between the drilling parameters and the sanding degree using various function forms to obtain multiple fitting models; Select the fitting model with the highest fitting degree as the sanding degree evaluation model.

[0016] According to the method for constructing a dolomite sanding degree evaluation model based on digital drilling provided by the present invention, based on the sample set composed of the drilling parameter data and the corresponding sanding degree quantification values, numerically fitting the functional relationship between the drilling parameters and the sanding degree includes: Determine the timing displacement data and the timing drilling parameter data in the digital drilling test; Determine the change rate of the drilling parameter data over time based on the timing drilling parameter data; Use the time points with a change rate greater than the set value as demarcation points to obtain multiple depth intervals; Perform numerical fitting on the functional relationship between the average value, standard deviation or coefficient of variation of the timing drilling parameter data within the depth interval and the corresponding quantification value of the sandification degree.

[0017] According to the method for constructing a dolomite sandification degree evaluation model based on digital drilling provided by the present invention, the drilling parameter data is the data obtained after preprocessing the original drilling data, and the preprocessing steps include: Use multiple outlier detection methods to detect outliers in the original drilling data; Evaluate the accuracy of each outlier detection method; Remove the outliers detected by the outlier detection method with the best outlier detection accuracy from the original drilling data.

[0018] According to the method for constructing a dolomite sandification degree evaluation model based on digital drilling provided by the present invention, the drilling parameter data is the data obtained after preprocessing the original drilling data, and the preprocessing steps include: Based on each filtering path in the filtering path set, filter the original drilling data to obtain multiple groups of filtered data; Select a group of filtered data with the best filtering effect as the preprocessed data.

[0019] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the method for constructing a rock dolomite sandification degree evaluation model as described in any one of the above.

[0020] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the method for constructing a dolomite sandification degree evaluation model based on digital drilling as described in any one of the above.

[0021] The present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the method for constructing a dolomite sandification degree evaluation model based on digital drilling as described in any one of the above.

[0022] The method and product for constructing a dolomite sandification degree evaluation model based on digital drilling provided by the present invention determine the functional relationship model between the drilling parameters and the dolomite sandification degree by numerically fitting the drilling parameter data obtained in the digital drilling test and the quantified value of the dolomite sandification degree. During the drilling operation in the target area, substituting the collected drilling parameter data into the above functional relationship model can obtain the evaluation result of the dolomite sandification degree of the rock. On the one hand, since the drilling parameters are objective, the evaluation result is also objective without the influence of human subjective factors. On the other hand, based on the dolomite sandification degree evaluation model of the present invention, only by substituting the drilling parameter data into the functional relationship model can the sandification degree be calculated, with strong real-time performance, and the effect of interpreting the sandification degree while drilling can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0024] Figure 1 FIG. is a schematic flow chart of the method for constructing a dolomite sandification degree evaluation model based on digital drilling provided by an embodiment of the present invention.

[0025] Figure 2 FIG. is a drilling time history curve graph in the original data of an embodiment of the present invention.

[0026] Figure 3 FIG. is a schematic diagram of time domain region division in an embodiment of the present invention.

[0027] Figure 4 FIG. is a schematic diagram of the dolomite sandification grade calibrated for each region of the rock sample taken by the drill bit in an embodiment of the present invention.

[0028] Figure 5 FIG. is a schematic structural diagram of a dolomite sandification degree evaluation model construction device provided by an embodiment of the present invention.

[0029] Figure 6 FIG. is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] To make the objectives, technical solutions, and advantages of the present invention more clear, the following will, in conjunction with the accompanying drawings in the present invention, clearly and completely describe the technical solutions in the present invention. Apparently, the described embodiments are some, rather than all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without making creative efforts fall within the scope of protection of the present invention.

[0031] The present invention combines digital drilling technology to solve the problems of strong subjectivity and poor real-time performance in the existing determination method for the degree of dolomite sandification. The provided method and product can be mounted on equipment such as geological drills and crawler drills. The following will, in conjunction with Figures 1-4 describe the method for constructing an evaluation model for the degree of dolomite sandification and the evaluation method provided by the embodiments of the present invention.

