A geological exploration sampling device and sampling analysis method

Abstract

The present invention relates to the field of geological exploration sampling and analysis technology, and specifically to a geological exploration sampling device and a sampling and analysis method. The present invention first obtains geological exploration data of a target sampling position; further obtains a local time domain window of the target acquisition moment; further obtains the local velocity change rate, the number of rock layers, and the torque accumulation parameter in the local time domain window at each acquisition moment, and forms a feature vector of the target sampling position in the time sequence of the sampling moment; further obtains the data anomaly coefficient of the target sampling position based on the difference characteristics of the feature vector of the target sampling position and the feature vectors of other sampling positions, combined with the distance between the target sampling position and the other sampling positions; finally, according to the outlier distribution characteristics of the data anomaly coefficients of all sampling positions, the abnormal sampling positions are screened out.

Description

Technical Field

[0001] The present invention relates to the technical field of geological exploration sampling and analysis, and in particular to a geological exploration sampling device and a sampling and analysis method. Background Art

[0002] When conducting geological exploration sampling, rock and soil samples at different locations and depths have different characteristics. Before actually conducting geological exploration sampling, we first obtain a three-dimensional map of the mission target location based on the mission objectives and geological conditions, and then use the grid point method on the topographic geological map to determine the sampling location based on geological conditions, mission objectives, etc. Considering the huge consumption of manpower, material resources, and financial resources in the geological exploration process, the layout of exploration points is usually sparse, so the sample data of each sampling location is very important.

[0003] However, in the actual survey process, the geological conditions in the field are changeable, and it is difficult to judge the representativeness of the samples after sampling at the arranged sampling points. For example, if a sampling point is located in a rock fissure, then the sample data collected is difficult to correctly represent the rock characteristics of the geographical location. The existing method is to compare the differences based on the characteristic data of other sampling points on the sampling grid. However, the geological conditions in the mountainous areas in the field are changeable, and the rock characteristics data of different sampling points themselves vary greatly. This method is difficult to make accurate judgments, and it is impossible to accurately judge abnormal sampling locations with low data representativeness. Summary of the invention

[0004] In order to solve the technical problem of inaccurate judgment of abnormal sampling positions in geological exploration, the purpose of the present invention is to provide a geological exploration sampling device and a sampling analysis method. The technical scheme adopted is as follows:

[0005] A geological exploration sampling and analysis method, the method comprising:

[0006] Select any sampling position as a target sampling position; obtain geological exploration data of the target sampling position; the geological exploration data at least includes drilling speed and rotary torque;

[0007] Select any acquisition time as the target acquisition time; take the target acquisition time as the end point of the local time domain window, select the time points to be analyzed from right to left in the historical acquisition time of the target acquisition time, until the change value of the rock drilling speed between the first time point to be analyzed and the first acquisition time is less than the average change value of the rock drilling speed between the time point to be analyzed and the target acquisition time, take the corresponding time point to be analyzed as the starting end point of the local time domain window of the target acquisition time, and obtain the local time domain window of the target acquisition time;

[0008] According to the variation characteristics of the rock drilling speed in the local time domain window at each acquisition moment, the local speed variation rate at each acquisition moment is obtained; according to the distribution characteristics of the local speed variation rates at all the acquisition moments, in combination with the corresponding local time domain window, the number of rock layers at each sampling moment is obtained; in the continuous time domain of the same rock layer, according to the variation characteristics of the rotary torque, the torque accumulation parameter at each acquisition moment is obtained; the rock drilling speed, the local speed variation rate, the torque accumulation parameter and the number of rock layers at all the sampling moments of the target sampling position are combined in the time sequence of the sampling moments to form a feature vector of the target sampling position;

[0009] According to the difference characteristics between the feature vector of the target sampling position and the feature vectors of other sampling positions, combined with the distance between the target sampling position and other sampling positions, the data anomaly coefficient of the target sampling position is obtained; according to the outlier distribution characteristics of the data anomaly coefficients of all sampling positions, the abnormal sampling positions are screened out.

[0010] Furthermore, the method for obtaining the data anomaly coefficient includes:

[0011] Obtaining difference parameters between the feature vector at the target sampling position and the feature vectors at other sampling positions by using a DTW algorithm;

[0012] The data anomaly coefficient of the target sampling position is obtained according to the distance between the target sampling position and other sampling positions and the difference parameter; the distance is negatively correlated with the data anomaly coefficient; and the difference parameter is positively correlated with the data anomaly coefficient.

