CN116910886A - An impact energy early warning method and system in karst areas - Google Patents

Abstract

The application discloses a method and a system for early warning impact energy in karst areas, wherein the flow comprises the following steps: collecting historical engineering geological data of a mine to be mined; arranging exploration points based on historical engineering geological data, and constructing a building foundation type; constructing a simulated mine based on the building foundation pattern; performing preliminary prediction on the simulated mine to obtain a preliminary prediction result; and constructing a prediction model based on the preliminary prediction result to obtain a final prediction result, and carrying out early warning on impact energy based on the final prediction result. According to the application, through simulating the change of the geological structure and the physical property, the release and accumulation process of impact energy is comprehensively mastered, and the geological disaster risk is accurately judged. Through data analysis and processing, the information originally needed to be monitored is converted into a feasible early warning standard, the early warning information is issued in time, and the early warning accuracy of geological disasters is improved.

Description

An impact energy early warning method and system in karst areas

Technical field

This application relates to the field of impact energy early warning, and specifically relates to an impact energy early warning method and system in karst areas.

Background technique

Karst areas refer to areas rich in caves and dissolved forms in the strata. Due to the special structure and mineral composition of rock layers, such areas are prone to various geological disasters, such as ground collapse, sudden outflow of groundwater, earthquakes, etc. Most of the current geological disaster monitoring methods use traditional geological exploration methods and lack timely and accurate impact energy early warning methods, which brings difficulties to effectively prevent and mitigate disasters in karst areas.

Contents of the invention

This application simulates changes in the geological structure and physical properties of karst areas and identifies the release and accumulation process of impact energy, thereby providing early warning of geological disasters in karst areas.

In order to achieve the above purpose, this application provides an impact energy early warning method in karst areas. The process includes:

Collect historical engineering geological data of mines to be mined;

Based on the historical engineering geological data, arrange exploration points and propose building foundation types;

Based on the building foundation type, construct a simulated mine;

Make preliminary predictions on the simulated mine and obtain preliminary prediction results;

Based on the preliminary prediction results, a prediction model is constructed to obtain the final prediction results, and an early warning of impact energy is performed based on the final prediction results.

Preferably, the proposed method for constructing the building foundation type includes:

Through the historical engineering geological data, obtain historical data on local control of karst, soil caves and collapse;

Based on the historical data, the load magnitude of the building foundation type is determined.

Preferably, when the mine conditions are complex, exploration points are arranged on each independent foundation, and the exploration points are arranged column by column in a column-by-pile manner.

Preferably, the method of constructing the simulated mine includes: establishing a mesoscopic particle flow model based on the exploration results of the exploration points, and constructing the simulated mine.

Preferably, the method for obtaining the prediction result includes: using an empirical analogy method and a microseismic monitoring method to obtain the prediction result.

Preferably, the method of constructing the prediction model includes: using a random forest model to construct the prediction model based on the prediction results.

This application also provides an impact energy early warning system for karst areas, including: a collection module, a simulation module, a calibration module, a prediction module and a building module;

The collection module is used to collect historical engineering geological data of the mine to be mined;

The simulation module is used to arrange exploration points and the foundation type of the proposed building based on the historical engineering geological data;

The calibration module is used to construct a simulated mine based on the building foundation type;

The prediction module is used to make preliminary predictions on the simulated mine and obtain preliminary prediction results;

The building module is used to construct a prediction model based on the preliminary prediction results, obtain the final prediction results, and provide early warning of impact energy based on the final prediction results.

Preferably, the impact energy early warning system in karst areas is characterized in that the workflow of the simulation module includes:

Through the historical engineering geological data mentioned above, we can understand the historical data of local control of karst, soil caves and collapse;

Based on the historical data, the load magnitude of the building foundation type is determined.

Compared with the existing technology, the beneficial effects of this application are as follows:

This application simulates changes in geological structure and physical properties to comprehensively grasp the release and accumulation process of impact energy and accurately determine geological disaster risks. Through data analysis and processing, the originally required monitoring information is transformed into feasible early warning standards, early warning information is released in a timely manner, and the accuracy of early warning of geological disasters is improved.

