Intelligent recognition and prediction system and method for submarine roadway water inrush structure

Abstract

The invention discloses an intelligent recognition and prediction system and method for a submarine roadway water inrush structure. According to the method, the conditions of the roadway, the structural plane features, the overlying strata structure, the regional structure and the mining influence are organically combined, various data are applied, coding is carried out according to a geologic structure model and a feature data chain, and accurate, rapid and efficient judgment and prediction of roadway water inrush are achieved through a system algorithm model, coding and deductive operation. The method not only overcomes the technical bottlenecks of large workload, low survey precision, possible induction of water inrush and the like of the existing roadway water inrush prediction in the aspects of monitoring and geological survey, but also can realize the water inrush evaluation, prediction and forecast of the roadway by means of real-time monitoring, stratum geology and other characteristic data without quantitative calculation and analysis, and has a wide application prospect. The method realizes efficient fusion utilization and intelligent prediction of test, geological data and mining data, and has the advantages of simple operation, high automation and intelligence degree and high identification precision.

Description

Technical Field

[0001] The present invention belongs to the technical field of mining, geology and disaster prevention and reduction, and specifically relates to an intelligent identification and prediction system for water inrush structures in submarine tunnels, and also relates to an intelligent identification and prediction method for water inrush structures in submarine tunnels. Background Art

[0002] In nature, many mineral resources are located under the seabed and other surface water bodies. When mining operations are carried out under these large water bodies, if a water inrush accident occurs, the mine will be flooded instantly, leading to catastrophic consequences. Therefore, the detection and evaluation of seabed water inrush structures is particularly important.

[0003] However, the existence of water bodies poses a huge challenge to geological exploration. Currently, it mainly relies on drilling and geophysical exploration. In actual operation, drilling technology may penetrate fractured rock mass, bring safety hazards and even directly induce water inrush accidents. In addition, the drilling results are affected by the exploration network, and it is difficult to fully and accurately grasp the detailed distribution and development of underwater geological structures and small faults. Although ground geophysical detection is non-destructive detection, it is greatly disturbed by human engineering activities, with poor signals and low signal-to-noise ratio, resulting in low interpretation accuracy, large errors, and insufficient detection accuracy and reliability.

[0004] On the other hand, whether it is drilling or geophysical exploration, due to the existence of large surface water bodies, the testing methods are limited, the construction of the survey platform is complex, the testing cycle is long, the efficiency is low, the investment is large, and the effect is poor. Therefore, innovating testing technology methods and improving the accuracy of identifying geological structures of water inrush have become key problems in the prediction and prevention of underground water inrush disasters.

[0005] At present, the research on water inrush disasters mainly focuses on advance detection and prediction in underground coal mines and tunnels. The technical methods are mainly various geophysical methods, including electrical method, magnetic method, electromagnetic method, seismic wave method, sound wave and infrared detection. These methods use the reflection time of electric current or electromagnetic wave on the structural surface to delineate weak structural surfaces or water-bearing bodies. However, electrical and magnetic detection are essentially the same. They are based on the differences in the electromagnetic properties of the constituent materials of the geological body. The detection profile and its distribution of resistivity are obtained through electromagnetic transformation, and then the weak structure and water-bearing area / zone are identified in combination with geological analysis.

[0006] In terms of submarine water inrush prediction, in addition to the above-mentioned geophysical methods, water chemical composition determination and isotope tracing methods are also widely used to identify the source of mine water inrush. Infrared detection technology is also applicable to tunnel water seepage detection. However, the electromagnetic field interference of the underground mining space itself, as well as factors such as the mine development system and the transportation and lifting system, make the underground space structure extremely complex and seriously disturb the electromagnetic field. When detecting in limited underground tunnels, the geophysical method is interfered by the earth's electromagnetic field, the signal is inaccurate, and the signal-to-noise ratio is low. In complex spatial structure strata, it is difficult to bypass or pass through weak structures such as complex fractured rock masses and aquifers, resulting in the failure to receive reflected signals. In addition, when the transmitted electromagnetic wave signal passes through the seabed strata into the seawater area, the reflected signal cannot be received. Therefore, the existing geophysical methods have difficult-to-overcome theoretical and technical bottlenecks in both ground and underground detection. Summary of the invention

[0007] The present invention proposes a submarine tunnel water inrush structure intelligent identification and prediction system and method, the purpose of which is to overcome the limitations of traditional water inrush structure prediction methods when dealing with low signal ratio data problems.