[0013] Furthermore, the method for obtaining the difference parameter includes:

[0014] Among other sampling positions of the target sampling position, select any other sampling position as a comparison sampling position;

[0015] The DTW distance and value between all the same-dimensional data of the feature vector of the target sampling position and the feature vector of the comparison sampling position are obtained as the difference parameter between the target sampling position and the comparison sampling position.

[0016] Furthermore, the method for obtaining the abnormal sampling position includes:

[0017] The outlier factor of each data anomaly coefficient is obtained by the LOF algorithm. When the outlier factor is greater than a preset outlier threshold, the corresponding sampling position is determined to be an abnormal sampling position.

[0018] Furthermore, the method for obtaining the local speed change rate includes:

[0019] In the local time domain window at the target acquisition moment, the absolute value of the difference between the rock drilling speed at the starting point and the rock drilling speed at the end point is taken as the speed change value corresponding to the target acquisition moment; and the local speed change rate at the target acquisition moment is obtained according to the speed change value and the duration of the local time domain window.

[0020] Furthermore, the method for obtaining the number of rock layers includes:

[0021] Among the local velocity change rates at all acquisition moments of the target sampling position, a local time domain window in which the local velocity change rate is greater than the average of all the local velocity change rates is recorded as a layered time domain window; a velocity change curve of all the rock drilling velocities at the target sampling position is obtained; and a time point corresponding to the maximum absolute value of the slope of the velocity change curve in each layered time domain window is taken as a layered time point;

[0022] The period between the first acquisition moment and the first stratification time point, the period between the last stratification time point and the last acquisition moment, and the period between the sequentially adjacent stratification time points constitutes a rock layer, and the number of rock layers at each acquisition moment is obtained.

[0023] Furthermore, the method for obtaining the torque accumulation parameter includes:

[0024] Select any rock layer as the rock layer to be analyzed; select any acquisition time in the rock layer to be analyzed as the time to be analyzed; in the local time domain window corresponding to the rock layer to be analyzed, the acquisition time before the time to be analyzed is used as the accumulation time of the time to be analyzed;

[0025] The torque accumulation parameter of the time to be analyzed is obtained according to the difference characteristics between the rotational torque at the time to be analyzed and the rotational torque at each accumulated time; the difference characteristics between the rotational torque at the time to be analyzed and the rotational torque at each accumulated time are positively correlated with the torque accumulation parameter.

[0026] Furthermore, the positive direction in the time domain is right.

[0027] Furthermore, the preset outlier threshold is 1.

[0028] An embodiment of the present invention further provides a geological exploration sampling device, comprising a device body, wherein the device implements any one of the steps of the geological exploration sampling and analysis method.

[0029] The present invention has the following beneficial effects:

[0030] The present invention first obtains geological exploration data at the target sampling position to provide a data basis for subsequent analysis; further utilizes the different physical properties of different rock layers, which causes large fluctuations in drilling speed, to obtain a local time domain window at the target acquisition time, which is convenient for subsequent distinction between different rock layers, and helps to analyze the characteristics of the rock layers and identify the distribution of the rock layers; further obtains the local velocity change rate at each acquisition time, which can more accurately determine the hierarchical structure and changes of the rock layers, and helps to analyze the distribution of underground rocks at the target sampling position, and at the same time prepares for obtaining the number of rock layers; further utilizes the distribution characteristics of the local velocity change rate at the acquisition time to reveal the distribution of geological layers, and combines the corresponding local time domain window to locate the stratification point of the rock layer, and obtain the number of rock layers at each sampling time; further obtains the torque cumulative parameter at each acquisition time The number characterizes the characteristic changes of the rock formation, and from the perspective of the accumulation of rotary torque, provides more quantitative information for the subsequent comparison of the sampling data between different sampling positions, and more accurately screens out abnormal sampling positions; further constructs the characteristic vector of the target sampling position, and in the form of a characteristic vector, the data of multiple dimensions are uniformly and intuitively represented, which is convenient for the subsequent comparison of the differences between different sampling positions; further, according to the difference characteristics between the characteristic vector of the target sampling position and the characteristic vectors of other sampling positions, combined with the distance between the target sampling position and other sampling positions, the data anomaly coefficient of the target sampling position is obtained, and the difference in sampling data between the target sampling position and other sampling positions is characterized, which provides a basis for finally obtaining the abnormal sampling position; according to the outlier distribution characteristics of the data anomaly coefficients of all sampling positions, the abnormal sampling positions are screened out. The present invention constructs the characteristic vector of the sampling position by analyzing the characteristics of the rock drilling speed, the local speed change rate, the torque accumulation parameter, etc., accurately represents the rock formation structure, screens out the abnormal sampling position by comparing the difference of the characteristic vector and combining the distance of the sampling position, and optimizes the identification of the abnormal position. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 A schematic diagram of a geological drilling rig sampling process provided by an embodiment of the present invention;

