Intelligent true triaxial mechanical test system and test method based on parameter self-adaption

Abstract

The invention relates to the technical field of rock indoor loading tests, in particular to an intelligent true triaxial mechanical test system and method based on parameter self-adaptation, and the method comprises the following steps: S1, testing a known rock sample based on known test data, historical parameter data, historical operation data, historical experiment data and rock characteristic data of the rock sample are obtained; s2, marking corresponding time points for the historical parameter data and the historical experiment data based on the operation time data to obtain a parameter subset and an experiment subset; s3, analyzing the parameter subset and the experiment subset to obtain an intervention relation; s4, determining a data relationship between the rock characteristics and test parameters of true triaxial test equipment to obtain a basic parameter relationship; s5, testing and data monitoring are carried out on a rock sample to be tested, rock experiment data are obtained, judgment and self-adaptive parameter adjustment are carried out based on the intervention relation, and the reliability and accuracy of a test result are guaranteed while the analysis rate is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of rock indoor loading test, and in particular to an intelligent true triaxial mechanical test system and a test method based on parameter self-adaptation. Background Art

[0002] The purpose of true triaxial rock testing is to understand the mechanical properties of deep engineering rock under triaxial stress, provide basic data for revealing deep rock failure mechanisms and evaluating rock mass engineering stability, and provide essential parameters for engineering design and construction. Existing true triaxial testing equipment typically requires personnel to evaluate the test parameters based on rock properties before the test. However, due to errors in the initial human evaluation and changes in rock data during actual testing, the true triaxial testing equipment can only perform tests mechanically based on the pre-determined test parameters, which can easily lead to test failures.

[0003] The Chinese invention patent with authorization announcement number CN108414346B discloses an intelligent true triaxial testing system and testing method with adaptive test parameters, including: using a self-learning model trained based on existing data samples to obtain the corresponding reasonable range of true triaxial test condition parameters according to the properties of the rock sample; setting the true triaxial test condition parameters based on the obtained reasonable range of the true triaxial test condition parameters, simulating the true triaxial test output, and determining whether the test results of the simulated true triaxial test output are within the load capacity of the current true triaxial testing equipment: if yes, outputting control instructions based on the set true triaxial test condition parameters to control the true triaxial testing equipment to perform a true triaxial test on the rock sample according to the set true triaxial test condition parameters, obtaining pressure, displacement and temperature data at each moment during the true triaxial test; calculating the real-time deformation rate of the rock in all directions in real time, and controlling the loading rate in real time to complete the true triaxial test; otherwise, resetting the true triaxial test condition parameters, and obtaining reasonable test parameters by adaptively adjusting the test rock sample to guide the test personnel to set the true triaxial test parameters, thereby improving the test efficiency and success rate. The present invention provides a test method that is different from the above-mentioned disclosed scheme. The present invention aims to improve the analysis rate while ensuring the reliability and accuracy of the test results. Summary of the Invention

[0004] In order to overcome the deficiencies in the prior art, the present invention provides an intelligent true triaxial mechanical testing system and a testing method based on parameter adaptation.

[0005] To achieve the above-mentioned purpose, the present invention discloses an intelligent true triaxial mechanical testing method based on parameter adaptation, which includes the following steps: S1: The true triaxial test equipment tests a known rock sample based on known test data, monitors the test process data of the true triaxial test equipment, and obtains historical parameter data, historical operation data, historical experimental data of the true triaxial test equipment, and rock property data of the rock sample; S2: Obtaining operation time data from the historical operation data, marking corresponding time points for the historical parameter data and historical experimental data based on the operation time data, and dividing the historical parameter data and historical experimental data according to the marked time points to obtain parameter subsets and experimental subsets; S3: Analyze the parameter subset and the experiment subset based on the operation time data, determine the operation instructions required under different experimental results and the operation data corresponding to the operation instructions, and obtain the intervention relationship; S4: Comprehensively analyzing the historical parameter data and rock property data, determining the data relationship between the rock properties and the test parameters of the true triaxial test equipment, and obtaining the basic parameter relationship; S5: Obtain rock characteristic data of the rock sample to be tested, and generate test parameter data required for the test based on the basic parameter relationship, conduct tests and data monitoring on the rock sample to be tested based on the test parameter data to obtain rock test data; judge and adaptively adjust the rock test data based on the intervention relationship.

[0006] Furthermore, step S2 includes the following steps: S21: dividing the historical experimental data based on the historical parameter data, and determining the execution instructions at each time point in the historical experimental data and the execution duration of the corresponding execution instructions; S22: determining an execution instruction corresponding to the operation time data based on the operation time data, determining a data selection interval for the historical parameter data according to the execution duration of the execution instruction, and selecting the historical parameter data according to the data selection interval to obtain a parameter subset; S23: Divide the historical experimental data based on the parameter subset to obtain an experimental subset.

[0007] Furthermore, in step S22, if the time point corresponding to the operation time data is located at the intersection between two execution instructions, the sum of the execution times of the two execution instructions is used as the data selection interval according to the time sequence, and then the historical parameter data is selected according to the data selection interval to obtain a parameter subset, and the intersection point is the intersection position where one instruction ends and another instruction begins.

[0008] Furthermore, the step S3 is specifically as follows: Determine the inflection points of the parameter subset and the experimental subset, and determine whether the operation time data is the inflection point of the parameter subset and the experimental subset based on the operation time data; the inflection point of the parameter subset is the start time data and end time data of each execution instruction in the parameter subset; the inflection point of the experimental subset is the time point when the experimental data collected in the experimental subset meets the experimental requirements and the time point when the experimental data undergoes a sudden change; Result A: If the time point of the operation time data is the inflection point of the parameter subset and the experimental subset, then the operation instruction corresponding to the historical operation data is determined to be an interrupt operation, and the corresponding intervention relationship is determined; Result B: If the time point of the operation time data is the inflection point of the parameter subset but not the inflection point of the experimental subset, then the operation instruction corresponding to the historical operation data is determined to be a cyclic operation, and the corresponding intervention relationship is determined; Result C: If the time point of the operation time data is the inflection point of the experimental subset but not the inflection point of the parameter subset, then the operation instruction corresponding to the historical operation data is determined to be a parameter modification operation, and the corresponding intervention relationship is determined; Result D: If the time point of the operation time data is not the inflection point of the experimental subset and the inflection point of the parameter subset, the operation instruction corresponding to the historical operation data is determined to be a load frequency modulation operation, and the corresponding intervention relationship is determined.

