Underground karst hidden danger intelligent detection system and method based on electric shock effect

Abstract

The invention discloses an underground karst hidden danger intelligent detection system and method based on an electric shock effect, and the method comprises the steps: transmitting an electromagnetic pulse signal to a detection region through an electromagnetic transmitting module, receiving the electromagnetic pulse signal through an electromagnetic collection module, and feeding back the electromagnetic pulse signal to an electromagnetic collection base station; the earthquake acquisition module receives earthquake wave signals and feeds back the earthquake wave signals to the earthquake acquisition base stations, and the intelligent control host carries out synchronous time synchronization with the two base stations, so that the process of simultaneously acquiring two kinds of data can be realized only through one-time electromagnetic pulse excitation without additional excitation equipment, the data acquisition is ensured to be under the same time-space condition, and the data acquisition efficiency is improved. And accurate data support is provided for subsequent non-destructive, high-efficiency and high-precision identification of urban underground hidden dangers. According to the method, the independently constructed model is adopted during data processing, whether the result of the abnormal body exists or not is judged and output, manual participation is not needed, the recognition accuracy is good, and therefore instant data support is provided for city safety management.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of geophysical exploration, and particularly to an underground karst hidden danger intelligent detection system and method based on electroseismic effect. BACKGROUND

[0002] Underground space karst safety hidden danger is one of the major risks threatening public safety and affecting road construction. The traditional detection methods (such as geological radar, shallow seismic) have obvious bottlenecks in detection: the geological radar attenuates rapidly in wet clay layer, and the detection depth is limited; the shallow seismic needs artificial source, which is difficult to implement in underground environment and has high cost. Moreover, these methods all need professional personnel to operate and process data on site, and cannot realize rapid and intelligent general survey and early warning. In addition, the above methods need different devices to collect data respectively, which not only needs a relatively complex operation process, but also cannot realize synchronous collection of two different detection data, and further cannot realize efficient and high-precision identification of underground hidden dangers.

[0003] Therefore, how to provide a new detection system and method, which can simultaneously collect seismic and electromagnetic geophysical parameters, realize non-destructive, efficient and high-precision identification of urban underground hidden dangers, and automatically identify and warn after processing data, can provide real-time data support for urban safety management, is the research direction of the present application. SUMMARY

[0004] In view of the problems existing in the prior art, the present application provides an underground karst hidden danger intelligent detection system and method based on electroseismic effect, which utilizes the reverse effect of electroseismic effect, collects seismic and electromagnetic geophysical parameters simultaneously by emitting an electromagnetic pulse, completes two collections with one excitation, thereby realizing non-destructive, efficient and high-precision identification of urban underground hidden dangers, and automatically identifying and warning after processing collected data by a constructed model, which provides real-time data support for urban safety management.

[0005] In order to achieve the above object, the technical scheme adopted by the present application is: an underground karst hidden danger intelligent detection system based on electric shock effect, comprising a mobile detection platform and a control center, wherein the mobile detection platform is provided with an intelligent control host, a remote communication module, an electromagnetic emission module, an electromagnetic collection module, an earthquake collection module, a battery and power control system and a movement control system; the electromagnetic emission module is arranged in the middle of the mobile detection platform and is used for exciting electromagnetic pulse signals below the mobile detection platform; the electromagnetic collection module is multiple, multiple electromagnetic collection modules are arranged on the mobile detection platform and are in the same horizontal plane with the electromagnetic emission module, and are used for receiving electromagnetic signals returned from the underground after the electromagnetic pulse signal emission and sending the electromagnetic signals to an electromagnetic collection base station; the earthquake collection module comprises multiple earthquake collection units arranged in the lower part of the mobile detection platform, and is used for receiving earthquake wave signals generated in the underground after the electromagnetic pulse signal emission and sending the earthquake wave signals to an earthquake collection base station; the intelligent control host is connected with the earthquake collection base station and the electromagnetic collection base station, the intelligent control host receives control instructions of the control center through the remote communication module, controls the movement of the mobile detection platform through the movement control system (i.e. motor and driving wheel), controls the excitation of the electromagnetic emission module, receives data fed back by the earthquake collection base station and the electromagnetic collection base station, automatically identifies underground abnormal bodies after model analysis and processing, and feeds back to the control center through the remote communication module to generate a three-dimensional abnormal body distribution map; and the battery and power control system is used for supplying power to the above modules.