[0033] Figure 2 A flow chart of a geological exploration sampling and analysis method provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a geological exploration sampling device and a sampling and analysis method proposed by the present invention, its specific implementation, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0035] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0036] The specific scheme of a geological exploration sampling device and a sampling and analysis method provided by the present invention is described in detail below with reference to the accompanying drawings.

[0037] The embodiment of the present invention provides a geological exploration sampling device, which includes a drilling parameter detection system, a geological drilling rig and a data analysis system.

[0038] The drilling parameter detection system is used to obtain various data of the drilling rig during the geological exploration process, including but not limited to drilling depth, drilling pressure, drilling speed, rotation speed, rotary torque, drilling rig electric power, in-well mud flow, mud pump pressure, out-well mud temperature, in-well mud temperature, total volume of mud pool, big hook load (drilling), outlet combustible gas and other parameters, and display the parameters in real time on the driller's display in digital or curve form. All data can be saved and recorded, and users can browse and print historical data in full.

[0039] The geological drilling rig includes the drilling rig body, drill rod, drill bit, drilling rig power system and cement slurry circulation system, which is used to perform geological exploration tasks. Figure 1 , which shows a schematic diagram of a geological drilling rig sampling process provided by an embodiment of the present invention, Figure 1 The data includes a drilling rig, a drill pipe, a first rock layer, a second rock layer, a third rock layer, a fourth rock layer, a rock layer density identifier, and a sampling location identifier.

[0040] The data analysis system is used to analyze the sampling data, implement the steps in a geological exploration sampling analysis method, and screen out abnormal sampling locations.

[0041] See also Figure 2 , which shows a flow chart of a geological exploration sampling and analysis method provided by an embodiment of the present invention, specifically comprising:

[0042] Step S1: select any sampling position as a target sampling position; obtain geological exploration data of the target sampling position; the geological exploration data at least includes drilling speed and rotary torque.

[0043] When conducting a mineral resource assessment, it is necessary to survey the mineral resources at the target location. During the survey, grid points of different densities will be arranged according to the different task objectives of the survey stage, and different sampling positions will be set. Any sampling position is selected as the target sampling position, and the sampling positions are analyzed one by one; the geological exploration data of the target sampling position is obtained to provide a data basis for subsequent analysis; the geological exploration data at least includes drilling speed and rotational torque.

[0044] It should be noted that the density requirements for grid points are different in different real-time scenarios. The specific grid point parameters are no longer limited here. When comparing the data anomaly characteristics between sampling locations in the subsequent comparison, the sampling locations of the same sampling batch are compared.

[0045] It should be noted that, in the embodiment of the present invention, the acquisition frequency of each geological exploration data is 1 Hz; in other embodiments of the present invention, the implementer can set other sampling frequencies by himself.

[0046] Step S2: select any acquisition time as the target acquisition time; take the target acquisition time as the end point of the local time domain window, select the time points to be analyzed from right to left in the historical acquisition time of the target acquisition time, until the change value of the rock drilling speed between the first time point to be analyzed and the first acquisition time is less than the average change value of the rock drilling speed between the time point to be analyzed and the target acquisition time, take the corresponding time point to be analyzed as the starting end point of the local time domain window of the target acquisition time, and obtain the local time domain window of the target acquisition time.

[0047] First, any acquisition time is selected as the target acquisition time, and the acquisition time of the target sampling point is analyzed one by one; considering that in geological exploration, the rock properties will change with the sampling depth. When it is at the boundary of rock layers, the drilling speed will fluctuate greatly due to the different physical properties such as density and hardness of different rock layers. Therefore, this feature is used to construct a local time domain window of the target acquisition time, which is convenient for distinguishing different rock layers in the future, helps to analyze the characteristics of rock layers, identify rock layer distribution, and provide more reference basis for the final comparison of abnormal characteristics of collected data;

[0048] Considering that the average change value of the rock drilling speed within the local time domain window at the target acquisition time is greater than the change value of the rock drilling speed on the left side of the local time domain window, it means that the rock drilling speed in the local time domain window at the target acquisition time changes greatly, which may include the boundary of rock formations. Therefore, the target acquisition time is taken as the end point of the local time domain window, and the time points to be analyzed are selected from right to left in the historical acquisition time of the target acquisition time, until the change value of the rock drilling speed between the first time point to be analyzed and the first acquisition time is less than the average change value of the rock drilling speed between the time point to be analyzed and the target acquisition time. The corresponding time point to be analyzed is taken as the starting end point of the local time domain window at the target acquisition time to obtain the local time domain window at the target acquisition time.

