Karst cave group detection method and device in shield construction

Abstract

The invention provides a karst cave group detection method and device in shield construction, and the method achieves the comprehensive detection of karst caves through a transient electromagnetic and sound wave array composite wave field detection mode, and improves the detection accuracy and detection efficiency of the karst cave group. And feature extraction is performed on the composite wave field detection data through a cross-physical coupling model, a karst cave probability model is established, and quantitative evaluation of a karst cave detection result is realized. When the karst cave possibly exists in the construction direction of the shield tunneling machine, the distribution prediction direction of the karst cave is further determined, and the shield tunneling machine is prevented from entering the karst cave area blindly in the tunneling process. Through the multi-source sensing array carried on the shield tunneling machine, equipment parameters during operation of the shield tunneling machine are collected in real time, potential construction anomalies can be found easily, and the accuracy and timeliness of karst cave detection are improved. Combined analysis is carried out on the composite wave field detection data and the operation parameters of the shield tunneling machine through the risk assessment model, a karst cave risk prediction value is determined, and the accuracy of karst cave risk assessment is improved.

Description

Technical Field

[0001] This application relates to the field of tunnel construction technology, and in particular to a method, device, computer equipment, and computer-readable storage medium for detecting karst cave groups during shield tunneling. Background Technology

[0002] Subsea tunnel boring machine (TBM) construction faces significant safety threats from karst cave systems in complex geological environments. Due to the high conductivity and electromagnetic shielding effect of seawater, traditional terrestrial detection methods (such as ground-penetrating radar and high-density electrical resistivity tomography) suffer a substantial decrease in penetration depth and resolution in the seabed environment. In particular, when the TBM is excavating, the strong electromagnetic interference generated by its metal structure can degrade the signal-to-noise ratio of the detection signal by more than 60%.

[0003] Currently, when shield tunnels pass through karst strata in the seabed and marine areas, the detection of karst cave groups mainly follows the "ground drilling + marine geophysical exploration" model: large exploration platforms are set up on the sea surface on both sides of the tunnel axis, and cross-hole acoustic CT, ground-penetrating radar or 3D seismic methods are used, supplemented by several advanced horizontal boreholes, to obtain spatial distribution information of karst (cavities) within a range of 30-100 m ahead, and grouting reinforcement or bypass schemes are formulated accordingly.

[0004] However, current mainstream forward detection technologies (such as the TSP seismic wave method) require frequent shutdowns to deploy sensors during dynamic tunneling, resulting in a 30%-50% loss in construction efficiency. Meanwhile, point detection methods based on borehole sampling (such as advanced drilling) have problems such as large blind spots and slow response, making it difficult to accurately capture the spatial distribution characteristics of irregular karst cave groups.

[0005] Therefore, improving the accuracy of detecting karst cave systems during shield tunneling has become an urgent technical problem to be solved. Summary of the Invention

[0006] This application provides a method and device for detecting karst cave groups during shield tunneling, aiming to improve the accuracy of karst cave group detection during shield tunneling of submarine tunnels.

[0007] Firstly, this application provides a method for detecting karst cave groups during shield tunneling construction, the method comprising: Based on a composite wave field detection method using transient electromagnetic and acoustic arrays, composite wave field detection data is obtained by detecting the detection area of ​​shield tunneling. Based on the cross-physics coupling model, feature extraction is performed on the composite wave field detection data to obtain the probability model of karst caves in the shield tunneling detection area; When the karst cave probability model indicates that there is a karst cave in the construction direction of the tunnel boring machine, the predicted karst cave distribution direction in the construction direction of the tunnel boring machine is determined based on the karst cave probability model. During the construction of the tunnel boring machine towards the predicted orientation of the karst cave, the equipment parameters of the tunnel boring machine during operation are collected based on the multi-source sensor array mounted on the tunnel boring machine to obtain the tunnel boring machine operating parameters; Based on the risk assessment model, the composite wave field detection data and the shield machine operating parameters are jointly analyzed to determine the predicted risk value of the karst cave corresponding to the predicted karst cave distribution location. When the predicted risk value of the karst cave is greater than or equal to the preset risk prediction threshold, a risk warning mechanism is triggered to send a warning to the central control terminal.

[0008] Secondly, this application also provides a karst cave group detection device for shield tunneling, the karst cave group detection device for shield tunneling includes: The data detection module is used to detect the shield tunneling detection area using a composite wave field detection method based on transient electromagnetic and acoustic arrays, and obtain composite wave field detection data. The feature analysis module is used to extract features from the composite wave field detection data based on the cross-physical coupling model to obtain the probability model of karst caves in the shield tunneling detection area. The cave distribution prediction module is used to determine the predicted location of the cave distribution in the construction direction of the tunnel boring machine based on the cave probability model when the cave probability model indicates that there is a cave in the construction direction of the tunnel boring machine. The operation parameter acquisition module is used to collect equipment parameters of the shield machine during the construction process of the shield machine towards the predicted orientation of the karst cave, based on the multi-source sensor array mounted on the shield machine, and obtain the shield machine operation parameters. The risk assessment module is used to perform joint analysis of the composite wave field detection data and the tunnel boring machine operating parameters based on the risk assessment model, and to determine the predicted risk value of the karst cave corresponding to the predicted karst cave distribution location. The risk warning module is used to trigger a risk warning mechanism to send a warning to the central control terminal when the predicted risk value of the karst cave is greater than or equal to a preset risk prediction threshold.

[0009] Thirdly, this application also provides a computer device, which includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, it implements the steps of the method for detecting karst cave groups in shield tunneling as described above.

[0010] Fourthly, this application also provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the steps of the method for detecting karst cave groups during shield tunneling as described above.