[0009] Furthermore, the method for determining the result A is: Determine a time interval of an experimental subset containing a corresponding time point based on the operation time data, and perform inflection point judgment on the experimental subset based on the time interval; If it is determined that there are multiple inflection points within the time interval, the operation instruction corresponding to the historical operation data is determined to be an interrupt operation, and a corresponding relationship between the inflection point data of the corresponding experimental subset and the operation instruction is established to obtain an intervention relationship.

[0010] Furthermore, the method for determining result B is: Compare historical experimental data with corresponding experimental requirement data to determine whether the historical experimental data meets the experimental requirement data; If the historical experimental data does not meet the experimental requirement data, the execution instructions before the corresponding time point and the execution instructions after the corresponding time point on the parameter subset are compared to determine whether the execution instructions before the corresponding time point and the execution instructions after the corresponding time point are repeated instructions; If it is determined to be a repeated instruction, the operation instruction is determined to be a loop operation, and a relationship is established between the operation instruction and the corresponding experimental subset to obtain an intervention relationship. Otherwise, the historical operation data is determined to be an invalid operation.

[0011] Furthermore, the result C is determined by: The historical experimental data before the inflection point are judged for data changes to obtain the data change rules of the corresponding experimental collection items; Match the historical operation data corresponding to the operation time data with the historical parameter data to determine the parameter item modified by the historical operation data; Based on the modified parameter items, the parameter item data before modification is compared with the historical experimental data before the inflection point, and the parameter item data after modification is compared with the historical experimental data after the inflection point, and the relationship between the data change of the corresponding parameter item and the data change of the experimental data is determined to obtain the first data change relationship; the correspondence between the data change law, the first data change relationship and the corresponding operation instruction is established to obtain the intervention relationship.

[0012] Furthermore, the result D is determined by: Based on the time point of the operation time data, the experimental subset within the corresponding time interval is divided into the first experimental sub-data before the time point and the second experimental sub-data after the time point; Determine the data volume change rate of the first experimental sub-data and the second experimental sub-data to obtain the data change rates of the two experimental sub-data before and after the time point, and record them as the first data change rate and the second data change rate respectively; Based on historical operation data, the adjustment result of the loaded frequency modulation parameter item in the corresponding operation instruction is determined to obtain the adjustment data volume, and based on the difference between the adjustment data volume and the first data change rate and the second data change rate, the relationship between the adjustment data volume and the data change rate is determined to obtain the second data change relationship; the corresponding relationship between the first data change rate, the second data change relationship and the corresponding operation instruction is constructed to obtain the intervention relationship.

[0013] Furthermore, the step S4 is: Obtain historical parameter data and corresponding historical operation data of the corresponding experiment, and modify the corresponding parameter values ​​of the historical parameter data based on the historical operation data to obtain the ideal parameter data of the corresponding experiment; Match the historical parameter data of each test with the rock property data of the corresponding test to determine the parameter items corresponding to the characteristics of different rocks; The ideal parameter data of the parameter items with the same rock characteristics and the ideal parameter data of the same parameter items with different rock characteristics are statistically analyzed to determine the relationship between the rock characteristics and the data values ​​of the parameter items and obtain the basic parameter relationship.

[0014] The present invention also discloses a test system based on the above-mentioned parameter-adaptive intelligent true triaxial mechanical test method, which includes a database establishment module, a data relationship construction module and an adaptive adjustment module; The database establishment module is configured to allow the true triaxial test equipment to test known rock samples based on known test data, monitor the data of the true triaxial test equipment during the test, and obtain historical parameter data, historical operation data, historical experimental data of the true triaxial test equipment, and rock property data of the rock samples; The data relationship construction module obtains operation time data of historical operation data, and divides historical parameter data and historical experimental data based on the operation time data to obtain parameter subsets and experimental subsets; analyzes the parameter subsets and experimental subsets to determine the operation instructions required under different test conditions and the operation data corresponding to the operation instructions, thereby obtaining an intervention relationship; and comprehensively analyzes the historical parameter data and rock property data to determine the data relationship between the rock properties and the test parameters of the true triaxial test equipment, thereby obtaining a basic parameter relationship; The adaptive adjustment module obtains rock characteristic data of the rock sample to be tested, generates test parameter data required for the test according to the basic parameter relationship, and makes judgments and data adaptive adjustments on the rock sample to be tested based on the intervention relationship.

[0015] The beneficial effects of the present invention are: 1. The present invention retests and monitors data on known rock samples to construct a database of parameter data, operation data, experimental data, and rock property data for different rock samples. By reading the operation time point of the operation data during each test, the data segments with human intervention are determined, thereby reducing the amount of data to be analyzed during data relationship analysis, improving the analysis rate, improving the data quality of the data to be analyzed, and improving the accuracy of the analysis results. By utilizing the change relationship between the experimental subset and the parameter subset, the execution or non-execution of each operation instruction in the operation data and the execution data volume are constructed. By utilizing the rock property data and the corresponding parameter data, the data obtained by human analysis are mechanized and intelligently converted, reducing human brain activity. By utilizing the intervention relationship and basic parameter relationship obtained by analysis, the rock samples to be tested are tested for test monitoring and data analysis, and the true triaxial test equipment during the test process is timely adjusted for data, ensuring the normal progress of the test and improving the test reliability. 2. With the help of historical reference data, the execution instructions and the execution time of the corresponding execution instructions at each time point in the test process are clarified. When selecting data at the corresponding time point based on the operation time data, the integrity of the selected data is guaranteed, the data quality of the selected data is improved, and the validity of the judgment results is improved. At the same time, by selecting historical parameter data and historical experimental data based on the operation time data, the corresponding relationship between historical operation data and historical parameter data and historical experimental data is clarified, thereby further improving the accuracy of the analysis and judgment results; 3. Comprehensively utilize the time dimension relationship between the operation time data and the parameter subset and the experimental subset in the historical operation data to clarify the operation purpose of the corresponding operation, and then adopt different judgment methods according to different operation purposes, thereby improving the judgment accuracy of the data relationship between the operation data and the parameter data and the experimental data under the corresponding operation purpose. Then, when the true triaxial test of rock is carried out according to the intervention relationship obtained by judgment and adaptive test monitoring is performed, the normal execution of the test can be effectively ensured, thereby improving the reliability of the test results. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flow chart of the overall method steps of the present invention; Figure 2 Flowchart of the steps for obtaining parameter subsets and experimental subsets in the present invention. DETAILED DESCRIPTION