Description of the drawings

In order to explain the technical solutions of the present application more clearly, the drawings required to be used in the embodiments are briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For ordinary people in the art, Technical personnel can also obtain other drawings based on these drawings without exerting creative labor.

Figure 1 is a schematic flow chart of the method of this application;

Figure 2 is a schematic diagram of the system structure of this application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

In order to make the above objects, features and advantages of the present application more obvious and understandable, the present application will be described in further detail below in conjunction with the accompanying drawings and specific implementation modes.

Embodiment 1

In geological research, the early warning of impact energy can essentially be attributed to the prediction of impact ground pressure. Therefore, this embodiment uses the method of predicting impact pressure to provide early warning of impact energy.

As shown in Figure 1, it is a schematic flow chart of the method in this embodiment. The process includes:

S1. Collect historical engineering geological data of the mine to be mined.

Karst areas are so-called karst landforms. Karst landforms are dominated by dissolution and also include mechanical erosion processes such as water erosion, submerged erosion, and collapse. Therefore, due to its particularity, the conventional method cannot be used to directly arrange survey points, which is prone to danger. Therefore, it is necessary to collect engineering geological data around the proposed site (i.e., historical engineering geological data) first.

S2. Based on historical engineering geological data, arrange exploration points and propose building foundation types.

Before surveying karst areas, engineering geological data around the proposed site should be fully collected, and historical data on local control of karst, soil caves and collapse should be obtained. When arranging the survey workload in karst areas, attention should be paid to the following issues:

Proposed building foundation type

In order to understand the load size of the proposed building, exploration points should be arranged according to the proposed foundation type. When conditions are complex, exploration points should be arranged on each independent foundation. When one column and one pile foundation is used, exploration holes should be arranged column by column.

Effects of karst stability

The layout of survey work should be diversified in methods and the process should be dynamic. The development of mountainous underground karst has the characteristics of concealment, diversity and large changes in spatial location. When arranging the workload, attention should be paid to using a combination of multiple methods.

For shallowly buried covered and exposed karst foundations, methods such as well exploration and trench exploration can be appropriately used to identify the shape of shallow karst caves. For buried karst foundations, technicians should obtain rock and soil layer information in a timely manner during the survey process. If there is a site with serious karst development, geotechnical testing, static exploration, geophysical exploration and other means should be added according to the actual situation to further identify Buried locations in loose fissures and hidden soil caves. At the same time, the depth of the exploration point should also be appropriately increased or decreased according to the actual situation to meet the needs of foundation design and stability verification.

S3. Construct a simulated mine based on the building foundation type.

Afterwards, based on the results obtained from the exploration points, a parallel bonding model was used to establish a uniaxial compression particle flow numerical model (simulated mine) of standard rock samples around the mine. The mesoscopic mechanical parameters were obtained through a trial and error method, and the calibration results were obtained. In this embodiment, the calibration results can be divided into: rock formation impact tendency and stope support stress zone characteristics. By predicting these two mesoscopic parameters, subsequent prediction results are obtained.

S4. Make preliminary predictions on the simulated mine and obtain preliminary prediction results.

Estimating rock formation impact tendency using empirical analogy method

The method of making regular summaries of past experiences and lessons and using them to guide the safe mining of percussion coal seams in this mine or other mines with similar conditions is called the experience analogy method.

In this embodiment, the following factors should be considered: the current status and development trend of rockbursts in this mine and adjacent mines; rockbursts have occurred in this coal seam or adjacent layers and adjacent areas; the old roof of the coal seam is more than 5m thick and single. Hard rock formations with axial compressive strength greater than 70MPa; island-shaped or peninsula-shaped coal pillars; support pressure affected areas; upper or lower remaining coal pillars or mining boundaries, areas with sudden changes in coal seam thickness or dip angle; fold or fault structural zones, etc.

