Method for constructing digital rock gene pool based on massive experimental data

Abstract

The invention discloses a method for constructing a digital rock gene pool based on mass experimental data, which comprises the following steps: collecting different typical rock samples, carrying out related experimental researches to obtain related experimental data of the typical rock samples, and processing the related experimental data to form rock standardized data; collecting typical geological disaster data occurring in history, and processing the typical geological disaster data to form historical disaster standardized data; collecting current real-time dynamic monitoring data of geological disasters; and performing fusion analysis and unified standardization processing on the three heterogeneous data sources to form the digital rock gene pool. The digital rock gene pool is constructed by adopting three heterogeneous data sources, the defect that analysis results are different due to incomplete data and single method in the existing rock mechanics field is overcome, the broad spectrum is high, and the accuracy is high.

Description

Technical Field

[0001] The invention belongs to the technical field of rock mechanics, and in particular relates to a method for constructing a digital rock gene library based on massive experimental data. Background Art

[0002] Geotechnical geological disasters include floods caused by rising groundwater levels, landslides, collapses and mudslides caused by rock stability problems. These disasters not only endanger the safety of construction facilities and equipment, but may also pose a serious threat to the safety of workers' lives and the environment. Geotechnical engineering geological disaster prediction is a technical method for predicting and warning geological disasters through comprehensive investigation, analysis and evaluation of the geological environment. It aims to quickly and accurately identify and analyze geological disasters, reduce the burden of manual investigation, and provide important support for geotechnical engineering planning, design and construction, thereby reducing the risks and losses of geological disasters.

[0003] Geological disasters are a multi-factor phenomenon. In recent years, with the development of the theoretical basis of rock mechanics and the innovation of experimental technology, we have a deeper understanding of geological disasters and their mechanisms. However, the formation mechanism of geological disasters is very complex, involving factors such as lithology, landforms, climate conditions, and human activities. The traditional method of studying it is to conduct field tests and indoor tests through on-site exploration and sampling. The disadvantage is that it is only targeted at a certain engineering site and repeated, which increases time and economic costs.

[0004] Therefore, it is urgent to use big data technology to collect, organize and reconstruct experimental data to predict the rock properties of the geological structure where a project is located, and provide a technical basis for the prediction and early warning of engineering geological disasters. Summary of the invention

[0005] The purpose of the present invention is to provide a method for constructing a digital rock gene library based on massive experimental data in order to solve the deficiencies of the prior art.

[0006] The purpose of the present invention is achieved by the following technical solutions:

[0007] A method for constructing a digital rock gene library based on massive experimental data includes the following steps:

[0008] Collect different typical rock samples, conduct relevant experimental research, obtain relevant experimental data of the typical rock samples, and process the relevant experimental data to form rock standardization data;

[0009] Collect typical geological disaster data that occurred in history and process them to form standardized data of historical disasters;

[0010] Collect current real-time dynamic monitoring data of geological disasters;

[0011] The rock standardized data, the historical disaster standardized data and the real-time dynamic monitoring data are fused and analyzed and uniformly standardized to form a digital rock gene library.

[0012] Preferably, the relevant experimental research includes at least one of optical microscope observation, scanning electron microscope observation, transmission electron microscope observation, X-ray diffraction analysis, CT scanning analysis, magnetic resonance imaging analysis, infrared spectroscopy analysis, relevant physical and chemical property detection experiments, sound wave experiments and relevant mechanical experiments.

[0013] Preferably, the relevant experimental data at least include microstructure parameters, basic physical and chemical property parameters and mechanical parameter data of the rock sample;

[0014] The microstructure parameters include at least one of microscopic crystal structure, grain boundary distribution, mineral particle morphology, mineral particle size, porosity, crack space distribution, crack number, crack length and crack area;

[0015] The basic physical and chemical property parameters include at least one of color, gloss, transparency, water content, water absorption, density, specific gravity, mineral type and composition, permeability, swelling, solubility and frost resistance;

[0016] The mechanical parameters include at least one of compressive strength, flexural strength, shear strength, impact strength, elastic modulus, deformation modulus and Poisson's ratio.