[0006] Further, the electromagnetic collection module is five in total, one of which is at the center of the electromagnetic emission module, and the other four are uniformly distributed around the electromagnetic emission module, the electromagnetic collection module is an electromagnetic expandable telescopic coil, and the four electromagnetic expandable telescopic coils around the electromagnetic emission module are respectively extended out of the mobile detection platform during detection and retracted into the mobile detection platform during non-detection through the first electric push rod.

[0007] Further, the earthquake collection unit comprises a second electric push rod and a seismic detector, the fixed end of the second electric push rod is connected with the lower part of the mobile detection platform, and the seismic detector is arranged at the telescopic end of the second electric push rod, the second electric push rod is in the retracted state when not detecting, at this time the seismic detector does not contact the ground; the second electric push rod is in the extended state when detecting, at this time the seismic detector is in close contact with the ground. This arrangement can arrange the earthquake collection units in an array under the mobile detection platform, and improve the accuracy of receiving earthquake data.

[0008] Further, the mobile detection platform is provided with a laser radar, a sound wave radar, a GPS positioning module, an electronic gyroscope and an IMU inertial measurement unit. The laser radar and the sound wave radar are used to monitor whether there is an obstacle around the mobile detection platform and feed back to the intelligent control host. The intelligent control host confirms the current position and attitude of the mobile detection platform through the GPS positioning module, the electronic gyroscope and the IMU inertial measurement unit after processing, and controls the mobile detection platform to take an obstacle avoidance measure. The mobile detection platform can be remotely controlled by the control center, or can adopt the above-mentioned automatic obstacle avoidance driving mode, and the above-mentioned mode is preferred. In this way, personnel remote control is not required, manpower is saved, and the delay caused by remote control is overcome.

[0009] Further, the mobile detection platform is provided with a Beidou clock module. The Beidou clock module is used to synchronize time with the electromagnetic collection base station, the seismic collection base station and the intelligent control host, so as to ensure the uniformity of the time stamp of the collected data.

[0010] Further, the mobile detection platform is provided with a touch control screen. The touch control screen is used to display the system state in real time, receive the input instructions of the field operator during field operation, and visually display the real-time processing results or alarm information of the current detection point.

[0011] The working method of the above-mentioned underground karst hidden danger intelligent detection system based on the electric seismic effect comprises the following steps:

[0012] Step one, detection system layout and transient electromagnetic excitation: the intelligent control host controls the mobile detection platform to travel along the planned travel route when the mobile detection platform travels to the first detection point, and then the intelligent control host receives the control instruction of the control center through the remote communication module, so that the electromagnetic emission module excites the electromagnetic pulse signal.

[0013] Step two, transient electromagnetic reception: the electromagnetic collection module collects the electromagnetic signal returned from the underground after the electromagnetic pulse signal is emitted and sends it to the electromagnetic collection base station.

[0014] Step three, seismic wave reception: when the electromagnetic pulse propagates to the karst area underground, it can act on the movable ions in the solid-liquid interface double electric layer of the karst area. The ions drag the pore fluid to move through the viscous force, thereby generating a body force on the surrounding rock mass, exciting a seismic wave. At this time, the seismic collection module collects the seismic wave signal returned from the underground and sends it to the seismic collection base station.

[0015] Step four, data acquisition of the detection area: the intelligent control host obtains the data fed back by the seismic acquisition base station and the electromagnetic acquisition base station, completes the data acquisition of the first detection point, then makes the mobile detection platform continue to travel along the planned travel route to each subsequent detection point, and repeats steps one to three respectively, so as to complete the data acquisition of all detection points in the detection area.

