Intelligent control method and system for post-peak failure process of hard rock under true triaxial surface disturbance

Abstract

The present invention provides an intelligent control method and system for the post-peak destruction process of hard rock under true triaxial surface disturbance, which divides the target rock into surface areas, and divides each surface area into a vertical coverage area, a first component force coverage area, and a second component force coverage area according to the force direction, and then analyzes the characteristic information change status in each area respectively, obtains the first influence correlation coefficient, the second influence correlation coefficient, and the third influence correlation coefficient respectively, and finally calculates the comprehensive control value based on each influence correlation coefficient, which is used to train the control parameter prediction model, so as to predict the optimal control parameters, which are used to effectively control the post-peak destruction process of hard rock of the target rock under true triaxial surface disturbance. In the above manner, the present invention can make full use of data resources to comprehensively count the changing laws of the internal structure of the rock, avoid the errors caused by analyzing only a single characteristic information, and improve the accuracy of data detection and analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of rock mechanics testing machine control, and in particular to an intelligent control method and system for a hard rock post-peak failure process under true triaxial surface disturbance. Background Art

[0002] The rock total stress-strain curve is a characteristic curve used to represent the change of stress with strain under the action of external forces or external factors. It can characterize the entire process of rock from initial stress deformation to the final loss of bearing capacity, and is of great significance for understanding the mechanical properties of rock. The rock total stress-strain curve can be obtained by rock mechanics testing using a rock mechanics testing machine. Based on the peak strength of the rock, the rock total stress-strain curve can be divided into a pre-peak curve and a post-peak curve. In practical applications, when conducting true triaxial surface disturbance tests, problems such as decreased control accuracy of the rock mechanics testing machine and inability to obtain the post-peak curve are likely to occur, affecting the analysis of the post-peak characteristics of the rock.

[0003] In the prior art, the patent with publication number CN110928181A provides an intelligent control method for the post-peak destruction process of hard rock under true triaxial surface disturbance. The patent detects and analyzes the disturbance force amplitude, disturbance force frequency and other parameters during the process of rock being disturbed by true triaxial surface disturbance, thereby dynamically optimizing the PID control parameters and the PID control parameter prediction model, thereby improving the accuracy and efficiency of PID control parameter prediction and the success rate of surface disturbance test. However, this method neither takes into account the change in water content caused by the structural changes at different positions inside the rock during the stress process, nor the degree of mutual correlation between the axial surfaces under true triaxial surface disturbance, resulting in the final parameter prediction result being not accurate enough, affecting the control effect of the post-peak destruction process of hard rock under true triaxial surface disturbance.

[0004] In view of this, it is necessary to design an improved intelligent control method and system for the post-peak failure process of hard rock under true triaxial surface disturbance to solve the above problems. Summary of the invention

[0005] In view of the defects of the above-mentioned prior art, the purpose of the present invention is to provide an intelligent control method and system for the post-peak destruction process of hard rock under true triaxial surface disturbance, which can more comprehensively statistically analyze the changing laws of the internal structure of the rock by subdividing the internal areas of the rock, avoid the errors caused by detecting and analyzing only single characteristic information, and improve the accuracy of data detection and analysis.

[0006] To achieve the above object, the present invention provides an intelligent control method for the post-peak failure process of hard rock under true triaxial surface disturbance, comprising the following steps:

[0007] Obtain a target rock and divide the target rock into surface areas; in each surface area, divide the surface area into a vertical coverage area, a first component force coverage area, and a second component force coverage area according to different force application directions;

[0008] According to a preset force threshold set, counting a first component action change difference value of an area covered by the first component force and a second component action change difference value of an area covered by the second component force;

[0009] Comparing the rock feature information before and after the force is applied in the vertical coverage area to obtain a first comparison result set; extracting a first influence correlation coefficient according to the first comparison result set;

[0010] Extracting a second influence correlation coefficient of changes between characteristic information within the area covered by the first component force according to the first component action change difference value; extracting a third influence correlation coefficient of changes between characteristic information within the area covered by the second component force according to the second component action change difference value;

[0011] According to the first influence correlation coefficient, the second influence correlation coefficient and the third influence correlation coefficient, a comprehensive control value is calculated; according to the comprehensive control value, a comprehensive control value data set is output, and a control parameter prediction model is trained; and the trained control parameter prediction model is used to output control parameter prediction data for controlling the post-peak destruction process of hard rock under true triaxial surface disturbance.