[0049] As an example, the judgment formula of the start and end points of the local time domain window at the target acquisition time includes:

[0050] ;

[0051] in, represents the drilling speed value at the first acquisition moment; x represents the sequence number of the time point to be analyzed on the left side of the time domain of the target acquisition moment, represents the drilling speed value of the xth time point to be analyzed at the target acquisition time; z represents the sequence number of the target acquisition time; Indicates the drilling speed value at the target acquisition time; It represents the time difference between the zth target acquisition moment and the xth time point to be analyzed; Indicates the change in drilling speed between the xth time point to be analyzed and the first acquisition moment; represents the average change in drilling speed between the xth time point to be analyzed and the zth target acquisition moment; Indicates taking the absolute value.

[0052] In the judgment formula of the start and end points of the local time domain window at the target acquisition moment, the time points to be analyzed are selected in sequence from the left side of the time domain at the target acquisition moment until the first time point to be analyzed meets the condition requirements of the judgment formula, and the corresponding time point to be analyzed is used as the start and end point of the local time domain window, and the target acquisition moment is used as the end point to obtain the local time domain window at the target acquisition moment.

[0053] It should be noted that the positive direction of the time domain is right, and the negative direction of the time domain is left; the collection time on the left side of the target collection time is the historical collection time of the target collection time.

[0054] It should be noted that when all the time points to be analyzed on the left side of the time domain of the target acquisition time do not meet the judgment formula, the first acquisition time is taken as the starting endpoint.

[0055] Step S3: According to the variation characteristics of the rock drilling speed in the local time domain window at each acquisition moment, the local speed variation rate at each acquisition moment is obtained; according to the distribution characteristics of the local speed variation rates at all acquisition moments, the number of rock layers at each sampling moment is obtained in combination with the corresponding local time domain window; in the continuous time domain of the same rock layer, according to the variation characteristics of the rotary torque, the torque accumulation parameter at each acquisition moment is obtained; the rock drilling speed, local speed variation rate, torque accumulation parameter and rock layer number at all sampling moments of the target sampling position are combined in the time sequence of the sampling moments to form a feature vector of the target sampling position.

[0056] Taking into account that the change of drilling speed reflects the change of physical properties such as hardness and density of different rock formations, by obtaining the local velocity change rate at each acquisition moment, the hierarchical structure and changes of the rock formation can be judged more accurately, which is helpful to analyze the distribution of underground rocks at the target sampling position and assist in the judgment of rock formation characteristics. At the same time, when the local velocity change rate is large, it can also reflect the stratification of the rock formation. Therefore, according to the change characteristics of the drilling speed in the local time domain window at each acquisition moment, the local velocity change rate at each acquisition moment is obtained, which provides a basis for the subsequent acquisition of rock formation data and provides more analysis basis for comparing the differences between the target sampling position and other sampling positions.

[0057] Preferably, in one embodiment of the present invention, within a local time domain window at the target acquisition moment, the absolute value of the difference between the rock drilling speed at the start point and the rock drilling speed at the end point is used as the speed change value corresponding to the target acquisition moment; and the local speed change rate at the target acquisition moment is obtained according to the speed change value and the duration of the local time domain window.

[0058] As an example, the ratio of the speed change value in the local time domain window at the target acquisition moment to the duration is taken as the local speed change rate at the target acquisition moment.

[0059] When the local velocity change rate is large, it means that the drilling may enter from one rock layer to another, so the distribution characteristics of the local velocity change rate at the time of acquisition can reveal the distribution of the geological layer; then the time point of the rock layer stratification is determined according to the local time domain window, and the number of rock layers at each sampling time is determined. Therefore, according to the distribution characteristics of the local velocity change rate at all acquisition times, combined with the corresponding local time domain window, the number of rock layers at each sampling time is obtained, which can provide more basis for subsequent data anomaly comparison, more accurately identify the abnormal sampling location, and ensure the representativeness of the sampling data for actual geological exploration.