[0032] Figure 1 is a schematic flowchart of the method for constructing an evaluation model for the degree of dolomite sandification based on digital drilling provided by the embodiments of the present invention. As Figure 1 shown, the method includes the following: Step 101: Determine the drilling parameter data corresponding to different drilling depths in the digital drilling experiment.

[0033] Step 102: Determine the quantification values of the degree of sandification of the dolomite rock samples taken by the drill bit at different drilling depths in the digital drilling experiment.

[0034] Step 103: Map and correspond the drilling parameter data and the quantification values of the degree of sandification corresponding to the same drilling depth.

[0035] Step 104: Based on the sample set composed of the drilling parameter data and the corresponding quantification values of the degree of sandification, perform numerical fitting on the functional relationship between the drilling parameters and the degree of sandification to obtain a sandification degree evaluation model.

[0036] By finding the internal law between the drilling parameters and the degree of dolomite sandification, the embodiments of the present invention evaluate the degree of dolomite sandification based on objective drilling parameter data, reducing the intervention of human factors and making the evaluation results objective. At the same time, the embodiments of the present invention analyze and summarize the relationship between the drilling parameters and the degree of dolomite sandification as a functional relationship, that is, by simply inputting the drilling parameter data into the function, the corresponding degree value of dolomite sandification can be directly obtained, enabling real-time recording and analysis of data during the formation drilling process, with fast feedback and strong real-time performance. Moreover, the digital drilling technology enables the present invention to directly collect drilling parameter data in the in-situ formation without going through complicated processes such as sampling, detection, and analysis to provide results.

[0037] In the embodiment of the present invention, the drilling parameter in step 101 may be at least one of drilling speed, drilling torque, and drilling pressure on bit, or may be at least one of data collected by sensors such as current sensors, acceleration sensors, and vibration sensors during the drilling process.

[0038] Since the present invention obtains the intrinsic relationship between the drilling parameters and the degree of dolomite sandification through numerical fitting, the basic drilling parameter data must be accurate. In order to ensure the authenticity and accuracy of the drilling parameter data, the embodiment of the present invention is designed from multiple aspects such as sensor selection and data cleaning, which are introduced below.

[0039] (1) Sensor design In terms of hardware, the present invention is equipped with a variety of sensors for the down-the-hole drill, which can monitor various physical parameters in the drilling process in real time. Specifically, according to different needs, a variety of sensors and corresponding mounting joints are provided, such as displacement sensors, pressure sensors, torque sensors, current sensors, acceleration sensors, vibration sensors, etc.

[0040] For each sensor, the present invention provides different types. For example, for displacement sensors, wire displacement sensors, laser displacement sensors, magnetostrictive sensors, etc. are provided to ensure that they can adapt to different drilling environments and engineering requirements.

[0041] Taking displacement sensors as an example, there are types of displacement sensors, laser displacement sensors, magnetostrictive sensors, etc. In order to ensure that the selected sensors are suitable for the current drilling environment and engineering requirements, the present invention uses each of the above displacement sensors to collect displacement data during the early drilling test. For the displacement data collected by different displacement sensors, the mean absolute deviation (MAD) is used to judge the degree of discreteness of the data. The larger the mean absolute deviation, the larger the fluctuation range of the data, and the worse the effect. Therefore, the present invention selects the displacement data with the smallest MAD as the basic data for subsequent applications.

[0042] In terms of sensor data transmission, the present invention adopts a unified data collection method, namely a wireless transmission method, to solve the problem of inconsistent data collection methods and interfaces of different sensors, thereby simplifying the complexity of data integration and equipment management.

[0043] (2) Data cleaning design The original drilling data is usually disorganized, so it is critical to remove abnormal data points caused by equipment failure, environmental interference and other factors and extract drilling data in the stable stage. In addition, the noise caused by mechanical vibration during the drilling process also needs to be filtered and reduced.

[0044] 1. Rejection of Abnormal Data Points The present invention uses a variety of outlier detection methods to detect outliers, such as embedding a variety of statistical methods and machine learning methods, including box plots, Z-score methods, DBSCAN clustering methods, isolation forest methods, etc.