As for the entire mine, through the analysis of geological conditions and mining technology conditions, the area where rockburst may occur can be delineated. As the mining depth increases, rockburst may occur when the stress of the coal and rock mass meets the strength condition. The depth of the initial impact is usually called the critical depth. From the initial depth, rockbursts may occur in coal pillars, coal seam protrusions, and upper and lower coal seam sections adjacent to coal pillars. As the mining level increases, the location and scope of rockbursts also expand. . All areas close to the mining face, areas with sudden changes in coal seam thickness and inclination, and geological structural zones may become dangerous areas for rockbursts.

Using microseismic monitoring method to estimate the characteristics of stope support stress zone

Use short-period seismometers to monitor and record the number of shocks above magnitude 0.5 and the energy released by the impact ground pressure. Use this trend prediction to predict the trend of recent geoburst occurrence and stress release. Until the positioning system is built, current seismometers will be used for current monitoring. Through the above process, preliminary prediction results are obtained.

After preliminary prediction, combined with the particularity of karst landforms, the above-mentioned critical depth should not exceed 668m; the advanced support pressure concentration range of the working face is 5 to 35m, and the stress concentration coefficient is 2.5, but the impact range of the advanced pressure of the upper conglomerate layer reaches 120m. Therefore, the concentrated stress of mining on the working face has a more obvious impact on the working face. During layered mining, the cyclic pressure intensity of the upper layered working face reaches a maximum of 510kn/m ² , and the pressure is relatively strong.

S5. Based on the preliminary prediction results, build a prediction model, obtain the final prediction results, and provide early warning of impact energy based on the final prediction results.

A random forest model is used to build a prediction model based on the preliminary prediction results.

First, collect a sample set composed of the above preliminary prediction results. After that, the mean substitution method is used to clean the collected sample set; in this embodiment, the main steps are to clean missing values, clean content inconsistent with the original data type, clean unnecessary data, and clean logical error data.

After that, feature data are selected from the sample set to form a feature set, and a random forest model is used based on the established feature set and the bootstrap method is used to randomly select m samples from the M samples in the sample set as a sub-training set to build a decision tree; Use the same method to simultaneously construct T (T>1) decision trees; randomly select p features from the n features in the feature set as a subset of node splits, and select the feature with the smallest error among the p features as the subset based on the square error. The node split feature keeps the node split until the decision tree cannot be split anymore; in this embodiment, the three features are the root node and the content node, and each prediction result is the output, which is a leaf node. Split T decision trees to form a random forest; based on the preliminary prediction results, each split decision tree is randomly trained on M samples to obtain the prediction results corresponding to each decision tree; the prediction results corresponding to each decision tree are The prediction results are averaged to obtain the final prediction result.

Finally, an early warning of impact energy is carried out based on the final prediction results.

Embodiment 2

As shown in Figure 2, it is a schematic diagram of the system structure of this embodiment, including: an acquisition module, a simulation module, a calibration module, a prediction module and a construction module. Among them, the acquisition module is used to collect historical engineering geological data of the mines to be mined; the simulation module is used to arrange exploration points and propose building foundation types based on historical engineering geological data; the calibration module is used to construct simulated mines based on the building foundation types. ; The prediction module is used to make preliminary predictions for the simulated mine and obtain preliminary prediction results; the construction module is used to build a prediction model based on the preliminary prediction results, obtain the final prediction results, and provide early warning of impact energy based on the final prediction results.

How this application solves technical problems in real life will be described in detail below in conjunction with this embodiment.

First, use the acquisition module to collect historical engineering geological data of the mine to be mined.

Karst areas are so-called karst landforms. Karst landforms are dominated by dissolution and also include mechanical erosion processes such as water erosion, submerged erosion, and collapse. Therefore, due to its particularity, the conventional method cannot be used to directly arrange survey points, which is prone to danger. Therefore, it is necessary to collect engineering geological data around the proposed site (i.e., historical engineering geological data) first.

After that, the simulation module arranges exploration points and proposes building foundation types based on historical engineering geological data.