[0060] Preferably, in one embodiment of the present invention, when the local velocity change rate at the target acquisition moment is greater than the average value of the local velocity change rates at all acquisition moments, it indicates that the local velocity change rate at the target acquisition moment is larger than the overall local velocity change rate, reflecting that the local time domain window at the target acquisition moment contains rock stratification; at the same time, the maximum absolute value of the slope of the velocity change curve of the drilling speed in the local time domain window reflects the time point when the stratification transition is most obvious, so it is determined that the rock stratum changes occur at the time point corresponding to the maximum absolute value of the slope;

[0061] Based on this, among the local velocity change rates at all acquisition moments of the target sampling position, the local time domain window whose local velocity change rate is greater than the average of all local velocity change rates is recorded as the layered time domain window; the velocity change curve of all rock drilling velocities at the target sampling position is obtained; the time point corresponding to the maximum absolute value of the slope of the velocity change curve in each layered time domain window is taken as the layered time point;

[0062] The period between the first acquisition moment and the first stratification time point, the period between the last stratification time point and the last acquisition moment, and the period between sequentially adjacent stratification time points constitutes a rock layer, and the number of rock layers at each acquisition moment is obtained.

[0063] It should be noted that, in one embodiment of the present invention, when the rock layers are divided at the acquisition time, the time domain interval corresponding to one rock layer includes the left endpoint but does not include the right endpoint, and the last acquisition time is divided into the last rock layer; for example, the serial number of the first acquisition time is 0, the serial number of the first stratification time point is 500, the serial number of the second stratification time point is 1350, the serial number of the third stratification time point is 2700, and the last acquisition time is 3800, then the first rock layer corresponds to [0,500), the second rock layer corresponds to [500,1350), the third rock layer corresponds to [1350,2700), and the fourth rock layer corresponds to [2700,3800]; the rock layer number of the acquisition time with serial number 500 is two, which belongs to the second rock layer.

[0064] In other embodiments of the present invention, the implementer may also set that when dividing the rock layers at the acquisition time, the time domain interval corresponding to one rock layer does not include the left endpoint but includes the right endpoint, and the first acquisition time is divided into the first rock layer, thereby avoiding the same acquisition time belonging to different rock layers.

[0065] Taking into account the evolution of different natural geological states during the formation of the same geological stratum, the thickness of different strata is not the same. In the process of stratum formation, taking sedimentary rocks as an example, as sediments continue to accumulate, the lower sediments will gradually be compressed, that is, in the same geological stratum, the deeper the depth, the greater the squeezing force of the stratum. During the drilling process, the lateral rotational torque exerted on the drill rig is greater. Therefore, in the continuous time domain of the same stratum, according to the changing characteristics of the rotational torque, the torque accumulation parameters at each acquisition moment are obtained to characterize the characteristic changes of the stratum, so as to provide more quantitative information for the subsequent comparison of sampling data between different sampling positions, and to more accurately screen out abnormal sampling positions.

[0066] Preferably, in one embodiment of the present invention, considering that within the continuous time domain of the same rock formation, the greater the difference in the rotary torque between the acquisition moment and the previous acquisition moment, the more obvious the cumulative characteristics of the rotary torque are, so any rock formation is selected as the rock formation to be analyzed; any acquisition moment in the rock formation to be analyzed is selected as the moment to be analyzed; in the local time domain window corresponding to the rock formation to be analyzed, the acquisition moment before the moment to be analyzed is used as the cumulative moment of the moment to be analyzed;

[0067] The torque accumulation parameter at the time to be analyzed is obtained according to the difference characteristics between the rotation torque at the time to be analyzed and the rotation torque at each accumulated time; the difference characteristics between the rotation torque at the time to be analyzed and the rotation torque at each accumulated time are positively correlated with the torque accumulation parameter.

[0068] As an example, the calculation formula of the torque accumulation parameter includes:

[0069] ;

[0070] Among them, r represents the torque accumulation parameter at the time to be analyzed; c represents the time sequence number of the time to be analyzed; b represents the accumulation time of the time to be analyzed; The number of accumulated moments representing the moments to be analyzed; Indicates the value of the rotation torque at the cth moment to be analyzed; Indicates the value of the rotational torque at the bth cumulative moment of the cth moment to be analyzed.