[0045] To determine the most suitable method, in the early stage of the embodiments of the present invention, a certain proportion of outliers are artificially set in the normal data. These outliers can represent the abnormal types that may be encountered in actual applications, and they are used to detect the ability of different methods to identify outliers. The present invention calculates the accuracy rate of each outlier detection method, that is, the proportion of correctly identified outliers. Finally, the outlier detection method with the highest accuracy rate is selected as the final abnormal data identification method, and the abnormal data points identified by it are removed.

[0046] 2. Extraction of Drilling Data in the Stable Stage The complete drilling process can be subdivided into three stages: the preparation stage, the stable drilling stage, and the drill rig lifting stage. The drilling displacement in the original data is as Figure 2 shown, Figure 2 where (a) in Figure 2 represents the preparation stage, Figure 2 where (b) in

[0047] represents the stable drilling stage, (1) (2) (3) (4) In the formula, N is the drilling speed of the drill rig in the real-time process, and P is the impact pressure of the drill rig in the real-time process; and are the drilling displacements corresponding to the starting point and the ending point of the stable drilling stage of the c-th displacement sensor (m), respectively, and c represents the number of displacement sensors; is the actual length of the drill pipe, m, which is adjusted according to the telescopic displacement range of the power head of the on-site equipment; is the maximum drilling speed, r / min, which is adjusted according to the rotational speed range of the power head of the on-site equipment; is the maximum impact pressure, MPa, which is adjusted according to the impact pressure range of the on-site equipment.

[0048] Taking the displacement sensor as an example, the drilling displacement in Equation (2) includes the following contents: 1) Whether the displacement data obtained by various types of displacement sensors, such as wire-drawing displacement sensors, laser displacement sensors, and magnetostrictive sensors, is monotonically decreasing or monotonically increasing, that is: Or (5) t is the drilling time.

[0049] 2) After they are all monotonically decreasing, the mean absolute deviation (MAD) is used to judge the dispersion degree of the data. The larger the mean absolute deviation, the larger the fluctuation range of the data, and the worse the effect.

[0050] (6) is the mean value; N is the number of data points collected by a single displacement sensor; x i is each data point collected by a single displacement sensor.

[0051] When the data of various types of displacement sensors are all monotonically decreasing and the mean absolute deviation of the data can be guaranteed within a certain range, it can be determined that this data segment is the data of the stable drilling stage.

[0052] 3. Filtering and noise reduction For the filtering and noise reduction of the data, the present invention adopts a multi-dimensional noise filtering technology, combines a variety of digital filtering technologies, realizes multi-dimensional suppression of noises such as mechanical vibration, improves the signal-to-noise ratio of the data, and can more effectively eliminate high-frequency noises and random interferences.

[0053] Specifically, first construct a set of filtering paths, then, based on each filtering path in the set of filtering paths, filter the original drilling data to obtain multiple groups of filtered data, and finally, select a group of filtered data with the best filtering effect as the basic data for subsequent applications, for example, as the drilling parameter data determined in step 101.

[0054] It should be noted that each filtering path includes multi-dimensional noise filtering, where multi-dimensional refers to multiple dimensions such as the time domain, frequency domain, and spatial domain. Each filtering path is the permutation and combination of the above multi-dimensional noise filtering methods. Specifically, the permutation and combination methods can be: time domain filtering - frequency domain filtering - spatial domain filtering; time domain filtering - spatial domain filtering - frequency domain filtering; frequency domain filtering - time domain filtering - spatial domain filtering; frequency domain filtering - spatial domain filtering - time domain filtering; spatial domain filtering - time domain filtering - frequency domain filtering; spatial domain filtering - frequency domain filtering - time domain filtering; time domain filtering - frequency domain filtering; frequency domain filtering - time domain filtering; frequency domain filtering - spatial domain filtering; spatial domain filtering - frequency domain filtering; time domain filtering - spatial domain filtering; spatial domain filtering - time domain filtering.

[0055] Taking the filtering path of time domain filtering - frequency domain filtering - spatial domain filtering as an example, the filtering of drilling displacement data will be described below.

[0056] 1) To establish a filtering and noise reduction method suitable for drilling data, Gaussian noise and Levy (non-Gaussian noise) with different signal-to-noise ratios are artificially added to the standard drilling data.