[0011] This application provides a method for detecting karst cave clusters during tunnel boring machine (TBM) construction. This method utilizes a composite wavefield detection approach combining transient electromagnetic and acoustic arrays to achieve comprehensive detection of different types of karst caves, effectively improving the accuracy and efficiency of karst cave detection. By extracting features from the composite wavefield detection data through a cross-physical coupling model, a karst cave probability model is established, enabling a quantitative assessment of the likelihood of karst caves existing within the TBM construction detection area, effectively improving the accuracy and reliability of karst cave detection. When the karst cave probability model indicates the potential presence of karst caves in the TBM's construction direction, the model is further used to determine the predicted distribution location of the karst caves, effectively preventing the TBM from blindly entering karst cave areas during tunneling. A multi-source sensor array mounted on the TBM collects equipment parameters in real time, dynamically monitoring the TBM's construction status, which helps to detect potential construction anomalies, further improving the accuracy and timeliness of karst cave detection. Combining the composite wavefield detection data with the TBM's operating parameters and conducting joint analysis through a risk assessment model allows for a more accurate determination of the predicted karst cave risk value, improving the accuracy of karst cave risk assessment. When the predicted risk value of the karst cave reaches or exceeds the preset threshold, the risk warning mechanism is triggered, and a warning message is sent to the central control terminal to achieve timely warning, ensure construction quality and safety, and improve construction efficiency. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart illustrating the first embodiment of a method for detecting karst cave groups during shield tunneling, as provided in this application. Figure 2 A flowchart illustrating a composite wave field detection method using transient electromagnetic and acoustic arrays provided in this application; Figure 3 A schematic diagram illustrating the process of constructing a three-dimensional model of a karst cave based on composite wave field detection data, provided in this application; Figure 4 This is a schematic diagram of the first embodiment of a karst cave group detection device provided in this application; Figure 5 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application.

[0014] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0015] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0016] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0017] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0018] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of a method for detecting karst cave groups during shield tunneling, as provided in this application.

[0019] like Figure 1 As shown, the method for detecting karst cave groups during shield tunneling includes steps S101 to S104.

[0020] S101. A composite wave field detection method based on transient electromagnetic and acoustic arrays is used to detect the shield tunneling detection area and obtain composite wave field detection data. In one embodiment, during the construction of a submarine shield tunnel, a composite wave field detection method combining TEM (Time Domain Electromagnetic Method) and acoustic array is used to comprehensively detect and acquire relevant data in front of and around the tunnel boring machine.

[0021] Specifically, a TEM transmitting coil and an acoustic transducer array can be embedded in the cutterhead of the tunnel boring machine (TBM) to sequentially transmit electromagnetic pulses and frequency-modulated acoustic signals via a pre-set frequency division multiplexing mechanism. Simultaneously, reflected and induced signals are received by a TEM receiving coil and an acoustic receiver arranged in the TBM's telescopic arm module, thereby collecting composite wavefield detection data containing geological features of the tunnel construction area.

[0022] For example, a TEM transmitting coil made of a rectangular hollow copper tube and an acoustic transducer array composed of piezoelectric ceramic units are set in the cutterhead area, while a μ-metal shield is used for electromagnetic shielding. For example, the μ-metal shield can adopt a segmented curved surface structure to cover the TEM transmitting coil and the acoustic transducer array, combined with a nanocrystalline absorbing coating, which has been measured to reduce the electromagnetic noise of the tunnel boring machine.

[0023] Furthermore, based on the preset frequency division multiplexing control timing, pulsed magnetic fields and linear frequency modulated sound waves are sequentially emitted to the shield tunneling detection area; the signals fed back by the receiving coil based on the square wave pulse and the signals fed back by the sound wave receiver based on the linear frequency modulated sound waves are simultaneously acquired to obtain composite wave field detection data.

[0024] In one embodiment, before data acquisition begins, the hardware devices, including the TEM transmitting coil, acoustic transducer array, μ-metal shield in the embedded detection unit of the tunnel boring machine, and the TEM receiving coil, acoustic receiver, and hydraulic propulsion mechanism in the telescopic arm module, are fully debugged and calibrated.

[0025] For example, an FPGA control board can be used to implement the coordinated operation of the TEM and the acoustic array, as well as the allocation of time slots. Figure 2 As shown, according to the set frequency division multiplexing control timing, pulsed magnetic fields (such as square wave pulses) and linear frequency modulated sound waves are emitted sequentially, while signals fed back from the receiving coil and sound wave receiver are acquired simultaneously. During data acquisition, the sampling rate is ensured to be high enough to accurately capture subtle changes in the signal, and the timestamps of the acquired data are recorded for subsequent data fusion and analysis.

[0026] By employing a composite wave field detection technology combining transient electromagnetic (TEM) and acoustic arrays, along with a frequency division multiplexing mechanism and a μ-metal electromagnetic shield, electromagnetic noise interference can be effectively reduced, and the detection range and signal-to-noise ratio can be improved.

[0027] In one embodiment, the acquired raw data undergoes preprocessing, including noise removal, filtering, and normalization. Due to the complex construction environment of underwater shield tunneling, the acquired data may be subject to various interferences, such as motor electromagnetic noise and tunneling vibration. Therefore, bandpass filters can be used to remove unwanted frequency components and retain effective signals; wavelet transform and other methods can be used to denoise the signals and improve the signal-to-noise ratio; and the data can be normalized to ensure it falls within the same dimension and numerical range, facilitating subsequent model processing and analysis.

[0028] For example, the LMS (Least Mean Square) adaptive filtering algorithm can be used to separate tunneling vibrations and cave reflections in real time, thereby improving the accuracy of cutterhead vibration signal identification.

[0029] Assume the received signal is: in, This is a wave reflected from the cave. The issue is related to cutterhead vibration. An adaptive filter is constructed using the LMS adaptive filtering algorithm, with the cutterhead speed synchronization signal as the reference input. Output estimated interference Final error signal: in, This refers to pure reflection waves from the cave.

[0030] In one embodiment, the composite wave field detection data may include key information such as electromagnetic field strength, attenuation characteristics, and acoustic wave reflection delay and amplitude.

[0031] This embodiment combines a TEM (Transmission Electromagnetic Sensor) and an acoustic array to achieve comprehensive detection of the area in front of and around the tunnel boring machine (TBM), acquiring richer geological information. By pre-setting a frequency division multiplexing (FDM) control sequence, electromagnetic pulses and frequency-modulated acoustic waves are emitted sequentially to avoid signal interference and improve detection efficiency. An FPGA control board enables the TEM and acoustic array to work collaboratively, synchronously acquiring reflected and induced signals to ensure data accuracy and timeliness. An LMS (Low Motion Filtering) adaptive filtering algorithm is used to separate tunneling vibrations and cavern reflections in real time, improving the identification accuracy of cutterhead vibration signals and enhancing detection accuracy.