[0017] The following is combined with Figure 1 To the attached Figure 2 The principles and features of the present invention are described, and the examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0018] Example 1: Intelligent true triaxial mechanical testing method based on parameter adaptation, such as Figure 1 As shown, the following steps are included: S1: The true triaxial test equipment tests a known rock sample based on known test data, monitors the test process data of the true triaxial test equipment, and obtains historical parameter data, historical operation data, historical experimental data of the true triaxial test equipment, and rock property data of the rock sample; S2: Obtain operation time data from the historical operation data, mark corresponding time points for the historical parameter data and historical experimental data based on the operation time data, and divide the historical parameter data and historical experimental data according to the marked time points to obtain parameter subsets and experimental subsets; S3: Analyze the parameter subset and the experiment subset based on the operation time data, determine the operation instructions required under different experimental results and the operation data corresponding to the operation instructions, and obtain the intervention relationship; S4: Comprehensively analyze historical parameter data and rock property data to determine the data relationship between rock properties and test parameters of true triaxial test equipment, and obtain basic parameter relationships; S5: Obtain rock property data of the rock sample to be tested, and generate test parameter data required for the test based on the basic parameter relationship. Perform tests and data monitoring on the rock sample to be tested based on the test parameter data to obtain rock test data; judge and adaptively adjust the rock test data based on the intervention relationship.

[0019] In this embodiment, by retesting and data monitoring known rock samples, a database of parameter data, operation data, experimental data, and rock property data for different rock samples is constructed. By reading the operation time point of the operation data during each test, the data segment with human intervention is determined, thereby reducing the amount of data to be analyzed when performing data relationship analysis, thereby improving the analysis rate, improving the data quality of the data to be analyzed, and improving the accuracy of the analysis results. Specifically, the reason for the reduction in the amount of data to be analyzed is that in this embodiment, only the data segment with human intervention needs to be analyzed. When no human intervention occurs, the test results are executed according to the pre-set operation data. When human intervention occurs, it indicates that the test results exceed the expectations of the test personnel, and therefore adjustments are required to achieve the target effect. Therefore, by analyzing the data segment with human intervention, the reason for the human intervention in the test process is determined, and then the corresponding automated operation can be performed according to the corresponding situation during the next test, thereby improving the test efficiency and the accuracy of the results. By utilizing the change relationship between the experimental subset and the parameter subset, the execution and execution data volume of each operation instruction in the operation data are constructed. By utilizing the rock characteristic data and the corresponding parameter data, the data obtained from manual analysis are mechanized and intelligently converted, reducing human mental activities. By utilizing the intervention relationship and basic parameter relationship obtained by analysis, the rock samples to be tested are monitored and data analyzed, and the data of the true triaxial test equipment during the test is adjusted in time to ensure the normal progress of the test and improve the reliability of the test.

[0020] Specifically, when conducting a true triaxial test, staff first cut and process the rock sample to be tested according to regulations. During this process, the basic properties of the rock sample can be determined, such as whether the rock is brittle or ductile, porous or low-permeability, soft or layered, etc. After determining the basic properties of the rock sample, staff manually judge the basic properties of the rock sample to develop a set of test data for the true triaxial test. Typical test data includes loading path and rate, confining pressure, and pore pressure. Based on the developed test data, various parameters of the true triaxial test equipment are adjusted. For example, if the test parameters are the order of loading paths X, Y, and Z, then the loading path items X, Y, and Z on the true triaxial test equipment are activated. If the loading rate is a, then the loading rate of loading paths X, Y, and Z on the true triaxial test equipment is adjusted to a. The equipment with adjusted parameters is tested. During the test, since the parameters are manually evaluated, some deviations may occur in the actual test process. In order to ensure the safety and effectiveness of the test, when some deviations occur or to speed up the test efficiency, it is necessary to manually readjust the equipment with adjusted parameters to ensure the effective conduct of the test. In this process, the manual readjustment constitutes the operation data, and the collection of a certain test result of the rock sample during the test constitutes the experimental data.

[0021] At the same time, since the rock samples to be tested are unknown, the normal progress of the test cannot be guaranteed when the initial parameter adjustment of the true triaxial test equipment is performed. At the same time, in order to improve the intelligence of the true triaxial test and reduce the number of human interventions, it is necessary to determine the relationship between rock property data and parameter data based on historical test data, and to determine the relationship between operation data, experimental data, and parameter data. Specifically, by retesting known rock samples, a sample database is constructed. By analyzing the sample database, the relationship between rock property data and parameter data, as well as the relationship between operation data, experimental data, and parameter data, is determined. Therefore, when performing true triaxial tests on unknown rock samples, a valid set of parameter data can be generated based only on the input rock property data. During the test, by monitoring the experimental data and the parameter data, it is determined whether parameter adjustment is required and the corresponding adjustment amount, ensuring the correct progress of the test. This realizes the adaptive adjustment of parameters in the true triaxial test, improves the intelligence level of the true triaxial test equipment, and thus improves the reliability of the test results.