[0057] The standard drilling data here is the smoothest, least noisy, and most ideal section of drilling displacement data selected from the drilling displacement data collected during the drilling process.

[0058] 2) Use moving average filtering or median filtering, etc. to perform time domain filtering on the noise.

[0059] 3) Perform Fourier transform to convert the time domain signal to the frequency domain, and then use a band-pass / band-stop filter to allow or block signals within a specific frequency range to pass through. Specifically, a Butterworth filter can be used.

[0060] 4) Perform wavelet transform on the data to analyze the signal at multiple scales and achieve spatial domain filtering of drilling parameters. It should be noted that for drilling displacement data, due to its linear or stationary characteristics, the signal changes do not occur at multiple scales, and wavelet transform may not provide additional useful information, so wavelet transform is no longer used for drilling displacement data.

[0061] After filtering the drilling parameter data based on each filtering path, it is compared with the standard drilling data, and the root mean square error (RMSE) is used to evaluate the noise reduction effect of each filtering path, and the optimal drilling parameter filtering path is selected.

[0062] The optimal noise reduction path of the finally selected drilling displacement data is as follows in the table.

[0063] For the drilling data collected by other sensors such as drilling speed data, drilling torque data, drilling pressure data, current data, and vibration data, the cleaning process is the same as that of the above-mentioned drilling displacement data, and thus will not be elaborated herein.

[0064] The above is the cleaning process of the original drilling data in the digital drilling test. After obtaining the cleaned drilling parameter data, it is necessary to determine the drilling parameter data corresponding to different drilling depths. Next, the mapping correspondence process between the drilling parameter data and the quantification value of dolomite sandification degree will be introduced.

[0065] It should be noted that the idea of this article is to map and correspond the drilling parameter data with the quantification value of dolomite sandification degree based on the drilling depth. Specifically, the present invention first corresponds the drilling parameter data with the drilling depth, and corresponds the drilling depth with the quantification value of dolomite sandification degree, and then finally maps and corresponds the drilling parameter data with the quantification value of dolomite sandification degree based on the drilling depth.

[0066] Taking the mapping correspondence between the drilling speed data and the drilling depth as an example, first, based on the time-series displacement data and time-series drilling speed data in the digital drilling test, the drilling speed data corresponding to different drilling depths can be determined; then, based on the rock samples drilled by the drill rig, the dolomite sandification grades corresponding to different drilling depths can be determined; finally, we map and correspond the drilling speed data and the dolomite sandification grade data at the same drilling depth.

[0067] It should be noted that when quantifying the dolomite sandification grade, the quantification rule adopted is that the weak sandification grade is represented by 1, the medium sandification grade is represented by 2, the strong sandification grade is represented by 3, and the extremely strong sandification grade is represented by 4, and so on. It can be understood that in terms of the drilling depth, each sandification grade of the rock sample has a certain depth interval, so that there will be an entire drilling depth interval corresponding to a sandification grade quantification value. For example, the entire drilling depth interval with a drilling depth of [40m, 43m] corresponds to the medium sandification grade quantification value 2, and there are countless drilling parameter data (such as drilling speed data) in this entire drilling depth interval. In this way, there will be countless drilling parameter data corresponding to one sandification grade quantification value, which is not conducive to the subsequent numerical fitting work. To solve this problem, the present invention takes the mean value, standard deviation or coefficient of variation of the drilling parameter data within the depth interval as the drilling parameter data mapped and corresponding to the sandification grade quantification value, so as to achieve a one-to-one mapping correspondence between the drilling parameter data and the sandification degree quantification value, which is convenient for subsequent numerical fitting.

[0068] The above-mentioned drilling depth interval was mentioned above. Next, the present invention will introduce the division of the drilling depth interval.

[0069] The division of the drilling depth interval is based on the drilling parameter data. Taking the drilling speed data as an example, the specific process is as follows: (1) Extract the drilling time history curve (time series displacement data) During a single drilling process, due to the limited length of the drill pipe, the drilling displacement is in segments. Therefore, it is necessary to splice the drilling displacement data of different drill pipes. First, ensure that the timestamps of the displacement data of each drill pipe are aligned. Then, dock the starting displacement data of the next drill pipe with the ending displacement data of the previous drill pipe. Check whether there are sudden changes or discontinuity points in the spliced displacement data, and perform necessary smoothing processing (such as moving average or low-pass filter) to reduce the noise caused by drill pipe replacement. Finally, rearrange the spliced data in chronological order to ensure the continuity of the time axis of the time history curve. According to the time series and the corresponding displacement data, a continuous drilling time history (drilling depth - time) curve is plotted.

