A lunar soil drilling sampling volume prediction method and system based on multi-dimensional telemetry data

Abstract

A method and system for predicting the sampling volume of lunar soil drilling based on multi-dimensional telemetry data. The present invention relates to the field of sampling exploration of extraterrestrial celestial bodies. It solves the defect that there is no method for predicting the real-time sampling volume during the drilling process in the prior art. The prediction method includes: collecting test data of the nominal working conditions of the drilling operation; dividing the operation types of each drilling pass into a stable stage and a disturbance stage; extracting and analyzing the load response characteristic signals of each drilling pass stage, and constructing a prediction model M for the sampling stable stage 稳定 ; performing multiple linear regression calculations on the time-domain characteristic parameters of the core extraction force in the disturbance stage and the sampling volume to obtain a prediction model M for the sampling disturbance stage 扰动 ; constructing a sampling volume prediction model M1 based on the core extraction force load; collecting the telemetry data of the drilling operation in real time through ground simulation tests and on-orbit operations, substituting them into the two types of prediction models respectively to obtain the real-time sampling volume, and combining the model prediction deviation to analyze the real-time sampling volume prediction range M3 to predict the sample state. The present invention is also applicable to the field of predicting the sampling volume of drilling

Description

Technical Field

[0001] The present invention relates to the technical field of extraterrestrial body sampling and detection. Background Art

[0002] Drilling sampling is a highly uncertain and nonlinear physical system. The lunar soil profile drilling operation is random, and it is difficult to estimate the sampling volume in real time under the condition of insufficient sensing resources. At the same time, human intervention and interpretation based on telemetry data alone make it impossible to accurately judge the current operation status. For lunar soil profile drilling sampling operations, in order to quantitatively obtain the real-time sampling volume and sampling status and then reasonably adjust the operating procedures, a drilling sampling status and sampling volume prediction method based on drilling operation telemetry data is constructed. This helps to visualize the on-orbit sampling operation status and ground operators to adjust the operation strategy in time, which plays an auxiliary decision-making role in the smooth progress of sampling and detection tasks. However, there is no method for predicting the real-time sampling volume of the drilling process in the prior art. Summary of the invention

[0003] The purpose of the present invention is to solve the defect that there is no method for predicting the real-time sampling amount during drilling in the prior art. To solve the above technical problem, the present invention is implemented by the following technical solutions:

[0004] Solution 1: A method for predicting the amount of lunar soil drilling sampling based on multi-dimensional telemetry data, the method comprising the following steps:

[0005] S1. Collect test data of nominal working conditions of drilling and production operations;

[0006] S2. Analyze the correlation between the sampling volume and the drilling load parameters based on the test data collected in S1, and obtain the correlation between the sampling volume and the core lifting force and the bottom hole drilling pressure in combination with the drilling coring mechanism; preliminarily judge and divide the operation type of each drilling into the stable stage and the disturbance stage;

[0007] S3, extract and analyze the load response characteristic signals of each drilling stage, and use machine learning to identify the sampling state to correct the classification results of the drilling operation type in S2;

[0008] S4. Based on the correlation between sampling volume and core lifting force in S2 and the classification results of the stable sampling state in S3, the core lifting force, drilling depth and sampling volume in the drilling process are multivariately fitted to construct the estimation model M of the stable sampling stage. 稳定 ;

[0009] S5, based on the classification results of the stable and disturbed phases of the sampling state in S4 and the stable phase prediction model M 稳定 Combined with the time-domain characteristic parameters of the core-lifting force in the disturbance stage, the time-domain characteristic parameters of the core-lifting force in the disturbance stage and the sampling amount are subjected to multivariate linear regression calculation to obtain the sampling disturbance stage estimation model M扰动 ;

[0010] S6, combining the fitting results of the stable stage and the disturbance stage in S5, constructing the sampling volume estimation model M1 based on the core lifting force load;

[0011] S7. According to the classification results of the stable and disturbed stages of the sampling state in S5 and the drilling depth, the area enclosed by the drilling pressure curve at the bottom of the hole is calculated to obtain the stable and disturbed stage area S 稳定 / S 扰动 , the stable and disturbed phase area S in S6 稳定 / S 扰动 Multivariate fitting is performed with the actual sampling volume data of each drilling test, and the effective sampling power consumption W is established to obtain the sampling volume estimation model M2 based on the bottom hole drilling pressure;

[0012] S8. Real-time telemetry data of drilling and production operations are collected during ground simulation tests and on-orbit operations, and the data are substituted into two types of estimation models to obtain the real-time sampling volume. The real-time sampling volume estimation range M3 is analyzed in combination with the model estimation deviation to predict the sample status.

[0013] Furthermore, a preferred embodiment is provided, wherein the test data in S1 includes historical data of drilling load parameters, and actual sampling volume and drilling depth data of each test drilling.

[0014] Further, a preferred embodiment is provided, in which the method for obtaining the correlation between the sampling volume and the core lifting force and the bottom hole drilling pressure in combination with the coring mechanism in S2 is:

[0015] F L +G=F TX +F D +F P

[0016] Among them, F L Indicates drilling pressure, G indicates the drilling rig's own gravity, F TX Indicates the core lifting force, F P Indicates chip removal resistance, F D Indicates the bottom hole drilling pressure.