[0012] As a further improvement of the present invention, the direction of force applied in the vertical coverage area is perpendicular to the core; the directions of force applied to the first component force coverage area and the second component force coverage area are not perpendicular to the core, and form an angle with the corresponding straight line direction of the contact surface; the corresponding straight line direction of the contact surface is the straight line direction formed by the contact surface between the surface area where the first component force coverage area and the second component force coverage area are located and the adjacent surface area.

[0013] As a further improvement of the present invention, the step of dividing the target rock into surface areas comprises:

[0014] The three axial surfaces of the target rock under the true triaxial surface disturbance are correspondingly divided into three surface regions, and a three-surface region set is obtained.

[0015] As a further improvement of the present invention, the step of dividing the force coverage area into a vertical coverage area, a first component force coverage area, and a second component force coverage area comprises:

[0016] Collecting original rock feature information of each surface area in the three surface area sets before being disturbed by the true triaxial surface, to obtain an original rock feature information set;

[0017] According to the original rock feature information set, extracting the area in each of the three-surface area sets where the force applied is perpendicular to the core as the vertical coverage area;

[0018] Counting the directions of straight lines formed by contact surfaces between every two adjacent surface areas in the three surface area sets to obtain a contact surface straight line direction set;

[0019] According to the contact surface straight line direction set and the original rock feature information set, the area in each face area of ​​the three-face area set whose force direction is not perpendicular to the core and forms an angle with the corresponding contact surface straight line direction is extracted, and the area close to the first adjacent face is divided into the first component force coverage area, and the area close to the second adjacent face is divided into the second component force coverage area.

[0020] As a further improvement of the present invention, the statistical method of the first partial action change value and the second partial action change value is the same; wherein the step of statistically calculating the first partial action change value comprises:

[0021] According to the force threshold set, force characteristic information of the area covered by the first component force is collected to obtain first component force characteristic information;

[0022] According to the first component force characteristic information, a comprehensive force result of the component force characteristic information corresponding to the surface area adjacent to the first component force coverage area and the first component force characteristic information is counted to obtain a first comprehensive force result;

[0023] According to the first force component characteristic information, the contact area between the first force component coverage area and the adjacent surface area is counted to obtain a first contact area;

[0024] A combination of the first comprehensive force result and the first contact area is used as the first component action change difference value.

[0025] As a further improvement of the present invention, the step of extracting the first impact correlation coefficient according to the first comparison result set includes:

[0026] Extracting the rock internal structure crack change difference value and the water content change difference value in different location areas from the first comparison result set to obtain the first characteristic change difference value;

[0027] Extracting the difference between the original characteristic data value of the rock and the new characteristic data value obtained after the rock is disturbed by true three-plane disturbance from the first comparison result set, to obtain the first parameter change difference value;

[0028] An influence correlation coefficient between the first feature change difference value and the first parameter change difference value is extracted to obtain a first influence correlation coefficient.

[0029] As a further improvement of the present invention, the second impact correlation coefficient and the third impact correlation coefficient are extracted in the same method; wherein the step of extracting the second impact correlation coefficient comprises:

[0030] Applying strain force to the rock according to the force threshold set, collecting rock internal feature information in the area covered by the first component force after the force is applied, and outputting a second new rock internal feature information set;

[0031] Comparing the second new rock feature information set with the original rock feature information set one by one, and calculating the difference value to obtain a second feature change difference value;

[0032] An influence correlation coefficient between the second feature change difference value and the first component effect change difference value is extracted to obtain a second influence correlation coefficient.

[0033] As a further improvement of the present invention, the step of calculating the comprehensive control value includes:

[0034] Predicting a first required control value, a second required control value, and a third required control value of a force threshold set respectively according to the first influence correlation coefficient, the second influence correlation coefficient, and the third influence correlation coefficient;

[0035] The first required control value, the second required control value and the third required control value are averaged to obtain a comprehensive control value.

[0036] As a further improvement of the present invention, the control parameter prediction model is trained using the original rock feature information set, the force threshold set and the comprehensive control value data set.

