Mechanical characterization methods and devices
By constructing a finite element geometric model and data-driven algorithm under the coal mine, processing the initial state value of the monitoring point, the problems of low accuracy and noise sensitivity of stress monitoring data in the coal mine are solved, the accuracy of mechanical characterization is improved, and the deep mining is supported Safety and efficiency.
Patent Information
- Application Number
- CN202410789326.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-18
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-06-18
AI Technical Summary
Underground stress monitoring of coal mines has problems such as low data accuracy and noise sensitivity, which leads to large errors in mechanical characterization results, limiting the development of deep mining.
A mechanical characterization method and device are proposed. By obtaining the initial state value of the monitoring point, a finite element geometric model is constructed, data preprocessing is performed, and the optimal state value is determined by using a data-driven algorithm to reduce data loss and noise influence.
It improves data accuracy, enhances the performance of digitally driven mechanical representation in actual engineering, and effectively supports coal mine impact ground pressure disaster prevention and control and deep mining.
Smart Images

Figure CN118657018B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of stress monitoring technology, and in particular to a mechanical characterization method and device. Background Art
[0002] Stress monitoring in coal mines has always been a key issue in the prevention and control of rock burst disasters and deep mining in coal mines. The stress distribution in coal mines is affected by many factors, such as burial depth, lithology, geological structure, etc. These factors lead to a certain degree of discreteness in stress data, which increases the difficulty of monitoring and analysis.
[0003] At present, although there are many stress measurement methods, such as stress relief method and hydraulic fracturing method, these methods may be limited by the underground environment in practical application, such as small space, high temperature and high humidity. Therefore, different monitoring methods are often used according to different actual conditions during monitoring, and the returned data format and accuracy are different. Stress monitoring in coal mines requires accurate and real-time data collection and analysis, but the current technology is not mature enough in data processing and analysis, which seriously restricts the development of deep underground resource mining. Summary of the invention
[0004] The present disclosure aims to solve one of the technical problems in the related art at least to some extent.
[0005] The present disclosure proposes a mechanical characterization method and device, which can pre-process the data of the initial state values of each monitoring point in advance to improve data accuracy, and can effectively enhance the performance of digitally driven mechanical characterization in actual engineering.
[0006] The first aspect of the present disclosure provides a mechanical characterization method, including:
[0007] For any of the monitoring points in the target mining area, a plurality of initial state pair values corresponding to the monitoring point are obtained; wherein the initial state pair values include a first value of the stress state and a second value of the strain state;
[0008] Constructing a finite element geometric model corresponding to the target mining area; wherein the finite element geometric model includes a plurality of nodes, and each of the monitoring points has a corresponding node in the finite element geometric model;
[0009] For any of the monitoring points, data preprocessing is performed on a plurality of initial state pairs of values corresponding to the monitoring point to obtain a plurality of first state pairs of values corresponding to the monitoring point;
[0010] Based on the finite element geometric model and a plurality of first state pairs corresponding to each of the monitoring points, a data-driven algorithm is used to determine the optimal state pair corresponding to each of the monitoring points.
[0011] The second aspect of the present disclosure provides a mechanical characterization device, comprising:
[0012] An acquisition module, for acquiring, for any of the plurality of monitoring points in the target mining area, a plurality of initial state pair values corresponding to the monitoring point; wherein the initial state pair values include a first value of the stress state and a second value of the strain state;
[0013] A construction module, used to construct a finite element geometric model corresponding to the target mining area; wherein the finite element geometric model includes a plurality of nodes, and each of the monitoring points has a corresponding node in the finite element geometric model;
[0014] A processing module, configured to perform data preprocessing on a plurality of initial state pairs of values corresponding to any of the monitoring points, so as to obtain a plurality of first state pairs of values corresponding to the monitoring point;
[0015] A determination module is used to determine the optimal state pair value corresponding to each of the monitoring points based on the finite element geometric model and multiple first state pair values corresponding to each of the monitoring points, using a data-driven algorithm.
[0016] The third aspect of the present disclosure proposes an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the mechanical characterization method proposed in the first aspect of the present disclosure is implemented.
[0017] The fourth aspect embodiment of the present disclosure proposes a non-temporary computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the mechanical characterization method proposed in the first aspect embodiment of the present disclosure.
[0018] The fifth aspect embodiment of the present disclosure proposes a computer program product. When the instructions in the computer program product are executed by a processor, the mechanical characterization method proposed in the first aspect embodiment of the present disclosure is executed.
[0019] Additional aspects and advantages of the present disclosure will be given in part in the following description and in part will be obvious from the following description or learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] The above and / or additional aspects and advantages of the present disclosure will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:
[0021] Figure 1 A schematic diagram of a flow chart of a mechanical characterization method provided in one embodiment of the present disclosure;
[0022] Figure 2A schematic flow chart of a mechanical characterization method provided by another embodiment of the present disclosure;
[0023] Figure 3 A schematic flow chart of a mechanical characterization method provided by another embodiment of the present disclosure;
[0024] Figure 4 A schematic flow chart of a mechanical characterization method provided by another embodiment of the present disclosure;
[0025] Figure 5 A schematic diagram of the process of the mechanical characterization method provided in the present disclosure;
[0026] Figure 6 A schematic structural diagram of a mechanical characterization method device provided in another embodiment of the present disclosure. DETAILED DESCRIPTION
[0027] Embodiments of the present disclosure are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limiting the present disclosure.
[0028] In the related art, the steps of the mechanical characterization method based on data-driven technology are:
[0029] (1) A “stress-strain” state pair value is specified for each material point (also called a monitoring point), where the “stress-strain” state pair value comes from a pre-specified material data set and needs to satisfy compatibility and equilibrium constraints.
[0030] (2) Define the penalty function: evaluate the degree of deviation of each material point state (such as stress state, strain state) from the corresponding data set.
[0031] (3) Define the constrained optimization problem: minimize the global penalty function while satisfying the balance and compatibility constraints.
[0032] (4) Problem transformation: Use the Lagrange multiplier method to transform the constrained optimization problem into an equivalent steady-state problem.
[0033] (5) Solve steady-state problems: obtain variables such as displacement and stress.
[0034] (6) Global search: Find the data point that is closest to its state for each material point in the data set.
[0035] (7) Repeat steps (5) and (6) until convergence.
[0036] As the dataset approaches the true material constitutive relations, the data-driven solution converges to the classical solution.
[0037] However, the above method has the following disadvantages:
[0038] 1. High data requirements. The existing technology requires that a "stress-strain" state pair value be specified for each material point. However, in actual engineering applications, depending on the different monitoring methods used, the values or data corresponding to the stress state and strain state may not be measured at the same time.
[0039] 2. Sensitive to data noise. In actual engineering, poor data accuracy is very common. However, the above method is completely dependent on the collected data. If the collected data is not accurate, it will greatly affect the characterization results, resulting in large errors in the results.
[0040] In response to at least one of the above problems, the present disclosure provides a mechanical characterization method and device.
[0041] The mechanical characterization method and device of the embodiments of the present disclosure are described below with reference to the accompanying drawings.
[0042] Figure 1 A schematic flow chart of a mechanical characterization method provided in accordance with an embodiment of the present disclosure.
[0043] The embodiment of the present disclosure takes the mechanical characterization method configured in a mechanical characterization method device as an example. The mechanical characterization method device can be applied to any electronic device so that the electronic device can execute the mechanical characterization method function.
[0044] Among them, the electronic device can be any device with computing capabilities, such as a PC (Personal Computer), a mobile terminal, a server, etc. The mobile terminal can be, for example, a mobile phone, a tablet computer, a personal digital assistant, a wearable device, and other hardware devices with various operating systems, touch screens and / or display screens.
[0045] like Figure 1 As shown, the mechanical characterization method may include the following steps:
[0046] Step 101: for any monitoring point among a plurality of monitoring points in a target mining area, obtain a plurality of initial state pair values corresponding to the monitoring point.
[0047] The initial state pair value may be a combination of a first value of the stress state and a second value of the strain state.
[0048] In the embodiment of the present disclosure, there may be multiple monitoring points in the target mining area, so that the multiple monitoring points in the target mining area can be monitored and multiple initial state pair values corresponding to any monitoring point can be obtained.
[0049] It should be noted that the first value in each initial state value pair may exist or may not exist; similarly, the second value in each initial state value pair may exist or may not exist.
