A radar-based tunnel wall leakage detection method and related apparatus

Abstract

This application discloses a radar-based method and related device for detecting tunnel wall leakage, relating to the field of tunnel engineering operation and maintenance. The method acquires echo signals reflected from the detection area of ​​the tunnel wall at multiple acquisition times; analyzes each echo signal to obtain wall surface state data; compares multiple wall surface state data with pre-acquired wall surface dryness data to obtain state difference values, which can eliminate background interference from the inherent material and structure of the tunnel wall, highlighting abnormal differences caused by leakage, and improving the stability and accuracy of detection; by inputting the state difference values ​​into a pre-constructed leakage discrimination model to determine whether the detection area is leaking, automated identification and judgment can be achieved. It can stably and accurately complete leakage detection in complex environments such as dim lighting, dust, fog, fluctuating temperature and humidity, and changing wind speeds, effectively solving the problems of poor environmental adaptability and low detection accuracy in existing technologies.

Description

Technical Field

[0001] This application relates to the field of tunnel engineering operation and maintenance technology, and in particular to a radar-based method and related device for detecting tunnel wall leakage. Background Technology

[0002] As a crucial component of transportation infrastructure, tunnels are highly susceptible to water seepage and leakage during long-term operation due to factors such as groundwater infiltration, lining cracks, and construction joint defects. This problem not only directly reduces tunnel driving safety and traffic environment safety but also triggers a chain reaction of damage, including lining concrete deterioration and steel reinforcement corrosion, seriously threatening the overall structural stability of the tunnel and shortening its service life. The leakage problem is particularly pronounced during rainy seasons or in high-humidity environments, posing a significant challenge to tunnel safety management and maintenance.

[0003] Currently, tunnel wall leakage detection methods include optical imaging and infrared thermal imaging. Optical imaging captures tunnel wall images using a high-definition camera and combines this with image processing algorithms to automatically identify leaks. Infrared thermal imaging utilizes the temperature difference between a leaking wall and a dry wall, capturing infrared thermal images of the tunnel wall to identify areas of abnormal temperature and thus determine the presence of leaks. Both existing methods have significant limitations. Optical imaging relies on lighting conditions to obtain tunnel wall images; in dimly lit, dusty, or foggy environments, it is difficult to capture clear images, making accurate leak detection impossible. Infrared thermal imaging can identify abnormal tunnel wall temperatures, but the complex environment inside a tunnel, such as the significant influence of ambient temperature and wind speed, leads to inaccurate leak detection.

[0004] In summary, as the safety requirements for tunnel construction and operation continue to increase, there is an urgent need for a leakage detection technology that can identify whether the tunnel wall is leaking in the complex environment of the tunnel. Summary of the Invention

[0005] In view of the above problems, this application provides a radar-based method and related device for detecting tunnel wall leakage, aiming to achieve the goal of leakage detection technology that can identify whether the tunnel wall is leaking in the complex environment of a tunnel. The specific solution is as follows:

[0006] The first aspect of this application provides a radar-based method for detecting tunnel wall leakage, comprising:

[0007] The echo signals acquired at multiple acquisition times are obtained, and the echo signals are the signals reflected by the transmitted signal through the detection area of ​​the tunnel wall;

[0008] For each of the echo signals, the echo signal is analyzed to obtain the wall surface state data of the detection area, so as to obtain multiple wall surface state data.

[0009] The state difference value is obtained by comparing the multiple wall surface state data with the pre-obtained wall surface drying data.

[0010] The state difference is input into a pre-built leakage discrimination model to determine whether the detection area is in a leakage state.

[0011] In one possible implementation, the step of parsing the echo signal to obtain the wall surface state data of the detection area for each echo signal includes:

[0012] Leakage data of the detection area at various acquisition times are obtained. The leakage parameters include the reflection intensity characterizing the dielectric constant of the detection area, the phase offset between the echo signal and the transmitted signal, and the Doppler frequency shift characterizing the relative motion between the liquid attached to the detection area and the detection area.

[0013] Statistical feature sets are obtained based on the leakage data at each collection time.

[0014] For each acquisition time, the energy distribution characteristics of the echo signal acquired at that acquisition time in different frequency ranges are obtained;

[0015] For each acquisition time, time-frequency energy parameters are obtained based on the time-frequency spectrum of the echo signal acquired at the acquisition time. The time-frequency energy parameters include the instantaneous energy peak value, peak frequency, and energy duration greater than or equal to a preset threshold in the time-frequency spectrum. The time-frequency spectrum is a three-dimensional distribution map of time, frequency, and energy.

[0016] In one possible implementation, obtaining the energy distribution characteristics of the echo signal acquired at the acquisition time in different frequency ranges includes:

[0017] The echo signal acquired at the acquisition time is subjected to frequency domain transformation to obtain a frequency domain signal;

[0018] The frequency domain signal is divided into multiple frequency ranges according to a preset frequency interval, and the signal energy corresponding to each frequency range is calculated.

[0019] Based on the signal energy corresponding to each of the aforementioned frequency ranges, the energy distribution characteristics of the echo signal in different frequency ranges are obtained.

[0020] In one possible implementation, obtaining the time-frequency energy parameters based on the time-spectrum diagram of the echo signal acquired at the acquisition time includes:

[0021] Time-frequency analysis is performed on the echo signals acquired at the acquisition time to construct a three-dimensional time-frequency spectrum with time as the horizontal axis, frequency as the vertical axis, and signal energy as the amplitude.

[0022] The maximum value of the signal energy in the three-dimensional time-frequency spectrum is determined to be the instantaneous energy peak value;

[0023] The frequency corresponding to the instantaneous energy peak value is determined as the peak frequency;

[0024] The duration of continuous time during which the signal energy in the three-dimensional time-spectrum graph is greater than or equal to a preset threshold is defined as the energy duration.

[0025] In one possible implementation, after the step of inputting the state difference into a pre-built leakage discrimination model to determine whether the detection area is in a leakage state, the method further includes:

[0026] The local leakage rate is obtained based on the wall surface state data corresponding to multiple acquisition times, and the local leakage rate characterizes the change law of the wall surface state data over time.

[0027] The seepage trend is obtained based on the change in the wall surface state data at each pair of adjacent acquisition times. The seepage trend characterizes the change pattern of the wall surface state data over time.

[0028] Based on each leakage area that includes the same local area at multiple acquisition times, the leakage range expansion rate of the leakage area is obtained. The leakage area includes the detection area that is spatially adjacent. The leakage range expansion rate characterizes the change law of the area of ​​the leakage area over time.

[0029] One possible implementation also includes:

[0030] The local leakage rate, the seepage trend, and the leakage range expansion rate are input into a pre-constructed leakage trend prediction model. The leakage rate of the detection area and the area of ​​the leakage area at future times are obtained through the leakage trend prediction model.

[0031] One possible implementation also includes:

[0032] The leakage level is determined based on the number of leaking detection areas, the local leakage rate corresponding to each detection area, the seepage trend corresponding to each detection area, the state difference corresponding to each detection area, and the leakage range expansion rate corresponding to each leaking area.

[0033] In one possible implementation, the step of obtaining the seepage trend based on the change in the wall surface state data at each pair of adjacent acquisition times includes:

[0034] The wall surface state data corresponding to the multiple acquisition times are sorted from earliest to latest according to the acquisition time to obtain a first sorting result;

[0035] The changes in each pair of adjacent wall surface state data in the first sorting result constitute time series data.

[0036] The seepage trend is determined based on the time series data.

[0037] In one possible implementation, the step of obtaining the leakage range expansion rate of a leakage area based on various leakage areas that contain the same local area at multiple acquisition times includes:

[0038] For any of the aforementioned acquisition times, the detection areas that are in a leaking state and have adjacent spatial coordinates at the acquisition time are divided into the same leaking area;

[0039] Each leakage area that contains the same local area at each of the aforementioned acquisition times is divided into the same leakage area set;

[0040] Obtain area change rate sequences corresponding to multiple sets of leakage areas, wherein the area change rate sequence includes: the area change rate of the leakage area set ordered sequentially from earliest to latest according to the collection time, and the area change rate is the ratio of the difference in area of ​​the leakage area corresponding to two adjacent collection times to the time interval between the two adjacent collection times;

[0041] The rate of expansion of the leakage range is obtained based on the area change rate sequence.