[0037] To achieve the above object, the present invention also provides an intelligent control system for the post-peak failure process of hard rock under true triaxial surface disturbance, comprising:

[0038] Area division unit: used to divide the target rock into surface areas, and divide each surface area into a vertical coverage area, a first component coverage area and a second component coverage area;

[0039] A variation difference statistics unit: used for counting variation difference values ​​of the first component effect in the area covered by the first component force and variation difference values ​​of the second component effect in the area covered by the second component force;

[0040] The first correlation coefficient extraction unit is used to compare the rock feature information before and after the force is applied in the vertical coverage area, and extract the first influencing correlation coefficient;

[0041] Correlation coefficient extraction unit 2: used for extracting a second influencing correlation coefficient of changes between characteristic information within the first force component coverage area and a third influencing correlation coefficient of changes between characteristic information within the second force component coverage area;

[0042] Control parameter prediction unit: used to calculate the comprehensive control value and train the control parameter prediction model; and use the trained control parameter prediction model to output control parameter prediction data to control the post-peak failure process of hard rock under true triaxial surface disturbance.

[0043] The beneficial effects of the present invention are:

[0044] The intelligent control method and system for the post-peak destruction process of hard rock under true triaxial surface disturbance provided by the present invention divides the rock into three-sided areas, and further divides each surface area into a vertical coverage area, a first component force coverage area, and a second component force coverage area according to different force application directions. This facilitates the detection and differential analysis of the change status of characteristic information of different position areas inside the rock corresponding to each axial surface under the true triaxial surface disturbance, thereby more comprehensively statistically analyzing the change law of the overall internal structure of the rock and improving the accuracy of data detection and analysis.

[0045] On this basis, the present invention further analyzes the characteristic information changes in the vertical coverage area, the first component coverage area, and the second component coverage area in a differentiated manner, obtains the first, second, and third influence correlation coefficients, and finally performs comprehensive processing to obtain a comprehensive control value, effectively avoiding the large errors caused by only detecting and analyzing single characteristic information in traditional technologies, and at the same time solving the problem that traditional methods do not fully utilize data resources. The control parameter prediction model obtained based on the method provided by the present invention can predict the optimal control parameters to achieve accurate control of the hard rock post-peak destruction process under true triaxial surface disturbance, and improve the work efficiency of hard rock post-peak destruction process control. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic flow chart of the intelligent control method for the post-peak destruction process of hard rock under true triaxial surface disturbance provided by the present invention.

[0047] Figure 2 This is a structural schematic diagram of the intelligent control system for the post-peak destruction process of hard rock under true triaxial surface disturbance provided by the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0049] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the scheme of the present invention are shown in the drawings, while other details that are not closely related to the present invention are omitted.

[0050] In addition, it should be noted that the terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or apparatus that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or apparatus.

[0051] like Figure 1 As shown, the present invention provides an intelligent control method for the post-peak failure process of hard rock under true triaxial surface disturbance, comprising the following steps:

[0052] S1. Obtain a target rock and divide the target rock into surface areas; in each surface area, divide the surface area into a vertical coverage area, a first component force coverage area, and a second component force coverage area according to different force application directions;

[0053] S2. According to a preset force threshold set, counting the first component effect change difference value of the first component force coverage area and the second component effect change difference value of the second component force coverage area;

[0054] S3. Compare the rock feature information before and after the force is applied in the vertical coverage area to obtain a first comparison result set; extract a first influence correlation coefficient based on the first comparison result set;

[0055] S4. extracting a second influence correlation coefficient of changes between characteristic information within the first component force coverage area according to the first component force change difference value; extracting a third influence correlation coefficient of changes between characteristic information within the second component force coverage area according to the second component force change difference value;

[0056] S5. Calculate a comprehensive control value based on the first influence correlation coefficient, the second influence correlation coefficient and the third influence correlation coefficient; output a comprehensive control value data set based on the comprehensive control value, and train a control parameter prediction model; use the trained control parameter prediction model to output control parameter prediction data for controlling the post-peak destruction process of hard rock under true triaxial surface disturbance.

[0057] Among them, the force direction in the vertical coverage area is perpendicular to the core; the force directions in the first component force coverage area and the second component force coverage area are not perpendicular to the core, and form an angle with the corresponding straight line direction of the contact surface; the corresponding straight line direction of the contact surface is the straight line direction formed by the contact surface between the surface area where the first component force coverage area and the second component force coverage area are located and the adjacent surface area.