[0008] The technical solution of the present invention is as follows: An intelligent identification and prediction system for water inrush structures in submarine tunnels, comprising a data acquisition unit, a water inrush structure deduction unit and a water inrush identification and prediction unit; The data acquisition unit is used to collect data on the tunnel structure surface and rock layer distribution; The water bursting structure deduction unit includes a digital image pickup device, a digital strain gauge, a data integration module and a computer, which is used to receive and process the data collected by the data acquisition unit and generate a characteristic data chain; The water inrush identification and prediction unit extracts the characteristic factors of the roadway according to the roadway water inrush factor characteristic database, and realizes water inrush identification and prediction by comparing the characteristic data chain of the roadway to be tested with the standard factor sequence of the water inrush characteristic database.

[0009] As a further improvement of the intelligent identification and prediction system for water inrush structures in submarine tunnels: the data acquisition unit collects data on the strata, lithology, structure, hydrology and mining data of the tunnel location and area, including data from early geological exploration, current status data revealed by the tunnel, and mining dynamics data.

[0010] As a further improvement of the intelligent identification and prediction system for water inrush structures in submarine tunnels, the data acquisition unit may collect the following contents: Tunnel location information: refers to the middle level of the tunnel, representing the tunnel depth and its associated underground facilities, indicating its role and status in the mine system; Plan layout: including the direction and length of the roadway and its relationship with the horizontal stope, chamber facilities and location; Lithology: refers to the distribution of lithology and its characteristic data along the axis of the roadway; Structural surface characteristics: refers to the distribution of main controlling structural surfaces such as faults, joints and fissures along the tunnel axis on the tunnel vault and wall, as well as the geometric structural characteristic data in terms of occurrence and scale; Rock structure: refers to the distribution of rock layers overlying the tunnel, including the distribution, lithology and thickness of rock layers from the tunnel to the seabed; Regional geological structure: including the type, size, distribution density, location of faults, whether the faults run through the tunnels, the positional relationship between the faults and the tunnels, and the regional earthquake intensity, magnitude and acceleration; Tunnel hydrology: including the distribution of water seepage points in the tunnel, the amount of water seepage and its chloride ion content, as well as the distribution and connectivity of water bodies from the seabed to the tunnel bottom; Mining disturbance data: including the maximum exposed area at the tunnel level and above the tunnel top and the maximum single-stage blasting charge.

[0011] As a further improvement of the intelligent identification and prediction system for submarine tunnel water inrush structures: the digital image pickup device is used to receive the digital image generated by the data acquisition unit and extract 3D scanning data of the tunnel structure surface, including shape, size and hydrological data.

[0012] As a further improvement of the intelligent identification and prediction system for water inrush structures in submarine tunnels: the digital strain gauge is a non-contact deformation test, which is used to obtain digital images from the digital image pickup device module and record hydrological data; when the monitoring signal is an electrical parameter, the monitoring signal is amplified by the amplifier and then enters the instrument, and then sent to the data integration module.

[0013] As a further improvement of the intelligent identification and prediction system for water inrush structures in submarine tunnels: the data integration module is used to analyze and integrate the input data, and then upload it to the identification and prediction unit.

[0014] As a further improvement of the intelligent identification and prediction system for submarine tunnel water inrush structures: the water inrush structure deduction unit also runs a deduction algorithm, which uses the research on the influence of tunnel structural surface characteristics and statistical analysis of literature case data to determine the characteristic factors that have a significant impact on the evolution of the structural surface from the many factors affecting the water inrush structure, and then extracts the characteristic parameters that affect the evolution of the structural surface from the characteristic database, and constructs a characteristic data chain based on certain principles.

[0015] As a further improvement of the intelligent identification and prediction system for water inrush structures in submarine tunnels, the process of the deductive algorithm is as follows: (1) Determine characteristic factors: Through the research on the influence of tunnel structural surface characteristics and statistical analysis of literature case data, determine the characteristic factors that have a significant impact on the evolution of structural surfaces from the factors affecting the water inrush structure; (2) Constructing a feature data chain based on the feature factors: The feature parameters in the currently acquired data that match the feature factors are encoded into a feature data chain according to the following rules: the value corresponding to each feature factor is 0 or 1, which is determined by the threshold of the feature factor; and finally a code sequence consisting of 0 and 1 is obtained as the feature data chain.