[0070] (2) Division of depth intervals The present invention divides the depth interval based on the rate of change of the drilling parameters over time. When the drilling parameter is the drilling speed, the drilling speed data can be obtained based on the slope of the drilling time history curve, and the rate of change of the drilling speed over time is the rate of change of the slope of the drilling time history curve over time.

[0071] The present invention uses an adaptive slope detection algorithm to detect the change in the slope of the drilling time history curve. This method can identify the regions where the slope changes significantly. The drilling process (drilling displacement or drilling depth) is divided at the significant change regions as the demarcation points.

[0072] Specifically: First, set an initial threshold (which can be determined according to the law of the short-distance drilling data in the early stage), and select a suitable time window (the time window is also calibrated according to the short-distance drilling law in the early stage) to calculate the rate of change of the drilling depth. Then, obtain the drilling displacement data in real time, and calculate the rate of change of the slope of the drilling displacement within the selected time window. Compare the difference between the slope calculated in real time and the current threshold (such as 5%). If the absolute value of the slope change exceeds the current threshold, this point is the demarcation point. The time window continues to roll the data and continues to detect new demarcation points until the drilling process stops.

[0073] The above is an illustration taking the drilling speed as an example. The same principle applies to other drilling parameters and will not be elaborated here.

[0074] The process of determining the quantization value of the degree of sandification of the dolomite rock sample taken by the drill bit at different drilling depths in the digital drilling test in step 102 of the present invention is actually the calibration process of the degree of sandification of dolomite. Figure 4 For Figure 3 the schematic diagram of the dolomite sandification grade calibrated for each region in, the following introduces this calibration process.

[0075] First, during the drilling process, a high-precision coring bit is used to ensure the acquisition of complete and undisturbed rock samples. The on-site workers quickly judge the sanding level based on the appearance, hardness and other characteristics of the dolomite rock samples. Then, the samples of different sanding levels judged by the workers are subjected to experiments such as indoor electron microscopy scanning to identify the sanding characteristics. Finally, based on the quantitative analysis of the indoor experiments and the qualitative judgment of the on-site workers, the accurate evaluation of the rock sanding level is realized. If there is a difference between the two judgments, different weights are assigned to the judgment results of the on-site workers and the indoor experiment results, or rock samples with other representative sanding levels are selected for re-evaluation. Finally, the weak sanding level corresponding to different drilling depths of the rock samples is obtained, and the sanding level is quantified. The quantification rule of this embodiment is that the weak sanding level is represented by 1, the medium sanding level is represented by 2, the strong sanding level is represented by 3, the extremely strong sanding level is represented by 4, and so on.

[0076] Next, the numerical fitting process of the drilling parameter data and the corresponding quantified sanding degree value in step 104 of the present invention will be introduced.

[0077] Taking the fitting of the drilling speed data and the quantified sanding degree value as an example, during numerical fitting, the drilling speed data is used as the independent variable, and the corresponding quantified sanding degree value is used as the dependent variable to fit the functional relationship between the drilling speed and the sanding degree, and a sanding degree evaluation model is obtained.

[0078] In some embodiments of the present invention, to improve the accuracy of the sanding degree evaluation model, not only one drilling parameter is used to determine the model, but two or more drilling parameters can be used to determine the model. The general idea is to perform numerical fitting between each drilling parameter data and the quantified sanding degree value respectively, and then perform weighted averaging on the fitting results to obtain the final sanding degree evaluation model.

[0079] Specifically, taking the two drilling parameters of drilling speed and drilling pressure as an example, the two drilling parameters are used as independent variables respectively, and the corresponding quantified sanding degree value is used as the dependent variable to construct sanding degree evaluation sub-models respectively.