[0050] It should also be noted that the present disclosure does not impose any restriction on the noise level of the first value in each initial state pair of values. For example, the noise level of the first value in the initial state pair of values is less than 5%, or the noise level of the first value in the initial state pair of values is greater than 5%. Similarly, the present disclosure does not impose any restriction on the noise level of the second value in each initial state pair of values.
[0051] It should be noted that, in actual applications, the number of initial state pair values corresponding to each monitoring point may be multiple, for example, 500, 1000, 2000, etc., and the present disclosure does not impose any limitation on this.
[0052] Step 102, constructing a finite element geometric model corresponding to the target mining area.
[0053] The finite element geometric model includes a plurality of nodes, and each monitoring point has a corresponding node in the finite element geometric model.
[0054] As an example, when constructing a finite element geometric model corresponding to the target mining area, each node, the topological relationship between the nodes, the shape functions between the nodes and the derivative matrix corresponding to the shape functions, and the weight corresponding to each node can be defined in the model, wherein each monitoring point has a corresponding node in the finite element geometric model.
[0055] Step 103 : for any monitoring point, perform data preprocessing on a plurality of initial state pair values corresponding to the monitoring point to obtain a plurality of first state pair values corresponding to the monitoring point.
[0056] Among them, data preprocessing may include, for example, missing value processing, data enhancement, etc., which is not limited in the present disclosure.
[0057] Step 104 , based on the finite element geometric model and the multiple first state pairs corresponding to the monitoring points, a data-driven algorithm is used to determine the optimal state pair corresponding to each monitoring point.
[0058] In the disclosed embodiment, based on the finite element geometric model and multiple first state pairs corresponding to each monitoring point, a data-driven algorithm can be used to determine the optimal state pair (also called global state pair) corresponding to each monitoring point.
[0059] The mechanical characterization method of the disclosed embodiment obtains multiple initial state pairs corresponding to any monitoring point among multiple monitoring points in the target mining area; wherein the initial state pairs include a first value of the stress state and a second value of the strain state; a finite element geometric model corresponding to the target mining area is constructed; wherein the finite element geometric model includes multiple nodes, and each monitoring point has a corresponding node in the finite element geometric model; for any monitoring point, data preprocessing is performed on the multiple initial state pairs corresponding to the monitoring point to obtain multiple first state pairs corresponding to the monitoring point; based on the finite element geometric model and the multiple first state pairs corresponding to each monitoring point, a data-driven algorithm is used to determine the optimal state pairs corresponding to each monitoring point. Thus, data preprocessing is performed on the initial state pairs of each monitoring point in advance to improve data accuracy, which can effectively enhance the performance of digital-driven mechanical characterization in actual engineering.
[0060] In order to clearly explain in the above embodiments of the present disclosure how data preprocessing is performed on multiple initial state pairs corresponding to any monitoring point to obtain multiple first state pairs corresponding to the monitoring point, the present disclosure also proposes a mechanical characterization method.
[0061] Figure 2 A schematic flow chart of a mechanical characterization method provided in another embodiment of the present disclosure.
[0062] like Figure 2 As shown, the mechanical characterization method may include the following steps:
[0063] Step 201: for any monitoring point among a plurality of monitoring points in a target mining area, obtain a plurality of initial state pair values corresponding to the monitoring point.
[0064] Step 202: construct a finite element geometric model corresponding to the target mining area.
[0065] The execution process of step 201 to step 202 can refer to the execution process of any embodiment of the present disclosure, and will not be repeated here.
[0066] Step 203 : for any monitoring point, according to the data category to which each initial state pair value corresponding to the monitoring point belongs, divide the multiple initial state pair values corresponding to the monitoring point to obtain at least one subset corresponding to the monitoring point.
[0067] In a possible implementation manner of the embodiment of the present disclosure, the data category may include at least one of a first category, a second category, a third category, and a fourth category, wherein:
[0068] The first category may indicate that both the first value and the second value exist in the corresponding initial state value pair, and the noise levels of the first value and the second value in the corresponding initial state value pair are both less than a set threshold;
[0069] The second category may indicate that the first value or the second value in the corresponding initial state pair of values is missing, and the noise level of the first value or the second value existing in the corresponding initial state pair of values is less than a set threshold;
[0070] The third category may indicate that both the first value and the second value exist in the corresponding initial state value pair, and the noise level of the first value and / or the second value in the corresponding initial state value pair is not less than a set threshold;
[0071] The fourth category may indicate that the first value or the second value in the corresponding initial state pair of values is missing, and the noise level of the first value or the second value that exists in the corresponding initial state pair of values is not less than a set threshold.
[0072] In the embodiment of the present disclosure, the threshold value may be preset, such as 5%, 10%, etc., and the present disclosure does not impose any limitation on this.
[0073] As an example, assume that the initial state pair corresponding to monitoring point 1 includes (a, b), and the threshold is set to 10%, where:
[0074] When the first value a and the second value b both exist in the initial state pair value (a, b), and the noise levels of the first value a and the second value b are both less than 10%, the data category of the initial state pair value (a, b) is determined to be the first category;
[0075] When the first value a in the initial state pair value (a, b) is missing and the noise level of the existing second value b is less than 10%, the data category of the initial state pair value (a, b) is determined to be the second category; when the second value b in the initial state pair value (a, b) is missing and the noise level of the existing first value a is less than 10%, the data category of the initial state pair value (a, b) is determined to be the second category;
[0076] When the first value a and the second value b both exist in the initial state pair value (a, b), and the noise level of the first value a and / or the second value b is not less than 10%, it is determined that the data category of the initial state pair value (a, b) is the third category;
[0077] When the first value a in the initial state pair of values (a, b) is missing and the noise level of the existing second value b is not less than 10%, the data category of the initial state pair of values (a, b) is determined to be the fourth category; when the second value b in the initial state pair of values (a, b) is missing and the noise level of the existing first value a is less than 10%, the data category of the initial state pair of values (a, b) is determined to be the fourth category.
[0078] In another possible implementation of the embodiment of the present disclosure, when neither the first value of the stress state nor the second value of the strain state in the initial state pair value corresponding to the monitoring point exists, the initial state pair value corresponding to the monitoring point may be deleted.
[0079] In the disclosed embodiment, for any monitoring point, multiple initial state pair values corresponding to the monitoring point can be divided according to the data category to which each initial state pair value corresponding to the monitoring point belongs, and at least one subset corresponding to the monitoring point can be obtained.
[0080] It should be noted that the number of subsets corresponding to the monitoring points may be, but is not limited to, 1, and the present disclosure does not impose any restrictions on this.
[0081] It should also be noted that each subset may have a corresponding data category.
[0082] Still using the above example for explanation, for monitoring point 1, after dividing the multiple initial state pair values corresponding to the monitoring point according to the data categories to which the initial state pair values corresponding to monitoring point 1 belong, the obtained subset corresponding to monitoring point 1 includes subset A and subset B, wherein the data category to which the initial state pair values in subset A belong is the first category, that is, the data category corresponding to subset A is the first category, and the data category to which the initial state pair values in subset B belong is the third category, that is, the data category corresponding to subset B is the third category.
[0083] Step 204 , for any subset corresponding to the monitoring point, based on the data category corresponding to the subset, a corresponding data preprocessing method is used to perform data preprocessing on the initial state pair values in the subset to obtain the first state pair values corresponding to the monitoring point.
[0084] In the disclosed embodiment, for any subset corresponding to the monitoring point, the initial state pair values in the subset can be preprocessed based on the data category corresponding to the subset using the corresponding data preprocessing method, so as to obtain the first state pair values corresponding to the monitoring point.
[0085] As an example, assume that the subsets corresponding to monitoring point 2 include subset A, subset B and subset C, and the data preprocessing methods include method 1, method 2, method 3 and method 4. Based on the data category corresponding to subset A, method 1 is used to perform data preprocessing on the initial state pair values in subset A to obtain the first state pair values corresponding to the monitoring point. Based on the data category corresponding to subset B, method 3 is used to perform data preprocessing on the initial state pair values in subset B to obtain the first state pair values corresponding to the monitoring point. Based on the data category corresponding to subset C, method 4 is used to perform data preprocessing on the initial state pair values in subset C to obtain the first state pair values corresponding to the monitoring point.
[0086] Step 205 , based on the finite element geometric model and the multiple first state pairs corresponding to the monitoring points, a data-driven algorithm is used to determine the optimal state pair corresponding to each monitoring point.
[0087] The execution process of step 205 may refer to the execution process of any embodiment of the present disclosure, and will not be described in detail here.