[0017] Preferably, the geological disaster data at least includes rock-related data that causes the disaster, hydrological and climate data, geological structure data, and topographic data.

[0018] Preferably, the real-time dynamic monitoring data includes at least satellite remote sensing data, aerial remote sensing data, and data collected by ground and underground monitoring equipment.

[0019] Preferably, the formation rock standardization data further includes:

[0020] Collect relevant data information of public and unpublished rock samples, and merge them with the relevant experimental data to form the rock standardization data.

[0021] A digital rock gene library constructed based on the method described above.

[0022] An application method based on the digital rock gene library as described above comprises the following steps:

[0023] According to the digital rock gene library, rock standard detection items are determined, and a relationship model between parameter values ​​of the rock standard detection items and other rock parameter values ​​is established;

[0024] For newly collected rock samples to be tested, the rock samples to be tested are tested according to the rock standard testing items, and the test results are substituted into the relationship model to obtain other parameter information of the rock samples to be tested.

[0025] Preferably, the determination of rock standard testing items further includes:

[0026] Determine the standard rock testing items corresponding to different geological disasters.

[0027] Preferably, if the detection result of the rock sample to be tested cannot correspond to the relationship model, relevant experimental research is carried out on the rock sample to be tested to obtain relevant experimental data of the rock sample to be tested, and the rock standardization data, the digital rock gene library and the relationship model are further updated.

[0028] This application uses three heterogeneous data sources to construct a digital rock gene library, which makes up for the defects of incomplete data and single methods in the existing rock mechanics field, resulting in different analysis results. It has strong broad spectrum and high accuracy. Based on this gene library, a small number of items can be quickly detected, and other basic properties of rocks can be analyzed and predicted, thus providing a technical basis for geological disaster prediction and early warning and guiding human activities. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a flow chart of the steps of the method for constructing a digital rock gene library provided by the present invention. DETAILED DESCRIPTION

[0030] The present invention provides a method for constructing a digital rock gene library based on massive experimental data. Figure 1 As shown, the following steps are included:

[0031] S1. Collect different typical rock samples, carry out various related experimental studies, obtain relevant experimental data of typical rock samples, and process the relevant experimental data to form rock standardized data.

[0032] Typical rock samples should be as comprehensive as possible, covering rocks from different regions, different geological structures (such as folds, joints, faults, etc.), different topography (such as basins, mountains, plains, etc.), different diagenetic ages (such as Quaternary, Jurassic, Triassic, Ordovician, Cambrian, etc.) and different types (such as igneous rocks, sedimentary rocks, metamorphic rocks, etc.). You can also collect relevant data and information of rock samples that are publicly available and unpublished through the Internet, books, newspapers and other channels verified by authoritative departments as a supplement. When the same typical rock samples are not collected, the relevant data of the collected rock samples can be directly adopted. When the same typical rock samples are collected, the experimental data can be compared and analyzed with the collected data to reduce error data and improve the accuracy of rock data.

[0033] The relevant experimental research mainly includes: (1) some rapid non-destructive testing methods such as optical microscope observation, scanning electron microscope observation, transmission electron microscope observation, CT scanning analysis, magnetic resonance imaging analysis, infrared spectrum analysis, X-ray diffraction analysis, etc., which can quickly obtain the micro-structure parameters related to rocks; (2) relevant physical and chemical property detection experiments: such as water content experiments, density experiments, elemental analysis experiments, etc., which can experimentally obtain the basic physical and chemical property parameters related to rocks; (3) acoustic wave experiments, various related mechanical experiments such as uniaxial compressive strength tests, direct shear tests, conventional triaxial tests, creep tests, hydraulic fracturing experiments, etc., which can explore the mechanical failure mechanism of rocks and collect various mechanical parameters of rocks.

[0034] The relevant experimental data include at least the microstructure parameters, basic physical and chemical property parameters and mechanical parameter data of the rock samples.