[0080] First, taking the drilling speed data as the independent variable and the corresponding quantified sanding degree value as the dependent variable, the functional relationship between the drilling speed and the sanding degree is fitted to obtain the first sanding degree evaluation sub-model. Taking the average drilling pressure data as the independent variable and the corresponding quantified sanding degree value as the dependent variable, the relationship between the average drilling pressure and the sanding degree is numerically fitted to obtain the second sanding degree evaluation sub-model. Taking the standard deviation data of the drilling pressure as the independent variable and the corresponding quantified sanding degree value as the dependent variable, the relationship between the standard deviation of the drilling pressure and the sanding degree is numerically fitted to obtain the third sanding degree evaluation sub-model.

[0081] Then, the above sanding degree evaluation sub-models are weighted and averaged to obtain the final sanding degree evaluation model.

[0082] It should be noted that, in some embodiments of the present invention, the selection of drilling parameters is based on the magnitude of their correlation with the sanding degree (i.e., the magnitude of their influence on the sanding degree evaluation result). The number of selected parameters can be an odd number to ensure the robustness of the decision-making and avoid the equalization of weight distribution. As for the magnitude of the influence of these parameters on the sanding degree evaluation result, based on the drilling parameter data collected from the digital drilling test, we can use two evaluation indicators, namely the mean square error (MSE) and the coefficient of determination (R²), to quantify the influence degree of each parameter on the sanding grade.

[0083] In addition, the weights of the sanding degree evaluation sub-models are also determined based on the magnitude of the influence of each drilling parameter on sanding. It can be understood that the weight is a value between 0 and 1, and the sum of the weights is equal to 1.

[0084] It should be noted that the present invention uses the drilling data of multiple sensors to separately fit the sanding degree, which also belongs to a redundant design, that is, when a certain sensor fails and stops working, the sanding degree can still be interpreted. For example, as described above, three sanding degree evaluation sub-models are respectively fitted based on the drilling speed, the mean drilling pressure, and the standard deviation of the drilling pressure. If the pressure sensor fails, the sanding degree can be interpreted only based on the sanding degree evaluation sub-model fitted by the drilling speed. If the displacement sensor fails, the sanding degree evaluation sub-models fitted by the mean drilling pressure and the standard deviation of the drilling pressure can be weighted to achieve the interpretation of the sanding degree.

[0085] In some embodiments of the present invention, in order to more objectively and accurately explore the relationship law between the drilling parameters and the sanding grade, when fitting the relationship between the drilling parameter data and the quantified value of the sanding degree, multiple function forms are used for fitting. Specifically, for each drilling parameter, with the drilling parameter data as the independent variable and the corresponding quantified value of the sanding degree as the dependent variable, multiple function forms are used to numerically fit the relationship between the drilling parameter and the sanding degree, obtaining multiple fitting models, and finally the fitting function with the highest fitting degree is selected as the sanding degree evaluation model of the drilling parameter.

[0086] It should be noted that the above function forms can be exponential form, power function form, logarithmic function form, multiple linear function form, etc. In order to objectively evaluate the effectiveness of these function forms, this embodiment introduces multiple scientific evaluation indicators such as the R² goodness of fit and the mean square error (MSE). Through the comparative analysis of these indicators, the function fitting model with the highest goodness of fit or the smallest mean square error is selected.

[0087] The above solution realizes the construction of the dolomite sanding degree evaluation model. During the drilling process, the drilling parameter data can be directly input into the dolomite sanding degree evaluation model, and the corresponding dolomite sanding degree value can be directly obtained, achieving the effect of interpreting the dolomite sanding degree while drilling.

[0088] For the output of the dolomite sanding degree evaluation model, the present invention adopts the interval division method. For example, if the output is between 0.0 - 1.5, it indicates that the sanding degree is weak sanding, representing the area with the lowest sanding degree; if the output is between 1.5 - 2.5, it indicates that the sanding degree is medium sanding, representing the medium level of sanding degree; if the output is between 2.5 - 3.0, it indicates that the sanding degree is strong sanding, representing the most severe sanding degree. It should be noted that the interval demarcation points of the above intervals can be adjusted adaptively according to the actual situation.