[0042] A second aspect of this application provides a radar-based tunnel wall leakage detection device, comprising:

[0043] The first acquisition module is used to acquire echo signals acquired at multiple acquisition times, wherein the echo signals are signals reflected by the transmitted signals through the detection area of ​​the tunnel wall;

[0044] The second acquisition module is used to parse the echo signal for each echo signal to obtain the wall surface state data of the detection area, so as to obtain multiple wall surface state data.

[0045] The third acquisition module is used to compare the multiple wall surface state data with the pre-obtained wall surface drying data to obtain the state difference value.

[0046] The judgment module is used to input the state difference into a pre-built leakage discrimination model to determine whether the detection area is in a leakage state.

[0047] A third aspect of this application provides a computer program product including computer-readable instructions that, when executed on an electronic device, cause the electronic device to implement the radar-based tunnel wall leakage detection method described in the first aspect or any implementation thereof.

[0048] A fourth aspect of this application provides an electronic device, including at least one processor and a memory connected to the processor, wherein:

[0049] The memory is used to store computer programs;

[0050] The processor is used to execute the computer program so that the electronic device can implement the radar-based tunnel wall leakage detection method of the first aspect or any implementation thereof.

[0051] The fifth aspect of this application provides a computer storage medium carrying one or more computer programs, which, when executed by an electronic device, enable the electronic device to implement the radar-based tunnel wall leakage detection method described in the first aspect or any implementation thereof.

[0052] By employing the above technical solution, this application provides a radar-based method for detecting tunnel wall leakage. By acquiring echo signals reflected from the detection area of ​​the tunnel wall at multiple acquisition times, it enables continuous temporal acquisition of the tunnel wall's condition, eliminating dependence on lighting conditions. By analyzing each echo signal to obtain wall surface condition data, it can accurately extract the physical characteristics and electromagnetic response information of the wall medium, unaffected by factors such as tunnel ambient temperature and wind speed, overcoming the shortcomings of infrared thermal imaging detection, which is easily affected by environmental factors leading to inaccurate detection. Furthermore, by comparing multiple wall surface condition data with pre-acquired wall surface interference data... By comparing the noise data to obtain the state difference value, the background interference of the inherent material and structure of the tunnel wall can be eliminated, highlighting the abnormal differences caused by leakage and improving the stability and accuracy of detection. By inputting the state difference value into the pre-built leakage discrimination model to determine whether the detection area is leaking, automated and intelligent identification and judgment can be achieved. It can stably and accurately complete leakage detection in complex environments such as dim lighting, dust, fog, temperature and humidity fluctuations, and wind speed changes in tunnels. It effectively solves the problems of poor environmental adaptability and low detection accuracy of existing technologies, and significantly improves the reliability and versatility of tunnel wall leakage identification. Attached Figure Description

[0053] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0054] Figure 1 A schematic diagram of a system architecture is provided for this application;

[0055] Figure 2 A schematic flowchart of a radar tunnel wall leakage detection method provided in an embodiment of this application;

[0056] Figure 3 A schematic diagram of a radar-based tunnel wall leakage detection device provided in an embodiment of this application;

[0057] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0058] The embodiments of this application are described below with reference to the accompanying drawings. The terminology used in the implementation section of this application is for explaining specific embodiments only and is not intended to limit the scope of this application.

[0059] The embodiments of this application will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are equally applicable to similar technical problems.

[0060] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of elements is not necessarily limited to those elements, but may include other elements not explicitly listed or inherent to those processes, methods, products, or apparatuses.

[0061] This application can be applied to the fields of tunnel engineering operation and maintenance, transportation infrastructure safety monitoring, or underground engineering quality inspection. The following will introduce several application scenarios for the product, taking highway tunnels, railway tunnels, and urban underground integrated pipe corridors as examples.

[0062] First, let's introduce the application scenarios of this application.

[0063] This application can be applied, but is not limited to, to applications with radar-based tunnel wall leakage detection capabilities or cloud services provided by cloud-side servers, which will be described in detail below:

[0064] See Figure 1 , Figure 1 A schematic diagram of a system architecture is shown. The system may include a terminal 100, a server 200, and multiple radars 300. The server 200 can provide the methods provided in the embodiments of this application to one or more terminals.

[0065] Among them, Radar 300 can be a millimeter-wave lidar or an FMCW (Frequency-Modulated Continuous Wave) radar.

[0066] Among them, the radar 300 can be designed with two deployment methods: fixed deployment and mobile mounting.

[0067] For example, the radar deployed can scan all vulnerable areas of the tunnel wall, such as sidewalls, arches, lining joints, bolt holes, and around pipelines.

[0068] For example, multiple radars can be fixedly installed on key parts such as both sides of the tunnel cross-section to achieve long-term online monitoring. For instance, the radars can be fixed to the embedded parts of the tunnel lining by brackets, and the brackets can be made vibration-damped (such as by adding rubber pads) to prevent the radars from shifting due to vibrations from passing vehicles.

[0069] For example, the radar housing is fitted with a waterproof and dustproof cover to adapt to the high humidity and dusty environment inside the tunnel.

[0070] For example, radar can be installed on mobile maintenance equipment to achieve full-line coverage scanning, such as on tunnel track inspection vehicles, highway tunnel inspection vehicles, and tunnel inspection robots.

[0071] For example, multiple radars can form a radar array, which can detect the tunnel wall at a preset period to acquire echo signals. The detection frequency of the preset period can be adjusted according to actual needs, satisfying both long-term trend monitoring and short-term high-frequency detection.

[0072] The terminal 100 may be equipped with a tunnel wall leakage detection and maintenance management application. The application and webpage can provide an interface. The terminal 100 can receive relevant parameters transmitted by the radar, such as echo signals, and send the parameters to the server 200. The server 200 can obtain the processing result based on the received parameters and return the processing result to the terminal 100.

[0073] It should be understood that in some optional implementations, the terminal 100 can also complete the action of obtaining the processing result based on the received parameters on its own, without the need for the server to cooperate. This application embodiment is not limited to this.

[0074] The following description Figure 1 The product form of the mid-terminal 100;

[0075] The terminal 100 in this application embodiment can be a mobile phone, tablet computer, wearable device, vehicle device, augmented reality (AR) / virtual reality (VR) device, laptop computer, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), etc., and this application embodiment does not impose any restrictions on it.

[0076] Terminal 100 may include a radio frequency unit, memory, input unit, display unit, camera (optional), audio circuitry (optional), speaker (optional), microphone (optional), headphone jack (optional), processor, external interface, power supply, and other components. Those skilled in the art will understand that the above-mentioned components are merely examples and do not constitute a limitation on the terminal or multifunctional device; it may include more or fewer components, or a combination of certain components, or different components.

[0077] The input unit can be used to receive input numeric or character information, and to generate key signal inputs related to user settings and function control of the portable multi-functional device. Specifically, the input unit may include a touchscreen (optional) and / or other input devices. Other input devices may include, but are not limited to, one or more of a physical keyboard, function keys (such as volume control buttons, power buttons, etc.), trackball, mouse, joystick, etc.

[0078] Among them, the input device can receive input data, etc.

[0079] The display unit can be used to display information input by the user or information provided to the user, various menus of the terminal, interactive interfaces, file display, and / or playback of any multimedia file. In the embodiments of this application, the display unit can be used to display the interface for data viewing and early warning handling, processing results, etc.

[0080] The memory can be used to store software code related to the radar tunnel wall leakage detection method, and the processor can execute the steps of the radar tunnel wall leakage detection method, and can also schedule other units (such as the above-mentioned input unit and display unit) to achieve the corresponding functions.

[0081] This radio frequency unit (optional) can be used to receive and send signals during information transmission or calls.