[0016] Step five: data processing and determination of urban karst hidden dangers: the intelligent control host denoises and restores the gain of the electromagnetic data and the seismic wave data obtained at each detection point, extracts time domain, frequency domain and time-frequency joint features, inputs the extracted features into the abnormal body recognition model constructed, the model outputs whether there is a geological abnormal body underground at each detection point, after the identification of all detection points is completed, the identification result is sent to the control center through the remote communication module, and a three-dimensional abnormal body distribution map of the underground of the detection area is generated.

[0017] Further, the abnormal body recognition model constructed in step five is specifically:

[0018] A, collect historical detection samples containing electromagnetic data E(t) and seismic wave data S(t) to form a data set , wherein is the electromagnetic and seismic wave data sequence of the i-th sample, and the label represents whether there is an abnormal body (1 represents existence, and 0 represents nonexistence); and the electromagnetic data and the seismic wave data are preprocessed respectively.

[0019] B, feature extraction is performed on the data set:

[0020] Electromagnetic signal features: exponential fitting is adopted to extract the decay time constant , the power spectral density is calculated, and the main frequency is extracted: , the time-frequency distribution is obtained through continuous wavelet transform .

[0021] Seismic wave signal features: the first arrival time t0, the maximum amplitude S max , and the root mean square amplitude: are extracted, the spectral envelope is calculated, the center frequency f c and the frequency bandwidth B S are extracted, and the frequency spectrum is obtained through short-time Fourier transform: .

[0022] C, the electromagnetic signal features and the seismic wave signal features extracted in step B are combined into a fusion feature vector: , principal component analysis (PCA) is adopted to reduce the dimension of the features: , wherein is the fusion feature vector; Wpca The transformation matrix is the principal component analysis, and 95% of the variance information is reserved.

[0023] D, a classification model based on deep learning is constructed, which includes an input layer, a hidden layer and an output layer, and a cross-entropy loss function is determined, and the feature vector F pca After the classification model is trained as the training data, an abnormal body recognition model is obtained.

[0024] Further, the received electromagnetic signal carries the electrical structure of the underground medium, and by analyzing it, it can be judged whether there is a cavity and water body in the underground; the received seismic wave signal carries the porosity, permeability and fluid saturation information of the underground medium, and by analyzing it, the compactness and wave impedance interface of the underground medium can be judged, and the combination of the two ultimately realizes the judgment of the underground abnormal body.

[0025] The core innovation principle of the application is: through research, it is found that the transient electromagnetic emission source emits a differential electromagnetic pulse to the underground, and on the ground, not only the returned electromagnetic pulse signal can be received, but also the returned seismic wave signal can be received, through further research, it is found that in the process of electromagnetic pulse propagation in the underground, it will act on the movable ions in the solid-liquid interface double electric layer of the karst area, the ions drag the pore fluid movement through the viscous force, thereby generating body force on the solid skeleton, and exciting seismic waves. This seismic wave carries the porosity, permeability and fluid saturation information of the medium below the excitation point; after the following formula derivation, the specific relationship between the electromagnetic pulse and the seismic wave can be obtained:

[0026] The electromagnetic emission source emits a transient electromagnetic pulse to the underground, and the current waveform is a differential pulse with fast turn-off, and the turn-off time τ_off=4μs. The magnetic moment of the transmitting coil is M(t)=N·I(t)·A. Wherein N is the number of turns of the coil, A is the area of the coil, and I is the current.

[0027] The eddy current electric field induced by the changing magnetic moment in the underground conductive medium is E_induced. The electric field strength is related to And attenuates with the increase of depth and medium conductivity. In the double electric layer existing in the solid-liquid interface of the karst area, the induced eddy current electric field E_induced produces coulomb force on the movable ions q in the diffusion layer. The movement of the ions is mainly dominated by the viscous force γ, and the movement speed satisfies γ·(dx / dt)≈q·E_induced. The movement of the ions drags the pore fluid through the viscous effect, and generates electroosmotic flow.