[0088] The mechanical characterization method of the disclosed embodiment is to divide the multiple initial state pair values corresponding to any monitoring point according to the data category to which each initial state pair value corresponding to the monitoring point belongs, so as to obtain at least one subset corresponding to the monitoring point; for any subset corresponding to the monitoring point, based on the data category corresponding to the subset, the initial state pair values in the subset are preprocessed using the corresponding data preprocessing method to obtain the first state pair value corresponding to the monitoring point. Thus, based on the data category corresponding to each subset, the corresponding data preprocessing method is used to preprocess the data, so that the data accuracy of each subset can be effectively improved.
[0089] In the present disclosure, when the data category includes at least one of the first category, the second category, the third category and the fourth category shown in step 203, for any subset corresponding to the monitoring point, in order to clearly explain how the initial state pair values in the subset are preprocessed based on the data category corresponding to the subset using the corresponding data preprocessing method to obtain the first state pair values corresponding to the monitoring point, the present disclosure also proposes a mechanical characterization method.
[0090] Figure 3 A schematic flow chart of a mechanical characterization method provided in another embodiment of the present disclosure.
[0091] like Figure 3 As shown, based on the above embodiments of the present disclosure, the mechanical characterization method may further include the following steps:
[0092] Step 301 : for any subset corresponding to the monitoring point, in response to the data category corresponding to the subset being the first category, determining the initial state pair value in the subset as the first state pair value corresponding to the monitoring point.
[0093] In the embodiment of the present disclosure, the number of initial state pair values in each subset may be, but is not limited to, one, and the present disclosure does not impose any limitation on this.
[0094] In the embodiment of the present disclosure, for any subset corresponding to the monitoring point, when the data category corresponding to the subset is the first category, the initial state pair value in the subset can be determined as the first state pair value corresponding to the monitoring point.
[0095] That is to say, when the data category corresponding to the subset is the first category, the initial state pair value in the subset can be directly determined as the first state pair value corresponding to the monitoring point.
[0096] Step 302 , in response to the data category corresponding to the subset being the second category, a method of filling missing values is used to perform first data filling on the initial state pair values in the subset to obtain first state pair values corresponding to the monitoring point.
[0097] In the embodiment of the present disclosure, when the data category corresponding to the subset is the second category, the method of filling missing values can be used to perform first data filling on the initial state pair values in the subset to obtain the first state pair values corresponding to the monitoring point.
[0098] Optionally, when the data category corresponding to the subset is the second category, the following steps may be used to perform first data filling on the initial state pair values in the subset using a method for filling missing values, so as to obtain first state pair values corresponding to the monitoring points:
[0099] Step 3021, in response to the data category corresponding to the subset being the second category, obtaining a first constitutive model and a first material parameter corresponding to the monitoring point in the target mining area.
[0100] Among them, the first material parameter matches the first constitutive model.
[0101] In the embodiment of the present disclosure, after obtaining the first constitutive model corresponding to the monitoring point in the target mining area, the first material parameters matching the first constitutive model can be obtained.
[0102] That is to say, the first material parameter depends on the first constitutive model obtained. For example, when the first constitutive model is a linear elastic constitutive model, the first material parameter may include an elastic modulus (such as Young's modulus) and Poisson's ratio, or the first material parameter may include a Lame constant and a shear modulus; when the first constitutive model is an elastic-plastic model, the first material parameter may include density, elastic model, and Poisson's ratio.
[0103] In the embodiment of the present disclosure, when the data category corresponding to the subset is the second category, the first constitutive model and the first material parameter corresponding to the monitoring point in the target mining area can be obtained.
[0104] Step 3022: construct a material stiffness matrix corresponding to the monitoring point according to the first constitutive model and the first material parameter corresponding to the monitoring point.
[0105] In the embodiment of the present disclosure, a material stiffness matrix corresponding to the monitoring point may be constructed according to the first constitutive model and the first material parameter corresponding to the monitoring point.
[0106] As an example, assuming that the first constitutive model corresponding to the monitoring point is a linear elastic constitutive model, the first material parameter includes the elastic modulus and Poisson's ratio, the elastic modulus is Young's modulus E, and the Poisson's ratio is ν, then the Lame constant λ and shear modulus μ of the material at the monitoring point are respectively:
[0107]
[0108] Then the material stiffness matrix C corresponding to the monitoring point is:
[0109]
[0110] Step 3023, for any initial state pair value in the subset, based on the material stiffness matrix corresponding to the monitoring point, determine the target value in the initial state pair value.
[0111] The target value can be the missing first value or the missing second value.
[0112] For example, assuming that there is an initial state pair value (a, b) in the subset, and the material stiffness matrix corresponding to the monitoring point is C, when the first value a in the initial state pair value (a, b) is missing, the first value a is the target value to be determined, and the target value a can be determined according to the following formula:
[0113] a=C -1 *b; (4)
[0114] When the second value b is missing in the initial state pair (a, b), the second value b is the target value to be determined. The target value b can be determined according to the following formula:
[0115] b=C -1 *a; (5)
[0116] Step 3024, based on the target value, the initial state pair value is padded, and the padded initial state pair value is determined as the first state pair value corresponding to the monitoring point.
[0117] In the embodiment of the present disclosure, the initial state pair value may be padded based on the target value, and the padded initial state pair value may be determined as the first state pair value corresponding to the monitoring point.
[0118] Still using the above example to illustrate, when the first value a in the initial state pair value (a, b) is missing, the initial state pair value (a, b) is padded based on the target value a determined by formula (4). The padded initial state pair value is (C -1 *b, b), can be (C -1 *b, b) is determined as the first state pair value corresponding to the monitoring point; when the second value b is missing in the initial state pair value (a, b), the initial state pair value (a, b) is filled based on the target value b determined by formula (5), and the filled initial state pair value is (a, C -1 *a), we can convert (a, C -1*a) Determine the first state pair value corresponding to the monitoring point.
[0119] Therefore, when the data category corresponding to the subset is the second category, the first data filling can be performed on the initial state pair values in the subset, thereby effectively obtaining the first state pair values corresponding to the monitoring points.
[0120] Step 303 , in response to the data category corresponding to the subset being the third category, a data construction method is used to perform first data supplement on the initial state pair values in the subset to obtain a first state pair value corresponding to the monitoring point.
[0121] In the embodiment of the present disclosure, when the data category corresponding to the subset is the third category, a data construction method can be used to perform first data supplement on the initial state pair values in the subset to obtain the first state pair values corresponding to the monitoring point.
[0122] Optionally, when the data category corresponding to the subset is the third category, the following steps may be used to supplement the initial state pair values in the subset with first data using a data construction method to obtain the first state pair values corresponding to the monitoring point:
[0123] Step 3031, in response to the data category corresponding to the subset being the third category, obtaining a second constitutive model and a second material parameter corresponding to the monitoring point in the target mining area, and obtaining a first boundary condition corresponding to the monitoring point.
[0124] The second material parameters are matched with the second constitutive model.
[0125] It should be noted that the second constitutive model may be the same as the first constitutive model in step 3021, or may be different, and the present disclosure does not impose any limitation on this.
[0126] It should also be noted that the second material parameter may be the same as the first material parameter in step 3021, or may be different, and the present disclosure does not impose any limitation on this.
[0127] The first boundary condition may be, for example, the magnitude of the applied force, the position information of the applied force, etc., which is not limited in the present disclosure.
[0128] In an embodiment of the present disclosure, when the data category corresponding to the subset is the third category, the second constitutive model and the second material parameters corresponding to the monitoring point in the target mining area can be obtained, and the first boundary condition corresponding to the monitoring point can be obtained. For example, in response to a user's selection operation, the second constitutive model can be obtained from a plurality of pre-configured constitutive models, and in response to a user's input operation, the second material parameters matching the second constitutive model and the first boundary condition corresponding to the monitoring point can be obtained.
[0129] Step 3032: Based on the second constitutive model, the second material parameter and the first boundary condition corresponding to the monitoring point, and the finite element geometric model, a finite element method is used to construct a second state pair value belonging to the subset.
[0130] In the disclosed embodiment, based on the second constitutive model, the second material parameter and the first boundary condition corresponding to the monitoring point, and the finite element geometric model, the finite element method can be used to construct the second state pair value belonging to the subset. For example, based on the second constitutive model, the second material parameter and the first boundary condition corresponding to the monitoring point, and the finite element geometric model, the finite element method is implemented through ANSYS computer-aided engineering software, ABAQUS engineering simulation finite element software, etc., to construct the second state pair value belonging to the subset.