[0032] S102. Based on the cross-physical coupling model, feature extraction is performed on the composite wave field detection data to obtain the probability model of karst caves in the shield tunneling detection area. In one embodiment, the acquired composite wave field detection data is input into a cross-physical coupling model that deeply integrates the electromagnetic induction principle of TEM detection and the reflection imaging mechanism of acoustic wave detection.

[0033] Furthermore, based on the cross-physical coupling model, feature extraction is performed on the composite wave field detection data to obtain the karst cave conductivity characteristics and karst cave morphology characteristics of at least one sub-region within the shield tunneling detection area; based on preset weighting coefficients, the karst cave conductivity characteristics and karst cave morphology characteristics corresponding to all sub-regions are weighted and fused to generate a karst cave probability model for the shield tunneling detection area.

[0034] In one embodiment, the extracted features include, but are not limited to, the TEM decay time constant. Sound wave reflection coefficient Among them, the TEM decay time constant The electromagnetic properties and size of the cave interface are reflected and can be obtained by fitting the secondary field attenuation curve obtained by the TEM receiving coil; the acoustic reflection coefficient... It reflects the acoustic characteristics of the cave interface and is calculated from the reflected sound waves received by the acoustic transducer array.

[0035] Specifically, calculate the sound wave reflection coefficient. : in, It is the wave impedance, which is related to the density and sound velocity of the rock and soil medium.

[0036] Specifically, transient electromagnetic (TEM) induces eddy currents at the cave interface by emitting pulsed magnetic fields. Subsequently, the decay curves of the secondary field generated by these eddy currents are received, and the voltage change over time follows a specific pattern. These curves reflect the electrical conductivity of the cave. By analyzing these curves, the electrical conductivity characteristics of the cave can be inferred, thus providing information about the cave.

[0037] Generally, transient electromagnetic (TEM) measurements in the deep sea are affected by fluctuations in seawater salinity and temperature, which may cause distortion in resistivity measurements. Therefore, a deep-sea dynamic compensation algorithm can be used to compensate for the conductivity data of caves acquired by TEM, thereby improving the accuracy of conductivity data detection. The deep-sea dynamic compensation algorithm can be expressed as follows: in, This represents the true resistivity after compensation. This represents the raw apparent resistivity measured by the TEM sensor. Indicates the temperature compensation coefficient (measured and calibrated). This indicates the actual temperature measured on-site. Indicates the salinity compensation coefficient. This represents the actual measured salinity. In the formula, 25 and 35 represent the standard seawater temperature of 25℃ and the standard seawater salinity of 35 PSU, respectively.

[0038] Acoustic detection works by emitting frequency-modulated sound waves, which are then reflected at the cave's interface. An array of receivers is used to capture the time delay difference of the reflected sound waves. : in, The dimension (distance) of the cave is represented by , and 'c' is the speed of sound propagation in the medium. Based on the time delay difference data of reflected sound waves captured by the array receiver, the morphological characteristics of the cave can be constructed.

[0039] Similarly, to address the velocity distortion and energy attenuation of sound waves propagating in seawater and strata, sound velocity profile correction and propagation loss compensation can be used to compensate for the sound wave detection results, thereby improving the accuracy of identifying cave morphological features.

[0040] For example, for sound velocity profile correction, temperature can be measured in real time. ,salinity ,depth Calculate the current sound speed profile: Based on ray acoustic reconstruction of sound wave propagation path, sound wave travel time is corrected: in, This represents the corrected total travel time of the sound wave. Indicates the depth along the propagation path The real-time speed of sound This indicates the actual distance the sound wave travels.

[0041] The following compensation formula can be used for compensation of transmission loss: in, This indicates the original received sound pressure level amplitude; Indicates the frequency-varying absorption coefficient; This represents the scattering loss factor (seabed sediment inversion), reflecting the seabed reflection / scattering characteristics; Indicates a reference distance (1m). This indicates the actual distance the sound wave travels.

[0042] like Figure 3 As shown, based on the electrical conductivity and morphological characteristics of karst caves, and combined with a large amount of historical geological data, the probability of karst caves existing in different sub-regions within the construction area is calculated using a karst cave probability model. This model presents the potential distribution of karst caves in the shield tunneling construction area in an intuitive and quantitative form.

[0043] To integrate information obtained from TEM and acoustic detection, a data fusion method can be used. Specifically, the probability of the existence of a cave is calculated using the following formula: in, Indicates the probability of a cave. Indicated based on the electrical conductivity of the cavern ( The electrical conductivity characteristics of the karst caves, Indicated based on cave size ( The morphological characteristics of the karst caves and These are weighting coefficients derived from historical data optimization, used to measure the relative importance of cave conductivity characteristics and cave morphology characteristics in cave exploration.

[0044] This embodiment effectively improves the accuracy of identifying the conductivity and morphological features of caves by integrating a cross-physical coupling model of TEM and acoustic detection with a deep-sea dynamic compensation algorithm to correct the detection data in real time.

[0045] S103. When the karst cave probability model indicates that there is a karst cave in the construction direction of the tunnel boring machine, the predicted karst cave distribution direction in the construction direction of the tunnel boring machine is determined based on the karst cave probability model. In one embodiment, when the cave probability model shows that there is a potential threat of caves in the predetermined construction direction of the tunnel boring machine, the cave distribution prediction and orientation determination process is initiated.

[0046] Specifically, by using the probability distribution data within the karst cave probability model, optimization algorithms can be employed to accurately locate the peak region of the karst cave probability. Furthermore, by comprehensively analyzing factors such as the attitude of the tunnel boring machine, the direction of excavation, and the orientation of the geological structure, and through three-dimensional spatial modeling and coordinate transformation, the predicted location of the karst cave distribution in the tunnel boring machine's construction direction can be accurately determined, providing a clear direction for subsequent construction decisions and risk avoidance.

[0047] Furthermore, based on the karst cave probability model, the predicted karst caves are spatially located and their morphology is constructed to generate a three-dimensional distribution map of the karst caves; the attitude data of the tunnel boring machine is obtained; the attitude data of the tunnel boring machine is converted into the three-dimensional distribution map of the karst caves to determine the predicted orientation of the karst cave distribution in the tunnel boring machine's construction direction.