[0082] In this embodiment of the application, the radio frequency unit can send data to the server 200 and receive the processing results sent by the server 200.

[0083] It should be understood that this radio frequency unit is optional and can be replaced with other communication interfaces, such as a network port.

[0084] Terminal 100 also includes a power source (such as a battery) for supplying power to the various components.

[0085] Terminal 100 also includes an external interface, which can be a standard Micro USB interface or a multi-pin connector, which can be used to connect terminal 100 to other devices for communication or to connect a charger to charge terminal 100.

[0086] Server 200 includes a bus, a processor, a communication interface, and memory. The processor, memory, and communication interface communicate with each other via the bus.

[0087] The memory can be used to store software code related to the radar tunnel wall leakage detection method, the processor can execute the steps of the radar tunnel wall leakage detection method of the chip, and can also schedule other units to achieve corresponding functions.

[0088] To address the aforementioned problems, this application provides a radar method for detecting tunnel wall leakage. The radar method for detecting tunnel wall leakage according to this application will be described in detail below with reference to the accompanying drawings.

[0089] Reference Figure 2 , Figure 2 A schematic flowchart of a radar tunnel wall leakage detection method provided in this application embodiment is shown below. Figure 2 As shown in the figure, the method for detecting leakage in the tunnel wall using radar provided in this application embodiment may include steps S201 to S205, which are described in detail below.

[0090] Step S201: Acquire echo signals collected at multiple acquisition times, wherein the echo signals are signals reflected by the transmitted signals through the detection area of ​​the tunnel wall.

[0091] The acquisition time refers to the point in time when the radar completes one signal transmission and reception at the tunnel wall.

[0092] For example, echo signals acquired at multiple acquisition times refer to radar signals acquired at different acquisition times t1, t2, ..., t... nThe system transmits and receives multiple sets of echo signals, each set corresponding to a sampling time. n is a positive integer greater than or equal to 1.

[0093] For example, the transmitted signal is a high-frequency electromagnetic wave signal actively emitted by the radar towards the tunnel wall.

[0094] For example, after the transmitted signal comes into contact with the tunnel wall, it is reflected and scattered by the tunnel wall (which includes concrete, lining, moisture, etc.), and the reflected electromagnetic wave signal captured by the radar receiver is the echo signal.

[0095] It is understandable that the characteristics (amplitude, phase, frequency) of the tunnel wall vary with the physical state of the tunnel wall (dry / seeping, whether there is dynamic water flow).

[0096] For example, the echo signal can be preprocessed to remove various interferences and standardize the signal, thereby ensuring the stability and reliability of the subsequent data extraction.

[0097] For example, preprocessing includes filtering, normalization, and time-domain alignment.

[0098] Filtering refers to using a combination of bandpass filters and "moving average or adaptive filters". First, the bandpass filter retains the energy belonging to the target frequency band that is relevant to leakage detection. Then, the moving average or adaptive filter suppresses instantaneous spike interference and random interference.

[0099] It is understandable that environmental noise (such as traffic flow in tunnels and electromagnetic noise generated by equipment operation) is mostly distributed in non-target frequency bands. Bandpass filters can accurately "filter" effective signals and eliminate noise. Instantaneous spike interference (such as sudden electromagnetic pulses) and random interference (such as signal fluctuations caused by dust scattering) are accidental and irregular. Moving average or adaptive filters can smooth signal fluctuations and avoid such interference from misleading subsequent feature analysis.

[0100] Normalization refers to the standardization of the filtered echo signal to unify the signal amplitude range.

[0101] It is understandable that echo signals from different acquisition times and different detection areas may have large amplitude differences due to slight fluctuations in radar transmission power and differences in detection distance, which directly affects the comparability of the data to be measured. Normalization can eliminate such amplitude differences, making all echo signals have the same dimension, and ensuring that the data to be measured from different detection areas and different acquisition times can be compared and analyzed.

[0102] Time-domain alignment refers to calibrating the timing deviation of the echo signal.

[0103] It is understandable that radar signals (such as transmitted or echo signals) are prone to multipath effects when propagating in tunnels. For example, radar signals may be superimposed after being reflected multiple times by the tunnel wall, resulting in timing deviations in the echo signals in the same detection area. If time-domain alignment is not performed, it will lead to errors in the extraction of features such as phase shift and Doppler frequency shift. Time-domain alignment can correct this timing deviation, reduce the negative impact of multipath effects, and ensure the accuracy of signal timing.

[0104] Step S202: For each echo signal, analyze the echo signal to obtain the wall surface state data of the detection area, so as to obtain multiple wall surface state data.

[0105] Wall surface condition data characterizes a multidimensional set of features representing the physical state of the wall medium in the detection area (such as dry / moist, dynamic / static).

[0106] For example, the detection area is the smallest independent spatial unit divided according to the spatial resolution of the radar.

[0107] For example, the detection area is grid-shaped, and its size is determined by radar hardware parameters (such as detection range and scanning angle).

[0108] Understandably, radar can perform a full-area scan of the tunnel wall according to a preset scanning angle and step size, and divide the tunnel wall into several smallest spatial units of uniform size, without overlap or blind spots, i.e., the detection area, based on the radar's spatial resolution.

[0109] Understandably, each detection area can be assigned unique spatial coordinates, for example, spatial coordinates are (tunnel mileage, wall lateral position, height above ground).

[0110] Step S203: Compare the multiple wall surface state data with the pre-obtained wall surface drying data to obtain the state difference value.

[0111] The surface dryness data of the tunnel wall refers to the baseline characteristic data of the detection area of ​​the tunnel wall under a dry condition without leakage.

[0112] For example, wall surface drying data includes, but is not limited to: reference data for the detection area in a leak-free mode. The reference data includes the reflection intensity characterizing the dielectric constant of the detection area, the phase shift between the echo signal and the transmitted signal corresponding to the detection area, and the Doppler frequency shift characterizing the tunnel wall when no liquid adheres to it.

[0113] The no-leakage mode refers to the normal physical state in which the tunnel wall is dry, free from water seepage, water accumulation, and adhering liquid. The dielectric constant and structural morphology of the tunnel wall remain stable, which is the benchmark state for leakage detection.

[0114] It is understandable that, after the tunnel is newly built and accepted but before it is put into operation, radar scanning of the tunnel wall is carried out during a period when the entire tunnel wall is dry in order to obtain reference data for each detection area under the no-leakage mode.

[0115] Understandably, during long-term monitoring, the detection area in the no-leakage mode can be automatically selected as the benchmark to update the reference data, thereby avoiding misjudgments caused by material aging of the tunnel wall or long-term changes in the external environment.

[0116] Understandably, during long-term monitoring, the temperature and humidity of the tunnel's operating environment will change. This change is seasonal and not real-time; that is, after a change in temperature and humidity, the environment will stabilize for a period of time. Increased humidity will increase the overall reflection intensity, and temperature fluctuations will affect the stability of the phase offset. Therefore, it is necessary to re-acquire reference data after changes in the temperature and humidity of the tunnel's operating environment. Alternatively, a correction factor can be obtained to correct the original reference data.

[0117] For example, the state difference can be a difference feature vector.

[0118] For example, a difference feature vector is an ordered set of quantified difference values ​​obtained by calculating the difference or ratio between wall surface condition data and wall surface dryness data for a single detection area. The difference feature vector characterizes the degree of leakage anomaly in the detection area.

[0119] Assuming the wall surface condition data includes reflection intensity, phase shift, and Doppler frequency shift, the difference feature vector can be [ΔI, ΔP, ΔF], where ΔI is the difference in reflection intensity, ΔP is the difference in phase shift, and ΔF is the difference in Doppler frequency shift.

[0120] For example, before performing step S203, the wall surface state data and the wall surface drying data can be normalized and aligned to ensure that they are comparable under the same dimensions, the same spatial coordinates, and the same signal extraction standard, thereby eliminating dimensional differences caused by equipment errors and environmental fluctuations.