[0022] Furthermore, if Figure 2 As shown, step S2 includes the following steps: S21: Divide the historical experimental data based on the historical parameter data, and determine the execution instructions at each time point in the historical experimental data and the execution duration of the corresponding execution instructions; wherein the execution duration is the execution duration of the execution instruction in the actual experiment.

[0023] S22: determining an execution instruction corresponding to the operation time data based on the operation time data, determining a data selection interval for the historical parameter data based on the execution duration of the execution instruction, and selecting the historical parameter data based on the data selection interval to obtain a parameter subset; If the time point corresponding to the operation time data is located at the intersection between two execution instructions, the sum of the execution time of the two execution instructions is used as the data selection interval according to the time sequence, and then the historical parameter data is selected according to the data selection interval to obtain a parameter subset; the intersection point is the intersection position when one instruction ends and another instruction begins. For example, there are execution instructions a and execution instructions b, the execution time interval of execution instruction a is (0, 10), and the execution time interval of execution instruction b is (10, 20). If the time point corresponding to the operation time data is 10, it is determined that the time point is located at the intersection of execution instructions a and b.

[0024] S23: Divide the historical experimental data based on the parameter subset to obtain an experimental subset.

[0025] In this embodiment, the execution instructions at each time point in the test process and the execution time of the corresponding execution instructions are clarified by utilizing historical reference data, so that when the data at the corresponding time point is selected according to the operation time data, the integrity of the selected data is guaranteed, the data quality of the selected data is improved, and the validity of the judgment result is improved. At the same time, the historical parameter data and historical experimental data are selected through the operation time data, and the correspondence between the historical operation data and the historical parameter data and the historical experimental data is clarified, thereby further improving the accuracy of the analysis and judgment results.

[0026] Specifically, when dividing historical parameter data and historical experimental data according to the operation time data in the historical operation data, it is necessary to ensure that the selected data can effectively identify the change pattern of the data, and the amount of data must be as small as possible, so that the divided data can be more concise and accurate.

[0027] Since the execution result corresponding to the same execution instruction in the parameter data should remain unchanged, and if there are no problems with the execution process, no human intervention is required. Conversely, when human intervention occurs, it indicates that the execution instruction is no longer suitable for the current test situation. Similarly, since human intervention is judged based on the actual test situation, when the actual test situation does not generate as expected, it indicates that intervention adjustment is needed. Therefore, by selecting the parameter data when human intervention occurs, it is possible to determine the experimental data that does not match the parameter data, thereby achieving both the accuracy of data selection and the simplicity of the selected data, thereby improving data quality.

[0028] For example, the historical parameter data are execution instructions a, b, and c, where the execution time of execution instructions a, b, and c is 10 minutes. Assuming that there is operation data p, if the operation data p is located in the execution process of execution instruction b, the execution process of execution instruction b will be selected to obtain the corresponding parameter subset. If it is assumed that the operation data p is located at the end position of execution instruction a and the starting position of execution instruction b, the execution processes of execution instructions a and execution instruction b will be selected to obtain the corresponding parameter subset.

[0029] For example, the parameter subset is execution instruction b, and the time interval corresponding to execution instruction b is (t, t+10), then the corresponding experimental data is read according to the time interval to obtain the corresponding experimental subset.

[0030] Furthermore, step S3 is specifically as follows: Determine the inflection points of the parameter subset and experimental subset. Based on the operation time data, determine whether the operation time data represents an inflection point. The definition of an inflection point is as follows: the start and end time data of each execution instruction in the parameter subset serve as the inflection point of the parameter subset. For the experimental subset, determine the time point when the experimental data collected in the experimental subset meets the test requirements and the time point when the experimental data undergoes a sudden change. For example, if the execution time interval of instruction a is (0, 10), then time points 0 and 10 are marked as inflection points.

[0031] Result A: If the time point of the operation time data is the inflection point of the parameter subset and the experimental subset, the operation instruction corresponding to the historical operation data is determined to be an interrupt operation, and the corresponding intervention relationship is determined; Result B: If the time point of the operation time data is the inflection point of the parameter subset but not the inflection point of the experimental subset, the operation instruction corresponding to the historical operation data is determined to be a cyclic operation, and the corresponding intervention relationship is determined; Result C: If the time point of the operation time data is the inflection point of the experimental subset rather than the inflection point of the parameter subset, then the operation instruction corresponding to the historical operation data is determined to be a parameter modification operation, and the corresponding intervention relationship is determined; Result D: If the time point of the operation time data is not the inflection point of the experimental subset and the inflection point of the parameter subset, the operation instruction corresponding to the historical operation data is determined to be a load frequency modulation operation, and the corresponding intervention relationship is determined.

[0032] In this embodiment, by determining the time dimension relationship between the operation time data and the parameter subset and the experimental subset in the historical operation data, the operation purpose of the corresponding operation is clarified, and then different judgment methods are adopted according to different operation purposes, thereby improving the judgment accuracy of the data relationship between the operation data and the parameter data and the experimental data under the corresponding operation purpose, so that when the true triaxial test of rock is carried out according to the intervention relationship obtained by judgment and adaptive test monitoring is performed, the normal execution of the test can be effectively ensured, thereby improving the reliability of the test results.

[0033] Specifically, when determining the operating instructions corresponding to different experimental phenomena based on the parameter subset and the experimental subset, different analysis methods are adopted by clarifying the data position of the operation time of the historical operation data to be analyzed in the parameter subset and the experimental subset, so that the results obtained by the final analysis are more targeted and accurate.