[0131] Optionally, the number of second state values corresponding to the constructed monitoring points may be a first set multiple of the number of initial state pair values in the corresponding subset.
[0132] The first set multiple may be pre-set, for example, any real number between 1 and 2, etc.
[0133] As an example, assuming that the data category corresponding to subset B of monitoring point 2 is the third category, the number of initial state pair values in subset B is n, and the first set multiple is 1.2, then the number of second state values corresponding to the constructed monitoring points is 1.2n.
[0134] Step 3033: Determine the constructed second state pair value and the initial state pair value in the subset as the first state pair value corresponding to the monitoring point.
[0135] In the embodiment of the present disclosure, the constructed second state pair value and the initial state pair value in the corresponding subset may both be determined as the first state pair value corresponding to the monitoring point.
[0136] Still using the above example to illustrate, the constructed second state pair values, which are 1.2n in number, and the initial state pair values, which are n in number, in subset B can all be determined as the first state pair values corresponding to the monitoring points.
[0137] Therefore, when the data category corresponding to the subset is the third category, the first data supplement can be performed on the initial state pair values in the subset, thereby effectively obtaining the first state pair values corresponding to the monitoring points.
[0138] Step 304 , in response to the data category corresponding to the subset being the fourth category, the initial state pair values in the subset are processed using a method of filling missing values and a method of data construction to obtain a first state pair value corresponding to the monitoring point.
[0139] In the embodiment of the present disclosure, when the data category corresponding to the subset is the fourth category, the method of filling missing values and the method of data construction can be used to process the initial state pair values in the subset to obtain the first state pair values corresponding to the monitoring point.
[0140] Optionally, when the data category corresponding to the subset is the fourth category, the following steps may be used to process the initial state pair values in the subset using a method of filling missing values and a method of data construction to obtain the first state pair values corresponding to the monitoring point:
[0141] Step 3041, in response to the data category corresponding to the subset being the fourth category, a method of filling missing values is used to perform second data filling on the initial state pair values in the subset to obtain third state pair values.
[0142] It should be noted that the method of performing the second data padding on the initial state pair values in the subset is similar to the method of performing the first data padding in step 302, and will not be described in detail here.
[0143] Step 3042: Use a data construction method to construct a fourth state pair value belonging to the subset.
[0144] Optionally, a third constitutive model and a third material parameter corresponding to the monitoring point in the target mining area are obtained, and a second boundary condition corresponding to the monitoring point is constructed, wherein the third material parameter matches the third constitutive model; based on the third constitutive model, the third material parameter and the second boundary condition corresponding to the monitoring point, and the finite element geometric model, a finite element method is used to construct a fourth state pair value belonging to the subset.
[0145] It should be noted that the third constitutive model may be the same as the first constitutive model in step 3021, or may be different, and the present disclosure does not impose any limitation on this.
[0146] It should also be noted that the third material parameter may be the same as the first material parameter in step 3021, or may be different, and the present disclosure does not impose any limitation on this.
[0147] The second boundary condition may be, for example, the magnitude of the applied force, the position information of the applied force, etc., which is not limited in the present disclosure.
[0148] In a possible implementation of the embodiment of the present disclosure, a first subset and a second subset exist simultaneously in the subsets corresponding to the monitoring points, wherein the data category corresponding to the first subset is the third category and the data category corresponding to the second subset is the fourth category, when a data construction method is used to respectively construct a second state pair value belonging to the first subset and a fourth state pair value belonging to the second subset, the number of constructed fourth state pair values is a second set multiple of the number of constructed second state pair values.
[0149] The second setting multiple may be pre-set, such as 0.8, 0.9, etc., and the present disclosure does not impose any limitation on this.
[0150] As an example, assuming that the first subset and the second subset exist simultaneously in the subset corresponding to monitoring point 1, the second set multiple is 0.9, and the number of second state pair values constructed using the data construction method is m, then the number of fourth state pair values constructed using the data construction method is 0.9m.
[0151] Step 3043: determine the third state pair value and the fourth state pair value as the first state pair value corresponding to the monitoring point.
[0152] In the embodiment of the present disclosure, the third state pair value and the fourth state pair value are both determined as the first state pair value corresponding to the monitoring point, that is, the third state pair value obtained by data filling and the fourth state pair value obtained by data construction are both determined as the first state pair value corresponding to the monitoring point.
[0153] As an example, assume that the subsets corresponding to monitoring point 3 include subset A, subset B, subset C, and subset D, where:
[0154] When the data category corresponding to subset A is the first category, the initial state pair value in subset A is determined as the first state pair value corresponding to the monitoring point;
[0155] When the data category corresponding to subset B is the second category, the method of filling missing values is used to perform first data filling on the initial state pair values in subset B to obtain the first state pair values corresponding to the monitoring point;
[0156] When the data category corresponding to the subset C is the third category, the method of filling missing values is used to perform first data filling on the initial state pair values in the subset to obtain the first state pair values corresponding to the monitoring point;
[0157] When the data category corresponding to the subset D is the fourth category, the initial state pair values in the subset are processed by using a method of filling missing values and a method of data construction to obtain a first state pair value corresponding to the monitoring point.
[0158] Therefore, by using different data processing methods to perform data preprocessing on the initial state pair values in each subset corresponding to the monitoring point, the first state pair values corresponding to the monitoring point can be effectively obtained.
[0159] It should be noted that at least one of step 301, step 302, step 303 and step 304 can be executed, and when multiple steps in the above steps are executed, they can be executed in parallel or sequentially, and the present disclosure does not impose any restrictions on this.
[0160] The mechanical characterization method of the disclosed embodiment is, for any subset corresponding to the monitoring point, in response to the data category corresponding to the subset being the first category, determining the initial state pair value in the subset as the first state pair value corresponding to the monitoring point; in response to the data category corresponding to the subset being the second category, using the method of filling missing values to perform first data filling on the initial state pair value in the subset to obtain the first state pair value corresponding to the monitoring point; in response to the data category corresponding to the subset being the third category, using the method of data construction to perform first data supplement on the initial state pair value in the subset to obtain the first state pair value corresponding to the monitoring point; in response to the data category corresponding to the subset being the fourth category, using the method of filling missing values and the method of data construction to process the initial state pair value in the subset to obtain the first state pair value corresponding to the monitoring point. Thus, on the one hand, the situation of missing data is effectively reduced, and on the other hand, the noise can be smoothed by constructing data by finite element, effectively reducing the data noise level.
[0161] It can be understood that the first state pair value corresponding to the monitoring point can have a corresponding physical state, wherein the physical state is, for example, a coal mining state, an external force disturbance state, an earthquake disturbance state, etc., that is, the first state pair value corresponding to the monitoring point is obtained when the monitoring point is in a corresponding physical state. Based on this, in the case where each first state pair value has a corresponding physical state, in order to clearly explain how, in any of the above embodiments of the present disclosure, a data-driven algorithm is used to determine the optimal state pair value corresponding to each monitoring point based on a finite element geometric model and multiple first state pairs corresponding to each monitoring point, the present disclosure also proposes a mechanical characterization method.
[0162] Figure 4 A schematic flow chart of a mechanical characterization method provided in another embodiment of the present disclosure.
[0163] like Figure 4 As shown, based on any of the above embodiments of the present disclosure, the mechanical characterization method may further include the following steps:
[0164] Step 401 , for any physical state, based on the finite element geometric model and multiple first state pairs of values of each monitoring point in the physical state, execute multiple rounds of iterations to obtain the optimal state pair of values of each monitoring point in the physical state.
[0165] The physical state may be, but is not limited to, a coal mining state, an external force disturbance state, an earthquake disturbance state, and the like.
[0166] As an example, assuming that the physical state is a coal mining state, for the coal mining state, multiple rounds of iterations can be performed based on the finite element geometric model and multiple first state pairs of each monitoring point under the coal mining state to obtain the optimal state pair of each monitoring point under the coal mining state.
[0167] Step 402: for any round of iteration, a candidate state pair value of each monitoring point in the physical state of this round is obtained from a plurality of first state pair values of each monitoring point in the physical state.
[0168] The candidate state pair values include a third value of the stress state and a fourth value of the strain state.