[0035] The microstructure parameters mainly include: microscopic crystal structure, grain boundary distribution, mineral particle morphology, mineral particle size, porosity, crack space distribution, crack number, crack length, crack area, etc., but are not limited to the above parameters;

[0036] Basic physical and chemical property parameters mainly include: color, gloss, transparency, moisture content, water absorption, density, specific gravity, mineral type and composition, permeability, swelling, solubility, frost resistance, etc., but not limited to the above parameters;

[0037] Mechanical parameters include: compressive strength, flexural strength, shear strength, impact strength, elastic modulus, deformation modulus, Poisson's ratio, etc., but are not limited to the above parameters.

[0038] In general, all experiments and all parameters should be covered as much as possible. However, due to the limitations of human and material resources, major experiments and key parameters can be selected for testing based on actual application conditions.

[0039] The mechanical parameter data of the above three experimental data directly reflect the elasticity, plasticity, toughness, rheology, brittleness and other mechanical properties of rocks in the geological process and under various stresses. They are important indicators for studying the deformation, destruction and change of rocks under stress. They play an important role in geological engineering such as tunnel engineering, retaining wall engineering, mining exploration, crushing machinery, geological disaster prediction and other aspects. Basic physical and chemical property parameters can reflect the stability and bearing capacity of rocks under external effects (such as hydraulics, temperature fields, and stress) to a certain extent. Micro-structural parameters can reveal the internal characteristics and mineral composition of rocks. These data can predict some basic physical and chemical properties and mechanical properties of rocks to a large extent.

[0040] Some basic physical and chemical properties and microstructure parameters can reflect the mechanical parameters of rocks and can be obtained more quickly and easily than mechanical parameters, which has important application value.

[0041] S2. Collect typical geological disaster data that have occurred in history and process them to form standardized data on historical disasters.

[0042] The geological disaster data include at least rock-related data (such as rock type, mechanical strength, cracks, etc.) that cause disasters, hydroclimatic data (such as groundwater, rainstorms, climate environment, human activities, etc.), geological structures (such as folds, joints, faults) and topographic data (slopes, side slopes, steep slopes), etc.

[0043] The formation mechanism of geological disasters is complex. There are many factors that cause geological disasters. For example, the main conditions for the formation of landslides are:

[0044] (1) Topographic conditions: slope (10-45 degrees);

[0045] (2) Lithological conditions: rock and soil with loose structure and low shear strength;

[0046] (3) Structural conditions: Landslides are more likely to occur when there are structural surfaces such as joints and faults;

[0047] (4) Groundwater: can reduce shear strength;

[0048] (5) Triggering conditions: earthquakes, heavy rains, and unreasonable human activities (such as excavation at the foot of the slope).

[0049] The main conditions for collapse are:

[0050] (1) Topographic conditions: steep terrain (greater than 45 degrees);

[0051] (2) Lithological conditions: The rock is hard;

[0052] (3) Structural conditions: with joints, faults and other structural surfaces;

[0053] (4) Climate conditions: large daily and annual temperature ranges;

[0054] (5) Trigger conditions: earthquakes, heavy rains, melting ice and snow, etc.

[0055] Historical geological disaster data can be used to analyze the causes, development trends, impacts, and the role of various factors such as rock properties in geological disasters. This plays an important role in predicting and preventing future geological disasters and guiding human activities.

[0056] S3. Collect current real-time dynamic monitoring data on geological disasters.

[0057] Real-time dynamic monitoring data includes at least satellite remote sensing data, aerial remote sensing data, and data collected by ground and underground monitoring equipment, namely, sky-air-ground-depth real-time dynamic monitoring data.

[0058] Satellite remote sensing data and aerial remote sensing data can respectively reflect the current surface information from the far end and near end, and identify the sensitive areas of geological disasters. The data collected by ground and underground monitoring equipment, such as the internal observation data of the slope, can reflect the current stratum lithology and internal geological information. The above three types of dynamic monitoring data reflect the current topography, stratum lithology, geological structure and other information in an all-round, multi-level and three-dimensional manner, which plays an important role in the early identification, early warning and prevention of geological disasters.