[0034] For example, if the operation time of the operation data is at the inflection point between executing instructions a and executing instruction b in the parameter subset, it indicates that the test process changed the next instruction to be executed after executing instruction a. Therefore, it is necessary to judge the experimental subset. If the operation time is also at the inflection point of the corresponding experimental subset, it means that the test requirements have been met during the execution of instruction a, and the next test requirement needs to be collected. Therefore, it can be determined that the purpose of the operation data is to interrupt the execution of instruction a. If the operation time is not the inflection point of the experimental subset, it indicates that the test process has not met the test requirements. For example, the test requirement that can be achieved by executing instruction a is to determine the A value of the rock. However, after executing instruction a, the A value of the rock cannot be determined. In this case, instruction a needs to be executed again until the A value of the rock is determined. Then, the next test requirement collection can be performed. It is then determined that the purpose of the operation data is to loop the execution of instruction a. If the operation time is not the inflection point of the parameter subset, but the inflection point of the experimental subset, there are two situations when the experimental subset has an inflection point, one is that the experimental data meets the experimental requirements, and the other is that the experimental data mutates; if the experimental data meets the experimental requirements, the next instruction should be executed to collect the next data. However, since the operation time is not the inflection point of the parameter subset, there is no situation where the experimental data meets the experimental requirements and the next instruction is not executed under intervention. Therefore, it can be judged that the reason for the inflection point of the experimental subset in this case is that the experimental data mutates, and the mutation is not the data required by the test. Therefore, it is necessary to adjust the parameter data of the true triaxial test equipment to achieve the collection of the test requirements. Therefore, it is judged that the purpose of the operation data is to adjust the parameter data of instruction a.

[0035] If the operation time data is not the inflection point of the parameter subset and the experimental subset, it indicates that after the intervention, the true triaxial test equipment continues to execute instruction a and continues to collect the A value of the rock, making the operation basically equivalent to no operation. Therefore, it can be judged that the operation is a loaded frequency modulation operation to speed up the test rate within the range that the rock can withstand.

[0036] Furthermore, the method for determining result A is: Based on the operation time data, a time interval of the experimental subset containing the corresponding time point is determined, and an inflection point judgment is performed on the experimental subset based on the time interval; wherein the time interval is the time interval between the start time of the time period containing the time point in the experimental subset and the time point.

[0037] If it is determined that multiple inflection points exist within the time interval, the operation instruction corresponding to the historical operation data is determined to be an interrupt operation, and a corresponding relationship between the inflection point data and the operation instruction of the corresponding experimental subset is established to obtain an intervention relationship. If it is determined that multiple inflection points do not exist within the time interval, the execution order of the instructions is judged based on the historical parameter data to determine whether the execution order of the instructions without the historical operation data intervention is the same as the execution order of the instructions with the historical operation data intervention.

[0038] If the execution instruction sequence is determined to be the same, the historical operation data is determined to be an invalid operation. Otherwise, the inflection point data of the corresponding experimental subset is bound to the operation instruction to establish an intervention relationship.

[0039] In this embodiment, an intervention relationship refers to the need to execute a corresponding instruction after a certain test result occurs. The corresponding relationship between the test result and the execution instruction is an intervention relationship. For example, in the test data, there are three instructions a, b, and c, each of which executes for 10 seconds. The three instructions correspond to three experimental phenomena A, B, and C, respectively. When executing instruction a, if phenomenon A occurs at the 5th second, instruction b will be executed directly. In this case, the relationship between experimental phenomenon A and instruction b is an intervention relationship.

[0040] In this embodiment, based on the logical characteristics of human intervention, by determining the time interval of the experimental subset containing the time point corresponding to the operation time data, an inflection point judgment is performed on the experimental subset within the corresponding time interval, thereby quickly judging the operation purpose of the operation data, and then establishing a correspondence between the inflection point data and the operation instruction for the effective operation. When a new rock sample is tested, the corresponding operation instruction is used when the experimental data shows a change in the experimental data, thereby realizing the intelligent operation of the true triaxial test equipment and improving the reliability of the test results.

[0041] Specifically, when the operation time data is the inflection point between the parameter subset and the experimental subset, there are two cases: the first case is that the desired A value is obtained before instruction a is completed; the second case is that the desired A value is obtained after instruction a is executed; The first case can be directly judged as an interruption, while the second case may be a modification of the execution steps of the parameter data or an erroneous or invalid intervention. Since the execution steps are based on the test steps of the true triaxial test equipment, the possibility of parameter data modification can be ruled out, and the erroneous or invalid intervention can be retained.

[0042] In the first case, during execution, the experimental subset will first experience a data change, forming an inflection point. The operator will then determine, based on the collected experimental data, that the experimental requirements have been met, and thus interrupt the current instruction and execute the next instruction. Therefore, by reading the inflection point of the experimental subset preceding the operation time data, the corresponding situation can be quickly determined. If an inflection point is present, it indicates that the operation data was intended to interrupt the execution of the instruction. If an inflection point is not present, it indicates that the operation data is invalid and has no effect on the intervention.

[0043] Furthermore, the method for determining result B is: Compare historical experimental data with corresponding experimental requirement data to determine whether the historical experimental data meets the experimental requirement data; If the historical experimental data does not meet the experimental requirement data, the execution instructions before the corresponding time point and the execution instructions after the corresponding time point on the parameter subset are compared to determine whether the execution instructions before the corresponding time point and the execution instructions after the corresponding time point are repeated instructions; wherein, repeated instructions mean that the two instructions are the same instruction, and the completeness of the execution instructions before the time point reaches 100%.

[0044] If it is determined to be a repeated instruction, the operation instruction is determined to be a loop operation, and a relationship is established between the operation instruction and the corresponding experimental subset to obtain an intervention relationship. Otherwise, the historical operation data is determined to be an invalid operation.

[0045] In this embodiment, the experimental subset at the corresponding time point is compared with the corresponding experimental requirement data to once again verify whether the time point is the turning point of the experimental subset, so as to make the judgment more accurate. Then, the execution instructions of the parameter subset before the time point and the execution instructions after the time point are repeated to further determine whether the historical operation data is a cyclic operation, and a correspondence is established between the operation data determined to be a cyclic operation and the corresponding experimental subset and parameter subset, so as to ensure that the execution result of each execution instruction can meet the test requirements, thereby ensuring the normal progress of the test and improving the reliability of the test results.