[0169] As an example, assuming that the monitoring points include monitoring point 1, monitoring point 2, and monitoring point 3, and the physical state is the coal mining state, the first state pair value of monitoring point 1 in the coal mining state includes (a 11 ,b 11 )、(a 12 ,b 12 ) and (a 13 ,b 13 ), the first state pair value of monitoring point 2 under coal mining state includes (a 21 ,b 21 ) and (a 22 ,b 22 ), the first state pair value of monitoring point 3 under coal mining state includes (a 31 ,b 31 )、(a 32 ,b 32 )、(a 33 ,b 33 ) and (a 34 ,b 34 ), then the candidate state pair values of each monitoring point in the current round under the coal mining state can be randomly obtained from the multiple first state pair values of each monitoring point in the physical state. For example, the candidate state pair values of each monitoring point in the current round under the coal mining state are (a 11 ,b 11 )、(a 22 ,b 22 )、(a 34 ,b 34 ).
[0170] Step 403, based on the candidate state pairs of the monitoring points in the current round under the physical state and the finite element geometric model and the finite element geometric model, a data-driven algorithm is used to determine the target state pairs of the monitoring points in the current round under the physical state.
[0171] The target state pair value includes a fifth value of the stress state and a sixth value of the strain state.
[0172] In the disclosed embodiment, a data-driven algorithm may be used to determine a target state pair value of each monitoring point in this round under a physical state based on candidate state pairs of each monitoring point in this round under a physical state and a finite element geometric model.
[0173] As an example, based on the candidate state pair values of each monitoring point in the current round under the physical state and the finite element geometric model, the target state pair values of each monitoring point in the current round under the physical state can be determined according to the following formula:
[0174]
[0175] Among them, the number of nodes in the finite element geometric model is n, the number of monitoring points is m, σ represents the vector composed of the third value of the stress state in the candidate state pair values of each monitoring point in this round, ε represents the vector composed of the fourth value of the strain state in the candidate state pair values of each monitoring point in this round, σ e represents the third value of the stress state in the candidate state pair value of the e-th monitoring point, ε e represents the fourth value of the strain state in the candidate state pair value of the e-th monitoring point, w e represents the weight of the corresponding node of the e-th monitoring point in the finite element geometric model, B ej represents the inverse matrix of the shape function between the e-th monitoring point and the j-th node, B ei Represents the inverse matrix of the shape function between the e-th monitoring point and the i-th node, C e is the preset intermediate matrix, u i is the displacement driven by the candidate state pair value of each monitoring point in the physical state to the i-th node in this round, η i is the Lagrange multiplier corresponding to the i-th node, f i represents the magnitude of the external force on the i-th node, σ * Represents the vector composed of the fifth value of the stress state in the target state pair value of each monitoring point, ε * Represents a vector composed of the sixth value of the strain state in the target state pair value of each monitoring point.
[0176] Step 404, based on the distance between the candidate state pair values of each monitoring point in the current round and the corresponding target state pair values in the physical state, determine whether the target state pair values of each monitoring point in the current round in the physical state are the optimal state pair values of the corresponding monitoring point in the physical state.
[0177] Optionally, the first coefficient is determined based on the distance between the third value of the candidate state pair values under the physical state of each monitoring point in this round and the fifth value of the corresponding target state pair values; the second coefficient is determined based on the distance between the fourth value of the candidate state pair values under the physical state of each monitoring point in this round and the sixth value of the corresponding target state pair values; in response to the first coefficient being less than the first set threshold and the second coefficient being less than the second set threshold, the target state pair values under the physical state of each monitoring point in this round are determined to be the optimal state pair values of the corresponding monitoring points in the physical state.
[0178] Still using the above example to illustrate, σ represents the vector composed of the third value of the stress state in the candidate state pair values of each monitoring point in this round, ε represents the vector composed of the fourth value of the strain state in the candidate state pair values of each monitoring point in this round, σ * Represents the vector composed of the fifth value of the stress state in the target state pair value of each monitoring point, ε * The vector consisting of the sixth value of the strain state in the target state pair value of each monitoring point, the first set threshold is l 1 , the second threshold is set to l 2 The first coefficient k can be determined according to the following formula: 1 :
[0179] k 1 =||σ-σ * ||; (7)
[0180] According to the following formula, the second coefficient k is determined 2 :
[0181] k 2 =||ε-ε * ||; (8)
[0182] When the first coefficient k 1 Less than the first set threshold value is l 1 , and the second coefficient k 2 Less than the second set threshold value is l 2 When , we can determine that the target state pair value of each monitoring point in the physical state in this round is the optimal state pair value of the corresponding monitoring point in the physical state, that is, (σ * , ε * ) is the optimal state pair value of each monitoring point under the corresponding physical state.
[0183] Optionally, when the first coefficient is not less than the first set threshold, and / or the second coefficient is not less than the second set threshold, it can be determined that the target state pair value of each monitoring point in the physical state in this round is not the optimal state pair value of the corresponding monitoring point in the physical state.
[0184] Step 405 , in response to the target state pair value of each monitoring point in the physical state in this round being the optimal state pair value of the corresponding monitoring point in the physical state, the iteration process is stopped.
[0185] Step 406 , in response to the target state pair value of each monitoring point in the physical state in this round not being the optimal state pair value of the corresponding monitoring point in the physical state, executing the next round of iteration process.
[0186] It should be noted that, in actual applications, step 405 and step 406 are performed alternatively.
[0187] The mechanical characterization method of the disclosed embodiment performs multiple rounds of iterations for any physical state based on a finite element geometric model and multiple first state pairs of values of each monitoring point in the physical state to obtain the optimal state pair of values of each monitoring point in the physical state; for any round of iterations, obtain the candidate state pair of values of each monitoring point in the physical state in this round from the multiple first state pairs of values of each monitoring point in the physical state; based on the candidate state pair of values of each monitoring point in the physical state in this round, use a data-driven algorithm to determine the target state pair of values of each monitoring point in the physical state in this round; based on the distance between the candidate state pair of values of each monitoring point in the physical state in this round and the corresponding target state pair of values, determine whether the target state pair of values of each monitoring point in the physical state in this round is the optimal state pair of values of the corresponding monitoring point in the physical state; in response to the target state pair of values of each monitoring point in the physical state in this round being the optimal state pair of values of the corresponding monitoring point in the physical state, stop the iteration; in response to the target state pair of values of each monitoring point in the physical state in this round not being the optimal state pair of values of the corresponding monitoring point in the physical state, perform the next round of iterations. Therefore, based on the theory of data-driven mechanics, the global optimal state pair, namely the global optimal stress state and the optimal strain state, is effectively determined.
[0188] In order to clearly illustrate the mechanical characterization method disclosed in the present invention, the above process is described in detail below with reference to examples.
[0189] As an example, Figure 5 As shown, the mechanical characterization method of the present disclosure may include the following steps:
[0190] Step 510, data collection and data classification
[0191] Specifically, the following steps 511 to 512 may be included:
[0192] Step 511: For any monitoring point of the plurality of monitoring points in the target mining area, monitoring data corresponding to the monitoring point is collected, wherein the monitoring data includes the position of the monitoring point, the measured stress (referred to as the first value in this disclosure), and the measured strain (referred to as the second value in this disclosure).
[0193] Step 512, data classification. The data is classified into four categories according to the actual situation, and marked as Class A, Class B, Class C, and Class D. Among them, Class A data refers to data in which both measured stress and measured strain exist, and the noise level of measured stress and measured strain is less than 10%; Class B data refers to data in which measured stress or measured strain is missing, and the noise level of measured stress and measured strain is less than 10%, wherein the data in which measured stress is missing is marked as Class B-1, and the data in which measured strain is missing is marked as Class B-2; Class C data refers to data in which both measured stress and measured strain exist, and the noise level of measured stress and / or measured strain is higher than 10%; Class D data refers to data in which measured stress or measured strain is missing, and the noise level of the existing measured stress or the existing measured strain is higher than 10%, wherein the data in which measured stress is missing is marked as Class D-1, and the data in which measured strain is missing is marked as Class D-2.
[0194] Step 520: Establish a finite element geometric model
[0195] Establish a finite element geometric model of the actual project (i.e., the target mining area), in which the nodes, the topological relationship between the nodes, the shape functions between the nodes, the derivative matrix corresponding to the shape functions between the nodes, and the weight of each node are defined in the finite element geometric model. When establishing the geometric model, each monitoring point has a corresponding node in the finite element geometric model.