[0058] The present invention first divides the target rock into three-sided areas, so as to facilitate further information analysis and processing for each surface area. Since the main force application direction is perpendicular to the core, but there is a contact surface between the two adjacent surface areas, and the straight line formed by the contact surface has a certain angle, the angle formed will affect the force situation of the area, and then affect a series of changes in the internal structure of the rock in the surrounding area. Therefore, the present invention further divides each surface area into a vertical coverage area, a first component force coverage area, and a second component force coverage area according to the direction of the applied strain force, and analyzes and counts the characteristic data change difference value and the force contact area of ​​the first component force coverage area and the second component force coverage area, respectively. A variety of influencing characteristic information is analyzed and counted, so as to more comprehensively consider the diversity of influencing factors, and effectively avoid the large errors caused by only detecting and analyzing the single characteristic information in the traditional technology while making full use of data resources, and effectively improve the accuracy of data detection and analysis, so as to obtain the optimal control parameter prediction data.

[0059] In some preferred embodiments of the present invention, step S1 specifically includes the following contents:

[0060] S11. Obtain a target rock, divide the three axial surfaces of the target rock under the true triaxial surface disturbance into three surface regions, and obtain a three-surface region set.

[0061] S12. Collect original rock feature information of each surface area in the three surface area set before being disturbed by the true triaxial surface, and obtain an original rock feature information set.

[0062] S13. Based on the original rock feature information set, extract the region in each of the three-surface region sets where the force direction is perpendicular to the rock core as the vertical coverage region.

[0063] S14. Count the directions of straight lines formed by the contact surfaces between every two adjacent surface areas in the three surface area sets to obtain a contact surface straight line direction set.

[0064] S15. Based on the contact surface straight line direction set and the original rock feature information set, extract the area in each surface area of ​​the three surface area sets where the force direction is not perpendicular to the core and forms an angle with the corresponding contact surface straight line direction, and divide the area close to the first adjacent surface into the first component force coverage area, and divide the area close to the second adjacent surface into the second component force coverage area.

[0065] Among them, the three axial planes of the true triaxial are perpendicular to each other, and the same or different principal stresses (including major principal stress, intermediate principal stress and minor principal stress) are applied to each axial plane, which is used to study the mechanical properties and constitutive relations of materials under complex stress states. In practical applications, since rocks are irregular, when analyzing different rocks, it is necessary to divide the rock surface into three surface areas according to the axial planes of the true triaxial, which can be represented as surface A area, surface B area and surface C area, and form a three-surface area set.

[0066] Specifically, the original rock characteristic information in step S12 may include multiple data information such as the distribution of cracks inside the rock, the water content at different structural positions, the disturbance force amplitude, the disturbance force frequency, the deformation control rate, and the rock brittleness index.

[0067] In step S14, the contact surfaces between every two adjacent surface areas specifically include the contact surfaces between the surface area A and the surface area B, the surface area A and the surface area C, and the surface area B and the surface area C. By counting the straight line directions of each contact surface, a set of straight line directions of the contact surface can be obtained. Among them, one end of the straight line direction of the contact surface is connected to the core.

[0068] When dividing the coverage area, taking the A surface area as an example, the vertical coverage area, the first component coverage area, and the second component coverage area in the surface area can be represented as area a, area b, and area c, respectively. Among them, the force direction applied to area a is perpendicular to the core. Except for area a, the remaining areas are not connected as a whole, but are mainly divided into two parts, one is adjacent to the B surface area, and the other is adjacent to the C surface area. Since the force directions applied to the B surface area and the C surface area are different, the force conditions and rock feature information of the partial area adjacent to the B surface area and the partial area adjacent to the C surface area are different. Based on this, the present invention divides the area adjacent to the B surface area where the force direction is not perpendicular to the core and forms an angle with the straight line direction of the contact surface into area b, and divides the area adjacent to the C surface area where the force direction is not perpendicular to the core and forms an angle with the straight line direction of the contact surface into area c. By processing and analyzing area b and area c separately, the rock feature information in different areas can be more comprehensively considered to perform distinctive data analysis and improve the accuracy of the final result.

[0069] In some preferred embodiments of the present invention, step S2 specifically includes the following steps:

[0070] S21. Pre-set a force threshold set; based on the force threshold set, respectively count the force characteristic information of the area covered by the first component force and the area covered by the second component force to obtain the first component force characteristic information and the second component force characteristic information.