[0046] Specifically, when the time point of the operation time data is the inflection point of the parameter subset, it indicates that before this time point represents the end of an execution instruction, and after this time point represents the beginning of an execution instruction; when the time point of the operation time data is not the inflection point of the experimental subset, it indicates that the experimental data does not meet the data required for the experiment.

[0047] Therefore, when it is determined that the experimental data at this time point does not meet the experimental requirement data, the historical parameter data is judged to determine whether the execution instruction before this time point is the same as the execution instruction after this time point. If they are the same, it indicates that the instruction is re-executed, thereby determining that the purpose of the operation data is a loop operation so that the experimental data can meet the experimental requirements.

[0048] Furthermore, the result C is determined by: The data change of the historical experimental data before the inflection point is judged to obtain the data change pattern of the corresponding experimental collection items; among which, the data change pattern can be judged by a mathematical statistics analysis algorithm or determined by a machine learning algorithm.

[0049] Match the historical operation data corresponding to the operation time data with the historical parameter data to determine the parameter item modified by the historical operation data; Based on the modified parameter items, the parameter item data before modification is compared with the historical experimental data before the inflection point, and the parameter item data after modification is compared with the historical experimental data after the inflection point, and the relationship between the data change of the corresponding parameter item and the data change of the experimental data is determined to obtain the first data change relationship; the correspondence between the data change law, the first data change relationship and the corresponding operation instruction is established to obtain the intervention relationship.

[0050] In this embodiment, after determining that the operation purpose of the historical operation data is to adjust the historical parameter data, the historical experimental data before the inflection point is changed, and based on the data change law of the experimental data, the data change law is used as a judgment indicator for starting the operation to clarify whether the parameter data needs to be modified; by determining the modified parameter item and the corresponding modified data amount, a data relationship between the data change of the modified parameter item and the data change of the experimental data is constructed, and then the data adjustment amount when adjusting the parameters is clarified, so that the true triaxial test equipment can adaptively adjust its own test parameters according to the experimental data collected in real time, thereby ensuring the smooth progress of the test and improving the reliability of the test results.

[0051] Specifically, when the time point of the operation time data is the inflection point of the experimental subset rather than the inflection point of the parameter subset, it indicates that there may be two situations at this time point during the experiment: the first situation is that the test results meet the test requirements; the second situation is that the experimental data mutates during the experiment; The first case: the test results meet the test requirements, and the corresponding execution instructions continue to be executed. If human intervention intervenes in the execution instructions that meet the test requirements, the next execution instruction should be executed, but the execution instructions have not changed, so it is inconsistent with the actual situation, which leads to the failure of the first case.

[0052] The second scenario involves a sudden change in experimental data, while the corresponding execution instruction continues to execute. If human intervention modifies the mutated experimental data, the execution instruction will not meet the desired test requirements, so the execution instruction needs to be re-executed. Furthermore, the sudden change in experimental data indicates that the parameter data corresponding to the execution instruction cannot meet the test requirements of the rock sample. Therefore, the operator needs to adjust the execution instruction data so that the adjusted instruction can meet the test requirements. This scenario is consistent with the actual situation, so it is determined that the purpose of the manipulation data is to modify the parameters of the execution instruction.

[0053] Furthermore, the result D is determined by: Based on the time point of the operation time data, the experimental subset within the corresponding time interval is divided into the first experimental sub-data before the time point and the second experimental sub-data after the time point; Determine the data volume change rate of the first experimental sub-data and the second experimental sub-data to obtain the data change rates of the two experimental sub-data before and after the time point, and record them as the first data change rate and the second data change rate respectively; Based on historical operation data, the adjustment result of the loaded frequency modulation parameter item in the corresponding operation instruction is determined to obtain the adjustment data volume, and based on the difference between the adjustment data volume and the first data change rate and the second data change rate, the relationship between the adjustment data volume and the data change rate is determined to obtain the second data change relationship; the corresponding relationship between the first data change rate, the second data change relationship and the corresponding operation instruction is constructed to obtain the intervention relationship.

[0054] In this embodiment, by taking the time point of the operation time data as a benchmark, the experimental subsets within the corresponding time interval are divided, and the data change rate before the intervention and the data change rate after the intervention are determined, thereby determining the start conditions of the intervention and the target data of the intervention, and then judging the data change relationship by judging the adjustment data volume in the historical operation data and the difference between the adjustment data volume and the first data change rate and the second data change rate, the intervention volume at the time of intervention can be calculated based on the data change relationship, making the intervention more accurate and effective, ensuring the normal progress of the experiment, and improving the reliability of the experiment results.

[0055] Specifically, when the time point of the operation time data is not the inflection point of the experimental subset and the inflection point of the parameter subset, it indicates that when the intervention is performed, the experimental data does not meet the experimental requirements and the execution instruction is not completed.

[0056] When the time point is not the inflection point of the parameter subset, there are two cases: the first case is to adjust the parameters; the second case is to adjust the loading rate.

[0057] If the parameters are adjusted in the first case, then after the data value of the parameter item in the historical parameter data is changed, the corresponding collected experimental data will also change, thus forming an inflection point. However, since this time point is not the inflection point of the experimental subset, it reflects that the collected experimental data does not have anomalies. In theory, a test process without anomalies does not require parameter adjustment intervention. Therefore, this is inconsistent with the actual situation, and it is determined that the purpose of the intervention is not to adjust the data value of the parameter item during the test process. If the loading rate is adjusted in the second case, then after changing the loading rate of the historical parameter data, since the historical parameter data remains unchanged, the changed experimental data will not change either, forming an inflection point. Therefore, this time point is not the inflection point of the experimental subset, which is consistent with the actual situation, and it is determined that the purpose of the intervention is to adjust the loading rate.