[0071] In the calculation formula of the torque accumulation parameter, the difference characteristics between the rotary torque at the time to be analyzed and the rotary torque at each cumulative time are expressed by the absolute value of the difference, thereby reflecting the cumulative characteristics of the rotary torque. The larger the value is, the greater the difference between the rotational torque at the time to be analyzed and the accumulation time is, and the more obvious the cumulative characteristics of the rotational torque are. The cumulative trend of the rotational torque from the first acquisition time to the time to be analyzed in the rock formation to be analyzed is expressed by summing up, and the torque accumulation parameter is obtained to quantitatively reflect the changes in rock properties in the same rock formation, providing more quantitative information for subsequent comparison of sampling data between different sampling positions.

[0072] In other embodiments of the present invention, a curve of the change of the rotational torque can be obtained, with the horizontal axis being the time axis and the vertical axis being the rotational torque value axis. Then, the curve segment contained in the time to be analyzed and the corresponding cumulative time is intercepted as the curve segment to be analyzed, and the left end point of the curve segment to be analyzed is translated to the origin of the coordinate system. The area of ​​the translated curve to be analyzed and the horizontal axis of the coordinate system is obtained, and the area is used as the torque accumulation parameter at the time to be analyzed.

[0073] Furthermore, the drilling speed, local speed change rate, torque accumulation parameter and number of rock layers at all sampling moments of the target sampling position are combined into a feature vector of the target sampling position in the chronological order of the sampling moments. In the form of feature vectors, data of multiple dimensions are represented in a unified and intuitive manner, which facilitates the subsequent comparison of differences between different sampling positions and the identification of data anomalies.

[0074] As an example, the order of data dimensions in the feature vector is: rock drilling speed, local speed change rate, torque accumulation parameter and number of rock layers. In other embodiments of the present invention, implementers can adjust the order of dimensions by themselves.

[0075] Step S4: According to the difference characteristics between the feature vector of the target sampling position and the feature vectors of other sampling positions, combined with the distance between the target sampling position and other sampling positions, the data anomaly coefficient of the target sampling position is obtained; according to the outlier distribution characteristics of the data anomaly coefficients of all sampling positions, the abnormal sampling positions are screened out.

[0076] After obtaining the characteristic vector of the sampling data representing the sampling location from multiple dimensions, the characteristic vectors of different sampling locations can be compared. Considering that the geological differences between sampling locations are greater as the distance between them increases, the data anomaly coefficient of the target sampling location is obtained based on the difference between the characteristic vector of the target sampling location and the characteristic vectors of other sampling locations, combined with the distance between the target sampling location and other sampling locations, to characterize the data anomaly characteristics of the target sampling location, providing a basis for finally obtaining the abnormal sampling location.

[0077] Preferably, in one embodiment of the present invention, considering that the DTW algorithm can effectively measure the difference characteristics between two groups of time series, the difference parameters between the feature vector of the target sampling position and the feature vectors of other sampling positions are obtained by the DTW algorithm;

[0078] Considering that the farther the distance between the sampling locations is, the more likely they are to be under different geological conditions, the greater the difference in the rock formation characteristics themselves is likely to be, and the larger difference between the characteristic vectors may be normal, the distance is used for negative correlation correction, and the data anomaly coefficient of the target sampling location is obtained according to the distance and difference parameters between the target sampling location and other sampling locations; the distance is negatively correlated with the data anomaly coefficient; the difference parameter is positively correlated with the data anomaly coefficient.

[0079] As an example, the difference parameter between the target sampling position and any other sampling position is used as the numerator, the distance is used as the denominator, the fractional ratio is used as the data anomaly sub-coefficient between the target sampling position and any other sampling position, and the sum of all data anomaly sub-coefficients of the target sampling position is used as the data anomaly coefficient of the target sampling position.

[0080] In other embodiments of the present invention, the implementer may also use a negative correlation mapping function, such as an exponential function exp(-x) with a natural constant e as the base, to perform negative correlation mapping, map the distance between the target sampling position and other sampling positions through exp(-x), and use the product of the mapping value and the difference parameter as the corresponding data anomaly sub-coefficient to obtain the data anomaly coefficient. Where x represents an independent variable.

[0081] Preferably, in one embodiment of the present invention, among other sampling positions of the target sampling position, any other sampling position is selected as the comparison sampling position;

[0082] Considering that the larger the DTW distance between two time series data is, the greater the difference between the two time series data is, the DTW distance and value between all data of the same dimension of the feature vector of the target sampling position and the feature vector of the comparison sampling position are obtained as the difference parameter between the target sampling position and the comparison sampling position.