[0048] The predicted location of the karst caves may include, but is not limited to, the relative distance between the karst caves and the current location of the tunnel boring machine, the azimuth angle, and the specific location range in the tunnel face or surrounding strata.

[0049] For example, a three-dimensional spatial analysis algorithm is used to spatially locate and morphologically construct the probability of the existence of karst caves in the karst cave probability model, generating a distribution map of the karst caves in three-dimensional space. This three-dimensional spatial analysis algorithm determines the center location, extension direction, and influence range of the karst caves by analyzing the probability density, spatial continuity, and relative positional relationship with the tunnel boring machine's construction axis.

[0050] Real-time acquisition of tunnel boring machine (TBM) attitude data, including key parameters such as position, azimuth angle, pitch angle, and roll angle. This data is provided by high-precision measurement sensors, such as total stations, GPS positioning systems, electronic compasses, and inertial measurement units (IMUs), ensuring the accuracy and real-time nature of the TBM attitude data.

[0051] The data on the distribution of karst caves generated by the three-dimensional spatial analysis algorithm are deeply integrated with the attitude data of the tunnel boring machine. Through coordinate transformation and spatial matching, the predicted orientation of the karst cave distribution relative to the construction direction of the tunnel boring machine is determined.

[0052] In one embodiment, the predicted orientation of the karst cave distribution can be represented in three-dimensional coordinates, which clarifies the positional relationship of the karst cave in front of, above, below or to the side of the tunnel boring machine, and calculates the minimum distance between the karst cave and the tunnel boring machine, providing intuitive orientation information for construction decisions.

[0053] In one embodiment, as the tunnel boring machine (TBM) advances and new detection data is acquired, the probability model of the karst caves and the TBM's attitude data can be updated in real time. The predicted azimuth is also dynamically adjusted accordingly, ensuring that construction personnel can promptly grasp the latest changes in the distribution of karst caves and take effective countermeasures in advance to ensure construction safety and efficiency.

[0054] This embodiment integrates the probability model of karst caves with the attitude data of the tunnel boring machine (TBM) and uses a three-dimensional spatial analysis algorithm to achieve accurate and dynamic determination of the predicted location of karst caves in the construction direction of the TBM. This provides a clear direction for construction decision-making and risk avoidance, and effectively improves construction safety and efficiency.

[0055] S104. During the construction of the tunnel boring machine towards the predicted orientation of the karst cave, the equipment parameters of the tunnel boring machine during operation are collected based on the multi-source sensor array mounted on the tunnel boring machine to obtain the tunnel boring machine operating parameters. As the tunnel boring machine (TBM) continues construction toward the predicted location of the karst caves, a multi-source sensor array mounted on the TBM body and cutterhead area can be used to collect key equipment parameters in real time. These parameters are then transmitted to the TBM central control system via a high-speed communication link and pre-processed, thereby accurately obtaining information on the TBM's operating status in complex geological formations.

[0056] In one embodiment, the multi-source sensor array includes a current sensor and a speed sensor mounted on the main drive motor of the tunnel boring machine (TBM), a vibration sensor mounted on the cutterhead, and a displacement sensor and a grouting pressure sensor mounted on the tail of the shield. Specifically, the current sensor and speed sensor are used to collect tunneling dynamic parameters; the vibration sensor is used to collect the vibration frequency and acceleration parameters of the cutterhead; and the displacement sensor and the grouting pressure sensor are used to collect the orientation parameters of the TBM and the grouting pressure parameters.

[0057] For example, the multi-source sensor array may include current and speed sensors of the main drive motor of the tunnel boring machine to accurately acquire tunneling power parameters; the multi-source sensor array may include vibration sensors installed on the cutterhead to sensitively capture the vibration frequency and acceleration of the cutterhead during tunneling in different strata; the multi-source sensor array may also include displacement sensors and grouting pressure sensors installed on the tail of the shield to provide real-time feedback on the attitude adjustment and grouting support status of the tunnel boring machine.

[0058] Specifically, the multi-source sensor array includes current, torque, and speed sensors for the cutterhead motor; pressure and displacement sensors for the propulsion cylinder; a cutterhead vibration accelerometer; a tunnel boring machine attitude sensor; and a grouting pressure sensor. In one embodiment, all sensors can be mounted on a hydraulic-air-float dual-stage vibration damping platform to attenuate cutterhead vibration.

[0059] By using high-precision torque sensors and speed encoders, the torque, speed, and power of the cutterhead motor are monitored in real time, thereby determining the stress and energy consumption of the cutterhead during the tunneling process and providing a basis for indirect analysis of geological hardness.

[0060] By using pressure and displacement sensors, the pressure, extension and retraction displacement, and total thrust of the tunnel boring machine's propulsion cylinders can be acquired in real time. Combined with the jack zoning of the tunnel boring machine, the propulsion difficulty and ground reaction force distribution of the tunnel boring machine in different geological strata can be analyzed to help identify potential geological anomalies.

[0061] Vibration acceleration sensors are installed on the cutterhead housing and main bearing to capture the spectral characteristics of the cutterhead vibration at a high sampling rate. The vibration frequency and amplitude are analyzed in real time using the Fast Fourier Transform (FFT) algorithm to identify the changes in vibration modes when the cutterhead comes into contact with different geological bodies, providing direct signals for the near-term prediction of karst caves.

[0062] By using multi-point displacement sensors and laser targets, combined with total station measurement data, the attitude parameters of the tunnel boring machine (TBM) are calculated in real time, including position, pitch angle, azimuth angle and roll angle. This ensures that the TBM advances accurately along the design axis while providing attitude compensation data for the inversion of the location of the karst cave.

[0063] Pressure transmitters are installed in the grouting pipeline to monitor the pressure changes during synchronous grouting. By analyzing the correlation between grouting pressure and the tunnel boring machine's advance speed and geological conditions, the grouting effect and geological stability can be determined, thus preventing ground subsidence caused by karst caves.

[0064] These sensors continuously collect data at a millisecond sampling frequency. After filtering, compensation, and other preprocessing steps to remove noise interference and outliers, the data is integrated to form a dataset of tunnel boring machine operating parameters, including key indicators such as tunneling speed, cutterhead torque, thrust, and tail grouting pressure. This dataset provides a detailed reflection of the tunnel boring machine's construction status under complex geological conditions.