[0058] When it is determined to adjust the loading rate, the starting factor and the final adjustment target of the loading rate adjustment are determined by determining the data change rate of the experimental data before adjustment and the data change rate of the experimental data after adjustment, thereby ensuring the normal start-up and normal adjustment of the loading rate; then determine the data change relationship between the operation amount of the operation data and the corresponding experimental sub-data after adjustment, thereby ensuring that the loading rate can be adjusted accurately and effectively when it is determined that the loading rate adjustment is necessary.

[0059] Furthermore, step S4 is: Obtain historical parameter data and corresponding historical operation data of the corresponding experiment, and modify the corresponding parameter values ​​of the historical parameter data based on the historical operation data to obtain the ideal parameter data of the corresponding experiment; Match the historical parameter data of each test with the rock property data of the corresponding test to determine the parameter items corresponding to the characteristics of different rocks; The ideal parameter data of the parameter items with the same rock characteristics and the ideal parameter data of the same parameter items with different rock characteristics are statistically analyzed to determine the relationship between the rock characteristics and the data values ​​of the parameter items and obtain the basic parameter relationship.

[0060] In this embodiment, the historical operation data during the test process is integrated into the historical parameter data, thereby maximizing the precision and accuracy of the parameter data. By utilizing big data analysis methods, the historical parameter data and rock property data of multiple tests are matched to clarify the parameter items corresponding to different rock properties. The ideal parameter data corresponding to the parameter items under the same rock property are statistically analyzed to determine the optimal data values ​​of the corresponding parameter items under the corresponding rock property. The optimal data values ​​of the same parameter items under different rock properties are statistically analyzed to clarify the relationship between different rock properties and parameter items. This makes the parameter data evaluated by the system more accurate when testing unknown rocks, reduces the number of system adaptive interventions, and thereby improves the efficiency of the test and improves the accuracy and reliability of the test.

[0061] Specifically, because the initial historical parameter data is estimated by staff based on rock property data, and because there are interventions during the test process, the initially estimated historical parameter data has a large error. To reduce this error, a data relationship can be established between the historical parameter data and historical operation data from multiple tests and the rock property data of the corresponding rock samples. This ensures that the true triaxial test equipment can directly estimate the parameter data required for the test based on the rock property data input by the staff, thereby improving accuracy.

[0062] Example 2: The present invention also discloses a testing system based on the above-mentioned parameter adaptive intelligent true triaxial mechanical testing method, including a database establishment module, a data relationship construction module and an adaptive adjustment module; A database establishment module is provided in which the true triaxial test equipment performs tests on known rock samples based on known test data, monitors the data of the true triaxial test equipment during the test, and obtains historical parameter data, historical operation data, historical experimental data of the true triaxial test equipment, and rock property data of the rock samples; The data relationship construction module obtains the operation time data of historical operation data and divides the historical parameter data and historical experimental data based on the operation time data to obtain parameter subsets and experimental subsets. The parameter subsets and experimental subsets are analyzed to determine the operation instructions required for different test conditions and the operation data corresponding to the operation instructions to obtain the intervention relationship. The historical parameter data and rock property data are comprehensively analyzed to determine the data relationship between rock properties and the test parameters of the true triaxial test equipment to obtain the basic parameter relationship. The adaptive adjustment module obtains the rock characteristic data of the rock sample to be tested, and generates the test parameter data required for the test based on the basic parameter relationship. The rock sample to be tested is tested and data monitoring is performed based on the test parameter data to obtain rock sample experimental data. The rock sample to be tested is judged and data adaptively adjusted based on the intervention relationship.

[0063] Compared with the existing true triaxial test system, the present invention improves the analysis rate and ensures the reliability of the test results.

[0064] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent true triaxial mechanical testing method based on parameter adaptation, characterized by: The following steps are involved: S1: The true triaxial test equipment tests a known rock sample based on known test data, monitors the test process data of the true triaxial test equipment, and obtains historical parameter data, historical operation data, historical experimental data of the true triaxial test equipment, and rock property data of the rock sample; S2: Obtaining operation time data from the historical operation data, marking corresponding time points for the historical parameter data and historical experimental data based on the operation time data, and dividing the historical parameter data and historical experimental data according to the marked time points to obtain parameter subsets and experimental subsets; S3: Analyze the parameter subset and the experiment subset based on the operation time data, determine the operation instructions required under different experimental results and the operation data corresponding to the operation instructions, and obtain the intervention relationship; S4: Comprehensively analyzing the historical parameter data and rock property data, determining the data relationship between the rock properties and the test parameters of the true triaxial test equipment, and obtaining the basic parameter relationship; S5: Obtain rock characteristic data of the rock sample to be tested, and generate test parameter data required for the test based on the basic parameter relationship, conduct tests and data monitoring on the rock sample to be tested based on the test parameter data to obtain rock test data; judge and adaptively adjust the rock test data based on the intervention relationship.

2. The intelligent true triaxial mechanical testing method based on parameter adaptation according to claim 1 is characterized in that: The step S2 includes the following steps: S21: dividing the historical experimental data based on the historical parameter data, and determining the execution instructions at each time point in the historical experimental data and the execution duration of the corresponding execution instructions; S22: determining an execution instruction corresponding to the operation time data based on the operation time data, determining a data selection interval for the historical parameter data according to the execution duration of the execution instruction, and selecting the historical parameter data according to the data selection interval to obtain a parameter subset; S23: Divide the historical experimental data based on the parameter subset to obtain an experimental subset.

3. The intelligent true triaxial mechanical testing method based on parameter adaptation according to claim 2 is characterized in that: In step S22, if the time point corresponding to the operation time data is located at the intersection between two execution instructions, the sum of the execution time of the two execution instructions is used as the data selection interval according to the time sequence, and then the historical parameter data is selected according to the data selection interval to obtain a parameter subset. The intersection point is the intersection position where one instruction ends and another instruction begins.