[0017] Furthermore, a preferred embodiment is provided, in which the method for preliminarily judging and dividing each drilling operation type into a stable stage and a disturbance stage in S2 is as follows: a drilling test with a coring rate exceeding 75% is defined as a sampling stable test, and a test with a coring rate below 75% has a coring disturbance stage; a test with a sampling volume exceeding 500g is defined as a sampling stable test, and a test with a sampling volume below 500g has a coring disturbance stage.

[0018] Further, a preferred embodiment is provided, in which the method for constructing the sampling amount estimation model M1 based on the core lifting force load in S6 is: M1=M 稳定+M 扰动 .

[0019] Furthermore, a preferred embodiment is provided, in which the load response characteristic signal of each drilling stage in S3 includes the load mean, peak-to-peak value, variance and slope.

[0020] Furthermore, a preferred embodiment is provided, in which the time-domain characteristic parameters of the core-lifting force in the disturbance stage in S5 include the peak-to-peak value, variance, standard deviation, and peak factor of the core-lifting force load.

[0021] Solution 2: A lunar soil drilling sampling volume prediction system based on multi-dimensional telemetry data, the system comprising:

[0022] A data collection module is used to collect test data of nominal working conditions of drilling and production operations;

[0023] The data analysis module is used to analyze the correlation between the sampling volume and the drilling load parameters based on the test data collected by the data collection module, and to obtain the correlation between the sampling volume and the core lifting force and the bottom hole drilling pressure in combination with the drilling coring mechanism; to make a preliminary judgment and divide the types of each drilling operation into a stable stage and a disturbance stage;

[0024] The data extraction module is used to extract and analyze the load response characteristic signals of each drilling stage, and to correct the classification results of the drilling operation type in the data analysis module by combining machine learning to identify the sampling state;

[0025] The data fitting module is used to perform multivariate fitting on the core lifting force, drilling depth and sampling volume during drilling according to the correlation between the sampling volume and the core lifting force in the data analysis module and the classification results of the stable sampling state in S3, and to construct an estimation model M for the stable sampling stage. 稳定 ;

[0026] The calculation module is used to estimate the model M in the stable phase according to the classification results of the sampling state stable phase and the disturbance phase in the data fitting module. 稳定 Combined with the time-domain characteristic parameters of the core-lifting force in the disturbance stage, the time-domain characteristic parameters of the core-lifting force in the disturbance stage and the sampling amount are subjected to multivariate linear regression calculation to obtain the sampling disturbance stage estimation model M 扰动 ;

[0027] The sampling volume estimation model construction module of the core lifting force load is used to combine the fitting results of the stable stage and the disturbance stage in the calculation module to construct the sampling volume estimation model M1 based on the core lifting force load;

[0028] The sampling volume estimation model module of the bottom hole drilling pressure is used to calculate the area enclosed by the bottom hole drilling pressure curve according to the classification results of the stable and disturbed stages of the sampling state in the calculation module combined with the drilling depth, and obtain the stable and disturbed stage area S 稳定 / S 扰动, the sampling volume of the core lifting force load estimation model construction module is used to estimate the area S of the stable and disturbed stages 稳定 / S 扰动 Multivariate fitting is performed with the actual sampling volume data of each drilling test, and the effective sampling power consumption W is established to obtain the sampling volume estimation model M2 based on the bottom hole drilling pressure;

[0029] The sample prediction module is used to collect telemetry data of drilling and production operations in real time during ground simulation tests and on-orbit operations, and substitutes it into two types of estimation models to obtain the real-time sampling volume. The real-time sampling volume estimation range M3 is analyzed in combination with the model estimation deviation to predict the sample status.

[0030] Solution three: a computer-readable storage medium, wherein the computer-readable storage medium is used to store a computer program, and the computer program executes the method for predicting the lunar soil drilling sampling volume based on multi-dimensional telemetry data as described in any one of Solution one.

[0031] Solution 4: A computer device includes a memory and a processor, wherein the memory stores a computer program. When the processor runs the computer program stored in the memory, the processor executes the method for predicting the lunar soil drilling sampling volume based on multi-dimensional telemetry data as described in any one of Solution 1.

[0032] The present invention is beneficial in that:

[0033] The method described in the present invention is a method for predicting the lunar soil drilling sampling volume based on multi-dimensional telemetry data. By collecting various data of drilling and production tests, the correlation analysis between the sampling volume and the drilling load parameters is carried out, and the drilling and coring mechanism is analyzed. The variation law between the sampling volume and the core lifting force and the bottom hole drilling pressure load parameters can be determined, and then a sampling volume estimation model is established with the core lifting force and the bottom hole drilling pressure load parameters as judgment indicators. Multi-faceted integrated analysis can more accurately predict the real-time sampling volume.

[0034] The method described in the present invention is a method for predicting the lunar soil drilling sampling volume based on multi-dimensional telemetry data. By extracting and analyzing the time domain characteristic signals of the drilling load parameters in the historical drilling test data, a sampling state classification algorithm is established, and the drilling stage is divided into a stable stage and a disturbance stage. Sampling volume prediction models are established for the two stages respectively. Multi-stage analysis can better predict the drilling sampling volume in the current stage, and the prediction result is more accurate.