[0121] Step S203 calculates the difference between the surface condition data and the dry surface data of the tunnel wall in the same dimension and detection area, amplifying the characteristic anomalies caused by leakage and offsetting the interference of the inherent characteristics of the tunnel wall. Inherent characteristics can include environmental background, such as the roughness of the tunnel wall material, the thickness of the lining, and the reflection of fixed structures, which exist in both dry and water-seeping states and are considered "inherent characteristics".

[0122] This application transforms the question of "whether there is leakage" into the question of "whether the value of the difference feature vector exceeds the normal range". For example, if there is no leakage in the detection area, the wall surface condition data is close to the wall surface dry data, and the values ​​of each component in the difference feature vector are close to 0; if there is leakage in the detection area, the wall surface condition data and the wall surface dry data show a significant deviation, and the values ​​of each component in the difference feature vector deviate significantly from 0.

[0123] Step S204: Input the state difference into the pre-built leakage discrimination model to determine whether the detection area is in a leakage state.

[0124] For example, the weights of each component in the state difference can be pre-set to obtain a weight vector; the product of the weight vector and the difference feature vector is then input into the leakage discrimination model. Weighting the multidimensional components can reduce the false alarm rate caused by a single feature.

[0125] A leakage detection model is constructed based on machine learning and deep learning algorithms. It takes differential feature vectors as input and the leakage status of the detection area as output. After training with a large number of labeled samples, it possesses the ability to automatically identify and determine leakage information in tunnel walls. Examples of commonly used machine learning or deep learning algorithms include Support Vector Machine (SVM), Convolutional Neural Network (CNN), and Graph Neural Network (GNN).

[0126] For example, the permeation status can include: leaked and non-leaking; leaked includes slight leakage, dynamic dripping, etc., and the leakage situation included in the leaked can be determined according to the severity level of the leakage.

[0127] The sample data used to train the leakage discrimination model can be the state difference of detection areas with different tunnels and different leakage levels, and the leakage state corresponding to each detection area can be labeled.

[0128] It is understandable that multiple radars are deployed on tunnel walls or mobile maintenance equipment to create spatial redundancy for observation. This ensures that each detection area is cross-covered by different radars. If at least two radars detect leakage in the same detection area, the area is confirmed as a "valid" detection area. If only a single radar detects leakage, but other radars do not, the area is marked as a "suspected" detection area. This avoids misjudgments caused by single radar failure or interference, ensuring the accuracy of leakage alarms.

[0129] This application provides a radar-based method for detecting tunnel wall leakage. By acquiring echo signals reflected from the detection area of ​​the tunnel wall at multiple acquisition times, it enables continuous temporal acquisition of the tunnel wall state, eliminating dependence on lighting conditions. By analyzing each echo signal to obtain wall surface state data, the physical characteristics and electromagnetic response information of the wall medium can be accurately extracted, unaffected by factors such as tunnel ambient temperature and wind speed, overcoming the shortcomings of infrared thermal imaging detection which is easily affected by the environment and thus leads to inaccurate detection. By comparing multiple wall surface state data with pre-acquired wall surface dryness data to obtain state difference values, it can eliminate background interference from the inherent material and structure of the tunnel wall, highlighting abnormal differences caused by leakage, and improving the stability and accuracy of detection. By inputting the state difference values ​​into a pre-constructed leakage discrimination model to determine whether the detection area is leaking, it can achieve automated and intelligent identification and judgment. It can stably and accurately complete leakage detection in complex environments such as dim lighting, dust, fog, fluctuating temperature and humidity, and wind speed changes in tunnels, effectively solving the problems of poor environmental adaptability and low detection accuracy of existing technologies, and significantly improving the reliability and versatility of tunnel wall leakage identification.

[0130] In one optional implementation, step S202 can be implemented in multiple ways. This application provides, but is not limited to, the following method, which includes the following steps A1 to A4.

[0131] Step A1: Obtain leakage data of the detection area at various acquisition times. The leakage parameters include the reflection intensity characterizing the dielectric constant of the detection area, the phase offset between the echo signal and the transmitted signal, and the Doppler frequency offset characterizing the relative motion between the liquid attached to the detection area and the detection area.

[0132] For example, reflection intensity can be characterized by the amplitude of the echo signal. The distribution of the echo signal amplitude allows for the basic distinction between seepage points and dry points. It is understandable that different material compositions and degrees of seepage in the detection area result in different reflection intensities. In other words, changes in the water content of the detection area lead to changes in the dielectric constant; that is, changes in the water content of the detection area will alter the scattering enhancement or attenuation pattern of the radar signal (i.e., the transmitted signal or the echo signal).

[0133] For example, phase offset characterizes the cumulative phase shift of a radar signal after propagation and reflection through the detection area. Its magnitude is related to the dielectric constant of the wall medium and the effective propagation path of the radar signal. Phase offset is related to the water seepage depth and the longitudinal distribution of water in the detection area. For instance, the water seepage depth determines the propagation path length of the transmitted signal in the water medium. In dry walls made of concrete or a single-medium lining, the penetration depth of the transmitted signal is fixed. When water seeps into the detection area, the water penetrates longitudinally into the wall. The deeper the seepage, the longer the path the transmitted signal must take in the water medium with a higher dielectric constant, resulting in a larger cumulative phase offset. For example, the longitudinal distribution of water determines the longitudinal gradient of the dielectric constant of the tunnel wall. Therefore, phase offset can characterize the degree of leakage within the detection area.

[0134] For example, Doppler frequency shift is the difference between the frequency of the echo signal and the frequency of the transmitted signal. It characterizes the relative velocity between the liquid (such as seepage) adhering to the detection area and the wall of the detection area. Dynamic water flow, such as dripping or flowing water, causes relative motion to the wall of the detection area, thus generating Doppler frequency shift. A dry detection area has no relative velocity, and the Doppler frequency shift is close to 0. The relationship between Doppler frequency shift and the aforementioned relative velocity is explained below.

[0135] The relationship between the Doppler frequency shift and the relative velocity is Δf = 2v × f0 / c, where the Doppler frequency shift of the echo signal is proportional to the relative velocity. Δf is the Doppler frequency shift, f0 is the frequency of the transmitted signal, c is the wave propagation speed, and v is the relative velocity.

[0136] Understandably, the greater the Doppler frequency shift or the greater the fluctuation of the Doppler frequency shift, the higher the leakage activity.

[0137] It is understandable that the echo signal at each acquisition moment reflects the medium state of the tunnel wall at that acquisition moment. Therefore, a set of leakage data can be extracted for each detection area at each acquisition moment, providing the original data basis for subsequent analysis of the dynamic changes, stability and activity of leakage.

[0138] Step A2: Obtain a statistical feature set based on the leakage data at each acquisition time.

[0139] The statistical feature set includes the mean of each reflection intensity in each leakage data set, the variance of each reflection intensity in each leakage data set, the energy entropy of each reflection intensity in each leakage data set, the variance of each phase shift in each leakage data set, the mean of each phase shift in each leakage data set, the energy entropy of each phase shift in each leakage data set, the mean of each Doppler frequency shift in each leakage data set, the variance of each Doppler frequency shift in each leakage data set, and the energy entropy of each Doppler frequency shift in each leakage data set.

[0140] The mean value refers to the average value, which can reflect the overall level and average intensity of leakage parameters in the detection area. It is used to determine whether the detection area is generally dry or wet, and whether there are any abnormalities.

[0141] Variance refers to the squared mean of the deviation from the mean, reflecting the magnitude and stability of the permeability parameter over time.

[0142] Energy entropy refers to the entropy value calculated from energy distribution, which characterizes the degree of disorder, complexity, and randomness of permeation parameters.

[0143] Understandably, statistical feature sets can comprehensively characterize the dynamic leakage characteristics of each detection area from three levels: overall intensity, fluctuation stability, and change complexity. This enables the leakage discrimination model not only to identify whether leakage has occurred, but also to distinguish between different states such as stable leakage, intermittent leakage, sudden leakage, and dynamic water flow. This significantly improves the leakage discrimination model's ability to identify leakage types and development trends, reduces misjudgments caused by environmental interference and random noise, and enhances the robustness, accuracy, and intelligence level of tunnel wall leakage detection.