[0016] As a further improvement of the intelligent identification and prediction system for water inrush structures in submarine tunnels, the processing process of the water inrush identification and prediction unit is as follows: (1) Feature extraction: Extract the water inrush factor features from the feature data chain of the roadway to be tested based on the roadway water inrush factor feature database; (2) Identification, prediction and forecasting: The extracted outburst factor features are compared with the standard factor sequence of the water inrush feature library. If they are consistent, it is judged that water inrush has occurred, and the evaluation and prediction of whether the tunnel structure surface has water inrush is completed; (3) Emergency response: Based on the hazard level of the identified prediction results, the emergency plan is activated according to the level.

[0017] The present invention also provides a method for intelligently identifying and predicting water inrush structures in submarine tunnels, which identifies and predicts water inrush based on the above-mentioned intelligent identification and prediction system for water inrush structures in submarine tunnels.

[0018] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention uses a variety of data (tunnel structure surface testing, stratum geology, regional geological structure, seismic network, tunnel hydrology and mining data) and can accurately, quickly and efficiently realize the identification and prediction of tunnel water inrush through system algorithm models, coding and deductive operations. Based on tunnel characteristic data, the present invention does not require quantitative calculation and analysis. With the help of real-time monitoring and stratum geology and other characteristic data, it can realize the evaluation and prediction of tunnel water inrush, overcome the limitations of traditional water inrush structure prediction methods in dealing with low signal ratio data problems, and has the advantages of simple operation, high degree of automation and intelligence, and high identification accuracy.

[0019] 2. The present invention organically combines the tunnel's own conditions, structural surface characteristics, overlying rock structure, regional structure and mining impact, and identifies and predicts the tunnel water inrush structure based on the geological structure model and characteristic data chain coding. This method overcomes the technical bottlenecks of existing tunnel water inrush prediction in monitoring and geological survey, such as large workload, low survey accuracy and possible induction of water inrush, and realizes the efficient integration and intelligent prediction of test, geological data and mining data. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the system composition of the present invention.

[0021] Figure 2 Schematic diagram of the basic composition of submarine tunnel characteristic data in the embodiment.

[0022] Figure 3 Schematic diagram of the logical relationship between identification and prediction of water inrush in tunnels. DETAILED DESCRIPTION

[0023] The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments.

[0024] like Figure 1 , an intelligent identification and prediction system for water inrush structures in submarine tunnels, the system includes a data acquisition unit, a water inrush structure deduction unit and a water inrush identification and prediction unit.

[0025] 1. Data acquisition unit.

[0026] The data acquisition unit is used to collect data on the tunnel structure surface and rock layer distribution. The unit collects images of the tunnel structure surface data to provide basic data for identification and prediction of tunnel water inrush structure. Image acquisition is automatically collected according to the time set by the acquisition system.

[0027] The data collection unit collects data on the strata, lithology, structure, hydrology and mining of the tunnel location and area, including data from early geological exploration, current status data revealed by the tunnel and mining dynamics data.

[0028] like Figure 2 , the data collection unit's collection content specifically includes: 1. Tunnel location information: refers to the middle level of the tunnel, representing the tunnel depth and its associated underground facilities, indicating its role and status in the mine system. It can be automatically imported through the satellite positioning system to obtain, or manually input to obtain geographic coordinates, orientation or tunnel engineering location information including the middle level.

[0029] 2. Plan layout: including the direction and length of the tunnel and its relationship with the horizontal mining area, chamber facilities and location.

[0030] 3. Lithology: refers to the lithology distribution and its characteristic data along the axis of the tunnel.

[0031] 4. Structural surface characteristics: refers to the distribution of main controlling structural surfaces such as faults, joints and fissures along the tunnel axis in the tunnel vault and wall, as well as their geometric structural characteristic data in terms of occurrence and scale.

[0032] 5. Rock structure: refers to the distribution of rock layers overlying the tunnel, including the distribution, lithology and thickness of rock layers from the tunnel to the seabed, such as the Quaternary alluvial layer, weathering layer, basement rock layer and other data on the seabed.

[0033] 6. Regional geological structure: including the type, scale, distribution density, location of faults, whether the faults run through tunnels, the positional relationship between faults and tunnels, as well as regional earthquake intensity, magnitude and acceleration. Earthquake data comes from the seismic network.

[0034] 7. Tunnel hydrology: including the distribution of seepage points in the tunnel, the amount of seepage and its chloride ion content, as well as hydrogeological data such as the distribution and connectivity of water bodies from the seabed to the bottom of the tunnel.

[0035] 8. Mining disturbance data: including the maximum exposed area at the tunnel level and above the tunnel top and the maximum single-stage blasting charge.