[0035] The method described in the present invention is a method for predicting the lunar soil drilling sampling volume based on multi-dimensional telemetry data. The sampling volume is real-time estimated and a model is established by utilizing the multi-dimensional telemetry data of the core lifting force and the bottom hole drilling pressure combined with the results of the sampling state classification algorithm. The parameters of the estimation model are corrected according to historical test data. The prediction results of the two estimation models are combined to finally obtain the real-time drilling sampling volume estimation range.

[0036] The present invention is also applicable to the field of drilling sampling volume estimation. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a flow chart of a method for predicting lunar soil drilling sampling volume based on multi-dimensional telemetry data as described in Implementation Method 1.

[0038] Figure 2 Schematic diagram of the force on the coring soft bag in Example 2 of a method for predicting the lunar soil drilling sampling volume based on multi-dimensional telemetry data described in Implementation Method 1.

[0039] Figure 3 It is a flow chart of dividing the sampling stabilization phase and the disturbance phase in the fourth implementation mode.

[0040] Figure 4 The sampling amount estimation model M is obtained in the fifth implementation mode 稳定 Schematic diagram of the process.

[0041] Figure 5 To obtain the sampling disturbance stage estimation model M in implementation mode 5 扰动 Schematic diagram of the process.

[0042] Figure 6 This is a flow chart of obtaining the sampling volume estimation model M2 based on the bottom hole drilling pressure in the fifth implementation mode.

[0043] Figure 7 This is the effect diagram described in Implementation Mode 5. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solutions and advantages of the implementation methods of the present application clearer, the technical solutions in the implementation methods of the present application will be clearly and completely described below in conjunction with the drawings in the implementation methods of the present application. Obviously, the described implementation methods are only part of the implementation methods of the present application, not all of the implementation methods.

[0045] Embodiment 1: This embodiment provides a method for predicting the amount of lunar soil drilling sampling based on multi-dimensional telemetry data, and the method for predicting the amount of lunar soil drilling sampling includes the following steps:

[0046] S1. Collect test data of nominal working conditions of drilling and production operations;

[0047] S2. Analyze the correlation between the sampling volume and the drilling load parameters based on the test data collected in S1, and obtain the correlation between the sampling volume and the core lifting force and the bottom hole drilling pressure in combination with the drilling coring mechanism; preliminarily judge and divide the operation type of each drilling into the stable stage and the disturbance stage;

[0048] S3, extract and analyze the load response characteristic signals of each drilling stage, and use machine learning to identify the sampling state to correct the classification results of the drilling operation type in S2;

[0049] S4. Based on the correlation between sampling volume and core lifting force in S2 and the classification results of the stable sampling state in S3, the core lifting force, drilling depth and sampling volume in the drilling process are multivariately fitted to construct the estimation model M of the stable sampling stage. 稳定 ;

[0050] S5, based on the classification results of the stable and disturbed phases of the sampling state in S4 and the stable phase prediction model M 稳定 Combined with the time-domain characteristic parameters of the core-lifting force in the disturbance stage, the time-domain characteristic parameters of the core-lifting force in the disturbance stage and the sampling amount are subjected to multivariate linear regression calculation to obtain the sampling disturbance stage estimation model M 扰动 ;

[0051] S6, combining the fitting results of the stable stage and the disturbance stage in S5, constructing the sampling volume estimation model M1 based on the core lifting force load;

[0052] S7. According to the classification results of the stable and disturbed stages of the sampling state in S5 and the drilling depth, the area enclosed by the drilling pressure curve at the bottom of the hole is calculated to obtain the stable and disturbed stage area S 稳定 / S 扰动 , the stable and disturbed phase area S in S6 稳定 / S 扰动 Multivariate fitting is performed with the actual sampling volume data of each drilling test, and the effective sampling power consumption W is established to obtain the sampling volume estimation model M2 based on the bottom hole drilling pressure;

[0053] S8. Real-time telemetry data of drilling and production operations are collected during ground simulation tests and on-orbit operations, and the data are substituted into two types of estimation models to obtain the real-time sampling volume. The real-time sampling volume estimation range M3 is analyzed in combination with the model estimation deviation to predict the sample status.

[0054] See also Figure 1 This embodiment specifically includes the following steps:

[0055] S1. Collect test data of nominal working conditions of drilling and production operations, including historical data of drilling load parameters (drilling pressure, core lifting force, rotary torque, etc.), as well as actual sampling volume and drilling depth data of each test drilling;

[0056] S2. Analyze the correlation between sampling volume and drilling load parameters using the collected historical test data, and obtain the correlation between sampling volume and core lifting force and bottom hole drilling pressure in combination with the drilling coring mechanism; and make preliminary judgments on the types of drilling operations, which are divided into stable stage and disturbance stage;

[0057] S3. Extract and analyze the load response characteristic signals of each drilling stage, including the load mean, peak-to-peak value, variance, and slope, combine machine learning to identify the sampling state, and correct the classification results of the drilling operation type;

[0058] S4. Based on the correlation between sampling volume and core lifting force and the classification results of the stable stage of sampling state, the core lifting force, drilling depth and sampling volume in the drilling process are multivariately fitted to construct the estimation model M of the stable stage of sampling. 稳定 ;

[0059] S5. Based on the classification results of the stable / disturbance phase of the sampling state and the stable phase prediction model M 稳定 Combined with the time-domain characteristic parameters of the core-lifting force in the disturbance stage, including the peak-to-peak value, variance, standard deviation, and peak factor of the core-lifting force load, the time-domain characteristic parameters of the core-lifting force in the disturbance stage and the sampling amount are subjected to multivariate linear regression calculation to obtain the sampling disturbance stage prediction model M 扰动 ;