[0144] Step A3: For each acquisition time, obtain the energy distribution characteristics of the echo signal acquired at that acquisition time in different frequency ranges.

[0145] For example, wavelet packet decomposition (WPD) can be used to perform multi-scale analysis on echo signals and extract energy distribution features at different frequency ranges.

[0146] For example, step A3 includes steps B11 to B13.

[0147] Step B11: Perform frequency domain transformation on the echo signal acquired at the acquisition time to obtain a frequency domain signal.

[0148] Frequency domain transformation refers to signal processing operations that convert time-domain waveforms into frequency-amplitude distributions.

[0149] Frequency domain signals are signal forms represented with frequency on the horizontal axis and amplitude on the vertical axis.

[0150] Step B12: Divide the frequency domain signal into multiple frequency ranges according to a preset frequency interval, and calculate the signal energy corresponding to each frequency range.

[0151] It is understandable that the echo signal, after undergoing Fourier transform or bandpass filtering, is divided into multiple continuous or segmented frequency ranges. Radar signals with different frequency components (such as echo signals or transmitted signals) penetrate to different depths and have different scattering mechanisms in the medium of the tunnel wall.

[0152] For example, the preset frequency range can be: 2GHz–4GHz, 4GHz–6GHz, 6GHz–8GHz (each segment is 2GHz). For example, multiple frequency ranges are [2GHz, 4GHz], [4GHz, 6GHz], [6GHz, 8GHz].

[0153] Step B13: Based on the signal energy corresponding to each of the frequency ranges, obtain the energy distribution characteristics of the echo signal in different frequency ranges.

[0154] Energy distribution characteristics refer to the energy magnitude, proportion, peak position, and distribution pattern of the echo signal in each frequency range, which are used to reflect the electromagnetic response of the wall medium in the detection area at different frequencies.

[0155] Step A4: For each acquisition time, obtain time-frequency energy parameters based on the time-frequency spectrum of the echo signal acquired at the acquisition time. The time-frequency energy parameters include the instantaneous energy peak value, peak frequency, and energy duration greater than or equal to a preset threshold in the time-frequency spectrum. The time-frequency spectrum is a three-dimensional distribution map of time, frequency, and energy.

[0156] For example, step A4 includes steps B21 to B24.

[0157] Step B21: Perform time-frequency analysis on the echo signal acquired at the acquisition time to construct a three-dimensional time-frequency spectrum with time as the horizontal axis, frequency as the vertical axis, and signal energy as the amplitude.

[0158] For example, the three-dimensional time-spectrum map of the detection area is a three-dimensional distribution map with time as the x-axis, frequency as the y-axis, and energy as the z-axis.

[0159] Step B22: Determine the maximum value of the signal energy in the three-dimensional time-frequency spectrum as the instantaneous energy peak value.

[0160] For example, the instantaneous energy peak value refers to the maximum energy value appearing at a specific frequency point at a certain moment in a three-dimensional time-frequency spectrum, reflecting the strongest response intensity in the detection area at that moment. The instantaneous energy peak value can reflect strong abnormal responses such as sudden changes in dielectric constant and water flow impact caused by leakage.

[0161] Step B23: Determine the frequency corresponding to the instantaneous energy peak as the peak frequency.

[0162] For example, peak frequency refers to the frequency value corresponding to the instantaneous energy peak. It characterizes the frequency component of the wall medium in the detection area that responds most strongly to radar signals (transmitted signals or echo signals) and is closely related to the water content and leakage depth of the tunnel wall. Peak frequency can reflect medium structure information such as leakage depth and moisture distribution.

[0163] Step B24: Determine the duration of continuous time during which the signal energy in the three-dimensional time spectrum is greater than or equal to a preset threshold as the energy duration.

[0164] For example, energy duration refers to the continuous length of time during which the signal energy in the three-dimensional time-spectrum is greater than or equal to a preset threshold. It is used to characterize the persistence and stability of abnormal signals caused by leakage. Energy duration can distinguish between transient interference and real leakage, with real leakage typically having a longer energy duration.

[0165] It is understandable that tunnel wall leakage, especially sudden seepage and dynamic water flow, is a non-stationary signal, and traditional Fourier transform cannot describe its frequency characteristics that change over time. This application uses short-time Fourier transform (STFT) to window and frame the echo signal, performing Fourier transform frame by frame to obtain the frequency energy distribution at each moment, and then stitching them together to form a three-dimensional time-frequency spectrum, achieving a joint representation of time, frequency, and energy.

[0166] It is understandable that if the dynamic changes of tunnel wall leakage cannot be grasped in time, the leakage hazards will not be dealt with in time and will gradually deteriorate, leading to serious problems such as tunnel lining spalling, steel corrosion, and structural cracking, affecting the tunnel's operational safety and service life. Therefore, it is necessary to grasp the dynamic changes in time. Based on this, this application provides the following method, which includes the following steps C1 to C3.

[0167] Step C1: Obtain the local leakage rate based on the wall surface state data corresponding to multiple acquisition times. The local leakage rate characterizes the change law of the wall surface state data over time.

[0168] Understandably, changes in reflection intensity are usually directly related to increases or decreases in leakage flow; the greater the leakage flow, the higher the reflection intensity, and vice versa. Phase shift and Doppler frequency shift primarily reflect changes in water flow velocity; different flow velocities will result in corresponding fluctuations in their values. Continuously observing the fluctuation patterns of reflection intensity, phase shift, and Doppler frequency shift over time allows for precise capture of dynamic changes, such as real-time alterations in leakage flow and water flow velocity, thereby obtaining the local leakage rate.

[0169] Step C2: Obtain the seepage trend based on the change in the wall surface state data at each pair of adjacent acquisition times. The seepage trend characterizes the change pattern of the wall surface state data over time.

[0170] Adjacent acquisition times refer to two consecutive acquisition times.

[0171] Assume the data collection times from earliest to latest are: t1, t2, ..., t n The changes in wall surface state data at each pair of adjacent acquisition times, arranged from earliest to latest acquisition time, are as follows: wall surface state data at acquisition time t2 - wall surface state data at acquisition time t1, wall surface state data at acquisition time t3 - wall surface state data at acquisition time t2, wall surface state data at acquisition time t4 - wall surface state data at acquisition time t3, ..., acquisition time t n Wall surface condition data - acquisition time t (n-1) The wall surface condition data.

[0172] Understandably, if the changes in the data collected from earliest to latest are consistently positive and gradually increase, it indicates a continuously worsening seepage trend, meaning the leakage is becoming increasingly severe. If the changes are consistently negative and their absolute values ​​gradually increase, it indicates a continuously mitigating seepage trend, meaning the leakage is gradually decreasing. If the changes fluctuate slightly around 0, it indicates a stable seepage trend with no significant change, meaning the leakage is stable. If the changes suddenly exhibit large positive or negative fluctuations, it indicates a sudden and abnormal seepage trend, such as sudden leakage or abrupt changes in water flow.

[0173] For example, by using a sliding window to smooth the changes in the wall surface state data of the detection area at each pair of adjacent acquisition times, instantaneous fluctuations are eliminated to obtain a stable trend curve of leakage changes, which is the seepage trend.

[0174] For example, weights are assigned to each change, with higher weights for changes more recent in time, and then the seepage trend is obtained through Exponentially Weighted Moving Average (EWMA).

[0175] For example, the changes in the wall surface state data of the detection area at each pair of adjacent acquisition times are input into the ARIMA (Autoregressive Integrated Moving Average) model, and the seepage trend of the detection area is obtained through the ARIMA model.

[0176] Step C3: Based on each leakage area that includes the same local area at multiple acquisition times, obtain the leakage range expansion rate of the leakage area. The leakage area includes the detection area that is spatially adjacent. The leakage range expansion rate characterizes the change law of the area of ​​the leakage area over time.