[0065] To ensure the consistency and comparability of multi-source data, all sensor data must undergo strict time synchronization. A high-precision clock chip provides a unified time reference for each sensor, ensuring that the acquired data are aligned according to the same time sequence. Simultaneously, the data acquisition process is time-calibrated, accurately recording the timestamp of each data point. The pre-processed tunnel boring machine (TBM) operating parameters are then spatiotemporally matched and dynamically analyzed with the real-time updated geological cave model. This is achieved by establishing mapping relationships between parameters and geological conditions, such as the correspondence between cutterhead torque and stratum hardness, and the correlation between propulsion pressure and the strength of the overlying strata of the cave roof.

[0066] This embodiment uses a multi-source sensor array mounted on the tunnel boring machine (TBM) to collect and preprocess the TBM's operating parameters in real time. Combined with time synchronization and spatiotemporal matching technologies, it provides construction personnel with accurate information on the TBM's operating status and geological conditions. This helps to identify potential risks in advance, improve construction safety, and optimize construction parameters to enhance construction efficiency and quality.

[0067] S105. Based on the risk assessment model, the composite wave field detection data and the shield machine operating parameters are jointly analyzed to determine the predicted risk value of the karst cave corresponding to the predicted karst cave distribution location. In one embodiment, the acquired composite wavefield detection data and tunnel boring machine operating parameters are input into a pre-constructed risk assessment model. The risk assessment model can be a model built based on machine learning algorithms such as Bayesian networks, neural networks, and support vector machines.

[0068] Specifically, the risk assessment model is based on the detection data such as the size and location of the karst cave and the stability of the surrounding rock. It combines the operating parameters such as the tunneling speed, cutterhead torque, and thrust, and uses a conditional probability matrix to characterize the relationship between the various factors. Through model reasoning and calculation, it accurately outputs the risk prediction value corresponding to the predicted location of the karst cave distribution.

[0069] For example, composite wave field detection data (such as the conductivity, shape, location, and size of karst caves) are deeply integrated with tunnel boring machine (TBM) operating parameters (such as tunneling speed, cutterhead torque, thrust, and TBM attitude). Based on geological engineering experience and data analysis, various risk factors closely related to karst cave risks are identified, such as the size of the karst cave, the stability of the strata, groundwater pressure, and the rate of change of TBM tunneling parameters. These risk factors are then quantified and transformed into quantifiable risk indicators.

[0070] Considering the spatiotemporal correlation between composite wavefield detection data and tunnel boring machine (TBM) operating parameters, spatiotemporal synchronous analysis is performed. Detection data from different locations and tunneling stages of the TBM are matched and aligned with corresponding operating parameters to construct a spatiotemporal data cube. This allows for a more accurate analysis of the spatiotemporal evolution of karst cave risk and the dynamic relationship between the TBM's operating status and karst cave risk.

[0071] The weights of different risk factors in risk assessment are determined using multi-criteria decision-making methods such as the analytic hierarchy process (AHP) and entropy weight method, reflecting the relative importance of each risk factor to the risk of karst caves. Then, by combining weighted summation, product, and other methods, the quantitative values ​​of various risk factors are combined with their corresponding weights to calculate the comprehensive risk prediction value corresponding to the predicted location of karst cave distribution, thus achieving the organic integration of multi-source information and comprehensive risk assessment.

[0072] Furthermore, based on the factor correlation between the composite wave field detection data and the tunnel boring machine operating parameters, a conditional probability matrix is ​​constructed; the composite wave field detection data, the tunnel boring machine operating parameters, and the conditional probability matrix are input into the risk assessment model, and the risk assessment model is used to predict the risk of the predicted karst cave distribution location, and outputs the predicted karst cave risk value corresponding to the predicted karst cave distribution location.

[0073] In one embodiment, the correlation between composite wavefield detection data (such as the conductivity, shape, location, and size of karst caves) and tunnel boring machine (TBM) operating parameters (such as tunneling speed, cutterhead torque, thrust, and TBM attitude) is analyzed. Larger karst caves generally result in poorer surrounding rock stability, leading to increased cutterhead torque and thrust, and a decrease in tunneling speed.

[0074] It can collect a large amount of historical construction data, including past cases of karst cave accidents and relevant data under normal construction conditions. Through statistical analysis, it can establish conditional probabilistic relationships between composite wave field detection data, tunnel boring machine (TBM) operating parameters, and karst cave risk. For example, it can determine the probability of a karst cave accident occurring when the TBM experiences high torque and low tunneling speed under specific conductivity and morphology conditions.

[0075] Based on historical data and expert experience, a conditional probability matrix is ​​initialized. The rows of the matrix represent different composite wavefield detection data states, and the columns represent different tunnel boring machine (TBM) operating parameter states. As new data is continuously collected and analyzed, the conditional probability matrix is ​​updated in real time to optimize the probabilistic relationships between various factors, thereby improving the model's accuracy and adaptability.

[0076] The preprocessed composite wavefield detection data, tunnel boring machine (TBM) operating parameters, and the constructed conditional probability matrix are used as input feature vectors and fed into the risk assessment model. Based on the input data and conditional probability matrix, the risk assessment model uses probabilistic inference algorithms (such as Bayesian inference) or the predictive capabilities of machine learning models to assess the risk of the predicted location of karst caves. The model comprehensively analyzes the correlation between the composite wavefield detection data and the TBM operating parameters, calculates the predicted risk value of the karst caves corresponding to their predicted locations, and quantitatively represents the magnitude of the karst cave risk.

[0077] This embodiment, by deeply integrating composite wavefield detection data with tunnel boring machine (TBM) operating parameters, utilizes a risk assessment model constructed with conditional probability matrices and advanced machine learning algorithms. This model can accurately and in real-time predict the multi-dimensional risks of karst caves during TBM construction, providing a scientific basis for construction decisions. The model effectively integrates geological and engineering information, and through dynamic updates and spatiotemporal synchronous analysis, significantly improves the accuracy and timeliness of risk prediction, thereby enhancing construction safety, reducing engineering risks, and ensuring the smooth progress of undersea tunnel TBM construction.