[0028] The movement of the fluid further produces stress F_seismic on the solid skeleton. The stress as the source term of the excited elastic wave (seismic wave) satisfies the wave equation:

[0029] .

[0030] The solution of the wave equation represents the propagation of seismic waves, and the P-wave and S-wave velocities are V_p = sqrt[(lambda + 2mu) / rho] and V_s = sqrt(mu / rho), respectively. By analyzing the characteristics of the received seismic wave signals, the elastic parameters (lambda, mu, rho) of the medium can be inverted, and it can be judged whether there is a geological anomaly body such as a karst cavity.

[0031] Through the above research, the technical scheme of the present application is generated, that is, only one electromagnetic pulse excitation is needed to obtain the electromagnetic signals reflected from the underground and the reverse effect of the electric shock effect caused by the electromagnetic pulse propagation, and then the generated seismic waves are excited, wherein the electromagnetic signals carry the electrical structure of the underground medium, and by analyzing the same, it can be judged whether there is a cavity and water body in the underground; the received seismic wave signals carry the porosity, permeability and fluid saturation information of the underground medium, and by analyzing the same, the compactness and wave impedance interface of the underground medium can be judged, and the combination of the two ultimately realizes the judgment of the underground anomaly body.

[0032] Compared with the prior art, the present application has the following advantages:

[0033] 1. The present application is found through research that an electromagnetic emission module is used to emit an electromagnetic pulse signal to the underground of the detection area, and an electromagnetic collection module is used to receive the electromagnetic pulse signal emitted from the underground and feed back to the electromagnetic collection base station, and a seismic collection module is used to receive the seismic wave signal generated in the underground after the electromagnetic pulse signal is emitted and feed back to the seismic collection base station, and the intelligent control host synchronously time-ensures with the two base stations to ensure that the time stamps of the collected data of the two are unified, so that only one electromagnetic pulse excitation can receive two kinds of detection data (i.e. electromagnetic data and seismic data); and the two kinds of data carry different geological information, without the need for additional excitation equipment, the process of simultaneously collecting two kinds of data is realized, the data collection is ensured under the same space-time condition, and accurate data support is provided for subsequent non-destructive, efficient and high-precision identification of urban underground hidden dangers.

[0034] 2. The present application adopts an abnormity body identification model constructed independently when processing data, which can analyze the two kinds of data respectively and comprehensively judge whether there is an abnormity body, and the identification accuracy is good, so as to provide instant data support for city safety management. DETAILED DESCRIPTION

[0035] Figure 1 is a schematic diagram of the overall external structure of the present application.

[0036] Figure 2 is a schematic diagram of the internal structure of the present application. Figure 1

[0037] Figure 3 is a schematic diagram of the internal structure of the present application. Figure 1 ​Structure diagram of the electromagnetic acquisition module.

[0038] Figure 4 is Figure 1 Structure diagram of the seismic acquisition module.

[0039] In the figure, 1, laser radar, 2, touch control screen, 3, car light, 4, electromagnetic emission module, 5, sound wave radar, 6, seismic acquisition module, 7, remote communication module, 8, GPS positioning module, 9, battery and power control system, 10, Beidou clock module, 11, electronic gyroscope, 12, electromagnetic acquisition base station, 13, seismic acquisition base station, 14, intelligent control host, 15, electromagnetic acquisition module, 16, first electric push rod, 17, seismic acquisition unit, 18, second electric push rod, 19, seismic detector, 20, mobile control system. DETAILED DESCRIPTION

[0040] The application will be further described below.