[0036] The overlying rock strata, lithology and rock strata structure are the geological characteristics of the strata between the seabed and the measured target tunnel.

[0037] The data of the data acquisition unit includes real-time monitoring data and existing survey data. The survey data and real-time monitoring data are stored in the database and are called for analysis by the water inrush structure deduction unit.

[0038] 2. Water bursting structure deduction unit.

[0039] The water inrush structure interpretation unit includes a digital image pickup device, a digital strain gauge, a data integration module and a computer.

[0040] in: 1. Digital image pickup device: Receives the digital image generated by the data acquisition unit and extracts the 3D scanning data of the tunnel structure surface, including shape, size and hydrological data. The extracted data can be obtained and retrieved in real time through the tunnel database and the underground mine monitoring system, or manually input.

[0041] 2. Digital strain gauge: It is a non-contact deformation test, which obtains digital images from the digital image pickup device module and records hydrological data. When the monitoring signal is an electrical parameter, the monitoring signal enters the instrument after being amplified by the amplifier and then sent to the data integration module.

[0042] 3. Data integration module: Analyze and integrate the input data, and then upload it to the identification and prediction unit. The monitoring data of tunnel water inrush structure can also be directly input into the data integration module from the system database.

[0043] The three-dimensional surface image of the tunnel extracted by the digital image pickup device is transmitted to the digital strain gauge through the CR-655 interface cable, and the digital strain gauge inputs the data into the data integration module through the RS-232C interface cable. At the same time, the digital signal in the data input by the water inrush structure deduction unit is also directly uploaded to the digital integration module through the RS-232C interface and cable. The data integration module is used to integrate the received data, input it into the computer through the CR-553B interface cable, and upload it to the water inrush identification and prediction unit by the computer. The digital strain gauge, data integration module and computer are integrated in the host.

[0044] The deductive unit of water inrush structure runs a deductive algorithm. It uses the research on the influence of the characteristics of the roadway structure surface and the statistical analysis of the literature case data to determine the characteristic factors that have a significant impact on the evolution of the structure surface from the many factors affecting the water inrush structure. Then, the characteristic parameters that affect the evolution of the structure surface are extracted from the characteristic database, and the characteristic data chain is constructed according to certain principles. The model of the deductive algorithm contains the combined relationship between the roadway water inrush structure and the roadway lithology, the overlying rock structure and its influencing factors. The model is determined by big data, machine learning and artificial intelligence. It is the basic mode of water inrush on the roadway structure surface under various combination conditions. The combined relationship is reflected in the characteristic data chain. Each feature in the characteristic data chain is a major factor affecting the roadway water inrush structure.

[0045] The process of deductive algorithm is: (1) Determine characteristic factors: Through the study of the influence of tunnel structural surface characteristics and statistical analysis of literature case data, the characteristic factors that have a significant impact on the evolution of the structural surface are determined from the factors affecting the water inrush structure.

[0046] The characteristic factors are stored in the factor characteristic library (a collection of factors constituting the characteristics of the tunnel structure surface) and are used to construct the characteristic data chain.

[0047] (2) Construct a characteristic data chain based on characteristic factors: The characteristic parameters in the currently acquired data that match the characteristic factors, including lithology, structural surface characteristics, regional geological structure, rock layer structure, tunnel hydrology and mining disturbance data, are combined into a characteristic data chain according to the following coding rules: The value corresponding to each characteristic factor is 0 or 1, which is determined by the threshold of the characteristic factor. Figure 3 First, the parameters that affect the characteristics of the tunnel structure surface are divided into those that have an impact on water inrush and those that have no impact. When the value of the characteristic parameter exceeds the threshold, it is characterized as having an impact on the water inrush structure, otherwise it is characterized as having no impact. Finally, a code sequence consisting of 0 and 1 is obtained. The sequence length is a multiple of 8, and 0 is used to fill in the gap when it is insufficient.

[0048] Table 1 - Interception of a characteristic data link

[0049] In this example, the system sequentially extracts characteristic factors from the distance from the tunnel to the seabed bedrock surface, structural surface, water-bearing body, rock layer structure, mining area structure and mining disturbance, judges and encodes the factor threshold, and generates code "1" if it is "greater than the threshold" and code "0" if it is "less than the threshold". All characteristic factors affecting tunnel water inrush are searched in turn, and a code sequence consisting of (0, 1) is formed by encoding. The code sequence is the system's internal operation code, which is used to obtain the diagnosis result of tunnel water inrush. In addition, the location information is the coordinates and burial depth of the key points, which can be expressed in 4 digits. The total number of digits in the code sequence is a multiple of 8, and the insufficient part is filled with 0. As shown in Table 1, the final code sequence is 0000 0010 1100.