[0071] S22. Based on the first component force characteristic information, the corresponding component force characteristic information in the surface area adjacent to the first component force coverage area and the comprehensive force result of the first component force characteristic information are counted to obtain a first comprehensive force result; based on the second component force characteristic information, the corresponding component force characteristic information in the surface area adjacent to the second component force coverage area and the comprehensive force result of the second component force characteristic information are counted to obtain a second comprehensive force result.

[0072] S23. According to the first component force characteristic information, the contact area between the first component force coverage area and the adjacent surface area is counted to obtain the first contact area; according to the second component force characteristic information, the contact area between the second component force coverage area and the adjacent surface area is counted to obtain the second contact area.

[0073] S24. Use the combination of the first comprehensive force result and the first contact area as the first partial action change difference value; use the combination of the second comprehensive force result and the second contact area as the second partial action change difference value.

[0074] In the above process, the method for obtaining the first and second sub-action change difference values ​​is the same. The following is a specific description of the process of obtaining the first sub-action change difference value as an example:

[0075] In step S21, the preset force threshold set refers to the applied strain force parameters set for the true triaxial surface disturbance force, that is, three applied strain force parameters. The first force characteristic information is the change of multiple data in the rock after the strain force is applied, among which the force direction and force magnitude can be obtained by detection, and the rock data can be obtained by combining the acoustic wave detection technology and the image detection technology.

[0076] In step S22, taking area b in the surface area A as an example, after obtaining the first component force characteristic information of area b, it is also necessary to count the corresponding component force characteristic information in the surface area B adjacent to area b. The surface area B is also divided into a vertical coverage area, a first component force coverage area, and a second component force coverage area, which can be represented as area d, area e, and area f, respectively. Among them, area e is the first component force coverage area adjacent to the surface area A, and has the component force characteristics corresponding to area b. Therefore, the corresponding component force characteristic information in the surface area adjacent to area b is the component force characteristic information of area e. By comprehensively processing the component force characteristic information of areas b and e, the comprehensive force direction and comprehensive force magnitude are obtained, which is the first comprehensive force result.

[0077] In step S23, according to the rock feature information in area b, the above-mentioned comprehensive force direction and the comprehensive force magnitude, the actual force contact area between area b and area e can be analyzed, which is the first contact area. For example, in the original state without force, the contact area between area b and area e is assumed to be 100%, and the contact surface is less squeezed; after the force is applied, the actual contact area subjected to a larger squeeze force is 80%, and the remaining 20% ​​is still in a state of less squeeze. By calculating the actual contact area, the present invention can avoid targeted analysis of the part that is still in a state of less squeeze, so as to reduce the processing of unnecessary information, improve the accuracy of data analysis, improve data analysis efficiency, and reduce costs.

[0078] Through the above method, the first partial action change difference value finally obtained includes both the first comprehensive force result and the first contact area, which can fully and accurately reflect the actual force situation.

[0079] In some preferred embodiments of the present invention, step S3 specifically includes the following steps:

[0080] S31. According to the force threshold set, rock feature information in a vertical coverage area of ​​the rock obtained after the strain force application operation is collected to obtain a first new rock feature information set.

[0081] S32. Compare the first new rock feature information set with the original rock feature information set one by one to obtain a first comparison result set.

[0082] S33. Extract the rock internal structure crack change difference value and the water content change difference value in different location areas from the first comparison result set as the first characteristic change difference value.

[0083] S34. Extract the difference between the original characteristic data value of the rock and the new characteristic data value obtained after the rock is disturbed by true three-plane disturbance from the first comparison result set as the first parameter change difference value.

[0084] S35. Extract the influence correlation coefficient between the first feature change difference value and the first parameter change difference value as the first influence correlation coefficient.

[0085] Specifically, in step S31, the information types of the rock characteristic information in the first new rock characteristic information set are consistent with the information types of the original rock characteristic information, and still refer to multiple data information such as the crack distribution status inside the rock, the water content status at different structural positions, the disturbance force amplitude, the disturbance force frequency, the deformation control rate, and the rock brittleness index. The only difference is that each piece of data information is collected under stress so as to be compared one by one with the corresponding data information before stress.

[0086] In step S33, the first characteristic change difference value obtained is the change of the crack distribution in area a and the increase, decrease and transfer of the water content. The change of the crack distribution can be counted by the distribution diffusion area data; the increase, decrease and transfer of the water content can be counted by the water content change value of the position area corresponding to the crack distribution change. For example, the increased crack coverage area corresponds to the increased or decreased water content data value. The difference value of these characteristic data and the original rock feature information before collection are calculated to obtain the first characteristic change difference value.