[0041] As Figure 1 and Figure 2 shown, an underground karst hidden danger intelligent detection system based on electric shock effect includes a mobile detection platform and a control center, wherein the mobile detection platform is provided with an intelligent control host 14, a remote communication module 7, an electromagnetic emission module 4, an electromagnetic acquisition module 15, a seismic acquisition module 6, and a battery and power control system 9; the electromagnetic emission module 4 is installed in the middle of the mobile detection platform and is used for exciting electromagnetic pulse signals below the mobile detection platform; an electromagnetic pulse emission source adopts a SiCMOSFET transmitter, the off time of which is controlled to be 4 μs to generate a higher di / dt and enhance the excited electric field intensity; and the excited pulse waveform adopts a differential pulse, which uses harmonic components to excite broadband seismic waves; the pulse emission power is 100 watts to 2 kilowatts, which can be switched in multiple gears; and the emission frequency is 100 Hz to 10 kHz.

[0042] As Figure 3As shown, the electromagnetic acquisition module 15 is multiple, multiple electromagnetic acquisition modules 15 are installed on the mobile detection platform, and are in the same horizontal plane with the electromagnetic emission module 4, for receiving the electromagnetic signal returned from the underground after the electromagnetic pulse signal is emitted and sending to the electromagnetic acquisition base station 12; the seismic acquisition module 6 includes multiple seismic acquisition units 17 arranged at the lower part of the mobile detection platform, for receiving the seismic wave signal generated in the underground after the electromagnetic pulse signal is emitted and sending to the seismic acquisition base station 13; the mobile detection platform is provided with a Beidou clock module 10, the Beidou clock module 10 is used for synchronizing the time of the electromagnetic acquisition base station 12, the seismic acquisition base station 13 and the intelligent control host 14, and ensuring the time stamp of the collected data uniform; the mobile detection platform is provided with a touch control screen 2, for real-time display of system state, input instructions received by the on-site operator during on-site operation, and visual display of real-time processing results or alarm information of the current detection point. The intelligent control host 14 is connected with the seismic acquisition base station 13 and the electromagnetic acquisition base station 12, the intelligent control host 14 receives the control instruction of the control center through the remote communication module, controls the movement of the mobile detection platform through the action control system 20 (i.e. motor and driving wheel), controls the excitation of the electromagnetic emission module 4, receives the data fed back by the seismic acquisition base station 13 and the electromagnetic acquisition base station 12, automatically identifies the underground abnormal body after model analysis and processing through the built-in model, and feeds back to the control center to generate a three-dimensional abnormal body distribution map through the remote communication module; the battery and power control system 9 are used for power supply for the above-mentioned modules.

[0043] As an improvement of the present application, the electromagnetic acquisition module 15 is five in total, one of which is at the center of the electromagnetic emission module 4, and the other four are uniformly distributed around the electromagnetic emission module 4, the electromagnetic acquisition module 15 is an electromagnetic expandable telescopic coil, and the four electromagnetic expandable telescopic coils around are respectively extended out of the mobile detection platform during detection and retracted into the mobile detection platform during non-detection through the first electric push rod 16. The standard mode of the electromagnetic expandable telescopic coil is designed as 0.5m*0.5m, and a modular interface is added, which can expand the size of the coil according to the needs.

[0044] As Figure 4As shown, the seismic acquisition units 17 are arranged in an array under the mobile detection platform, each seismic acquisition unit 17 includes a second electric push rod 18 and a geophone 19, the fixed end of the second electric push rod 18 is connected with the lower part of the mobile detection platform, and the geophone 19 is installed at the telescopic end of the second electric push rod 18; when not detecting, the second electric push rod 18 is in a retracted state, and at this time the geophone 19 does not contact the ground; when detecting, the second electric push rod 18 is in an extended state, and at this time the geophone 19 is in close contact with the ground. Since the earthquake caused by the electromagnetic pulse is relatively weak, the geophone 19 adopts a three-component MEMS accelerometer. Range: ±2g or ±5g. Noise density: <100µg / √Hz. Bandwidth: DC (0Hz to 800Hz). In this way, the accuracy of receiving the seismic wave signal can be effectively ensured.