[0060] S6. Combining the fitting results of the stable stage and the disturbance stage, a sampling volume estimation model M1 based on the core lifting force load is constructed;

[0061] S7. Calculate the area enclosed by the bottom hole drilling pressure curve based on the classification results of the sampling state stable / disturbance stage combined with the drilling depth to obtain the stable / disturbance stage area S 稳定 / S 扰动 , the stable / disturbance phase area S 稳定 / S 扰动 Multivariate fitting is performed with the actual sampling volume data of each drilling test, and the effective sampling power consumption W is established to obtain the sampling volume estimation model M2 based on the bottom hole drilling pressure;

[0062] S8. Ground simulation test / on-orbit operation collects drilling and production telemetry data in real time, obtains real-time sampling volume based on two types of estimation models, analyzes the real-time sampling volume estimation range M3 in combination with model estimation deviation, and predicts sample status.

[0063] The data collected in S1 should be drilling sampling test data of a fixed-point lunar soil condition with the drilling object being nominal lunar soil, or based on nominal lunar soil.

[0064] Figure 2 In the process of coring, since the coring bag is in a uniform inward-turned state during the drilling and coring process, it can be regarded as being in a static equilibrium state. When the lunar soil sample is filled into the soft bag, the friction caused by the lateral stress will affect the size of the core-lifting force. At the same time, the change of the lateral stress is linearly related to the sampling amount. Therefore, it can be considered that the size of the core-lifting force is correlated with the sampling quality.

[0065] Implementation method 2. This implementation method is a further limitation of the method for predicting the lunar soil drilling sampling volume based on multi-dimensional telemetry data described in implementation method 1. The test data described in S1 includes historical data of drilling load parameters, as well as actual sampling volume and drilling depth data of each test drilling.

[0066] S1 collects test data of nominal working conditions of drilling operations, including historical data of drilling load parameters (drilling pressure, core lifting force, rotary torque, etc.), as well as actual sampling volume and drilling depth data of each test drilling.

[0067] Implementation method 3: This implementation method is a further limitation of the method for predicting the lunar soil drilling sampling volume based on multi-dimensional telemetry data described in implementation method 1. The method for obtaining the correlation between the sampling volume and the core lifting force and the bottom hole drilling pressure in S2 in combination with the drilling coring mechanism is:

[0068] F L +G=F TX +F D +F P

[0069] Among them, F L Indicates drilling pressure, G indicates the drilling rig's own gravity, F TX Indicates the core lifting force, F P Indicates chip removal resistance, F D Indicates the bottom hole drilling pressure.

[0070] See also Figure 2 This embodiment describes the coring process of lunar soil. Based on the "slip-free" coring mechanism, the mechanical transmission link of the coring process is analyzed. According to the uniform inward turning state of the soft bag during the coring process, the bottom hole drilling pressure model is established by combining the drilling pressure, the core lifting force, the drilling rig's own gravity and the chip removal resistance:

[0071] F L +G=F TX +F D +F P

[0072] Among them, F L Indicates drilling pressure, G indicates the drilling rig's own gravity, F TX Indicates the core lifting force, F P Indicates chip removal resistance, F D Indicates the bottom hole drilling pressure.

[0073] Implementation method four. This implementation method is a further limitation of the method for predicting the lunar soil drilling sampling volume based on multi-dimensional telemetry data described in implementation method one. The method for preliminarily judging and dividing each drilling operation type into a stable stage and a disturbance stage in S2 is as follows: a drilling test with a coring rate exceeding 75% is defined as a sampling stable test, and a test with a coring rate below 75% has a coring disturbance stage; a test with a sampling volume exceeding 500g is defined as a sampling stable test, and a test with a sampling volume below 500g has a coring disturbance stage.

[0074] See also Figure 3 This embodiment is described. This embodiment further divides the sampling stable stage and the disturbance stage by extracting the time domain characteristic signals of the core lifting force and the drilling pressure response and combining the machine learning method. Specifically, the following steps are included:

[0075] (1) Determination of sample learning library: From the perspective of sampling volume, each drilling test is divided into a sample library for the test data in the sampling stability phase and a sample library for the test data in the sampling disturbance phase;

[0076] (2) Determination of judgment indicators: Based on the correlation analysis between sampling volume and drilling load parameters in S2, the core lifting force is determined as the main indicator and the drilling pressure is used as the auxiliary indicator to estimate the sampling volume and judge the sampling status;

[0077] (3) Model construction and correction: Model 1 uses the core lifting force as the main indicator, and divides every 10 data points into a data segment. The fluctuation degree of the core lifting force is used to preliminarily judge whether there is a disturbance in the sampling state. That is, if the core lifting force is disturbed, the current data segment is the disturbance stage, otherwise it is the stable stage; Model 2 uses drilling pressure as the main indicator, and performs secondary judgment and refinement on the stable stage screened out by Model 1. That is, if the drilling pressure is disturbed, the current data segment is regarded as the disturbance stage, otherwise it is the stable stage; the time domain characteristics (mean, peak-to-peak value, variance and fitting slope of the drilling force load in the current data segment) of the training learning sample are used as input features, and the output features are the sampling stable stage (represented by 0) and the sampling disturbance stage (represented by 1). Finally, the classification results of the sampling state are output. The classification algorithm is tested and corrected using the experimental data of the test sample.