[0059] S4. The above three heterogeneous data sources are integrated, analyzed and standardized to form a digital rock gene library.

[0060] Modern methods such as artificial intelligence, machine learning, and deep learning can be used to extract useful information data from the above-mentioned standard multi-source data, form a digital rock gene library, realize data sharing and services, and provide the best data platform for geological disaster prediction and human activity guidance.

[0061] Through the digital rock gene bank, we can not only query the relevant data information of different rocks, historical geological disaster data and dynamic monitoring data, but also analyze the rock gene information, extract useful data, and establish a rock property prediction model, so as to provide a technical basis for geological disaster prediction and early warning, and guide human activities. Specifically:

[0062] When the digital rock gene library is used, firstly, the rock standard detection items are determined according to the digital rock gene library, and the relationship model between the parameter values ​​of the rock standard detection items and the values ​​of other rock parameters is established.

[0063] As mentioned above, the digital rock gene bank includes a large amount of relevant experimental data information of typical rock samples, and some parameters have a certain correspondence with other parameters, especially the mineral composition and micro-structure parameters of rocks have an important influence on the mechanical properties of rocks. Through modern data processing methods, useful data can be extracted from the digital rock gene bank information, rock standard detection items can be determined, and a relationship model between the parameter values ​​of rock standard detection items and other rock parameter values ​​can be established. For newly collected rock samples to be tested (such as geological disaster warning for a certain mountain, or survey work for a certain mining area, local rock samples are collected), the rock samples to be tested are tested according to the rock standard detection items, and the test results are substituted into the relationship model to obtain other parameter information of the rock samples to be tested, without the need to conduct comprehensive experiments on the rock samples, which significantly saves experimental time and cost. After obtaining the other parameter information data of the rock, the rock properties of the collection area can be basically mastered, which provides an important basis for geological disaster prediction, early warning and human activity guidance in the rock sample collection area.

[0064] Standard rock testing items should preferably be representative items that can be tested by rapid non-destructive testing methods. For example, standard rock testing items may include: rock density testing, infrared spectrum analysis, transmission electron microscopy observation, mineral composition analysis, etc. Of course, destructive testing methods may also be used when necessary, such as rock fracturing experiments.

[0065] After the initial construction of the digital rock gene library, as time goes by, real-time dynamic monitoring data and historical geological disaster data are constantly updated, and the digital rock gene library and relationship model also need to be continuously iterated and updated.

[0066] If the test results of the rock sample to be tested cannot correspond to the relationship model, then the relevant experimental research will be carried out on the rock sample to be tested, just like the typical rock sample, to obtain the relevant experimental data of the rock sample to be tested, and further update the rock standardization data, digital rock gene library and relationship model.

[0067] The digital rock gene bank of this application includes three types of heterogeneous data, among which the relevant experimental data of typical rock sample descriptions directly reveal the relationship between the standard test items of rocks and other parameters. However, there are many types of rock property parameters, and other parameters reflected by a standard test item are also limited. By testing a few items, it is impossible to reflect all the properties of rocks. Through historical geological disaster data, it is possible to analyze which specific rock properties have caused the corresponding geological disasters. Therefore, for different geological disasters, different rock standard test items can be determined to reflect the rock properties that have the greatest impact on the geological disaster. For example, for landslides, when determining the standard rock test items, you can choose a test item that can reflect the shear resistance of the rock. For collapse, when determining the standard rock test items, you can choose a test item that can reflect the hardness of the rock. This can not only improve the detection efficiency, but also accurately obtain the key properties of the rock. However, historical disaster data can only reflect historical geological disaster information, but cannot reflect current geological information. Moreover, geological disasters are somewhat sporadic and intermittent. Historical disaster data are incomplete in both time and space. The rock properties and other conditions that cause current geological disasters may have changed significantly. By collecting real-time dynamic monitoring data, we can understand the relationship between current rock properties and other conditions and geological disasters, make up for the lack of timeliness of historical disaster data, and improve the reliability and applicability of newly collected rock sample testing.