[0196] Step 530, Class B data processing
[0197] Specifically, the following steps 531 to 534 may be included:
[0198] Step 531: For any monitoring data in the B-type data, obtain the constitutive model (referred to as the first constitutive model in this disclosure) and material parameters (referred to as the first material parameters in this disclosure) corresponding to the monitoring point corresponding to the monitoring data in the target mining area, wherein the relevant staff can estimate the possible constitutive model and material parameters of the monitoring point based on engineering experience. The constitutive model may be, for example, a linear elastic constitutive model, and the material parameters may include, for example, Young's modulus and Poisson's ratio.
[0199] Step 532: Construct a material stiffness matrix according to the constitutive model and material parameters.
[0200] For example, assuming that the constitutive model is linear elastic, the Young's modulus of the material parameters is E, and the Poisson's ratio is ν, then the Lame constant λ of the material at the monitoring point is as shown in formula (1), the shear modulus μ is as shown in formula (2), and the material stiffness matrix C is as shown in formula (3).
[0201] Step 533: Process the missing data of measured strain. Traverse the B-1 data, assume that the measured stress is a, determine and fill in the corresponding measured strain b as:
[0202] b=C -1 *a; (9)
[0203] Step 534: Process the data with missing measured stress. Traverse the B-2 data, assume that the measured strain is b, and determine and fill in the corresponding measured stress a as:
[0204] a=C -1 *b; (10)
[0205] Step 540, Class C data processing
[0206] Specifically, the following steps 541 to 543 may be included:
[0207] Step 541, construction: using the material stiffness matrix established in step 530, and carrying out finite element calculation based on the finite element geometric model established in step 520, random boundary conditions can be constructed within a reasonable range, so that the "stress-strain" pairs corresponding to the monitoring points in the C-type data can be constructed based on the material stiffness matrix, the finite element geometric model and the boundary conditions. It should be noted that, for any monitoring point, the number of "stress-strain" pairs constructed for the monitoring point can be determined according to the number of C-type data corresponding to the monitoring point, for example, the number of "stress-strain" pairs constructed for the monitoring point is 1 to 2 times the number of C-type data corresponding to the monitoring point; it should also be noted that the specific value of the multiple is determined according to the noise level of the data, and the greater the data noise, the higher the value of the multiple.
[0208] Step 542, data appending (also called data supplementation): for any monitoring point, append the constructed "stress-strain" pair of the monitoring point to the C-type data of the corresponding monitoring point.
[0209] Step 543, random arrangement: randomly arrange the C-type data to which the construction data is appended, so that the data is evenly mixed and arranged in the storage space of the C-type data.
[0210] Step 550: Class D data processing
[0211] First, the D-1 data and D-2 data are processed respectively according to the processing method for the B-1 data and the B-2 data in step 530, and then the construction data is added according to the processing method for the C data in step 540. For any monitoring point, the amount of construction data for the D data of the monitoring point is generally 0.8 to 0.9 times the amount of construction data for the C data of the monitoring point.
[0212] Step 560, data driven calculation
[0213] Specifically, the following steps 561 to 564 may be included:
[0214] Step 561, rearrange data: for any monitoring point, establish a corresponding data set. Specifically, merge and store the four types of data A, B, C, and D corresponding to the monitoring point after data processing in the corresponding data set, and randomly rearrange them so that the four types of data are randomly and evenly arranged in the corresponding data set.
[0215] It should be noted that, in the present disclosure, each "stress-strain" pair in the data set is recorded as a first state pair value.
[0216] Step 562: For a certain physical state, randomly select a "stress-strain" pair of the corresponding monitoring point in the physical state from the data set corresponding to any monitoring point, and use the selected "stress-strain" pair of the corresponding monitoring point in the physical state as a candidate state pair value of the monitoring point in the physical state.
[0217] The candidate state pair values include a third value of the stress state and a fourth value of the strain state.
[0218] Step 563: Based on the finite element geometric model and the candidate state pair values of each monitoring point in the physical state, according to formula (6), determine the target state pair value of each monitoring point in the physical state.
[0219] The target state pair value includes a third value of the stress state and a fourth value of the strain state.
[0220] Step 564, error check:
[0221] Based on the distance between the third value of the candidate state pair values of each monitoring point in the physical state and the fifth value of the corresponding target state pair values, a first coefficient is determined; based on the distance between the fourth value of the candidate state pair values of each monitoring point in the physical state and the sixth value of the corresponding target state pair values, a second coefficient is determined; in response to the first coefficient being less than the first set threshold value, and the second coefficient being less than the second set threshold value, it can be determined that the target state pair value of each monitoring point in the physical state is the optimal state pair value (also called the global state pair value) of the corresponding monitoring point in the physical state; at this time, the determination process of the optimal state pair value in the physical state can be exited, and the determination process of the optimal state pair value in the next physical state can be executed until the determination process of the optimal state pair value in all physical states is completed;
[0222] Optionally, when the first coefficient is not less than the first set threshold, and / or the second coefficient is not less than the second set threshold, it can be determined that the target state pair value of each monitoring point in the physical state is not the optimal state pair value of the corresponding monitoring point in the physical state; at this time, steps 562-564 can be executed in a loop until the optimal state pair value under the physical state is determined.
[0223] In the field of coal mine stress monitoring, there are currently methods such as fuzzy mathematics, artificial neural networks, big data analysis, and multi-source data fusion to reduce monitoring data noise and reduce data missing. However, the above methods lack the support of mechanical principles and cannot completely solve practical engineering problems.
[0224] In summary, the mechanical characterization method disclosed in the present invention can effectively reduce the noise level of monitoring data and effectively reduce the situation of data missing. It uses the basic theory of data-driven mechanics as support, so that the calculated global stress state conforms to the basic principles of computational mechanics, and can effectively support the prevention and control of coal mine rock burst disasters and the development of deep mining.
[0225] The mechanical characterization method disclosed in the present invention, on the one hand, is suitable for scenarios with missing data and has strong applicability; on the other hand, it reduces the error caused by data noise by constructing data through finite elements, thereby enhancing the performance of digitally driven mechanical characterization in actual engineering.
[0226] With the above Figures 1 to 4 Corresponding to the mechanical characterization method provided in the embodiment, the present disclosure also provides a mechanical characterization device. Figures 1 to 4 The mechanical characterization method provided in the embodiment corresponds to the embodiment, so the implementation method of the mechanical characterization method is also applicable to the mechanical characterization device provided in the embodiment of the present disclosure, and will not be described in detail in the embodiment of the present disclosure.
[0227] Figure 6 A schematic structural diagram of a mechanical characterization device provided in another embodiment of the present disclosure.
[0228] like Figure 6 As shown, the mechanical characterization device 600 may include: an acquisition module 601 , a construction module 602 , a processing module 603 and a determination module 604 .
[0229] The acquisition module 601 is used to acquire a plurality of initial state pair values corresponding to any one of a plurality of monitoring points in the target mining area; wherein the initial state pair values include a first value of a stress state and a second value of a strain state.
[0230] The construction module 602 is used to construct a finite element geometric model corresponding to the target mining area; wherein the finite element geometric model includes a plurality of nodes, and each monitoring point has a corresponding node in the finite element geometric model.
[0231] The processing module 603 is used to perform data preprocessing on a plurality of initial state pair values corresponding to any monitoring point to obtain a plurality of first state pair values corresponding to the monitoring point.
[0232] The determination module 604 is used to determine the optimal state pair value corresponding to each monitoring point by using a data-driven algorithm based on the finite element geometric model and multiple first state pair values corresponding to each monitoring point.
[0233] In a possible implementation of the embodiments of the present disclosure, the processing module 603 is used to: for any monitoring point, divide the multiple initial state pair values corresponding to the monitoring point according to the data category to which the initial state pair values corresponding to the monitoring point belong, so as to obtain at least one subset corresponding to the monitoring point; for any subset corresponding to the monitoring point, based on the data category corresponding to the subset, use the corresponding data preprocessing method to perform data preprocessing on the initial state pair values in the subset, so as to obtain the first state pair value corresponding to the monitoring point.
[0234] In a possible implementation manner of the embodiment of the present disclosure, the data category includes at least one of a first category, a second category, a third category, and a fourth category, wherein:
[0235] The first category indicates that both the first value and the second value exist in the corresponding initial state pair value, and the noise levels of the first value and the second value in the corresponding initial state pair value are both less than the set threshold;
[0236] The second category indicates that the first value or the second value in the corresponding initial state pair of values is missing, and the noise level of the first value or the second value in the corresponding initial state pair of values is less than a set threshold;
[0237] The third category indicates that both the first value and the second value exist in the corresponding initial state pair value, and the noise level of the first value and / or the second value in the corresponding initial state pair value is not less than the set threshold;
[0238] The fourth category indicates that the first value or the second value in the corresponding initial state pair of values is missing, and the noise level of the first value or the second value that exists in the corresponding initial state pair of values is not less than a set threshold.