[0078] Implementation 5: This implementation is a further limitation of the lunar soil drilling sampling volume prediction method based on multi-dimensional telemetry data described in Implementation 1. The method for constructing the sampling volume estimation model M1 based on the core lifting force load in S6 is: M1=M 稳定 +M 扰动 .

[0079] Combination Figure 4This embodiment describes the present invention. After the current data segment is identified as the sampling stable stage according to the sampling state classification algorithm, the sampling volume estimation model M is obtained by combining the core lifting force, drilling depth and actual sampling volume with multivariate linear fitting. 稳定 , specifically including the following steps:

[0080] (6) Sample selection: Select several groups of core drilling tests with a sampling mass of more than 500 g as learning samples, and record the actual sampling volume of each sample as m 实际 ;

[0081] (7) Calculation of theoretical sampling volume: Assuming that the density and coring rate of the drilled sample are uniform and do not change with the drilling depth, every 10 data points are divided into a data segment, and the theoretical sampling mass m of the learning sample is calculated. 理论 and Δm within each data segment 理论 ;

[0082] (8) Sampling volume difference calculation: Calculate the difference m between the actual sampling volume and the theoretical sampling volume 差值 , and the difference m 差值 As the depth is linearly distributed into each data segment within Δm, that is, for Δm 理论 make corrections;

[0083] (9) Estimation model construction: The multivariate linear fitting method is used to calculate the drilling depth z, the core lifting force F TX As the independent variable, Δm as the dependent variable, obtain Δm and z, F TX The fitting equation is then calculated; and the Δm of each data segment is accumulated to obtain the stable stage sampling volume estimation model M 稳定 :

[0084]

[0085] (10) Model testing and correction: Select several groups of core drilling tests with a sampling mass of more than 500 g as test samples, calculate the fitting values ​​of Δm and m of the test samples, and compare them with the actual sampling mass m. 实际 Comparison and error analysis were performed.

[0086] Combination Figure 5 This embodiment is described. This embodiment is based on the sampling state classification result and the stable stage prediction model M 稳定 The time domain characteristic parameters of the core lifting force load in the disturbance stage and the sampling amount are subjected to multivariate linear regression calculation to obtain the sampling disturbance stage estimation model M 扰动 , specifically including the following steps:

[0087] (6) Sample selection: Select several groups of core drilling tests with a sampling mass of less than 500 g as learning samples, and record the actual sampling volume of each sample as m 实际 ;

[0088] (7) Calculation of stable sampling volume: Based on the sampling volume estimation model M in the stable sampling stage described in S4 稳定 , calculate the estimated quality m of the learning sample in sampling stability 稳定 ;

[0089] (8) Sampling volume difference calculation: Calculate the difference m between the actual sampling volume and the stable estimated sampling volume 差值 , similarly divide 10 data points into a data segment, and divide the difference m 差值 Evenly distributed into each sampling disturbance section Δm;

[0090] (9) Estimation model construction: The multivariate linear fitting method is used, with the time domain characteristic parameters of the core lifting force in the disturbance stage (peak-to-peak value pk, variance var, standard deviation std, peak factor C) as independent variables and Δm as the dependent variable, to obtain the fitting equation between Δm and pk, var, std, and C; and the data segments are accumulated to obtain the sampling amount estimation model M in the disturbance stage. 扰动 :

[0091]

[0092] (10) Model testing and correction: Several groups of core drilling tests with a sampling mass of less than 500 g were selected as test samples. The fitting values ​​of Δm of the test samples in the stable stage and the disturbance stage were calculated respectively, and the estimated value m of the total sampling volume was accumulated and compared with the actual sampling volume m. 实际 Comparison and error analysis were performed.

[0093] This implementation adds the sampling amount estimation models in the stable / disturbance phase to obtain the sampling amount estimation model M1 based on load feedback:

[0094] M1=M 稳定 +M 扰动 .

[0095] See also Figure 6 to Figure 7 To explain this embodiment, Figure 7 This is the effect diagram of this implementation method. This implementation method calculates the area enclosed by the bottom hole pressure curve based on the classification results of the sampling state stable / disturbance stage combined with the drilling depth to obtain the stable / disturbance stage area S 稳定 / S 扰动 , the stable / disturbance phase area S 稳定 / S 扰动 The actual sampling volume data of each drilling test is multivariately fitted to obtain the sampling volume estimation model M2 based on the bottom hole drilling pressure, which specifically includes the following steps:

[0096] (1) Sample selection: Select several groups of drilling and coring tests with known operating conditions as learning samples, and record the actual sampling volume of each sample as m 实际 ;

[0097] (2) Data segment area calculation: Combine the bottom hole drilling pressure model in S7 and the drilling depth data to obtain the bottom hole drilling pressure F with drilling depth z as the independent variable. D is the bottom hole drilling pressure variation curve of the dependent variable, and the area ΔS of each data segment is calculated by the trapezoidal integration method;

[0098] (3) Calculation of the area in the stable / disturbance phase: Based on the sampling state classification results, the area ΔS of each data segment is accumulated in sequence to obtain the stable / disturbance phase area S 稳定 / S 扰动 ;

[0099] (4) Estimation model construction: The multivariate linear fitting method is used to calculate the stable stage area S 稳定 , disturbance stage area S 扰动 is the actual sampling size m of the independent variable and sample 实际 As the dependent variable, obtain m and S 稳定 , S 扰动 The fitting equation is used; at the same time, since the area enclosed by the load and the depth can be regarded as the operating power consumption, the fitted S 稳定 , S 扰动 The effective sampling power consumption W is integrated, and the sampling volume estimation model M2 based on the bottom hole drilling pressure is obtained:

[0100]

[0101] In this embodiment, the drilling and production telemetry data collected in real time during ground simulation tests / on-orbit operations are substituted into two types of estimation models to obtain the real-time sampling volume respectively, and then the real-time sampling volume estimation range M3 is analyzed by combining the estimation deviation of the two types of models to predict the sample status.