[0087] In step S34, the first parameter change difference value refers to the statistics of the change difference values ​​of other items of data except the two items of crack distribution status and water content status at different structural positions.

[0088] In step S35, the first impact correlation coefficient can be calculated in different ways according to the actual impact situation. The following is an example of the calculation method in one embodiment of the present invention:

[0089] The first average ratio is obtained by calculating the ratio of the crack distribution change area data to the disturbance force amplitude change data, the disturbance force frequency, the deformation control rate, and the rock brittleness index, and then averaging the four ratios. The second average ratio is obtained by calculating the average of the ratios of the water content status at different structural positions with respect to the other four data information in a similar manner. The comprehensive average ratio is obtained by averaging the first average ratio and the second average ratio, which is the first influencing correlation coefficient.

[0090] In some preferred embodiments of the present invention, step S4 specifically includes the following steps:

[0091] S41. Apply strain force to the rock according to the force threshold set, and collect rock feature information in the area covered by the first component force after the force is applied as a second new rock feature information set; collect rock feature information in the area covered by the second component force after the force is applied as a third new rock feature information set.

[0092] S42. Compare the second new rock feature information set with the original rock feature information set one by one, and calculate the difference value to obtain the second feature change difference value; compare the third new rock feature information set with the original rock feature information set one by one, and calculate the difference value to obtain the third feature change difference value.

[0093] S43. Extract the influence correlation coefficient between the second characteristic change difference value and the first sub-action change difference value to obtain the second influence correlation coefficient; extract the influence correlation coefficient between the third characteristic change difference value and the second sub-action change difference value to obtain the third influence correlation coefficient.

[0094] In the above process, the methods for obtaining the second impact correlation coefficient and the third impact correlation coefficient are the same. The following is a specific description of the process of obtaining the second impact correlation coefficient as an example:

[0095] In step S41 , the information type of the rock feature information in the second new rock feature information set is consistent with the rock feature information in the first new rock feature information set in step S31 , and will not be described in detail here.

[0096] In step S42, the second characteristic change difference value specifically refers to the change difference data of six data before and after the force is applied, namely, the crack distribution condition inside the rock, the water content condition at different structural positions, the disturbance force amplitude, the disturbance force frequency, the deformation control rate, and the rock brittleness index.

[0097] In step S43, the second impact correlation coefficient can be calculated in different ways according to the actual impact situation. The following is an example of the calculation method in one embodiment of the present invention:

[0098] The third average ratio is obtained by respectively calculating the ratio of the six data items in the second characteristic change difference value to the first comprehensive force result in the first partial action change difference value, and then averaging the six ratios; then the fourth average ratio is obtained by respectively calculating the ratio of the six data items in the second characteristic change difference value to the first contact area in the first partial action change difference value, and averaging the ratios in a similar manner; finally, the third average ratio and the fourth average ratio are averaged, and the obtained comprehensive average ratio is the second influence correlation coefficient.

[0099] In some preferred embodiments of the present invention, step S5 specifically includes the following steps:

[0100] S51. Predict the first required control value, the second required control value and the third required control value of the force threshold set respectively according to the first influence correlation coefficient, the second influence correlation coefficient and the third influence correlation coefficient.

[0101] S52. Take an average of the first required control value, the second required control value and the third required control value to obtain a comprehensive control value.

[0102] S53. The control parameter prediction model is trained using the original rock feature information set, the force threshold set and the comprehensive control value data set to obtain a trained control parameter prediction model.

[0103] S54. Obtain the rock to be processed, collect the real-time rock characteristic information set of the rock to be processed, and preset the real-time force parameters; then input the real-time rock characteristic information set and the real-time force parameters into the trained control parameter prediction model, and output the optimal control parameter prediction data; according to the optimal control parameter prediction data, control the post-peak destruction process of the hard rock to be processed under the true triaxial surface disturbance.

[0104] Specifically, in step S51, taking the first impact correlation coefficient and the first required control value as an example, assuming that y=kx, where y is the first required control value, k is the first impact correlation coefficient, and x is the force threshold, the first required control value of the force threshold set can be predicted according to the first impact correlation coefficient. The prediction method of the second required control value and the third required control value is consistent with the above content and will not be repeated here.