[0239] In a possible implementation of the embodiment of the present disclosure, the processing module 603 is used to: for any subset corresponding to the monitoring point, in response to the data category corresponding to the subset being the first category, determine the initial state pair value in the subset as the first state pair value corresponding to the monitoring point; in response to the data category corresponding to the subset being the second category, use a method of filling missing values to perform first data filling on the initial state pair value in the subset to obtain the first state pair value corresponding to the monitoring point; in response to the data category corresponding to the subset being the third category, use a data construction method to perform first data supplement on the initial state pair value in the subset to obtain the first state pair value corresponding to the monitoring point; in response to the data category corresponding to the subset being the fourth category, use a method of filling missing values and a data construction method to process the initial state pair value in the subset to obtain the first state pair value corresponding to the monitoring point.
[0240] In a possible implementation of the embodiment of the present disclosure, the processing module 603 is used to: in response to the data category corresponding to the subset being the second category, obtain the first constitutive model and the first material parameter corresponding to the monitoring point in the target mining area; wherein the first material parameter matches the first constitutive model; construct the material stiffness matrix corresponding to the monitoring point according to the first constitutive model and the first material parameter corresponding to the monitoring point; for any initial state pair value in the subset, determine the target value in the initial state pair value based on the material stiffness matrix corresponding to the monitoring point; wherein the target value is the missing first value or the missing second value; based on the target value, fill in the initial state pair value, and determine the filled initial state pair value as the first state pair value corresponding to the monitoring point.
[0241] In a possible implementation of the embodiment of the present disclosure, the processing module 603 is used to: in response to the data category corresponding to the subset being the third category, obtain the second constitutive model and the second material parameter corresponding to the monitoring point in the target mining area, and construct the first boundary condition corresponding to the monitoring point; wherein the second material parameter matches the second constitutive model; based on the second constitutive model, the second material parameter and the first boundary condition corresponding to the monitoring point, and the finite element geometric model, use the finite element method to construct a second state pair value belonging to the subset; determine the constructed second state pair value and the initial state pair value in the subset as the first state pair value corresponding to the monitoring point.
[0242] In a possible implementation of the embodiment of the present disclosure, the processing module 603 is used to: in response to the data category corresponding to the subset being the fourth category, perform second data filling on the initial state pair values in the subset using a method of filling missing values to obtain a third state pair value; use a data construction method to construct a fourth state pair value belonging to the subset; and determine both the third state pair value and the fourth state pair value as the first state pair value corresponding to the monitoring point.
[0243] In a possible implementation of the embodiment of the present disclosure, each first state pair value has a corresponding physical state, and there is at least one physical state; a determination module is used to: for any physical state, based on a finite element geometric model and a plurality of first state pairs of values of each monitoring point in the physical state, perform multiple rounds of iterations to obtain the optimal state pair value of each monitoring point in the physical state; for any round of iterations, obtain the candidate state pair values of each monitoring point in the physical state in this round from the plurality of first state pairs of values of each monitoring point in the physical state; wherein the candidate state pair values include a third value of the stress state and a fourth value of the strain state; based on the candidate state pair values of each monitoring point in the physical state in this round and the finite element geometric model, adopt A data-driven algorithm is used to determine a target state pair value of each monitoring point in the current round under a physical state; wherein the target state pair value includes a fifth value of the stress state and a sixth value of the strain state; based on the distance between the candidate state pair value of each monitoring point in the current round under a physical state and the corresponding target state pair value, determine whether the target state pair value of each monitoring point in the current round under a physical state is an optimal state pair value of the corresponding monitoring point in the physical state; in response to the target state pair value of each monitoring point in the current round under a physical state being the optimal state pair value of the corresponding monitoring point in the physical state, stop the iteration process; in response to the target state pair value of each monitoring point in the current round under a physical state not being the optimal state pair value of the corresponding monitoring point in the physical state, execute the next iteration process.
[0244] In a possible implementation of the embodiment of the present disclosure, the processing module 603 is used to: determine a first coefficient based on the distance between a third value in the candidate state pair values of each monitoring point in the current round under a physical state and a fifth value in the corresponding target state pair values; determine a second coefficient based on the distance between a fourth value in the candidate state pair values of each monitoring point in the current round under a physical state and a sixth value in the corresponding target state pair values; in response to the first coefficient being less than a first set threshold value and the second coefficient being less than a second set threshold value, determine that the target state pair values of each monitoring point in the current round under a physical state are the optimal state pair values of the corresponding monitoring point in the physical state.
[0245] The mechanical characterization device of the disclosed embodiment obtains multiple initial state pairs corresponding to any monitoring point among multiple monitoring points in the target mining area; wherein the initial state pairs include a first value of the stress state and a second value of the strain state; constructs a finite element geometric model corresponding to the target mining area; wherein the finite element geometric model includes multiple nodes, and each monitoring point has a corresponding node in the finite element geometric model; for any monitoring point, data preprocessing is performed on the multiple initial state pairs corresponding to the monitoring point to obtain multiple first state pairs corresponding to the monitoring point; based on the finite element geometric model and the multiple first state pairs corresponding to each monitoring point, a data-driven algorithm is used to determine the optimal state pairs corresponding to each monitoring point. Thus, data preprocessing is performed on the initial state pairs of each monitoring point in advance to improve data accuracy, which can effectively enhance the performance of digital-driven mechanical characterization in actual engineering.
[0246] In order to implement the above embodiments, the present disclosure further proposes an electronic device, wherein the electronic device can be the server or detection device in the aforementioned embodiments; it includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the program, it implements the mechanical characterization method proposed in any of the aforementioned embodiments of the present disclosure.
[0247] In order to implement the above embodiments, the present disclosure further proposes a non-temporary computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the mechanical characterization method proposed in any of the aforementioned embodiments of the present disclosure is implemented.
[0248] In order to implement the above embodiments, the present disclosure further proposes a computer program product. When the instructions in the computer program product are executed by a processor, the mechanical characterization method proposed in any of the above embodiments of the present disclosure is executed.
[0249] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.
[0250] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, "plurality" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0251] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present disclosure includes additional implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present disclosure belong.
[0252] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.
[0253] It should be understood that the various parts of the present disclosure can be implemented in hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0254] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.
[0255] In addition, each functional unit in each embodiment of the present disclosure may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.
[0256] The storage medium mentioned above may be a read-only memory, a disk or an optical disk, etc. Although the embodiments of the present disclosure have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations of the present disclosure. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present disclosure.
Claims
1. A mechanical characterization method, It is characterized in that The method comprises: For any of the monitoring points in the target mining area, a plurality of initial state pair values corresponding to the monitoring point are obtained; wherein the initial state pair values include a first value of the stress state and a second value of the strain state; Constructing a finite element geometric model corresponding to the target mining area; wherein the finite element geometric model includes a plurality of nodes, and each of the monitoring points has a corresponding node in the finite element geometric model; For any of the monitoring points, data preprocessing is performed on a plurality of initial state pairs of values corresponding to the monitoring point to obtain a plurality of first state pairs of values corresponding to the monitoring point; Based on the finite element geometric model and the plurality of first state pairs corresponding to the monitoring points, a data-driven algorithm is used to determine the optimal state pair corresponding to each of the monitoring points; Wherein, each of the first state pair values has a corresponding physical state, and there is at least one physical state; based on the finite element geometric model and the multiple first state pair values corresponding to each of the monitoring points, a data-driven algorithm is used to determine the optimal state pair value corresponding to each of the monitoring points, including: For any of the physical states, based on the finite element geometric model and a plurality of first state pairs of the monitoring points under the physical state, performing multiple rounds of iterations to obtain an optimal state pair of the monitoring points under the physical state; For any round of the iteration process, a candidate state pair value of each monitoring point in the physical state in this round is obtained from a plurality of the first state pair values of each monitoring point in the physical state; wherein the candidate state pair value includes a third value of the stress state and a fourth value of the strain state; Based on the candidate state pair values of each monitoring point in this round under the physical state and the finite element geometric model, a data-driven algorithm is used to determine the target state pair values of each monitoring point in this round under the physical state; wherein the target state pair values include the fifth value of the stress state and the sixth value of the strain state; Based on the distance between the candidate state pair value of each monitoring point in the current round under the physical state and the corresponding target state pair value, determining whether the target state pair value of each monitoring point in the current round under the physical state is the optimal state pair value of the corresponding monitoring point in the physical state; In response to the target state pair value of each monitoring point in the current round being the optimal state pair value of the corresponding monitoring point in the physical state, stopping the iteration process; In response to the target state pair value of each monitoring point in the current round under the physical state not being the optimal state pair value of the corresponding monitoring point in the physical state, executing the next round of iteration process; Wherein, determining whether the target state pair value of each monitoring point in the current round under the physical state is the optimal state pair value of the corresponding monitoring point in the physical state based on the distance between the candidate state pair value of each monitoring point in the current round under the physical state and the corresponding target state pair value includes: Determine a first coefficient based on a distance between a third value in the candidate state pair value under the physical state and a fifth value in the corresponding target state pair value of each monitoring point in this round; Determine a second coefficient based on a distance between a fourth value in the candidate state pair values and a sixth value in the corresponding target state pair values of each monitoring point in the physical state in this round; In response to the first coefficient being less than a first set threshold and the second coefficient being less than a second set threshold, the target state pair value of each monitoring point in the current round under the physical state is determined to be the optimal state pair value of the corresponding monitoring point under the physical state.