Before surveying karst areas, engineering geological data around the proposed site should be fully collected, and historical data on local control of karst, soil caves and collapse should be obtained. When arranging the survey workload in karst areas, attention should be paid to the following issues:

Proposed building foundation type

In order to understand the load size of the proposed building, exploration points should be arranged according to the proposed foundation type. When conditions are complex, exploration points should be arranged on each independent foundation. When one column and one pile foundation is used, exploration holes should be arranged column by column.

Effects of karst stability

The layout of survey work should be diversified in methods and the process should be dynamic. The development of mountainous underground karst has the characteristics of concealment, diversity and large changes in spatial location. When arranging the workload, attention should be paid to using a combination of multiple methods.

For shallowly buried covered and exposed karst foundations, methods such as well exploration and trench exploration can be appropriately used to identify the shape of shallow karst caves. For buried karst foundations, technicians should obtain rock and soil layer information in a timely manner during the survey process. If there is a site with serious karst development, geotechnical testing, static exploration, geophysical exploration and other means should be added according to the actual situation to further identify Buried locations in loose fissures and hidden soil caves. At the same time, the depth of the exploration point should also be appropriately increased or decreased according to the actual situation to meet the needs of foundation design and stability verification.

The calibration module builds a simulated mine based on the basic building type.

Afterwards, based on the results obtained from the exploration points, a parallel bonding model was used to establish a uniaxial compression particle flow numerical model (simulated mine) of standard rock samples around the mine. The mesoscopic mechanical parameters were obtained through a trial and error method, and the calibration results were obtained. In this embodiment, the calibration results can be divided into: rock formation impact tendency and stope support stress zone characteristics. By predicting these two mesoscopic parameters, subsequent prediction results are obtained.

The prediction module makes preliminary predictions for the simulated mine and obtains preliminary prediction results.

Estimating rock formation impact tendency using empirical analogy method

The method of making regular summaries of past experiences and lessons and using them to guide the safe mining of percussion coal seams in this mine or other mines with similar conditions is called the experience analogy method.

In this embodiment, the following factors should be considered: the current status and development trend of rockbursts in this mine and adjacent mines; rockbursts have occurred in this coal seam or adjacent layers and adjacent areas; the old roof of the coal seam is more than 5m thick and single. Hard rock formations with axial compressive strength greater than 70MPa; island-shaped or peninsula-shaped coal pillars; support pressure affected areas; upper or lower remaining coal pillars or mining boundaries, areas with sudden changes in coal seam thickness or dip angle; fold or fault structural zones, etc.

As for the entire mine, through the analysis of geological conditions and mining technology conditions, the area where rockburst may occur can be delineated. As the mining depth increases, rockburst may occur when the stress of the coal and rock mass meets the strength condition. The depth of the initial impact is usually called the critical depth. From the initial depth, rockbursts may occur in coal pillars, coal seam protrusions, and upper and lower coal seam sections adjacent to coal pillars. As the mining level increases, the location and scope of rockbursts also expand. . All areas close to the mining face, areas with sudden changes in coal seam thickness and inclination, and geological structural zones may become dangerous areas for rockbursts.

Using microseismic monitoring method to estimate the characteristics of stope support stress zone

Use short-period seismometers to monitor and record the number of shocks above magnitude 0.5 and the energy released by the impact ground pressure. Use this trend prediction to predict the trend of recent geoburst occurrence and stress release. Until the positioning system is built, current seismometers will be used for current monitoring. Through the above process, preliminary prediction results are obtained.

After preliminary prediction, combined with the particularity of karst landforms, the above-mentioned critical depth should not exceed 668m; the advanced support pressure concentration range of the working face is 5 to 35m, and the stress concentration coefficient is 2.5, but the impact range of the advanced pressure of the upper conglomerate layer reaches 120m. Therefore, the concentrated stress of mining on the working face has a more obvious impact on the working face. During layered mining, the cyclic pressure intensity of the upper layered working face reaches a maximum of 510kn/m ² , and the pressure is relatively strong.