[0078] S106. When the predicted risk value of the karst cave is greater than or equal to the preset risk prediction threshold, a risk warning mechanism is triggered to send a warning to the central control terminal.

[0079] In one embodiment, the predicted risk value of the karst cave corresponding to the predicted location of the karst cave is obtained through reasoning and calculation of the risk assessment model. The predicted risk value of the karst cave is presented as a quantitative score, including multiple dimensions such as the risk of karst cave collapse, the risk of tunnel boring machine jamming, and the risk of ground subsidence.

[0080] In one embodiment, risk prediction thresholds for different risk types (such as karst cave collapse, tunnel boring machine jamming, and ground subsidence) can be determined based on historical construction data and expert experience. Threshold setting needs to consider factors such as the severity of the risk, the likelihood of its occurrence, the construction stage, and geological conditions to ensure that the thresholds effectively reflect the severity of the risk.

[0081] A dynamic adjustment mechanism for risk prediction thresholds can be established, adjusting the thresholds in a timely manner based on real-time construction information and risk conditions. For example, when construction enters a high-risk area or encounters special geological conditions, the threshold can be lowered to improve the sensitivity of early warnings; conversely, in low-risk areas, the threshold can be appropriately raised to reduce unnecessary early warnings.

[0082] Furthermore, when the risk warning mechanism is triggered, the risk prediction type and risk prediction level are determined based on the predicted risk value of the karst cave; risk warning information is generated based on the risk prediction type and the risk prediction level; and the risk warning information is sent to the central control terminal to realize risk warning.

[0083] In one embodiment, a risk warning mechanism is triggered when the predicted risk value of a karst cave is greater than or equal to a preset risk prediction threshold. The risk assessment model identifies risk types, including karst cave collapse risk, tunnel boring machine jamming risk, and ground subsidence risk. The identification is based on the characteristics and performance of each risk in the prediction model, combined with real-time data to determine which type of risk is more likely to occur.

[0084] Based on the magnitude of the risk prediction, a clear risk level classification standard is established, dividing risks into four levels: low, medium, high, and extremely high. Each level corresponds to a specific risk description and possible consequences; for example, low risk indicates a low probability and minor consequences, while extremely high risk indicates a high probability and serious consequences. The classification comprehensively considers both the probability of the risk occurring and the degree of its impact to ensure that the level accurately reflects the severity of the risk.

[0085] Detailed risk warning information is generated based on the risk level and risk type.

[0086] The risk warning information can include the predicted value of the karst cave risk, the type of risk prediction, the level of risk prediction, the location of the risk, the predicted time of occurrence, and a brief analysis of the causes of the risk. It can also provide corresponding decision support suggestions, such as adjusting tunneling parameters, strengthening support, and grouting, to help construction personnel take effective risk response measures and ensure the safety and smooth progress of shield tunneling.

[0087] Risk warning information is transmitted in real time to the human-machine interface in the shield machine's control room via the internal communication link of the shield machine. It attracts the attention of the operators in the form of audible and visual alarms, and is simultaneously sent to the monitoring screen of the ground control center and the terminal equipment of the engineering management personnel through a dedicated communication network. This ensures that relevant personnel receive risk warnings in a timely manner, allowing sufficient time for subsequent emergency evacuation, construction adjustments and other response measures.

[0088] This embodiment sets and dynamically adjusts risk prediction thresholds, uses a risk assessment model to quantify karst cave risks from multiple dimensions, and accurately determines the risk type and level after triggering the risk warning mechanism. It generates detailed warning information and conveys it to operators and managers through multiple channels, significantly improving the timeliness, accuracy, and effectiveness of karst cave risk warnings during shield tunneling. This provides construction personnel with sufficient time to take countermeasures, reduces the probability of accidents, enhances construction safety, and ensures the quality and efficiency of tunnel construction.

[0089] This application provides a method for detecting karst cave clusters during tunnel boring machine (TBM) construction. This method utilizes a composite wavefield detection approach combining transient electromagnetic and acoustic arrays to achieve comprehensive detection of different types of karst caves, effectively improving the accuracy and efficiency of karst cave detection. By extracting features from the composite wavefield detection data using a cross-physical coupling model, a karst cave probability model is established, enabling a quantitative assessment of the likelihood of karst caves existing within the TBM construction detection area, effectively improving the accuracy and reliability of karst cave detection. When the karst cave probability model indicates the potential presence of karst caves in the TBM's construction direction, the model is further used to determine the predicted distribution location of the karst caves, effectively preventing the TBM from blindly entering karst cave areas during tunneling. A multi-source sensor array mounted on the TBM collects equipment parameters in real time, dynamically monitoring the TBM's construction status, which helps to identify potential construction anomalies, further improving the accuracy and timeliness of karst cave detection. Combining the composite wavefield detection data with the TBM's operating parameters and conducting joint analysis through a risk assessment model allows for a more accurate determination of the predicted karst cave risk value, improving the accuracy of karst cave risk assessment. When the predicted risk value of the karst cave reaches or exceeds the preset threshold, the risk warning mechanism is triggered, and a warning message is sent to the central control terminal to achieve timely warning, ensure construction quality and safety, and improve construction efficiency.

[0090] Please see Figure 4 , Figure 4 This is a schematic diagram of the first embodiment of a karst cave group detection device provided in this application. The karst cave group detection device is used to perform the aforementioned karst cave group detection method in shield tunneling.

[0091] like Figure 4 As shown, the karst cave group detection device 200 used in shield tunneling includes: a data detection module 201, a feature analysis module 202, a karst cave distribution prediction module 203, an operation parameter acquisition module 204, a risk assessment module 205, and a risk early warning module 206.

[0092] The data detection module 201 is used to detect the shield tunneling detection area based on a composite wave field detection method of transient electromagnetic and acoustic arrays, and obtain composite wave field detection data. Feature analysis module 202 is used to extract features from the composite wave field detection data based on the cross-physical coupling model to obtain a probability model of karst caves in the shield tunneling detection area. The cave distribution prediction module 203 is used to determine the predicted location of the cave distribution in the construction direction of the tunnel boring machine based on the cave probability model when the cave probability model indicates that there is a cave in the construction direction of the tunnel boring machine. The operation parameter acquisition module 204 is used to acquire equipment parameters of the shield machine during the construction of the shield machine towards the predicted orientation of the karst cave based on the multi-source sensor array mounted on the shield machine, and obtain the shield machine operation parameters. Risk assessment module 205 is used to perform joint analysis of the composite wave field detection data and the shield machine operating parameters based on the risk assessment model, and determine the predicted risk value of the karst cave corresponding to the predicted karst cave distribution location. The risk warning module 206 is used to trigger a risk warning mechanism to send a warning to the central control terminal when the predicted risk value of the karst cave is greater than or equal to a preset risk prediction threshold.