[0102] Implementation method six: This implementation method is a further limitation of the method for predicting the lunar soil drilling sampling volume based on multi-dimensional telemetry data described in implementation method one. The load response characteristic signal of each drilling stage in S3 includes the force mean, peak-to-peak value, variance, and slope.

[0103] Implementation method seven. This implementation method is a further limitation of the method for predicting the lunar soil drilling sampling volume based on multi-dimensional telemetry data described in implementation method one. The time domain characteristic parameters of the core lifting force in the disturbance stage in S5 include the peak-to-peak value, variance, standard deviation, and peak factor of the core lifting force load.

[0104] Embodiment 8: This embodiment proposes a lunar soil drilling sampling amount prediction system based on multi-dimensional telemetry data, the system comprising:

[0105] A data collection module is used to collect test data of nominal working conditions of drilling and production operations;

[0106] The data analysis module is used to analyze the correlation between the sampling volume and the drilling load parameters based on the test data collected by the data collection module, and to obtain the correlation between the sampling volume and the core lifting force and the bottom hole drilling pressure in combination with the drilling coring mechanism; to make a preliminary judgment and divide the types of each drilling operation into a stable stage and a disturbance stage;

[0107] The data extraction module is used to extract and analyze the load response characteristic signals of each drilling stage, and to correct the classification results of the drilling operation type in the data analysis module by combining machine learning to identify the sampling state;

[0108] The data fitting module is used to perform multivariate fitting on the core lifting force, drilling depth and sampling volume during drilling according to the correlation between the sampling volume and the core lifting force in the data analysis module and the classification results of the stable sampling state in S3, and to construct an estimation model M for the stable sampling stage. 稳定 ;

[0109] The calculation module is used to estimate the model M in the stable phase according to the classification results of the sampling state stable phase and the disturbance phase in the data fitting module. 稳定 Combined with the time-domain characteristic parameters of the core-lifting force in the disturbance stage, the time-domain characteristic parameters of the core-lifting force in the disturbance stage and the sampling amount are subjected to multivariate linear regression calculation to obtain the sampling disturbance stage estimation model M 扰动 ;

[0110] The sampling volume estimation model construction module of the core lifting force load is used to combine the fitting results of the stable stage and the disturbance stage in the calculation module to construct the sampling volume estimation model M1 based on the core lifting force load;

[0111] The sampling volume estimation model module of the bottom hole drilling pressure is used to calculate the area enclosed by the bottom hole drilling pressure curve according to the classification results of the stable and disturbed stages of the sampling state in the calculation module combined with the drilling depth, and obtain the stable and disturbed stage area S 稳定 / S 扰动 , the sampling volume of the core lifting force load estimation model construction module is used to estimate the area S of the stable and disturbed stages 稳定 / S 扰动 Multivariate fitting is performed with the actual sampling volume data of each drilling test, and the effective sampling power consumption W is established to obtain the sampling volume estimation model M2 based on the bottom hole drilling pressure;

[0112] The sample prediction module is used to collect telemetry data of drilling and production operations in real time during ground simulation tests and on-orbit operations, and substitutes it into two types of estimation models to obtain the real-time sampling volume. The real-time sampling volume estimation range M3 is analyzed in combination with the model estimation deviation to predict the sample status.

[0113] Embodiment 9: A computer-readable storage medium, wherein the computer-readable storage medium is used to store a computer program, wherein the computer program executes the method for predicting the lunar soil drilling sampling volume based on multi-dimensional telemetry data as described in any one of embodiments 1 to 7.

[0114] Embodiment 10: A computer device comprises a memory and a processor, wherein the memory stores a computer program. When the processor runs the computer program stored in the memory, the processor executes the method for predicting the lunar soil drilling sampling amount based on multi-dimensional telemetry data as described in any one of embodiments 1 to 7.

[0115] Implementation eleven: The examples proposed in this implementation are used to explain the above implementations, wherein:

[0116] Example 1: The ultra-high time resolution flow field reconstruction method based on a spatiotemporal coupled deep network described in the present invention is as follows:

[0117] The first step is to measure the flow field through the PIV system to obtain 400Hz high-frequency flow field data and 2000Hz ultra-high-frequency random discrete flow field data. At the same time, wind pressure sensors are arranged on the surface of the object to obtain 2000Hz ultra-high time resolution pressure data at the local measurement point. Figure 1 As shown, the first, second, third, and fourth pulse light and exposure time intervals are equal, that is, dt1 = dt3, and dt2 is adjustable. Dt is the duty cycle, which is the inverse of the frequency, that is, 2.5×10-3s. By adjusting dt2, 400Hz high-frequency flow field data and 2000Hz ultra-high-frequency random discrete flow field data can be obtained.