[0050] 3. Water inrush identification and prediction unit.

[0051] The water inrush identification and prediction unit extracts the characteristic factors of the roadway according to the roadway water inrush factor characteristic database, and realizes water inrush identification and prediction by comparing the characteristic data chain of the roadway to be tested with the standard factor sequence of the water inrush characteristic database.

[0052] The processing process of the water inrush identification and prediction unit is as follows: (1) Feature extraction: Extract the water inrush factor features from the feature data chain of the roadway to be tested based on the roadway water inrush factor feature database.

[0053] The innate conditions of the environmental factors that breed water inrush in tunnels are divided into factor types according to their functions and properties. The characteristics of water inrush factors are the factors that are significantly expressed in the breeding environment.

[0054] (2) Identification, prediction and forecasting: The extracted prominent factor features are compared with the standard factor sequences in the water inrush feature library. If they are consistent, it is judged that water inrush has occurred.

[0055] This process specifically evaluates and predicts whether the tunnel structure surface will experience water inrush: Evaluation refers to the assessment of water inrush conditions under disturbances such as tunnel structural surface, stratum geology, geological structure, lithology and mining, to determine water inrush factors and conditions for their occurrence. Water inrush occurs when the characteristic factor state is 1 and matches the standard data link, and no water inrush occurs when the state is 0 and does not match. This is used to determine the water inrush breeding environment and the impact of external factors on tunnel water inrush safety.

[0056] Prediction and forecast refers to the estimation of the possibility and time of water inrush in the tunnel. The conditions of the factor characteristics corresponding to the water inrush characteristic factor data link and the time when the conditions occur are the time when the water inrush occurs, and an alarm is issued according to the degree of water inrush hazard.

[0057] (3) Emergency response: Based on the hazard level of the identified prediction results, the emergency plan is activated according to the level.

[0058] The unit also includes a central control system and corresponding terminals for data storage, display, analysis, printing, transmission and issuing instructions.

[0059] Typical case analysis: Taking the mining of a certain seabed metal deposit as an example, the thickness of the Quaternary sedimentary layer on the seabed is 38 meters, the thickness of the weathering layer is 45 meters, the uppermost tunnel is located 150 meters below sea level, and there is a fault fracture zone.

[0060] The tunnel structure surface code and rock layer structure code 0000 indicate that there is no fault crossing the tunnel or there is a fault but the depth to the top bedrock is limited and there is no threat of water seepage. The structural surface is downward or tilted, the strike length deviates from the water body and is small in scale. The structural surface lithology code 0000 indicates that the bedrock layer at the top of the tunnel is a special rock such as non-carbonate rock. The geological structure code 1001 indicates that there are faults in the area but they do not cross the tunnel. The scale is small and the density is low. The regional earthquake intensity is VII degrees and there are active faults. The tunnel hydrology code 1000 indicates that there are seepage points, a certain amount of seepage and stable water inflow, no chloride ions, and no penetration with seawater. The mining disturbance code 0001 indicates that the maximum exposure area above the tunnel level and the top of the tunnel will not cause the goaf to collapse, but the vibration generated by the maximum single-stage blasting charge may induce fault movement and cause water inrush. The resulting judgment prediction code sequence 00000000 1001 1000 0001 can be used to find the cause and solve it.

[0061] The data input of the present invention can adopt online test data or manually input data, and can be applicable to the identification and prediction of tunnel water inrush.

[0062] It should be noted that it is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. The scope of the present invention is defined by the claims rather than the above description.

Claims

1. An intelligent identification and prediction system for water inrush structures in submarine tunnels, characterized by: It includes data acquisition unit, water inrush structure deduction unit and water inrush identification and prediction unit; The data acquisition unit is used to collect data on the tunnel structure surface and rock layer distribution; The water bursting structure deduction unit includes a digital image pickup device, a digital strain gauge, a data integration module and a computer, which is used to receive and process the data collected by the data acquisition unit and generate a characteristic data chain; The water inrush identification and prediction unit extracts the characteristic factors of the roadway according to the roadway water inrush factor characteristic database, and realizes water inrush identification and prediction by comparing the characteristic data chain of the roadway to be tested with the standard factor sequence of the water inrush characteristic database.