[0045] As another improvement of the present application, the mobile detection platform is provided with a laser radar 1, a sound wave radar 5, a GPS positioning module 8, an electronic gyroscope 11 and an IMU inertial measurement unit; the laser radar 1 and the sound wave radar 5 are used to monitor whether there is an obstacle around the mobile detection platform and feed back to the intelligent control host 14, the intelligent control host 14 confirms the current position and attitude of the mobile detection platform through the GPS positioning module 8, the electronic gyroscope 11 and the IMU inertial measurement unit after processing, and controls the mobile detection platform to take obstacle avoidance measures.

[0046] The working method of the above-mentioned underground karst hidden danger intelligent detection system based on the electroseismic effect includes the following steps:

[0047] Step one, detection system layout and transient electromagnetic excitation: drive the planned route and detection points on the ground where city karst hidden danger detection is required, the intelligent control host 14 controls the mobile detection platform to drive along the planned route, stops when driving to the first detection point, and then the intelligent control host 14 receives the control instruction of the control center through the remote communication module 7, so that the electromagnetic emission module 4 excites the electromagnetic pulse signal;

[0048] Step two, transient electromagnetic reception: the electromagnetic acquisition module 15 collects the electromagnetic signal returned from the underground after the electromagnetic pulse signal is emitted and sends it to the electromagnetic acquisition base station 12;

[0049] Step three, seismic wave reception: when the electromagnetic pulse propagates to the karst area underground, it can act on the movable ions in the double electric layer of the solid-liquid interface of the karst area, the ions drag the pore fluid to move through the viscous force, thereby generating a body force on the surrounding rock mass, and exciting a seismic wave, at this time the seismic acquisition module 6 collects the seismic wave signal returned from the underground and sends it to the seismic acquisition base station 13;

[0050] Step four, data acquisition of the detection area: the intelligent control host 14 acquires the data fed back by the seismic acquisition base station 13 and the electromagnetic acquisition base station 12, completes the data acquisition of the first detection point, then makes the mobile detection platform continue to travel along the planned travel route to each subsequent detection point, and repeats steps one to three respectively, so as to complete the data acquisition of all detection points in the detection area; wherein the received electromagnetic signal carries the electrical structure of the underground medium, and through analysis thereof, it can be judged whether there is a cavity and water body in the underground; the received seismic wave signal carries the porosity, permeability and fluid saturation information of the underground medium, and through analysis thereof, the compactness and wave impedance interface of the underground medium can be judged, and the combination of the two finally realizes the judgment of the underground abnormal body.

[0051] Step five: data processing and determination of urban karst hidden dangers: the intelligent control host 14 denoises and restores the gain of the electromagnetic data and the seismic wave data acquired at each detection point, extracts time domain, frequency domain and time-frequency joint features, constructs an abnormal body recognition model, and the specific process is as follows:

[0052] A, collect historical detection samples containing electromagnetic data E(t) and seismic wave data S(t) to form a data set , wherein is the electromagnetic and seismic wave data sequence of the i-th sample, and the label indicates whether there is an abnormal body (1 indicates that there is, and 0 indicates that there is not); and the electromagnetic data and the seismic wave data are preprocessed respectively, and the specific process is as follows:

[0053] Electromagnetic signal preprocessing , wherein μ E and σ E are the mean and standard deviation of the electromagnetic signal respectively; the seismic wave signal preprocessing is as follows: , wherein T is the signal length.

[0054] B, feature extraction of the data set:

[0055] Electromagnetic signal features: exponential fitting is adopted to extract the decay time constant , calculate the power spectral density , and extract the main frequency: , get the time-frequency distribution through continuous wavelet transform ; wherein E0 is the initial amplitude of the electromagnetic signal; F{·} is the Fourier transform operator; is the mother wavelet function; a is the wavelet scale parameter; and b is the wavelet translation parameter.