2. The method according to claim 1, It is characterized in that For any of the monitoring points, data preprocessing is performed on a plurality of initial state pairs corresponding to the monitoring point to obtain a plurality of first state pairs corresponding to the monitoring point, including: For any of the monitoring points, according to the data category to which each of the initial state pair values corresponding to the monitoring point belongs, the multiple initial state pair values corresponding to the monitoring point are divided to obtain at least one subset corresponding to the monitoring point; For any of the subsets corresponding to the monitoring points, based on the data category corresponding to the subset, a corresponding data preprocessing method is used to perform data preprocessing on the initial state pair values in the subset to obtain the first state pair values corresponding to the monitoring points.
3. The method according to claim 2, It is characterized in that The data category includes at least one of a first category, a second category, a third category, and a fourth category, wherein: The first category indicates that both the first value and the second value exist in the corresponding initial state pair, and the noise levels of the first value and the second value in the corresponding initial state pair are both less than a set threshold; The second category indicates that the first value or the second value in the corresponding initial state pair of values is missing, and the noise level of the first value or the second value that exists in the corresponding initial state pair of values is less than the set threshold; The third category indicates that both the first value and the second value exist in the corresponding initial state pair value, and the noise level of the first value and / or the second value in the corresponding initial state pair value is not less than the set threshold; The fourth category indicates that the first value or the second value in the corresponding initial state pair of values is missing, and the noise level of the first value or the second value that exists in the corresponding initial state pair of values is not less than the set threshold.
4. The method according to claim 3, It is characterized in that For any of the subsets corresponding to the monitoring points, based on the data category corresponding to the subset, using a corresponding data preprocessing method to perform data preprocessing on the initial state pair values in the subset to obtain the first state pair values corresponding to the monitoring point, including: For any of the subsets corresponding to the monitoring points, in response to the data category corresponding to the subset being the first category, determining the initial state pair value in the subset as the first state pair value corresponding to the monitoring point; In response to the data category corresponding to the subset being the second category, performing first data filling on the initial state pair values in the subset using a method for filling missing values to obtain a first state pair value corresponding to the monitoring point; In response to the data category corresponding to the subset being the third category, using a data construction method to perform first data supplement on the initial state pair values in the subset to obtain a first state pair value corresponding to the monitoring point; In response to the data category corresponding to the subset being the fourth category, the method for filling missing values and the method for data construction are used to process the initial state pair values in the subset to obtain the first state pair values corresponding to the monitoring point.
5. The method according to claim 4, It is characterized in that In response to the data category corresponding to the subset being the second category, performing first data filling on the initial state pair values in the subset using a method for filling missing values to obtain the first state pair values corresponding to the monitoring point, including: In response to the data category corresponding to the subset being the second category, obtaining a first constitutive model and a first material parameter corresponding to the monitoring point in the target mining area; wherein the first material parameter matches the first constitutive model; Constructing a material stiffness matrix corresponding to the monitoring point according to the first constitutive model and the first material parameter corresponding to the monitoring point; For any of the initial state pairs in the subset, based on the material stiffness matrix corresponding to the monitoring point, determine a target value in the initial state pair; wherein the target value is the missing first value or the missing second value; Based on the target value, the initial state pair value is padded, and the initial state pair value after the padded process is determined as the first state pair value corresponding to the monitoring point.
6. The method according to claim 4, It is characterized in that In response to the data category corresponding to the subset being the third category, using a data construction method to perform first data supplementation on the initial state pair values in the subset to obtain the first state pair values corresponding to the monitoring point, including: In response to the data category corresponding to the subset being the third category, obtaining a second constitutive model and a second material parameter corresponding to the monitoring point in the target mining area, and constructing a first boundary condition corresponding to the monitoring point; wherein the second material parameter matches the second constitutive model; Based on the second constitutive model, the second material parameter and the first boundary condition corresponding to the monitoring point, and the finite element geometric model, a finite element method is used to construct a second state pair value belonging to the subset; The constructed second state pair value and the initial state pair value in the subset are both determined as the first state pair value corresponding to the monitoring point.
7. The method according to claim 4, It is characterized in that In response to the data category corresponding to the subset being the fourth category, using the method for filling missing values and the method for data construction to process the initial state pair values in the subset to obtain the first state pair values corresponding to the monitoring point, including: In response to the data category corresponding to the subset being the fourth category, performing second data filling on the initial state pair values in the subset using the missing value filling method to obtain a third state pair value; Using a data construction method, construct a fourth state pair value belonging to the subset; The third state pair value and the fourth state pair value are both determined as the first state pair value corresponding to the monitoring point.
8. A mechanical characterization device, It is characterized in that The device comprises: An acquisition module, for acquiring, for any of the plurality of monitoring points in the target mining area, a plurality of initial state pair values corresponding to the monitoring point; wherein the initial state pair values include a first value of the stress state and a second value of the strain state; A construction module, used to construct a finite element geometric model corresponding to the target mining area; wherein the finite element geometric model includes a plurality of nodes, and each of the monitoring points has a corresponding node in the finite element geometric model; A processing module, configured to perform data preprocessing on a plurality of initial state pairs of values corresponding to any of the monitoring points, so as to obtain a plurality of first state pairs of values corresponding to the monitoring point; A determination module, configured to determine the optimal state pair value corresponding to each of the monitoring points by using a data-driven algorithm based on the finite element geometric model and a plurality of first state pair values corresponding to each of the monitoring points; Wherein, each of the first state pair values has a corresponding physical state, and there is at least one physical state; and the determining module is further used to: For any of the physical states, based on the finite element geometric model and a plurality of first state pairs of the monitoring points under the physical state, performing multiple rounds of iterations to obtain an optimal state pair of the monitoring points under the physical state; For any round of the iteration process, a candidate state pair value of each monitoring point in the physical state in this round is obtained from a plurality of the first state pair values of each monitoring point in the physical state; wherein the candidate state pair value includes a third value of the stress state and a fourth value of the strain state; Based on the candidate state pair values of each monitoring point in this round under the physical state and the finite element geometric model, a data-driven algorithm is used to determine the target state pair values of each monitoring point in this round under the physical state; wherein the target state pair values include the fifth value of the stress state and the sixth value of the strain state; Based on the distance between the candidate state pair value of each monitoring point in the current round under the physical state and the corresponding target state pair value, determining whether the target state pair value of each monitoring point in the current round under the physical state is the optimal state pair value of the corresponding monitoring point in the physical state; In response to the target state pair value of each monitoring point in the current round being the optimal state pair value of the corresponding monitoring point in the physical state, stopping the iteration process; In response to the target state pair value of each monitoring point in the current round under the physical state not being the optimal state pair value of the corresponding monitoring point in the physical state, executing the next round of iteration process; Wherein, the determining module is further used for: Determine a first coefficient based on a distance between a third value in the candidate state pair value under the physical state and a fifth value in the corresponding target state pair value of each monitoring point in this round; Determine a second coefficient based on a distance between a fourth value in the candidate state pair values and a sixth value in the corresponding target state pair values of each monitoring point in the physical state in this round; In response to the first coefficient being less than a first set threshold and the second coefficient being less than a second set threshold, the target state pair value of each monitoring point in the current round under the physical state is determined to be the optimal state pair value of the corresponding monitoring point under the physical state.