[0083] It should be noted that the DTW algorithm is a technical means well known to those skilled in the art and will not be described in detail here.

[0084] In another embodiment of the present invention, the implementer may also align the lengths of the feature vectors of all sampling positions, take the longest feature vector in the time domain as the reference, fill the other feature vectors with 0, and then express the difference characteristics of the feature vectors by the absolute value of the difference between the feature vectors. The calculation formula of the data anomaly coefficient includes:

[0085] ;

[0086] Where i represents the serial number of the target sampling position; represents the data anomaly coefficient of the i-th target sampling position; a represents the serial number of the comparison sampling position; A represents the number of the comparison sampling positions; represents the distance between the i-th target sampling position and the a-th comparison sampling position; The feature vector representing the i-th target sampling position; The feature vector representing the a-th comparison sampling position; Indicates obtaining the absolute value; Represents the modulo function.

[0087] In the calculation formula of the data anomaly coefficient, the difference characteristics of the eigenvectors are expressed by the absolute value of the difference between the eigenvectors, and then by The absolute value of the difference between the eigenvectors is corrected, and finally the data anomaly coefficient is obtained through modulo operation.

[0088] It should be noted that, in other embodiments of the present invention, the implementer may not perform the modulo operation, and express the difference between the target sampling position and the other sampling positions in the form of a vector, such as ;in Data anomaly vector representing the target sampling location.

[0089] By comparing the differences between the feature vectors and correcting them with the help of distance factors, we can obtain the data anomaly coefficient that characterizes the difference between the data of each sampling position and other sampling positions. Then, we can identify the abnormal data anomaly coefficients and screen out the abnormal sampling positions based on the outlier distribution characteristics of the data anomaly coefficients of all sampling positions.

[0090] Preferably, in one embodiment of the present invention, considering that the LOF algorithm can identify outliers or abnormal points in the data, the outlier factor of each data anomaly coefficient is obtained by the LOF algorithm, and the outlier distribution characteristics of the data anomaly coefficient are represented by the outlier factor. When the outlier factor is greater than the preset outlier threshold, the corresponding sampling position is determined to be an abnormal sampling position.

[0091] As an example, the preset outlier threshold is 1.

[0092] In other embodiments of the present invention, the implementer may also analyze the outlier distribution characteristics of the data anomaly coefficient by clustering methods, such as clustering by K-means algorithm, and use the sampling positions corresponding to the isolated data anomaly coefficients after clustering as abnormal sampling positions.

[0093] It should be noted that the LOF algorithm and the K-means algorithm are both technical means well known to those skilled in the art and will not be described in detail here.

[0094] After screening out abnormal sampling locations, relevant personnel can be prompted that the sampling data at the abnormal sampling locations are less representative of the sampling area. Relevant personnel can choose the sampling data at the abnormal sampling locations based on actual conditions, and decide whether to re-sample at a fixed point.

[0095] In summary, in response to the technical problem of inaccurate judgment of abnormal sampling positions in geological exploration, the present invention proposes a geological exploration sampling device and sampling analysis method. The present invention first obtains geological exploration data of the target sampling position; further obtains the local time domain window of the target acquisition moment; further obtains the local velocity change rate, the number of rock layers, and the torque accumulation parameter at each acquisition moment in the local time domain window, and forms a feature vector of the target sampling position in the time sequence of the sampling moment; further, according to the difference characteristics between the feature vector of the target sampling position and the feature vectors of other sampling positions, combined with the distance between the target sampling position and other sampling positions, the data anomaly coefficient of the target sampling position is obtained; finally, according to the outlier distribution characteristics of the data anomaly coefficients of all sampling positions, the abnormal sampling positions are screened out. The present invention constructs the feature vector of the sampling position by analyzing the characteristics of the drilling speed, the local velocity change rate, the torque accumulation parameter, etc., accurately represents the rock structure, and screens out the abnormal sampling position by comparing the difference of the feature vectors and combining the distance of the sampling positions, thereby optimizing the identification of abnormal positions.