[0068] Therefore, this application uses three heterogeneous data sources to construct a digital rock gene library, which makes up for the defects of incomplete data and single methods in the existing rock mechanics field, resulting in different analysis results. It has strong broad spectrum and high accuracy. Based on this gene library, a small number of items can be quickly detected, and other basic properties of rocks can be analyzed and predicted, thus providing a technical basis for geological disaster prediction and early warning and guiding human activities.

[0069] The digital rock gene bank can share data and services among multiple units and multiple platforms, realizing "big system, big platform, big data, and big integration" under the cloud architecture, breaking the data gap between units, integrating various rock information services from various units, and forming a unified, orderly, large-scale, and authoritative unified information service platform.

[0070] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for constructing a digital rock gene library based on massive experimental data, characterized in that: The following steps are involved: Collect different typical rock samples, conduct relevant experimental research, obtain relevant experimental data of the typical rock samples, and process the relevant experimental data to form rock standardization data; Collect typical geological disaster data that occurred in history and process them to form standardized data of historical disasters; Collect current real-time dynamic monitoring data of geological disasters; The rock standardized data, the historical disaster standardized data and the real-time dynamic monitoring data are fused and analyzed and uniformly standardized to form a digital rock gene library.

2. The method for constructing a digital rock gene library based on massive experimental data according to claim 1, characterized in that: The relevant experimental research includes at least one of optical microscope observation, scanning electron microscope observation, transmission electron microscope observation, X-ray diffraction analysis, CT scanning analysis, magnetic resonance imaging analysis, infrared spectrum analysis, relevant physical and chemical property detection experiments, sound wave experiments and relevant mechanical experiments.

3. The method for constructing a digital rock gene library based on massive experimental data according to claim 2, characterized in that: The relevant experimental data at least include microstructure parameters, basic physical and chemical property parameters and mechanical parameter data of the rock sample; The microstructure parameters include at least one of microscopic crystal structure, grain boundary distribution, mineral particle morphology, mineral particle size, porosity, crack space distribution, crack number, crack length and crack area; The basic physical and chemical property parameters include at least one of color, gloss, transparency, water content, water absorption, density, specific gravity, mineral type and composition, permeability, swelling, solubility and frost resistance; The mechanical parameters include: at least one of compressive strength, flexural strength, shear strength, impact strength, elastic modulus, deformation modulus and Poisson's ratio.

4. The method for constructing a digital rock gene library based on massive experimental data according to claim 1, characterized in that: The geological disaster data at least includes rock-related data that cause the disaster, hydrological and climate data, geological structure data, and topographic data.

5. The method for constructing a digital rock gene library based on massive experimental data according to claim 1, characterized in that: The real-time dynamic monitoring data at least includes satellite remote sensing data, aerial remote sensing data, and data collected by ground and underground monitoring equipment.

6. The method for constructing a digital rock gene library based on massive experimental data according to claim 1, characterized in that: The formation rock normalization data further includes: Collect relevant data information of public and unpublished rock samples, and merge them with the relevant experimental data to form the rock standardization data.

7. A digital rock gene library constructed based on the method according to any one of claims 1 to 6.

8. An application method based on the digital rock gene library according to claim 7, characterized in that: The following steps are involved: According to the digital rock gene library, determine the rock standard detection items, and establish a relationship model between the parameter values ​​of the rock standard detection items and other rock parameter values; For newly collected rock samples to be tested, the rock samples to be tested are tested according to the rock standard testing items, and the test results are substituted into the relationship model to obtain other parameter information of the rock samples to be tested.

9. The method for applying the digital rock gene library according to claim 8, characterized in that: The determination of rock standard testing items further includes: Determine the standard rock testing items corresponding to different geological disasters.

10. An application method based on the digital rock gene library according to claim 8, characterized in that: If the test result of the rock sample to be tested cannot correspond to the relationship model, relevant experimental research is carried out on the rock sample to obtain relevant experimental data of the rock sample to be tested, and the rock standardization data, the digital rock gene library and the relationship model are further updated.

Images