[0089] Since the sanding degree evaluation model is obtained by fitting the data from digital drilling tests, the geological characteristics of the actual target area may be different from those of the test site. To ensure the applicability of the sanding degree evaluation model, in some embodiments of the present invention, the above sanding degree evaluation model is continuously iteratively optimized. For example, based on the field data (drilling parameter data of the target area and the corresponding quantified value of the actual sanding degree of the rock sample), the sanding degree evaluation model is fitted and adjusted.

[0090] Specifically, when we actually conduct borehole drilling at the target area site, by monitoring the evaluation results of the sanding degree evaluation model and comparing them with the actual sanding grades, if there is a large error, it indicates that the evaluation model has performance shortcomings under different geological conditions. At this time, we add the drilling parameter data collected on-site currently and its corresponding actual sanding grade to the parameter sample library when constructing the sanding degree evaluation model (composed of the drilling parameter data collected in digital drilling tests and their corresponding quantified sanding degree values), and then perform numerical fitting again according to the updated parameter sample library to retrain the sanding degree evaluation model. Then, the optimized model is redeployed to the site for the next round of real-time data analysis.

[0091] Next, the device for constructing the rock dolomite sanding degree evaluation model provided by the present invention will be described. The device for constructing the rock dolomite sanding degree evaluation model described below can be correspondingly referred to the method for constructing the rock dolomite sanding degree evaluation model described above. Refer to Figure 5 , the device for constructing the rock dolomite sanding degree evaluation model includes: A drilling parameter determination module 501, configured to determine the drilling parameter data corresponding to different drilling depths in digital drilling tests.

[0092] The sanding degree determination module 502 is configured to determine the quantified values of the sanding degrees of the dolomite rock samples taken by the drill bit in the digital drilling test at different drilling depths.

[0093] The data mapping module 503 is configured to map and correspond the drilling parameter data and the quantified values of the sanding degrees corresponding to the same drilling depth.

[0094] The numerical fitting module 504 is configured to perform numerical fitting on the drilling parameter data and the corresponding quantified values of the sanding degrees to obtain a sanding degree evaluation model; the sanding degree evaluation model is used to describe the functional relationship between the drilling parameters and the sanding degree.

[0095] Figure 6 An entity structure schematic diagram of an electronic device is exemplified, as Figure 6 shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630, and a communication bus 640. Among them, the processor 610, the communication interface 620, and the memory 630 complete communication with each other through the communication bus 640. The processor 610 may call the logical instructions in the memory 630 to execute the method for constructing a sanding degree evaluation model of dolomite rock.

[0096] In addition, when the logical instructions in the above-mentioned memory 630 are implemented in the form of software functional units and sold or used as an independent product, they may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc that can store program codes.

[0097] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the method for constructing a sanding degree evaluation model of dolomite rock provided by the above-mentioned various methods.

[0098] The method for constructing an evaluation model for the degree of dolomite sandification of the rock includes: determining the drilling parameter data corresponding to different drilling depths in the digital drilling test; determining the quantification values of the degree of sandification of the dolomite rock samples taken by the drill bit at different drilling depths in the digital drilling test; performing a mapping correspondence between the drilling parameter data and the quantification values of the degree of sandification corresponding to the same drilling depth; performing numerical fitting on the drilling parameter data and the corresponding quantification values of the degree of sandification to obtain an evaluation model for the degree of sandification; and the evaluation model for the degree of sandification is used to describe the functional relationship between the drilling parameters and the degree of sandification.

[0099] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it is configured to execute the method for constructing an evaluation model for the degree of dolomite sandification of the rock provided by the above-mentioned various methods.

[0100] The method for constructing an evaluation model for the degree of dolomite sandification of the rock includes: determining the drilling parameter data corresponding to different drilling depths in the digital drilling test; determining the quantification values of the degree of sandification of the dolomite rock samples taken by the drill bit at different drilling depths in the digital drilling test; performing a mapping correspondence between the drilling parameter data and the quantification values of the degree of sandification corresponding to the same drilling depth; performing numerical fitting on the drilling parameter data and the corresponding quantification values of the degree of sandification to obtain an evaluation model for the degree of sandification; and the evaluation model for the degree of sandification is used to describe the functional relationship between the drilling parameters and the degree of sandification.