[0118] Step 2, see Figure 2 To illustrate this implementation, the flow field data obtained by PIV is input into the convolutional autoencoder to construct a 13-layer deep convolutional autoencoder model. The encoder consists of three convolutional layers plus three pooling layers, and the decoder consists of three upsampling layers plus three convolutional layers, as well as a hidden vector layer in the middle. The convolution kernel size of the convolutional layer of the entire model is selected as 3×3, the pooling layer uses average pooling, and all activation functions use the ReLU function. The model is trained according to the loss function used in formula (3). Finally, the modal coefficients of the flow field data and the decoder are obtained.

[0119] Step 3, see Figure 3To illustrate this implementation, a surface pressure data sequence within a suitable time range is selected and input into the Transformer model, the connection between the surface pressure of the object and the nonlinear characteristics of the flow field is established, and the flow field characteristics with ultra-high time resolution are reconstructed. The loss function is trained using the PIV flow field data at a known time according to formula (5), and finally the complete 2000Hz ultra-high time resolution flow field modal coefficients are obtained.

[0120] The fourth step is to put the modal coefficients with ultra-high temporal resolution reconstructed by the Transformer model into the decoder in the deep convolutional autoencoder, and reconstruct the flow field with ultra-high temporal resolution through the decoder. Figure 4 By describing this embodiment, it can be obtained that the ultra-high time resolution flow field reconstruction method based on the spatiotemporal coupled deep network can reconstruct complete ultra-high time resolution flow field data.

[0121] In this embodiment, the Pytorch deep learning framework is used to build a model, which can be used to reconstruct the flow field with ultra-high time resolution. A four-way beam-combining design is adopted. The 400Hz high-frequency flow field data obtained by the dual-light-guide-arm PIV and the 2000Hz ultra-high frequency random discrete flow field data can be used to reconstruct the complete 2000Hz ultra-high time resolution flow field data.

[0122] Figure 2 and Figure 3Any process or method description in the flowchart described in or otherwise described herein can be understood as representing a module, fragment or portion of a code including one or more executable instructions for implementing the steps of a custom logic function or process, and the scope of the preferred embodiment of the present invention includes other implementations, in which the functions may be performed in a substantially simultaneous manner or in a reverse order according to the functions involved, which should be understood by a person skilled in the art of the present invention. The logic and / or steps represented in the flowchart or otherwise described herein illustrate the possible implementation architecture, functions and operations of the apparatus and methods according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logic function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It is also noted that each block in the block diagram or flowchart, and combinations of blocks in the block diagram or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions. For example, what may be considered as an ordered list of executable instructions for implementing a logical function may be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other system that can fetch instructions from and execute instructions in an instruction execution system, apparatus, or device).

[0123] As far as the ninth embodiment of the present specification is concerned, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use in an instruction execution system, device or equipment or in combination with these instruction execution systems, devices or equipment. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion (electronic device) with one or N wirings, a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and editable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting or processing in other suitable ways as necessary, and then stored in a computer memory. It should be understood that the various parts of the present invention can be implemented in hardware, software, firmware or a combination thereof. In the above embodiment, N steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one of the following technologies known in the art or a combination thereof: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0124] Those skilled in the art will appreciate that the above are only preferred embodiments of the present invention, and the various embodiments of the present disclosure and / or the features described in the claims may be combined or combined in various ways, even if such combinations or combinations are not explicitly described in the present disclosure. It is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art may still modify the technical solutions described in the aforementioned embodiments, or perform equivalent substitutions on some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

[0125] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for predicting lunar soil drilling sampling volume based on multi-dimensional telemetry data, characterized in that: The method for predicting the lunar soil drilling sampling amount comprises the following steps: S1. Collect test data of nominal working conditions of drilling and production operations; S2. Analyze the correlation between the sampling volume and the drilling load parameters based on the test data collected in S1, and obtain the correlation between the sampling volume and the core lifting force and the bottom hole drilling pressure in combination with the drilling coring mechanism; preliminarily judge and divide the operation type of each drilling into the stable stage and the disturbance stage; S3, extract and analyze the load response characteristic signals of each drilling stage, and use machine learning to identify the sampling state to correct the classification results of the drilling operation type in S2; S4. Based on the correlation between sampling volume and core lifting force in S2 and the classification results of the stable sampling state in S3, the core lifting force, drilling depth and sampling volume in the drilling process are multivariately fitted to construct the estimation model M of the stable sampling stage. 稳定 ; S5, based on the classification results of the stable and disturbed phases of the sampling state in S4 and the stable phase prediction model M 稳定 Combined with the time-domain characteristic parameters of the core-lifting force in the disturbance stage, the time-domain characteristic parameters of the core-lifting force in the disturbance stage and the sampling amount are subjected to multivariate linear regression calculation to obtain the sampling disturbance stage estimation model M 扰动 ; S6. Combining the fitting results of the stable stage and the disturbance stage in S5, a sampling volume estimation model M1 based on the core lifting force load is constructed; S7. According to the classification results of the stable and disturbed stages of the sampling state in S5 and the drilling depth, the area enclosed by the drilling pressure curve at the bottom of the hole is calculated to obtain the stable and disturbed stage area S 稳定 / S 扰动 , the stable and disturbed phase area S in S6 稳定 / S 扰动 Multivariate fitting is performed with the actual sampling volume data of each drilling test, and the effective sampling power consumption W is established to obtain the sampling volume estimation model M2 based on the bottom hole drilling pressure; S8. Real-time telemetry data of drilling and production operations are collected during ground simulation tests and on-orbit operations, and the data are substituted into two types of estimation models to obtain the real-time sampling volume. The real-time sampling volume estimation range M3 is analyzed in combination with the model estimation deviation to predict the sample status.