[0093] In one embodiment, the data detection module 201 includes: The frequency division multiplexing unit is used to sequentially transmit pulsed magnetic fields and linear frequency modulated sound waves to the shield tunneling detection area based on a preset frequency division multiplexing control timing sequence. The signal acquisition unit is used to synchronously acquire the signal fed back by the receiving coil based on the square wave pulse and the signal fed back by the acoustic receiver based on the linear frequency modulated acoustic wave, so as to obtain composite wave field detection data.

[0094] In one embodiment, the feature analysis module 202 includes: The feature extraction unit is used to extract features from the composite wave field detection data based on the cross-physical coupling model, and obtain the conductivity characteristics and morphological characteristics of the karst caves in at least one sub-region within the shield tunneling detection area. The feature fusion unit is used to perform weighted fusion of the electrical conductivity features and morphological features of the karst caves corresponding to all sub-regions based on preset weight coefficients, so as to generate a karst cave probability model of the shield tunneling detection area.

[0095] In one embodiment, the multi-source sensor array includes a current sensor and a speed sensor located on the main drive motor of the tunnel boring machine, a vibration sensor located on the cutterhead, and a displacement sensor and a grouting pressure sensor located at the tail of the shield. The operating parameter acquisition module 204 includes: The tunneling power parameter acquisition unit is used to acquire tunneling power parameters based on the current sensor and the speed sensor. The vibration data acquisition unit is used to acquire the vibration frequency and acceleration parameters of the cutter head based on the vibration sensor. The grouting parameter acquisition unit is used to acquire the position parameters and grouting pressure parameters of the tunnel boring machine based on the displacement sensor and the grouting pressure sensor.

[0096] In one embodiment, the risk assessment module 205 includes: The conditional probability matrix construction unit is used to construct a conditional probability matrix based on the factor correlation between the composite wave field detection data and the tunnel boring machine operating parameters; The risk prediction unit is used to input the composite wave field detection data, the tunnel boring machine operating parameters, and the conditional probability matrix into the risk assessment model, perform risk prediction on the predicted azimuth distribution location of the karst cave through the risk assessment model, and output the predicted karst cave risk value corresponding to the predicted azimuth distribution location.

[0097] In one embodiment, the risk warning module 206 includes: The risk type and level determination unit is used to determine the risk prediction type and risk prediction level based on the predicted risk value of the karst cave when the risk warning mechanism is triggered. A risk warning information generation unit is used to generate risk warning information based on the risk prediction type and the risk prediction level. The risk warning information sending unit is used to send the risk warning information to the central control terminal to realize risk warning.

[0098] In one embodiment, the cave distribution prediction module 203 includes: The three-dimensional distribution map generation unit is used to spatially locate and morphologically construct the predicted caves based on the cave probability model, and generate a three-dimensional distribution map of the caves. The attitude data acquisition unit is used to acquire the attitude data of the tunnel boring machine. The cave location prediction unit is used to convert the attitude data of the tunnel boring machine into the three-dimensional distribution map of the caves, and determine the predicted location of the cave distribution in the tunnel boring machine's construction direction.

[0099] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the device and each module described above can be referred to the corresponding process in the aforementioned embodiment of the method for detecting karst cave groups in shield tunneling construction, and will not be repeated here.

[0100] The apparatus provided in the above embodiments can be implemented as a computer program, which can be used in, for example... Figure 5 It runs on the computer device shown.

[0101] Please see Figure 5 , Figure 5 This is a schematic block diagram illustrating the structure of a computer device according to an embodiment of this application. The computer device may be a server.

[0102] See Figure 5The computer device includes a processor, memory, and network interface connected via a system bus, wherein the memory may include non-volatile storage media and internal memory.

[0103] Non-volatile storage media can store operating systems and computer programs. These computer programs include program instructions that, when executed, cause the processor to perform any method for detecting karst caves during tunnel boring machine (TBM) construction.

[0104] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0105] The internal memory provides an environment for the execution of computer programs in non-volatile storage media. When the computer program is executed by the processor, it enables the processor to perform any method for detecting karst cave groups during shield tunneling.

[0106] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0107] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0108] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps: Based on a composite wave field detection method using transient electromagnetic and acoustic arrays, composite wave field detection data is obtained by detecting the detection area of ​​shield tunneling. Based on the cross-physics coupling model, feature extraction is performed on the composite wave field detection data to obtain the probability model of karst caves in the shield tunneling detection area; When the karst cave probability model indicates that there is a karst cave in the construction direction of the tunnel boring machine, the predicted karst cave distribution direction in the construction direction of the tunnel boring machine is determined based on the karst cave probability model. During the construction of the tunnel boring machine towards the predicted orientation of the karst cave, the equipment parameters of the tunnel boring machine during operation are collected based on the multi-source sensor array mounted on the tunnel boring machine to obtain the tunnel boring machine operating parameters; Based on the risk assessment model, the composite wave field detection data and the shield machine operating parameters are jointly analyzed to determine the predicted risk value of the karst cave corresponding to the predicted karst cave distribution location. When the predicted risk value of the karst cave is greater than or equal to the preset risk prediction threshold, a risk warning mechanism is triggered to send a warning to the central control terminal.

[0109] The embodiments of this application also provide a computer-readable storage medium storing a computer program, the computer program including program instructions, and the processor executing the program instructions to implement any of the methods for detecting karst cave groups during shield tunneling provided in the embodiments of this application.

[0110] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, SmartMediaCard (SMC), SecureDigital (SD) card, or FlashCard equipped on the computer device.