[0177] A leakage area is a continuous region consisting of multiple detection areas that are spatially adjacent and all are in the "leaking" state. The leakage range is the area defined by the boundaries of the leakage area.

[0178] The rate of expansion of a leak is a physical quantity that quantifies how quickly the area of ​​a leak changes over time, using a single leaking region as the unit. It primarily reflects the expansion or contraction pattern of the leak area, and its unit is usually "area / time" (e.g., m²). 2 / h); a positive rate of leakage range expansion indicates that the leakage range is expanding, a negative rate of leakage range expansion indicates that the leakage range is shrinking, and a rate of leakage range expansion close to 0 indicates that the leakage range is stable.

[0179] Understandably, leaking areas containing the "same local area" at different collection times are grouped into a single leaking area set. This means that the leaking area set corresponds to the same leaking area, and the range only expands or shrinks over time, thus avoiding misjudging the same leaking area as multiple independent leaking areas.

[0180] For example, the area of ​​the leakage area can be calculated based on the number of detection areas contained in the leakage area and the area of ​​a single detection area.

[0181] Understandably, once the dynamic change pattern is understood, the leakage level can be determined. The specific steps are as follows: Based on the number of detection areas that have already leaked, the local leakage rate corresponding to each detection area, the seepage trend corresponding to each detection area, the state difference corresponding to each detection area, and the leakage range expansion rate corresponding to each leakage area, the leakage level is determined.

[0182] For example, the method for determining the level of leakage is as follows:

[0183] Leakage level "Level 1 (Slight)": The number of leaking detection areas is less than or equal to the first threshold, the components of the state difference of the leaking detection areas are small (e.g., each component is less than or equal to the corresponding threshold A), the seepage trend is stable with no significant change, the rate of expansion of the leakage range is close to zero, and the local leakage rate is close to zero.

[0184] Leakage level "Level 2 (General)": The leakage status is that the number of leaking detection areas is greater than the first threshold and less than or equal to the second threshold. The leakage status is that each component of the state difference of the leaking detection areas has a certain deviation (such as each component is greater than the corresponding threshold A and each component is less than or equal to the corresponding threshold B). The seepage trend is stable with no signs of aggravation. The rate of expansion of the leakage range is slow and the local leakage rate is low.

[0185] Leakage level "Level 3 (Significant)": The number of leaking detection areas is greater than the second threshold and less than or equal to the third threshold; the difference in reflection intensity in the state difference of the leaking detection areas is significant; the Doppler frequency shift in the difference feature vector of the leaking detection areas is large (e.g., greater than threshold B); the seepage trend is slowly aggravated; the rate of expansion of the leakage range is moderate; and the local leakage rate is moderate.

[0186] Leakage level "Level 4 (Severe)": The number of leaking detection areas is greater than the second threshold and less than or equal to the third threshold. The rate of expansion of the leakage range is increasing rapidly. The difference in the status of the leaking detection areas is relatively large. The water seepage trend is rapidly aggravated. The local leakage rate is high.

[0187] Leakage level "Level 5 (Extremely Severe)": The leakage status is that the number of leaking detection areas is greater than the third threshold, the rate of expansion of the leakage range is rapidly increasing, the seepage trend continues to deteriorate, and both the local leakage rate and the rate of expansion of the leakage range are significantly higher.

[0188] For example, when the leakage level is greater than or equal to a preset threshold, an early warning signal can be automatically triggered. After the early warning signal is triggered, the data to be measured, leakage level, local leakage rate, seepage trend, differential feature vector, and leakage range expansion rate can be uploaded to the remote monitoring center to facilitate timely handling by maintenance personnel.

[0189] Understandably, if tunnel wall leakage is not predicted in time, it can rapidly escalate from minor to severe leakage, leading to safety hazards such as lining spalling, steel reinforcement corrosion, and structural cracking. By predicting the leakage rate in the future (e.g., predicting a significant increase in the leakage rate of a certain detection area in the next hour) and the area of ​​the leakage area (e.g., predicting that the area of ​​a leakage area will double in the next two hours), the risk of leakage worsening can be identified in advance, triggering an early warning before the hazard expands, buying time for maintenance personnel to handle the situation, avoiding damage to the tunnel structure, and ensuring the safe operation of the tunnel. The specific method is as follows: The local leakage rate, the seepage trend, and the leakage range expansion rate are input into a pre-constructed leakage trend prediction model. The leakage trend prediction model then obtains the future leakage rate of the detection area and the future area of ​​the leakage area.

[0190] For example, the leakage trend prediction model is a prediction model trained based on machine learning algorithms or deep learning algorithms, such as LSTM (Long Short-Term Memory) network and CNN-LSTM (Convolutional Neural Network-Long Short-Term Memory). It takes local leakage rate, leakage trend and leakage range expansion rate as input, and uses the leakage rate of each detection area and the area of ​​each leakage area at future time as training targets. After training with a large number of time series samples, it has the ability to accurately predict the future dynamics of leakage.

[0191] It is understandable that the leakage rate of the detection area at future time represents how fast the leakage of the detection area will change at future time; the area of ​​the leakage area at future time represents the scale of expansion or contraction of the leakage range at future time.

[0192] The development of tunnel wall leakage is continuous (e.g., a continuously worsening leakage will not suddenly ease, and a slowly expanding area will not suddenly shrink significantly). Three types of parameters, namely, local leakage rate, seepage trend, and leakage range expansion rate, completely record the historical dynamic pattern of leakage. By learning this pattern, the leakage trend prediction model can accurately predict the leakage rate of each detection area and the area of ​​each leakage area at future time, providing forward-looking support for operation and maintenance.

[0193] It is understood that there are multiple ways to implement step C2. The embodiments of this application provide, but are not limited to, the following method, which includes the following steps D1 to D3.

[0194] Step D1: Sort the wall surface state data corresponding to the multiple acquisition times from earliest to latest according to the acquisition time to obtain the first sorting result.

[0195] Assume multiple acquisition times are t1, t2, t3, and t4. Assume the four acquisition times are ordered from earliest to latest as t1, t2, t3, and t4. Assume the first sorting result of the detection area is: wall surface state data A11 obtained from the echo signal of the detection area acquired at acquisition time t1; wall surface state data A12 obtained from the echo signal of the detection area acquired at acquisition time t2; wall surface state data A13 obtained from the echo signal of the detection area acquired at acquisition time t3; and wall surface state data A14 obtained from the echo signal of the detection area acquired at acquisition time t4.

[0196] Step D2: Determine the changes in the state data of each pair of adjacent wall surfaces in the first sorting result to form time series data.

[0197] Taking the above example, the time series data of the detection area are: wall surface state data A12-wall surface state data A11, wall surface state data A13-wall surface state data A12, and wall surface state data A14-wall surface state data A13.

[0198] Step D3: Determine the seepage trend based on the time series data.

[0199] It is understood that there are multiple ways to implement step C3. The embodiments of this application provide, but are not limited to, the following method, which includes the following steps E1 to E4.

[0200] Step E1: For any of the acquisition times, the detection areas that are in a leaking state and have adjacent spatial coordinates at the acquisition time are divided into the same leakage area.

[0201] Understandably, at each data acquisition moment, the leakage status (leaking / not leaking) of each detection area is first determined using a leakage discrimination model, and its spatial coordinates are recorded. Subsequently, spatial coordinates are represented by markers, with consecutive markers indicating adjacent spatial coordinates and discontinuous markers indicating non-adjacent spatial coordinates.

[0202] For the acquisition time t1, assuming that the number of leaking detection areas is 3 and they are labeled as ①, ②, ③ respectively; and the labels of the non-leaking detection areas are ④, ⑤, ⑥..., then a leaking area (denoted as A1) is obtained, which includes the detection areas labeled as ①, ②, ③.

[0203] For the data collection time t2: the number of leaking detection areas is 4, and they are labeled as ①, ②, ③, ④ respectively; the labels of the non-leaking detection areas are ⑤, ⑥..., then a leaking area (denoted as A2) is obtained, which includes the detection areas labeled as ①, ②, ③, ④.