2. The method for predicting lunar soil drilling sampling volume based on multi-dimensional telemetry data according to claim 1 is characterized in that: The test data described in S1 include historical data of drilling load parameters, as well as actual sampling volume and drilling depth data of each test drilling.

3. The method for predicting lunar soil drilling sampling volume based on multi-dimensional telemetry data according to claim 1 is characterized in that: The method for obtaining the correlation between the sampling volume, the core lifting force, and the bottom hole drilling pressure in S2 by combining the drilling and coring mechanism is: F L +G=F TX +F D +F P Among them, F L Indicates drilling pressure, G indicates the drilling rig's own gravity, F TX Indicates the core lifting force, F P Indicates chip removal resistance, F D Indicates the bottom hole drilling pressure.

4. The method for predicting lunar soil drilling sampling volume based on multi-dimensional telemetry data according to claim 1 is characterized in that: In S2, the preliminary judgment and classification of each drilling operation type into a stable stage and a disturbance stage are as follows: a drilling test with a coring rate exceeding 75% is defined as a sampling stable test, and a test with a coring rate below 75% has a coring disturbance stage; a test with a sampling volume exceeding 500g is defined as a sampling stable test, and a test with a sampling volume below 500g has a coring disturbance stage.

5. The method for predicting lunar soil drilling sampling volume based on multi-dimensional telemetry data according to claim 1 is characterized in that: The method for constructing the sampling volume estimation model M1 based on the core lifting force load in S6 is: M1 = M 稳定 +M 扰动 .

6. The method for predicting lunar soil drilling sampling volume based on multi-dimensional telemetry data according to claim 1 is characterized in that: The load response characteristic signals of each drilling stage in S3 include the load mean, peak-to-peak value, variance, and slope.

7. The method for predicting lunar soil drilling sampling volume based on multi-dimensional telemetry data according to claim 1 is characterized in that: The time domain characteristic parameters of the core lifting force in the disturbance stage in S5 include the peak-to-peak value, variance, standard deviation, and peak factor of the core lifting force load.

8. A lunar soil drilling sampling volume prediction system based on multi-dimensional telemetry data, characterized in that: The system comprises: A data collection module, used to collect test data of nominal working conditions of drilling and production operations; The data analysis module is used to analyze the correlation between the sampling volume and the drilling load parameters based on the test data collected by the data collection module, and to obtain the correlation between the sampling volume and the core lifting force and the bottom hole drilling pressure in combination with the drilling coring mechanism; to make a preliminary judgment and divide the types of each drilling operation into a stable stage and a disturbance stage; The data extraction module is used to extract and analyze the load response characteristic signals of each drilling stage, and to correct the classification results of the drilling operation type in the data analysis module by combining machine learning to identify the sampling state; The data fitting module is used to perform multivariate fitting on the core lifting force, drilling depth and sampling volume during drilling according to the correlation between the sampling volume and the core lifting force in the data analysis module and the classification results of the stable sampling state in S3, and to construct an estimation model M for the stable sampling stage. 稳定 ; The calculation module is used to estimate the model M in the stable phase according to the classification results of the sampling state stable phase and the disturbance phase in the data fitting module. 稳定 Combined with the time-domain characteristic parameters of the core-lifting force in the disturbance stage, the time-domain characteristic parameters of the core-lifting force in the disturbance stage and the sampling amount are subjected to multivariate linear regression calculation to obtain the sampling disturbance stage estimation model M 扰动 ; The sampling volume estimation model construction module of the core lifting force load is used to combine the fitting results of the stable stage and the disturbance stage in the calculation module to construct the sampling volume estimation model M1 based on the core lifting force load; The sampling volume estimation model module of the bottom hole drilling pressure is used to calculate the area enclosed by the bottom hole drilling pressure curve according to the classification results of the stable and disturbed stages of the sampling state in the calculation module combined with the drilling depth, and obtain the stable and disturbed stage area S 稳定 / S 扰动 , the sampling volume of the core lifting force load estimation model construction module is used to estimate the area S of the stable and disturbed stages 稳定 / S 扰动 Multivariate fitting is performed with the actual sampling volume data of each drilling test, and the effective sampling power consumption W is established to obtain the sampling volume estimation model M2 based on the bottom hole drilling pressure; The sample prediction module is used to collect telemetry data of drilling and production operations in real time during ground simulation tests and on-orbit operations, and substitutes it into two types of estimation models to obtain the real-time sampling volume. The real-time sampling volume estimation range M3 is analyzed in combination with the model estimation deviation to predict the sample status.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium is used to store a computer program, and the computer program executes the method according to any one of claims 1 to 7.

10. A computer device comprising a memory and a processor, characterized in that A computer program is stored in the memory. When the processor runs the computer program stored in the memory, the processor implements the method according to any one of claims 1 to 7.

Images