[0105] In step S52, for each rock to be processed, the first required control value, the second required control value and the third required control value corresponding to each surface area can obtain a comprehensive control value, and the three surface areas obtain a total of one comprehensive control value as a set of comprehensive control values. By processing different target rocks in a similar manner, the multiple sets of comprehensive control value data formed can form a comprehensive control value data set for training the control parameter prediction model.

[0106] In step S53, by pairing information of each data set, a group of rock characteristic information and a group of force thresholds are combined into a group of characteristic information, corresponding to a group of comprehensive control values ​​in the comprehensive control value data set, to form several paired information groups, i.e., pairing tables, which are used to train the control parameter prediction model.

[0107] Through the above method, the present invention can match the information based on the real-time rock feature information and real-time force parameters collected with the matching table formed in the trained control parameter prediction model to obtain a set of comprehensive control value data, which is the optimal control parameter prediction data. According to the optimal control parameter prediction data, the hard rock post-peak destruction process of the treated rock under true triaxial surface disturbance can be controlled in real time. Parameter control can be achieved, and further parameter verification effects can be achieved, which improves the accuracy of the analysis results and the work efficiency of the hard rock post-peak destruction process control, and improves the utilization rate of resources.

[0108] like Figure 2 As shown, the present invention also provides an intelligent control system for the post-peak destruction process of hard rock under true triaxial surface disturbance, comprising:

[0109] Area division unit: used to divide the target rock into surface areas, and divide each surface area into a vertical coverage area, a first component coverage area and a second component coverage area;

[0110] A variation difference statistics unit: used for counting variation difference values ​​of the first component effect in the area covered by the first component force and variation difference values ​​of the second component effect in the area covered by the second component force;

[0111] The first correlation coefficient extraction unit is used to compare the rock feature information before and after the force is applied in the vertical coverage area, and extract the first influencing correlation coefficient;

[0112] Correlation coefficient extraction unit 2: used for extracting a second influencing correlation coefficient of changes between characteristic information within the first force component coverage area and a third influencing correlation coefficient of changes between characteristic information within the second force component coverage area;

[0113] Control parameter prediction unit: used to calculate the comprehensive control value and train the control parameter prediction model; and use the trained control parameter prediction model to output control parameter prediction data to control the post-peak failure process of hard rock under true triaxial surface disturbance.

[0114] In summary, the present invention provides an intelligent control method and system for the post-peak destruction process of hard rock under true triaxial surface disturbance, by dividing the target rock into surface areas, and dividing each surface area into a vertical coverage area, a first component force coverage area, and a second component force coverage area according to the force direction, and then performing a differential analysis on the characteristic information change status in each area, respectively, and obtaining the first influence correlation coefficient, the second influence correlation coefficient, and the third influence correlation coefficient, and finally calculating the comprehensive control value based on each influence correlation coefficient for training the control parameter prediction model, so as to predict the optimal control parameters, which are used to effectively control the post-peak destruction process of hard rock of the target rock under true triaxial surface disturbance. Through the above method, the present invention can make full use of data resources to comprehensively count the changing laws of the internal structure of the rock, avoid the large errors caused by the detection and analysis of only single characteristic information in the traditional technology, and effectively improve the accuracy of data detection and analysis.