[0101] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0102] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solutions, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disc, etc., including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a dolomite sandification degree evaluation model based on digital drilling, characterized in that: include: Determine the drilling parameter data corresponding to different drilling depths in the digital drilling test; Determine the quantitative value of the degree of sandification of the dolomite rock sample taken by the drill bit at different drilling depths in the digital drilling test; Mapping the drilling parameter data and the quantified value of the degree of sandification corresponding to the same drilling depth; Based on the sample set consisting of the drilling parameter data and the corresponding sandification degree quantified values, numerical fitting is performed on the functional relationship between the drilling parameter and the sandification degree to obtain a sandification degree evaluation model.

2. The method for constructing a dolomite sandification degree evaluation model based on digital drilling according to claim 1, characterized in that: The drilling parameters include at least one of drilling speed, drilling torque, and drilling pressure on bit.

3. The method for constructing a dolomite sandification degree evaluation model based on digital drilling according to claim 1, characterized in that: The drilling parameters include at least two types; Based on the sample set consisting of the drilling parameter data and the corresponding sandification degree quantified values, numerical fitting is performed on the functional relationship between the drilling parameter and the sandification degree, including: Different types of drilling parameter data are used as independent variables, and the corresponding sandification degree quantification value is used as the dependent variable. The relationship between drilling parameters and sandification degree is numerically fitted to obtain multiple sandification degree evaluation sub-models. The sandification degree evaluation model is obtained by performing weighted averaging on each of the sandification degree evaluation sub-models.

4. The method for constructing a dolomite sandification degree evaluation model based on digital drilling according to claim 1, characterized in that: Based on the sample set consisting of the drilling parameter data and the corresponding sandification degree quantified values, numerical fitting is performed on the functional relationship between the drilling parameter and the sandification degree, including: Taking the drilling parameter data as the independent variable and the corresponding sandification degree quantified value as the dependent variable, a plurality of function forms are used to numerically fit the relationship between the drilling parameter and the sandification degree to obtain a plurality of fitting models; The fitting model with the highest fitting degree is selected as the sandification degree evaluation model.

5. The method for constructing a dolomite sandification degree evaluation model based on digital drilling according to claim 1, characterized in that: Based on the sample set consisting of the drilling parameter data and the corresponding sandification degree quantified values, numerical fitting is performed on the functional relationship between the drilling parameter and the sandification degree, including: Determine the time series displacement data and time series drilling parameter data in the digital drilling test; determining a rate of change of the drilling parameter data over time based on the time-series drilling parameter data; The time point when the rate of change is greater than the set value is used as the dividing point to obtain multiple depth intervals; A numerical fit is performed on the functional relationship between the mean value, standard deviation or coefficient of variation of the time series drilling parameter data within the depth interval and the corresponding quantified value of the degree of sandification.

6. The method for constructing a dolomite sandification degree evaluation model based on digital drilling according to claim 1, characterized in that: The drilling parameter data is the data obtained by preprocessing the original drilling data, and the preprocessing step includes: Using a variety of outlier detection methods to detect outliers on the original drilling data; Evaluate the accuracy of each outlier detection method; The abnormal points detected by the abnormal value detection method with the best abnormal point detection accuracy are removed from the original drilling data.

7. The method for constructing a dolomite sandification degree evaluation model based on digital drilling according to claim 1, characterized in that: The drilling parameter data is the data obtained by preprocessing the original drilling data, and the preprocessing step includes: Based on each filter path in the filter path set, the original drilling data is filtered to obtain multiple groups of filtered data; A set of filtered data with the best filtering effect is selected as the preprocessed data.

8. The method for constructing a dolomite sandification degree evaluation model based on digital drilling according to claim 1, characterized in that: Also includes: Obtain the drilling parameter data of the target area and the corresponding quantitative value of the actual sandification degree of the rock sample to obtain the field data; The field data is added to the sample set, and the sandification degree evaluation model is optimized based on the fitting of the updated sample set.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the program, the method for constructing a dolomite sandification degree evaluation model based on digital drilling as described in any one of claims 1 to 8 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for constructing a dolomite sandification degree evaluation model based on digital drilling as claimed in any one of claims 1 to 8 is implemented.

Images