[0096] It should be noted that the sequence of the above embodiments of the present invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0097] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A geological exploration sampling and analysis method, characterized in that: The method comprises: Select any sampling position as a target sampling position; obtain geological exploration data of the target sampling position; the geological exploration data at least includes drilling speed and rotary torque; Select any acquisition time as the target acquisition time; take the target acquisition time as the end point of the local time domain window, select the time points to be analyzed from right to left in the historical acquisition time of the target acquisition time, until the change value of the rock drilling speed between the first time point to be analyzed and the first acquisition time is less than the average change value of the rock drilling speed between the time point to be analyzed and the target acquisition time, take the corresponding time point to be analyzed as the starting end point of the local time domain window of the target acquisition time, and obtain the local time domain window of the target acquisition time; According to the variation characteristics of the rock drilling speed in the local time domain window at each acquisition moment, the local speed variation rate at each acquisition moment is obtained; according to the distribution characteristics of the local speed variation rates at all the acquisition moments, in combination with the corresponding local time domain window, the number of rock layers at each sampling moment is obtained; in the continuous time domain of the same rock layer, according to the variation characteristics of the rotary torque, the torque accumulation parameter at each acquisition moment is obtained; the rock drilling speed, the local speed variation rate, the torque accumulation parameter and the number of rock layers at all the sampling moments of the target sampling position are combined in the time sequence of the sampling moments to form a feature vector of the target sampling position; According to the difference characteristics between the feature vector of the target sampling position and the feature vectors of other sampling positions, combined with the distance between the target sampling position and other sampling positions, the data anomaly coefficient of the target sampling position is obtained; according to the outlier distribution characteristics of the data anomaly coefficients of all sampling positions, the abnormal sampling positions are screened out; The method for obtaining the data anomaly coefficient includes: Obtaining difference parameters between the feature vector at the target sampling position and the feature vectors at other sampling positions by using a DTW algorithm; According to the distance between the target sampling position and other sampling positions and the difference parameter, a data anomaly coefficient of the target sampling position is obtained; the distance is negatively correlated with the data anomaly coefficient; and the difference parameter is positively correlated with the data anomaly coefficient; The method for obtaining the local velocity change rate includes: In the local time domain window at the target acquisition time, the absolute value of the difference between the rock drilling speed at the start point and the rock drilling speed at the end point is used as the speed change value corresponding to the target acquisition time; according to the speed change value and the duration of the local time domain window, the local speed change rate at the target acquisition time is obtained; The method for obtaining the number of rock layers includes: Among the local velocity change rates at all acquisition moments of the target sampling position, a local time domain window in which the local velocity change rate is greater than the average of all the local velocity change rates is recorded as a layered time domain window; a velocity change curve of all the rock drilling velocities at the target sampling position is obtained; and a time point corresponding to the maximum absolute value of the slope of the velocity change curve in each layered time domain window is taken as a layered time point; The period between the first acquisition moment and the first layering time point, the period between the last layering time point and the last acquisition moment, and the period between the layering time points that are sequentially adjacent constitutes a rock layer, and the number of rock layers at each acquisition moment is obtained; The method for obtaining the torque accumulation parameter includes: Select any rock layer as the rock layer to be analyzed; select any acquisition time in the rock layer to be analyzed as the time to be analyzed; in the local time domain window corresponding to the rock layer to be analyzed, use the acquisition time before the time to be analyzed as the accumulation time of the time to be analyzed; The torque accumulation parameter of the time to be analyzed is obtained according to the difference characteristics between the rotational torque at the time to be analyzed and the rotational torque at each accumulated time; the difference characteristics between the rotational torque at the time to be analyzed and the rotational torque at each accumulated time are positively correlated with the torque accumulation parameter.

2. A geological exploration sampling and analysis method according to claim 1, characterized in that: The method for obtaining the difference parameter includes: Among other sampling positions of the target sampling position, select any other sampling position as a comparison sampling position; The DTW distance and value between all the same-dimensional data of the feature vector of the target sampling position and the feature vector of the comparison sampling position are obtained as the difference parameter between the target sampling position and the comparison sampling position.

3. A geological exploration sampling and analysis method according to claim 1, characterized in that: The method for obtaining the abnormal sampling position includes: The outlier factor of each data anomaly coefficient is obtained by the LOF algorithm. When the outlier factor is greater than a preset outlier threshold, the corresponding sampling position is determined to be an abnormal sampling position.

4. A geological exploration sampling and analysis method according to claim 1, characterized in that: The positive direction in the time domain is right.

5. A geological prospecting sampling and analysis method according to claim 3, characterized in that: The preset outlier threshold is 1.

6. A geological prospecting sampling device, comprising a device body, characterized in that: The device implements the steps of a geological exploration sampling and analysis method as described in any one of claims 1 to 5.

Images