[0204] For the data collection time t3: the number of leaking detection areas is 5, and they are labeled as ①, ②, ③, ④, ⑤ respectively; the labels of the non-leaking detection areas are ⑥, ⑦, ⑧..., then a leaking area (denoted as A3) is obtained, which includes the detection areas labeled as ①, ②, ③, ④.

[0205] For data collection time t4: the number of leaking detection areas is 6, labeled ①, ②, ③, ④, ⑤, ⑨; the labels of non-leaking detection areas are ⑥, ⑦, ⑧, ⑩, ..., resulting in two leaking areas (denoted as A4 and A5). Leaking area A4 includes the detection areas labeled ①, ②, ③, and ④, and leaking area A5 includes the detection areas labeled ⑤ and ⑨.

[0206] Step E2: Divide the various leakage areas that contain the same local area at each of the aforementioned collection times into the same leakage area set.

[0207] Taking the above example again, since leakage areas A1, A2, A3, and A4 all contain detection areas marked ①, ②, and ③, two sets of leakage areas are obtained. One set of leakage areas is {leakage area A1, leakage area A2, leakage area A3, and leakage area A4}, and the other set of leakage areas is {none, none, none, and leakage area A5}.

[0208] Step E3: Obtain the area change rate sequence corresponding to each of the multiple sets of leakage areas. The area change rate sequence includes: the area change rate of the leakage area set ordered from earliest to latest according to the collection time. The area change rate is the ratio of the difference in area of ​​the leakage area corresponding to two adjacent collection times to the time interval between the two adjacent collection times.

[0209] Taking the above example again, the area change rate sequence of the leakage area set {leakage area A1, leakage area A2, leakage area A3, leakage area A4} is {(leakage area A2 - leakage area A1) / time interval, (leakage area A3 - leakage area A2) / time interval, (leakage area A4 - leakage area A3) / time interval}.

[0210] Step E4: Obtain the rate of expansion of the leakage range based on the area change rate sequence.

[0211] The above describes a radar-based tunnel wall leakage detection method provided by the embodiments of this application. The following describes the apparatus for performing the radar-based tunnel wall leakage detection method described above.

[0212] Please see Figure 3 , Figure 3 This is a schematic diagram of a radar-based tunnel wall leakage detection device provided in an embodiment of this application. Figure 3 As shown, the radar-based tunnel wall leakage detection device includes:

[0213] The first acquisition module 301 is used to acquire echo signals acquired at multiple acquisition times, wherein the echo signals are signals reflected by the transmitted signals through the detection area of ​​the tunnel wall;

[0214] The second acquisition module 302 is used to parse the echo signal for each echo signal to obtain the wall surface state data of the detection area, so as to obtain multiple wall surface state data.

[0215] The third acquisition module 303 is used to compare the multiple wall surface state data with the pre-obtained wall surface drying data to obtain the state difference value.

[0216] The judgment module 304 is used to input the state difference into a pre-built leakage discrimination model to determine whether the detection area is in a leakage state.

[0217] In one optional implementation, the second acquisition module includes:

[0218] The first acquisition unit is used to acquire leakage data of the detection area at various acquisition times. The leakage parameters include the reflection intensity characterizing the dielectric constant of the detection area, the phase offset between the echo signal and the transmitted signal, and the Doppler frequency offset characterizing the relative motion between the liquid attached to the detection area and the detection area.

[0219] The second acquisition unit is used to acquire a statistical feature set based on the leakage data at each acquisition time.

[0220] The third acquisition unit is used to acquire the energy distribution characteristics of the echo signal acquired at each acquisition time in different frequency ranges.

[0221] The fourth acquisition unit is used to acquire time-frequency energy parameters based on the time-frequency spectrum of the echo signal acquired at each acquisition time. The time-frequency energy parameters include the instantaneous energy peak value, peak frequency, and energy duration greater than or equal to a preset threshold in the time-frequency spectrum. The time-frequency spectrum is a three-dimensional distribution map of time, frequency, and energy.

[0222] In one optional implementation, the third acquisition unit includes:

[0223] The frequency domain transformation subunit is used to perform frequency domain transformation processing on the echo signal acquired at the acquisition time to obtain a frequency domain signal;

[0224] The calculation subunit is used to divide the frequency domain signal into multiple frequency ranges according to a preset frequency interval, and calculate the signal energy corresponding to each frequency range respectively.

[0225] The acquisition sub-unit is used to obtain the energy distribution characteristics of the echo signal in different frequency ranges based on the signal energy corresponding to each of the frequency ranges.

[0226] In one optional implementation, the fourth acquisition unit includes:

[0227] A sub-unit is constructed to perform time-frequency analysis based on the echo signal acquired at the acquisition time, and to construct a three-dimensional time-frequency spectrum with time as the horizontal axis, frequency as the vertical axis, and signal energy as the amplitude.

[0228] The first determining subunit is used to determine that the maximum value of the signal energy in the three-dimensional time-spectrum diagram is the instantaneous energy peak value;

[0229] The second determining subunit is used to determine the frequency corresponding to the instantaneous energy peak as the peak frequency;

[0230] The third determining subunit is used to determine the continuous time length during which the signal energy in the three-dimensional time spectrum is greater than or equal to a preset threshold as the energy duration.

[0231] In one alternative implementation, it also includes:

[0232] The fourth acquisition module is used to acquire the local leakage rate based on the wall surface state data corresponding to multiple acquisition times, wherein the local leakage rate characterizes the change law of the wall surface state data over time.

[0233] The fifth acquisition module is used to obtain the seepage trend based on the change in the wall surface state data at each pair of adjacent acquisition times. The seepage trend characterizes the change pattern of the wall surface state data over time.

[0234] The sixth acquisition module is used to acquire the leakage range expansion rate of each leakage area that includes the same local area at multiple acquisition times. The leakage area includes the detection area that is spatially adjacent, and the leakage range expansion rate characterizes the change law of the area of ​​the leakage area over time.

[0235] In one alternative implementation, it also includes:

[0236] The seventh acquisition module is used to input the local leakage rate, the seepage trend, and the leakage range expansion rate into a pre-constructed leakage trend prediction model, and obtain the leakage rate of the detection area and the area of ​​the leakage area at future times through the leakage trend prediction model.

[0237] In one alternative implementation, it also includes:

[0238] The first determining module is used to determine the leakage level based on the number of detection areas that have already leaked, the local leakage rate corresponding to each detection area, the seepage trend corresponding to each detection area, the state difference corresponding to each detection area, and the leakage range expansion rate corresponding to each leakage area.

[0239] In one optional implementation, the fifth acquisition module includes:

[0240] The sorting unit is used to sort the wall surface state data corresponding to the multiple acquisition times from earliest to latest according to the acquisition time to obtain a first sorting result;

[0241] The first determining unit is used to determine that the change in each pair of adjacent wall surface state data in the first sorting result constitutes time series data.

[0242] The second determining unit is used to determine the seepage trend based on the time series data.

[0243] In one optional implementation, the sixth acquisition module includes:

[0244] The first division unit is used to divide, for any one of the acquisition times, the detection areas that are in a leaking state at the acquisition time and whose spatial coordinates are adjacent into the same leaking area.

[0245] The second division unit is used to divide the various leakage areas that contain the same local area at each of the acquisition times into the same leakage area set.

[0246] The first acquisition subunit is used to acquire the area change rate sequence corresponding to the multiple sets of leakage areas respectively. The area change rate sequence includes: the area change rate of the leakage area set ordered from earliest to latest according to the acquisition time. The area change rate is the ratio of the difference in area of ​​the leakage area corresponding to two adjacent acquisition times to the time interval between the two adjacent acquisition times.

[0247] The fifth acquisition unit is used to acquire the leakage range expansion rate based on the area change rate sequence.