[0056] Seismic wave signal features: extract the first arrival time t0, the maximum amplitude S max , and the root mean square amplitude: , calculate the spectral envelope, and extract the center frequency f cAnd the frequency band width B S , the frequency spectrum is obtained by short-time Fourier transform: ; Where w is the window function.

[0057] C, the electromagnetic signal features and seismic wave signal features extracted in step B are combined into a fusion feature vector: , principal component analysis (PCA) is used to reduce the dimension of the features: , wherein, is the fusion feature vector; W pca is the transformation matrix of principal component analysis, which retains 95% of the variance information.

[0058] D, a classification model based on deep learning is constructed, which includes an input layer, a hidden layer and an output layer;

[0059] Input layer: receive the reduced dimension feature vector F pca ;

[0060] Hidden layer: two fully connected networks are used;

[0061]

[0062]

[0063] Output layer: the classification probability is output by the softmax function;

[0064]

[0065] The cross-entropy loss function is determined as:

[0066]

[0067] In the above formulas, W i is the weight vector, b i is the bias vector, ReLU is the linear unit activation function, N is the total number of training samples, y i is the true label, is the predicted probability, and L is the cross-entropy loss value.

[0068] The reduced dimension feature vector F pca is used as training data and the classification model is trained using the Adam optimizer. After testing and verification, an abnormal body recognition model is obtained.

[0069] The extracted features are input into the constructed abnormal body recognition model, and the model outputs whether there is a geological anomaly body at each detection point. After identifying all detection points, the recognition results are sent to the control center through the remote communication module 7, and a three-dimensional anomaly body distribution map of the detection area is generated.

[0070] The abnormal body recognition model can add each newly collected and manually verified data to the training set, periodically performs incremental learning on the model, and updates the model, and the specific formula is:

[0071]

[0072] In the formula, is an old loss function value; is a new loss function value; and λ is a regularization hyperparameter.

[0073] The above only describes the preferred embodiments of the present application, and it should be noted that those skilled in the art can make several improvements and refinements without departing from the principles of the present application, and these improvements and refinements should also be considered within the scope of protection of the present application.

Claims

1. An underground karst hidden danger intelligent detection system based on an electric shock effect, characterized in that, The mobile detection platform and the control center are connected, the intelligent control host receives the control instruction of the control center through the remote communication module, controls the movement of the mobile detection platform through the action control system, controls the excitation of the electromagnetic emission module, receives the data feedback by the seismic acquisition base station and the electromagnetic acquisition base station, automatically identifies the underground abnormal body after the built-in model analysis and processing, and feeds back to the control center to generate a three-dimensional abnormal body distribution map through the remote communication module. The battery and power control system is used for supplying power to the above-mentioned modules. The electromagnetic acquisition module includes five electromagnetic expandable telescopic coils, one of which is at the center of the electromagnetic emission module, and the other four are uniformly distributed around the electromagnetic emission module. The seismic acquisition unit includes a second electric push rod and a seismic detector, the fixed end of the second electric push rod is connected with the lower part of the mobile detection platform, and the seismic detector is installed at the telescopic end of the second electric push rod. The mobile detection platform is equipped with a laser radar, a sound wave radar, a GPS positioning module, an electronic gyroscope and an IMU inertial measurement unit, the laser radar and the sound wave radar are used to monitor whether there is an obstacle around the mobile detection platform and feed back to the intelligent control host, the intelligent control host processes and confirms the current position and attitude of the mobile detection platform through the GPS positioning module, the electronic gyroscope and the IMU inertial measurement unit, and controls the mobile detection platform to take obstacle avoidance measures. The mobile detection platform is equipped with a Beidou clock module, which is used to synchronize the time of the electromagnetic acquisition base station, the seismic acquisition base station and the intelligent control host, and ensure the uniformity of the data time stamp.

2. The underground karst hidden trouble intelligent detection system based on electric shock effect according to claim 1, characterized in that, The mobile detection platform is equipped with a touch control screen.