Finally, the building module builds a prediction model based on the preliminary prediction results, obtains the final prediction results, and provides early warning of impact energy based on the final prediction results.

A random forest model is used to build a prediction model based on the preliminary prediction results.

First, collect a sample set composed of the above preliminary prediction results. After that, the mean substitution method is used to clean the collected sample set; in this embodiment, the main steps are to clean missing values, clean content inconsistent with the original data type, clean unnecessary data, and clean logical error data.

After that, feature data are selected from the sample set to form a feature set, and a random forest model is used based on the established feature set and the bootstrap method is used to randomly select m samples from the M samples in the sample set as a sub-training set to build a decision tree; Use the same method to simultaneously construct T (T>1) decision trees; randomly select p features from the n features in the feature set as a subset of node splits, and select the feature with the smallest error among the p features as the subset based on the square error. The node split feature keeps the node split until the decision tree cannot be split anymore; in this embodiment, the three features are the root node and the content node, and each prediction result is the output, which is a leaf node. Split T decision trees to form a random forest; based on the preliminary prediction results, each split decision tree is randomly trained on M samples to obtain the prediction results corresponding to each decision tree; the prediction results corresponding to each decision tree are The prediction results are averaged to obtain the final prediction result.

Finally, an early warning of impact energy is carried out based on the final prediction results.

The above-described embodiments are only descriptions of preferred modes of the present application and do not limit the scope of the present application. Without departing from the design spirit of the present application, those of ordinary skill in the art may make various modifications to the technical solutions of the present application. All modifications and improvements shall fall within the protection scope determined by the claims of this application.

Claims (8)

1. An impact energy early warning method in karst areas, characterized in that the process includes: Collect historical engineering geological data of mines to be mined; Based on the historical engineering geological data, arrange exploration points and propose building foundation types; Based on the building foundation type, construct a simulated mine; Make preliminary predictions on the simulated mine and obtain preliminary prediction results; Based on the preliminary prediction results, a prediction model is constructed to obtain the final prediction results, and an early warning of impact energy is performed based on the final prediction results. 2. The impact energy early warning method in karst areas according to claim 1, characterized in that the proposed method for constructing the building foundation type includes: Through the historical engineering geological data, obtain historical data on local control of karst, soil caves and collapse; Based on the historical data, the load magnitude of the building foundation type is determined. 3. The impact energy early warning method in karst areas according to claim 2, characterized in that when the mine conditions are complex, exploration points are arranged on each independent foundation, and the exploration points are arranged column by column in a column-by-pile manner. 4. The impact energy early warning method in karst areas according to claim 1, characterized in that the method of constructing the simulated mine includes: establishing a mesoscopic particle flow model based on the exploration results of the exploration point, and constructing the simulated mine. 5. The impact energy early warning method in karst areas according to claim 1, characterized in that the method for obtaining the prediction result includes: obtaining the prediction result using an empirical analogy method and a microseismic monitoring method. 6. The impact energy early warning method in karst areas according to claim 1, characterized in that the method of constructing the prediction model includes: based on the prediction results, using a random forest model to construct the prediction model. 7. An impact energy early warning system in karst areas, characterized by including: a collection module, a simulation module, a calibration module, a prediction module and a construction module; The collection module is used to collect historical engineering geological data of the mine to be mined; The simulation module is used to arrange exploration points and the foundation type of the proposed building based on the historical engineering geological data; The calibration module is used to construct a simulated mine based on the building foundation type; The prediction module is used to make preliminary predictions on the simulated mine and obtain preliminary prediction results; The building module is used to construct a prediction model based on the preliminary prediction results, obtain the final prediction results, and perform early warning of impact energy based on the final prediction results. 8. The impact energy early warning system for karst areas according to claim 7, characterized in that the workflow of the simulation module includes: Through the historical engineering geological data mentioned above, we can understand the historical data of local control of karst, soil caves and collapse; Based on the historical data, the load magnitude of the building foundation type is determined.