[0248] This application also provides an electronic device in its embodiments. (See reference...) Figure 4 The diagram illustrates a structural schematic suitable for implementing the electronic device in the embodiments of this application. The electronic device in the embodiments of this application may include, but is not limited to, fixed terminals such as mobile phones, laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 4 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0249] like Figure 4As shown, the electronic device may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 401, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 402 or a program loaded from a storage device 408 into a random access memory (RAM) 403. When the electronic device is powered on, the RAM 403 also stores various programs and data required for the operation of the electronic device. The processing unit 401, ROM 402, and RAM 403 are interconnected via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0250] Typically, the following devices can be connected to I / O interface 405: input devices 406 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 407 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 408 including, for example, memory cards, hard drives, etc.; and communication devices 409. Communication device 409 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or have alternatively.

[0251] This application also provides a computer program product including computer-readable instructions, which, when executed on an electronic device, cause the electronic device to implement any of the radar-based tunnel wall leakage detection methods provided in this application.

[0252] This application also provides a computer-readable storage medium carrying one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any of the radar-based tunnel wall leakage detection methods provided in this application.

[0253] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the device embodiment drawings provided in this application, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.

[0254] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, or it can be implemented by special-purpose hardware including application-specific integrated circuits, special-purpose CPUs, special-purpose memory, special-purpose components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, training equipment, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0255] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0256] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, training device, or data center to another website, computer, training device, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

Claims

1. A radar-based tunnel wall leakage detection method, characterized by, include: The echo signals acquired at multiple acquisition times are obtained, and the echo signals are the signals reflected by the transmitted signal through the detection area of ​​the tunnel wall; For each of the echo signals, the echo signal is analyzed to obtain the wall surface state data of the detection area, so as to obtain multiple wall surface state data. The state difference value is obtained by comparing the multiple wall surface state data with the pre-obtained wall surface drying data. The state difference is input into a pre-built leakage discrimination model to determine whether the detection area is in a leakage state.

2. The millimeter wave radar-based tunnel wall leakage detection method according to claim 1, characterized in that, The step of parsing the echo signal to obtain the wall surface state data of the detection area for each echo signal includes: Leakage data of the detection area at various acquisition times are obtained. The leakage parameters include the reflection intensity characterizing the dielectric constant of the detection area, the phase offset between the echo signal and the transmitted signal, and the Doppler frequency shift characterizing the relative motion between the liquid attached to the detection area and the detection area. Statistical feature sets are obtained based on the leakage data at each collection time. For each acquisition time, the energy distribution characteristics of the echo signal acquired at that acquisition time in different frequency ranges are obtained; For each acquisition time, time-frequency energy parameters are obtained based on the time-frequency spectrum of the echo signal acquired at the acquisition time. The time-frequency energy parameters include the instantaneous energy peak value, peak frequency, and energy duration greater than or equal to a preset threshold in the time-frequency spectrum. The time-frequency spectrum is a three-dimensional distribution map of time, frequency, and energy. 3.The tunnel wall leakage detection method based on millimeter wave radar according to claim 2, characterized in that, The process of acquiring the energy distribution characteristics of the echo signal acquired at the acquisition time in different frequency ranges includes: The echo signal acquired at the acquisition time is subjected to frequency domain transformation to obtain a frequency domain signal; The frequency domain signal is divided into multiple frequency ranges according to a preset frequency interval, and the signal energy corresponding to each frequency range is calculated. Based on the signal energy corresponding to each of the aforementioned frequency ranges, the energy distribution characteristics of the echo signal in different frequency ranges are obtained. 4.The tunnel wall leakage detection method based on millimeter wave radar according to claim 2, characterized in that, The acquisition of time-frequency energy parameters based on the time-spectrum diagram of the echo signal acquired at the acquisition time includes: Time-frequency analysis is performed on the echo signals acquired at the acquisition time to construct a three-dimensional time-frequency spectrum with time as the horizontal axis, frequency as the vertical axis, and signal energy as the amplitude. The maximum value of the signal energy in the three-dimensional time-frequency spectrum is determined to be the instantaneous energy peak value; The frequency corresponding to the instantaneous energy peak value is determined as the peak frequency; The duration of continuous time during which the signal energy in the three-dimensional time-spectrum graph is greater than or equal to a preset threshold is defined as the energy duration.

5. The millimeter wave radar-based tunnel wall leakage detection method according to any one of claims 1 to 4, characterized in that, After the step of inputting the state difference into a pre-built leakage discrimination model to determine whether the detection area is in a leakage state, the method further includes: The local leakage rate is obtained based on the wall surface state data corresponding to multiple acquisition times, and the local leakage rate characterizes the change law of the wall surface state data over time. The seepage trend is obtained based on the change in the wall surface state data at each pair of adjacent acquisition times. The seepage trend characterizes the change pattern of the wall surface state data over time. Based on each leakage area that includes the same local area at multiple acquisition times, the leakage range expansion rate of the leakage area is obtained. The leakage area includes the detection area that is spatially adjacent. The leakage range expansion rate characterizes the change law of the area of ​​the leakage area over time. 6.The tunnel wall leakage detection method based on millimeter wave radar according to claim 5, wherein, Also includes: The local leakage rate, the seepage trend, and the leakage range expansion rate are input into a pre-constructed leakage trend prediction model. The leakage rate of the detection area and the area of ​​the leakage area at future times are obtained through the leakage trend prediction model. 7.The tunnel wall leakage detection method based on millimeter wave radar according to claim 5, wherein, Also includes: The leakage level is determined based on the number of leaking detection areas, the local leakage rate corresponding to each detection area, the seepage trend corresponding to each detection area, the state difference corresponding to each detection area, and the leakage range expansion rate corresponding to each leaking area. 8.The tunnel wall leakage detection method based on millimeter wave radar according to claim 5, wherein, The step of obtaining the seepage trend based on the change in the wall surface state data at each pair of adjacent acquisition times includes: The wall surface state data corresponding to the multiple acquisition times are sorted from earliest to latest according to the acquisition time to obtain a first sorting result; The changes in each pair of adjacent wall surface state data in the first sorting result constitute time series data. The seepage trend is determined based on the time series data. 9.The tunnel wall leakage detection method based on millimeter wave radar according to claim 5, wherein, The step of obtaining the leakage range expansion rate of each leakage area based on multiple leakage areas containing the same local area at multiple acquisition times includes: For any of the aforementioned acquisition times, the detection areas that are in a leaking state and have adjacent spatial coordinates at the acquisition time are divided into the same leaking area; Each leakage area that contains the same local area at each of the aforementioned acquisition times is divided into the same leakage area set; Obtain area change rate sequences corresponding to multiple sets of leakage areas, wherein the area change rate sequence includes: the area change rate of the leakage area set ordered sequentially from earliest to latest according to the collection time, and the area change rate is the ratio of the difference in area of ​​the leakage area corresponding to two adjacent collection times to the time interval between the two adjacent collection times; The rate of expansion of the leakage range is obtained based on the area change rate sequence. 10.A radar-based tunnel wall leakage detection apparatus, characterized by, include: The first acquisition module is used to acquire echo signals acquired at multiple acquisition times, wherein the echo signals are signals reflected by the transmitted signals through the detection area of ​​the tunnel wall; The second acquisition module is used to parse the echo signal for each echo signal to obtain the wall surface state data of the detection area, so as to obtain multiple wall surface state data. The third acquisition module is used to compare the multiple wall surface state data with the pre-obtained wall surface drying data to obtain the state difference value. The judgment module is used to input the state difference into a pre-built leakage discrimination model to determine whether the detection area is in a leakage state.

11. A computer program product, characterised in that, It includes computer-readable instructions that, when executed on an electronic device, cause the electronic device to implement the radar-based tunnel wall leakage detection method as described in any one of claims 1 to 9.

12. An electronic device, comprising: It includes at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program to enable the electronic device to implement the radar-based tunnel wall leakage detection method as described in any one of claims 1 to 9.

13. A computer storage medium, characterized in that The storage medium carries one or more computer programs that, when executed by an electronic device, enable the electronic device to implement the radar-based tunnel wall leakage detection method as described in any one of claims 1 to 9.

Images