4. The intelligent true triaxial mechanical testing method based on parameter adaptation according to claim 1 is characterized in that: The step S3 is specifically as follows: determining an inflection point of the parameter subset and the experimental subset, and determining whether the operation time data is an inflection point of the parameter subset and the experimental subset based on the operation time data; The inflection points of the parameter subset are the start time data and the end time data of each execution instruction in the parameter subset; The inflection point of the experimental subset is the time point when the experimental data collected in the experimental subset meets the experimental requirements and the time point when the experimental data mutates; Result A: If the time point of the operation time data is the inflection point of the parameter subset and the experimental subset, then the operation instruction corresponding to the historical operation data is determined to be an interrupt operation, and the corresponding intervention relationship is determined; Result B: If the time point of the operation time data is the inflection point of the parameter subset but not the inflection point of the experimental subset, then the operation instruction corresponding to the historical operation data is determined to be a cyclic operation, and the corresponding intervention relationship is determined; Result C: If the time point of the operation time data is the inflection point of the experimental subset but not the inflection point of the parameter subset, then the operation instruction corresponding to the historical operation data is determined to be a parameter modification operation, and the corresponding intervention relationship is determined; Result D: If the time point of the operation time data is not the inflection point of the experimental subset and the inflection point of the parameter subset, the operation instruction corresponding to the historical operation data is determined to be a load frequency modulation operation, and the corresponding intervention relationship is determined.

5. The intelligent true triaxial mechanical testing method based on parameter adaptation according to claim 4 is characterized in that: The method for determining the result A is: Determine a time interval of an experimental subset containing a corresponding time point based on the operation time data, and perform inflection point judgment on the experimental subset based on the time interval; If it is determined that there are multiple inflection points within the time interval, the operation instruction corresponding to the historical operation data is determined to be an interrupt operation, and a corresponding relationship between the inflection point data of the corresponding experimental subset and the operation instruction is established to obtain an intervention relationship.

6. The intelligent true triaxial mechanical testing method based on parameter adaptation according to claim 4 is characterized in that: The method for determining result B is: Compare historical experimental data with corresponding experimental requirement data to determine whether the historical experimental data meets the experimental requirement data; If the historical experimental data does not meet the experimental requirement data, the execution instructions before the corresponding time point and the execution instructions after the corresponding time point on the parameter subset are compared to determine whether the execution instructions before the corresponding time point and the execution instructions after the corresponding time point are repeated instructions; If it is determined to be a repeated instruction, the operation instruction is determined to be a loop operation, and a relationship is established between the operation instruction and the corresponding experimental subset to obtain an intervention relationship. Otherwise, the historical operation data is determined to be an invalid operation.

7. The intelligent true triaxial mechanical testing method based on parameter adaptation according to claim 4 is characterized in that: The method for determining the result C is: The historical experimental data before the inflection point are judged for data changes to obtain the data change rules of the corresponding experimental collection items; Match the historical operation data corresponding to the operation time data with the historical parameter data to determine the parameter item modified by the historical operation data; Based on the modified parameter items, the parameter item data before modification is compared with the historical experimental data before the inflection point, and the parameter item data after modification is compared with the historical experimental data after the inflection point, and the relationship between the data change of the corresponding parameter item and the data change of the experimental data is determined to obtain the first data change relationship; the correspondence between the data change law, the first data change relationship and the corresponding operation instruction is established to obtain the intervention relationship.

8. The intelligent true triaxial mechanical testing method based on parameter adaptation according to claim 4 is characterized in that: The method for determining the result D is: Based on the time point of the operation time data, the experimental subset within the corresponding time interval is divided into the first experimental sub-data before the time point and the second experimental sub-data after the time point; Determine the data volume change rate of the first experimental sub-data and the second experimental sub-data to obtain the data change rates of the two experimental sub-data before and after the time point, and record them as the first data change rate and the second data change rate respectively; Based on historical operation data, the adjustment result of the loaded frequency modulation parameter item in the corresponding operation instruction is determined to obtain the adjustment data volume, and based on the difference between the adjustment data volume and the first data change rate and the second data change rate, the relationship between the adjustment data volume and the data change rate is determined to obtain the second data change relationship; the corresponding relationship between the first data change rate, the second data change relationship and the corresponding operation instruction is constructed to obtain the intervention relationship.

9. The intelligent true triaxial mechanical testing method based on parameter adaptation according to claim 1, characterized in that: The step S4 is: Obtain historical parameter data and corresponding historical operation data of the corresponding experiment, and modify the corresponding parameter values ​​of the historical parameter data based on the historical operation data to obtain the ideal parameter data of the corresponding experiment; Match the historical parameter data of each test with the rock property data of the corresponding test to determine the parameter items corresponding to the characteristics of different rocks; The ideal parameter data of the parameter items with the same rock characteristics and the ideal parameter data of the same parameter items with different rock characteristics are statistically analyzed to determine the relationship between the rock characteristics and the data values ​​of the parameter items and obtain the basic parameter relationship.

10. A testing system based on the parameter-adaptive intelligent true triaxial mechanical testing method according to any one of claims 1 to 9, characterized in that: It includes database establishment module, data relationship construction module and adaptive adjustment module; The database establishment module is configured to allow the true triaxial test equipment to test known rock samples based on known test data, monitor the data of the true triaxial test equipment during the test, and obtain historical parameter data, historical operation data, historical experimental data of the true triaxial test equipment, and rock property data of the rock samples; The data relationship construction module obtains operation time data of historical operation data, and divides historical parameter data and historical experimental data based on the operation time data to obtain parameter subsets and experimental subsets; analyzes the parameter subsets and experimental subsets to determine the operation instructions required under different test conditions and the operation data corresponding to the operation instructions, thereby obtaining an intervention relationship; and comprehensively analyzes the historical parameter data and rock property data to determine the data relationship between the rock properties and the test parameters of the true triaxial test equipment, thereby obtaining a basic parameter relationship; The adaptive adjustment module obtains rock characteristic data of the rock sample to be tested, generates test parameter data required for the test according to the basic parameter relationship, and makes judgments and data adaptive adjustments on the rock sample to be tested based on the intervention relationship.

Images