2. The intelligent identification and prediction system for water inrush structures in submarine tunnels according to claim 1, characterized in that: The data collection unit collects data on the strata, lithology, structure, hydrology and mining of the tunnel location and area, including data from early geological exploration, current status data revealed by the tunnel and mining dynamics data.

3. The intelligent identification and prediction system for water inrush structures in submarine tunnels according to claim 2, characterized in that: The data collection unit collects the following information: Tunnel location information: refers to the middle level of the tunnel, representing the tunnel depth and its associated underground facilities, indicating its role and status in the mine system; Plan layout: including the direction and length of the roadway and its relationship with the horizontal stope, chamber facilities and location; Lithology: refers to the distribution of lithology and its characteristic data along the axis of the roadway; Structural surface characteristics: refers to the distribution of main controlling structural surfaces such as faults, joints and fissures along the tunnel axis on the tunnel vault and wall, as well as the geometric structural characteristic data in terms of occurrence and scale; Rock structure: refers to the distribution of rock layers overlying the tunnel, including the distribution, lithology and thickness of rock layers from the tunnel to the seabed; Regional geological structure: including the type, size, distribution density, location of faults, whether the faults run through the tunnels, the positional relationship between the faults and the tunnels, and the regional earthquake intensity, magnitude and acceleration; Tunnel hydrology: including the distribution of water seepage points in the tunnel, the amount of water seepage and its chloride ion content, as well as the distribution and connectivity of water bodies from the seabed to the tunnel bottom; Mining disturbance data: including the maximum exposed area at the tunnel level and above the tunnel top and the maximum single-stage blasting charge.

4. The intelligent identification and prediction system for water inrush structures in submarine tunnels according to claim 1, characterized in that: The digital image pickup device is used to receive the digital image generated by the data acquisition unit and extract the 3D scanning data of the tunnel structure surface, including the shape, size and hydrological data.

5. The intelligent identification and prediction system for water inrush structures in submarine tunnels according to claim 1, characterized in that: The digital strain gauge is a non-contact deformation tester, which is used to obtain digital images from the digital image pickup device module and record hydrological data; when the monitoring signal is an electrical parameter, the monitoring signal enters the instrument after being amplified by the amplifier and then sent to the data integration module.

6. The intelligent identification and prediction system for water inrush structures in submarine tunnels according to claim 1, characterized in that: The data integration module is used to analyze and integrate the input data, and then upload it to the identification and prediction unit.

7. The intelligent identification and prediction system for water inrush structures in submarine tunnels according to claim 1, characterized in that: The water inrush structure deduction unit also runs a deduction algorithm, which uses the research on the influence of tunnel structural surface characteristics and statistical analysis of literature case data to determine the characteristic factors that have a significant impact on the evolution of the structural surface from the many factors affecting the water inrush structure, and then extracts the characteristic parameters that affect the evolution of the structural surface from the characteristic database, and constructs a characteristic data chain based on certain principles.

8. The intelligent identification and prediction system for water inrush structures in submarine tunnels according to claim 7, characterized in that: The process of deductive algorithm is: (1) Determine characteristic factors: Through the research on the influence of tunnel structural surface characteristics and statistical analysis of literature case data, determine the characteristic factors that have a significant impact on the evolution of structural surfaces from the factors affecting the water inrush structure; (2) Constructing a feature data chain based on the feature factors: The feature parameters in the currently acquired data that match the feature factors are encoded into a feature data chain according to the following rules: the value corresponding to each feature factor is 0 or 1, which is determined by the threshold of the feature factor; and finally a code sequence consisting of 0 and 1 is obtained as the feature data chain.

9. The intelligent identification and prediction system for water inrush structures in submarine tunnels according to claim 1, characterized in that: The processing process of the water inrush identification and prediction unit is as follows: (1) Feature extraction: Extract the water inrush factor features from the feature data chain of the roadway to be tested based on the roadway water inrush factor feature database; (2) Identification, prediction and forecasting: The extracted outburst factor features are compared with the standard factor sequence of the water inrush feature library. If they are consistent, it is judged that water inrush has occurred, and the evaluation and prediction of whether the tunnel structure surface has water inrush is completed; (3) Emergency response: Based on the hazard level of the identified prediction results, the emergency plan is activated according to the level.

10. A method for intelligent identification and prediction of water inrush structures in submarine tunnels, characterized in that: Water inrush is identified and predicted based on the intelligent identification and prediction system for water inrush structures in submarine tunnels as described in any one of claims 1 to 9.

Images