[0111] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for detecting karst cave systems during shield tunneling, characterized in that, The method includes: Based on a composite wave field detection method using transient electromagnetic and acoustic arrays, composite wave field detection data is obtained by detecting the detection area of ​​shield tunneling. Based on the cross-physics coupling model, feature extraction is performed on the composite wave field detection data to obtain the probability model of karst caves in the shield tunneling detection area; When the karst cave probability model indicates that there is a karst cave in the construction direction of the tunnel boring machine, the predicted karst cave distribution direction in the construction direction of the tunnel boring machine is determined based on the karst cave probability model. During the construction of the tunnel boring machine towards the predicted orientation of the karst cave, the equipment parameters of the tunnel boring machine during operation are collected based on the multi-source sensor array mounted on the tunnel boring machine to obtain the tunnel boring machine operating parameters; Based on the risk assessment model, the composite wave field detection data and the shield machine operating parameters are jointly analyzed to determine the predicted risk value of the karst cave corresponding to the predicted karst cave distribution location. When the predicted risk value of the karst cave is greater than or equal to the preset risk prediction threshold, a risk warning mechanism is triggered to send a warning to the central control terminal.

2. The method for detecting karst cave groups during shield tunneling construction according to claim 1, characterized in that, The composite wave field detection method based on transient electromagnetic and acoustic arrays is used to detect the shield tunneling detection area and obtain composite wave field detection data, including: Based on the preset frequency division multiplexing control timing, pulsed magnetic fields and linear frequency modulated sound waves are sequentially emitted to the shield tunneling detection area; Synchronously acquire the signal fed back by the square wave pulse from the receiving coil and the signal fed back by the linear frequency modulated acoustic wave from the acoustic receiver to obtain composite wave field detection data.

3. The method for detecting karst cave groups during shield tunneling construction according to claim 1, characterized in that, The method of extracting features from the composite wavefield detection data based on the cross-physical coupling model to obtain a probability model of karst caves in the shield tunneling detection area includes: Based on the cross-physical coupling model, feature extraction is performed on the composite wave field detection data to obtain the conductivity characteristics and morphological characteristics of karst caves in at least one sub-region within the shield tunneling detection area. Based on preset weighting coefficients, the electrical conductivity characteristics and morphological characteristics of the karst caves corresponding to all sub-regions are weighted and fused to generate a karst cave probability model for the shield tunneling detection area.

4. The method for detecting karst cave groups during shield tunneling construction according to claim 1, characterized in that, The multi-source sensor array includes a current sensor and a speed sensor installed on the main drive motor of the tunnel boring machine, a vibration sensor installed on the cutterhead, and a displacement sensor and a grouting pressure sensor installed on the tail of the shield. The method of acquiring equipment parameters during tunnel boring machine (TBM) operation based on a multi-source sensor array mounted on the TBM includes: Based on the current sensor and speed sensor, tunneling power parameters are collected; Based on the vibration sensor, the vibration frequency and acceleration parameters of the cutter head are collected; Based on the displacement sensor and the grouting pressure sensor, the position parameters and grouting pressure parameters of the tunnel boring machine are collected.

5. The method for detecting karst cave groups during shield tunneling construction according to claim 1, characterized in that, The method, based on a risk assessment model, involves jointly analyzing the composite wave field detection data and the tunnel boring machine operating parameters to determine the predicted risk value of the karst cave corresponding to the predicted location of the karst cave distribution. This includes: Based on the correlation between the composite wave field detection data and the tunnel boring machine operating parameters, a conditional probability matrix is ​​constructed; The composite wave field detection data, the tunnel boring machine operating parameters, and the conditional probability matrix are input into the risk assessment model. The risk assessment model is used to predict the risk of the predicted azimuth distribution location of the karst cave and outputs the predicted karst cave risk value corresponding to the predicted azimuth distribution location.

6. The method for detecting karst cave groups during shield tunneling construction according to claim 1, characterized in that, When the predicted risk value of the karst cave is greater than or equal to a preset risk prediction threshold, a risk warning mechanism is triggered to send a warning to the central control unit, including: When the risk warning mechanism is triggered, the risk prediction type and risk prediction level are determined based on the predicted risk value of the karst cave. Based on the risk prediction type and the risk prediction level, risk warning information is generated; The risk warning information is sent to the central control terminal to achieve risk warning.

7. The method for detecting karst cave groups during shield tunneling construction according to claim 1, characterized in that, The determination of the predicted karst cave distribution location based on the karst cave probability model, along with the determination of the karst cave distribution location along the tunnel boring machine's construction direction, includes: Based on the cave probability model, the predicted caves are spatially located and their morphology is constructed to generate a three-dimensional distribution map of the caves. Obtain the attitude data of the tunnel boring machine; The attitude data of the tunnel boring machine is converted into the three-dimensional distribution map of the karst caves to determine the predicted orientation of the karst cave distribution in the tunnel boring machine's construction direction.

8. A device for detecting karst cave systems during shield tunneling construction, characterized in that, The karst cave detection device used in shield tunneling includes: The data detection module is used to detect the shield tunneling detection area using a composite wave field detection method based on transient electromagnetic and acoustic arrays, and obtain composite wave field detection data. The feature analysis module is used to extract features from the composite wave field detection data based on the cross-physical coupling model to obtain the probability model of karst caves in the shield tunneling detection area. The cave distribution prediction module is used to determine the predicted location of the cave distribution in the construction direction of the tunnel boring machine based on the cave probability model when the cave probability model indicates that there is a cave in the construction direction of the tunnel boring machine. The operation parameter acquisition module is used to collect equipment parameters of the shield machine during the construction of the shield machine towards the predicted orientation of the karst cave, based on the multi-source sensor array mounted on the shield machine, and obtain the shield machine operation parameters. The risk assessment module is used to perform joint analysis of the composite wave field detection data and the tunnel boring machine operating parameters based on the risk assessment model, and to determine the predicted risk value of the karst cave corresponding to the predicted karst cave distribution location. The risk warning module is used to trigger a risk warning mechanism to send a warning to the central control terminal when the predicted risk value of the karst cave is greater than or equal to a preset risk prediction threshold.

9. A computer device, characterized in that, The computer device includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, it implements the steps of the method for detecting karst cave groups during shield tunneling as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, it implements the steps of the method for detecting karst cave groups during shield tunneling as described in any one of claims 1 to 7.

Images