3. The underground karst hidden trouble intelligent detection system based on electric shock effect according to claim 1, characterized in that, The following steps are included:

4. The underground karst hidden trouble intelligent detection system based on electric shock effect according to claim 1, characterized in that, ​ 5. The underground karst hidden trouble intelligent detection system based on electric shock effect according to claim 1, characterized in that, ​ 6. The underground karst hidden trouble intelligent detection system based on electric shock effect according to claim 1, characterized in that, ​ 7. The working method of the underground karst hidden trouble intelligent detection system based on the electric shock effect according to claim 1, characterized in that, ​ Step one, detection system layout and transient electromagnetic excitation: drive along the planned route and detection points on the ground where urban karst hidden danger detection is required, the intelligent control host controls the mobile detection platform to drive along the planned route, stops when it reaches the first detection point, then the intelligent control host receives the control instructions from the control center through the remote communication module, and the electromagnetic emission module emits electromagnetic pulse signals; Step two, transient electromagnetic reception: the electromagnetic acquisition module collects the electromagnetic signals returned from the underground after the electromagnetic pulse signal emission and sends them to the electromagnetic acquisition base station; Step three, seismic wave reception: when the electromagnetic pulse propagates to the karst area underground, it can act on the movable ions in the solid-liquid interface double electric layer of the karst area, and the ions drag the pore fluid to move through the viscous force, thereby generating body force on the surrounding rock mass, and exciting seismic waves. At this time, the seismic acquisition module collects the seismic wave signals returned from the underground and sends them to the seismic acquisition base station; Step four, detection area data acquisition: the intelligent control host obtains the data fed back by the seismic acquisition base station and the electromagnetic acquisition base station, completes the data acquisition of the first detection point, then makes the mobile detection platform continue to drive along the planned route to each subsequent detection point, and repeats steps one to three respectively, thereby completing the data acquisition of all detection points in the detection area; Step five: data processing and determination of urban karst hidden danger: the intelligent control host denoises and restores the gain of the electromagnetic data and seismic wave data obtained at each detection point, extracts time domain, frequency domain and time-frequency joint features, inputs the extracted features into the constructed abnormal body recognition model, and the model outputs whether there is a geological abnormal body underground at each detection point. After identifying all detection points, the recognition results are sent to the control center through the remote communication module to generate a three-dimensional abnormal body distribution map of the underground in the detection area.

8. The method of working according to claim 7, characterized in that, The abnormal body recognition model constructed in step five is as follows: A, collect historical detection samples containing electromagnetic data E(t) and seismic wave data S(t) to form a data set; and preprocess the electromagnetic data and seismic wave data respectively; B, feature extraction on the data set: Electromagnetic signal features: exponential fitting is used to extract the decay time constant, power spectral density is calculated, the main frequency is extracted, and the time-frequency distribution is obtained through continuous wavelet transform; Seismic wave signal features: extract the first arrival time, maximum amplitude, root mean square amplitude, calculate the spectral envelope, extract the center frequency and frequency bandwidth, and obtain the frequency spectrum through short-time Fourier transform; C. combine the electromagnetic signal features and the seismic wave signal features extracted in step B into a fusion feature vector, and perform dimension reduction on the features by using principal component analysis: wherein, is the fusion feature vector; W pca is a transformation matrix of the principal component analysis; D、constructing a classification model based on deep learning, which includes an input layer, a hidden layer and an output layer, and determining a cross-entropy loss function, the feature vector F after dimensionality reduction pca After training the classification model as training data, an abnormal body recognition model is obtained.

9. The method of claim 7, wherein, The received electromagnetic signal carries the electrical structure of the underground medium, which can be used to judge whether there is a cavity and water body underground; the received seismic wave signal carries the porosity, permeability and fluid saturation information of the underground medium, which can be used to judge the compactness and wave impedance interface of the underground medium. The combination of the two ultimately realizes the judgment of the underground abnormal body.

Images