[0115] The above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. An intelligent control method for the post-peak failure process of hard rock under true triaxial surface disturbance, characterized in that: The steps include: Obtaining a target rock, and performing surface area division on the target rock: dividing the three axial surfaces of the target rock under the true triaxial surface disturbance into three surface areas, and obtaining a three-surface area set; In each face region, the face region is divided into a vertical coverage region, a first component force coverage region, and a second component force coverage region according to different force application directions: the original rock feature information of each face region in the three face region sets before being subjected to true triaxial face disturbance is collected to obtain an original rock feature information set; based on the original rock feature information set, the region in each face region in the three face region sets where the force application direction is perpendicular to the rock core is extracted as the vertical coverage region; the straight line direction formed by the contact surface between every two adjacent face regions in the three face region sets is counted to obtain a contact surface straight line direction set; based on the contact surface straight line direction set and the original rock feature information set, the region in each face region in the three face region sets where the force application direction is not perpendicular to the rock core and forms an angle with the corresponding contact surface straight line direction is extracted, and the region close to the first adjacent face is divided into the first component force coverage region, and the region close to the second adjacent face is divided into the second component force coverage region; According to a preset force threshold set, counting a first component action change difference value of an area covered by the first component force and a second component action change difference value of an area covered by the second component force; The statistical method of the first component action change difference value and the second component action change difference value is the same; wherein, the step of statistically analyzing the first component action change difference value comprises: according to the force threshold set, statistically analyzing the force characteristic information of the area covered by the first component force to obtain the first component force characteristic information; according to the first component force characteristic information, statistically analyzing the comprehensive force result of the corresponding component force characteristic information in the surface area adjacent to the first component force coverage area and the first component force characteristic information to obtain the first comprehensive force result; according to the first component force characteristic information, statistically analyzing the contact area between the first component force coverage area and the adjacent surface area to obtain the first contact area; and using the combination of the first comprehensive force result and the first contact area as the first component action change difference value; Compare the rock feature information before and after the force is applied in the vertical coverage area to obtain a first comparison result set; extract a first influence correlation coefficient based on the first comparison result set: extract the difference value of the rock internal structure crack change and the difference value of the water content change in different position areas from the first comparison result set to obtain a first feature change difference value; extract the difference value between the original rock feature data value and the new feature data value obtained after the rock is subjected to true three-plane disturbance from the first comparison result set to obtain a first parameter change difference value; extract the influence correlation coefficient between the first feature change difference value and the first parameter change difference value to obtain a first influence correlation coefficient; According to the first component action change difference value, extract the second influence correlation coefficient of the change between the characteristic information in the area covered by the first component force; according to the second component action change difference value, extract the third influence correlation coefficient of the change between the characteristic information in the area covered by the second component force; the second influence correlation coefficient and the third influence correlation coefficient are extracted in the same method; wherein the step of extracting the second influence correlation coefficient comprises: applying strain force to the rock according to the force threshold set, collecting the rock characteristic information in the area covered by the first component force after the force is applied, and outputting a second new rock characteristic information set; comparing the second new rock characteristic information set with the original rock characteristic information set one by one, and calculating the difference value to obtain the second characteristic change difference value; extracting the influence correlation coefficient between the second characteristic change difference value and the first component action change difference value to obtain the second influence correlation coefficient; calculating the comprehensive control value according to the first influence correlation coefficient, the second influence correlation coefficient and the third influence correlation coefficient; outputting the comprehensive control value data set according to the comprehensive control value, and training the control parameter prediction model; using the trained control parameter prediction model to output control parameter prediction data for controlling the post-peak destruction process of hard rock under true triaxial surface disturbance.

2. The intelligent control method for the post-peak failure process of hard rock under true triaxial surface disturbance according to claim 1 is characterized by: The step of calculating the comprehensive control value comprises: Predicting a first required control value, a second required control value, and a third required control value of a force threshold set respectively according to the first influence correlation coefficient, the second influence correlation coefficient, and the third influence correlation coefficient; The first required control value, the second required control value and the third required control value are averaged to obtain a comprehensive control value.

3. The intelligent control method for the post-peak failure process of hard rock under true triaxial surface disturbance according to claim 1 is characterized by: The control parameter prediction model is trained using the original rock feature information set, the force threshold set and the comprehensive control value data set.

4. An intelligent control system for the post-peak failure process of hard rock under true triaxial surface disturbance, characterized in that: The method for intelligently controlling the post-peak destruction process of hard rock under true triaxial surface disturbance as claimed in any one of claims 1 to 3 comprises: Area division unit: used to divide the target rock into surface areas, and divide each surface area into a vertical coverage area, a first component coverage area and a second component coverage area; A variation difference statistics unit: used for counting variation difference values ​​of the first component effect in the area covered by the first component force and variation difference values ​​of the second component effect in the area covered by the second component force; The first correlation coefficient extraction unit is used to compare the rock feature information before and after the force is applied in the vertical coverage area, and extract the first influencing correlation coefficient; Correlation coefficient extraction unit 2: used for extracting a second influencing correlation coefficient of changes between characteristic information within the first force component coverage area and a third influencing correlation coefficient of changes between characteristic information within the second force component coverage area; Control parameter prediction unit: used to comprehensively calculate the comprehensive control value according to the first influence correlation coefficient, the second influence correlation coefficient and the third influence correlation coefficient, and train the control parameter prediction model; and use the trained control parameter prediction model to output control parameter prediction data to control the post-peak failure process of hard rock under true triaxial surface disturbance.

Images