Variable array type piping detection radar device, piping monitoring device and identification method

Abstract

The invention discloses a variable array type piping detection radar device, a piping monitoring device and an identification method. Comprising a radar support which is adjustable in a longitudinal-transverse three-level mode, 16 radar units are distributed in a 4 * 4 matrix mode, an edge calculation module is integrated to achieve local noise reduction and preprocessing of signals, the device is carried on a self-propelled platform, and a multi-mode monitoring system is constructed by combining a visible infrared double-spectrum probe of an acoustic-optical-electric integrated mast, a voiceprint recognition module and a Beidou positioning system. According to the monitoring method, day and night monitoring weight is dynamically adjusted, bidirectional LSTM is combined with an attention mechanism to extract features, noise robustness is optimized through adversarial training, and piping risk factor fusion calculation is achieved. According to the invention, a closed-loop monitoring system with environment self-adaption, signal multi-source perception and intelligent decision is constructed, and high-efficiency and high-precision monitoring and early warning of piping are realized; and comprehensive breakthrough is realized on core indexes such as positioning precision, identification accuracy and environmental adaptability.

Description

Technical Field

[0001] The present invention relates to the technical field of levee engineering safety monitoring, and particularly to a variable array type piping detection radar device, a piping monitoring device and an identification method. Background Art

[0002] In recent years, extreme weather has occurred frequently globally. As a typical form of levee seepage damage, piping has become the primary hidden danger threatening flood control safety due to its strong suddenness, high concealment, and fast breach speed. Traditional monitoring methods have significant defects: manual inspections rely on a large amount of manpower, facing problems such as lagging response, low positioning accuracy, and insufficient continuous monitoring ability. Moreover, affected by the change of the labor force structure, the number of personnel with professional identification experience is becoming increasingly scarce; the existing single-point detection technologies such as piezometers have a false alarm rate as high as 35% under complex working conditions such as turbid water flow and vegetation coverage, and have insufficient ability to capture early abnormal signals such as micro-vibrations and low seepage flow rates, making it difficult to achieve full-domain real-time monitoring; while the early warning models based on static geological parameters do not effectively integrate real-time hydrological data and intelligent algorithms, and the prediction accuracy drops by more than 40% under extreme working conditions, unable to meet the needs of dynamic risk assessment.

[0003] The core bottlenecks of the existing technologies are concentrated in three aspects: First, the global perception ability is insufficient. Single-point monitoring devices can only obtain local seepage data and cannot construct a dynamic model of a complex seepage field, and the risk of missed detection increases significantly with the change of water conditions; second, the multi-source data fusion efficiency is low. Multi-dimensional data such as hydrology, geology, and environment have not formed effective associations, and the early warning model lacks the ability to perform non-linear analysis on the piping evolution process, making it difficult to predict the development path of the danger; third, the engineering adaptability is limited. Traditional monitoring devices are large in size, high in power consumption, and high in deployment cost, and lack modular design, making it difficult to be applied on a large scale in long-distance levees and remote areas, and also unable to adapt to mobile loading platforms such as unmanned aerial vehicles and inspection vehicles.

[0004] In response to the above problems, the present invention relies on technological breakthroughs such as the Internet of Things, edge computing, and machine learning to develop an active variable array type detection radar device and an intelligent identification method. Through high-sensitivity array sensors, real-time monitoring of the seepage field with a decimeter-level spatial resolution is achieved. By integrating the Beidou high-precision positioning technology and the deep neural network algorithm, a full-chain technology system for data acquisition, intelligent analysis, precise positioning, and dynamic early warning is constructed, reducing the early warning response time from the hour level to the minute level, and improving the prediction reliability to more than 95%. The device adopts a miniaturized and modular design, supports flexible deployment in various levee scenarios, effectively fills the technical gap of all-weather intelligent monitoring, and provides an innovative solution for flood control safety under extreme climates. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides a variable array type piping detection radar device, a piping monitoring device and an identification method, which combines variable array radar structure design, multi-modal sensor fusion and dynamic weight deep learning to construct a closed-loop monitoring system with environment adaptability, multi-source signal perception and intelligent decision-making, realizing efficient and high-precision monitoring and early warning of piping; comprehensive breakthroughs are achieved in core indicators such as positioning accuracy, identification accuracy rate, and environmental adaptability.

[0006] To achieve the above object, the present invention discloses an active variable array type piping detection radar device, which includes a radar support and a plurality of radar units arranged on the radar support;

[0007] The radar support includes a support base plate, a transverse telescopic support and a longitudinal telescopic support. The longitudinal telescopic support is installed on the support base plate. The longitudinal telescopic support includes a longitudinal guide rail and a longitudinal linear bearing slidably connected to the longitudinal guide rail, and a longitudinal limit buckle is arranged on the longitudinal linear bearing. The longitudinal linear bearing is relatively fixed to the longitudinal guide rail through the longitudinal limit buckle;

[0008] The transverse telescopic support is installed on the longitudinal linear bearing of the longitudinal telescopic support. The transverse telescopic support includes a transverse guide rail and a transverse linear bearing slidably connected to the transverse guide rail, and a transverse limit buckle is arranged on the transverse linear bearing. The transverse linear bearing is relatively fixed to the transverse guide rail through the transverse limit buckle;

[0009] Each of the radar units is installed on the transverse linear bearing. The radar units are slidably adjusted in the longitudinal direction by the longitudinal telescopic support and fixed and limited by the longitudinal limit buckle. Then, the radar units are adjusted in multiple levels within the unit spacing in the longitudinal direction;

[0010] Each of the radar units is slidably adjusted in the transverse direction by the transverse telescopic support and fixed and limited by the transverse limit buckle. Then, the radar units are adjusted in multiple levels within the unit spacing in the transverse direction.

[0011] Furthermore, there are 16 radar units, which are arranged in a regular 4×4 matrix distribution on the radar support; the radar units adapt to complex dam detection environments with different soil qualities and heights by setting different transmitting and receiving arrays in the longitudinal and transverse directions; the radar units are expanded in multiple levels in the longitudinal and transverse directions to adapt to different dam widths and improve the monitoring range and accuracy.

[0012] Furthermore, the radar units are divided into four groups A1, A2, A3, and A4 in the longitudinal direction. The four groups of radar units use the same frequency, or rows A1 and A3 are combined, and rows A2 and A4 are combined, so as to synchronously transmit and collect two groups of signals with different frequency characteristics. The collected signals are analyzed in groups through digital filtering technology.

[0013] Further, the radar units are divided into four groups, namely B1, B2, B3, and B4, in the lateral direction. The four groups of radar units use the same frequency, or B1 and B3 rows are combined, and B2 and B4 rows are combined, so as to synchronously transmit and collect two groups of signals with different frequency characteristics. The collected signals are analyzed by grouping through digital filtering technology.

[0014] Further, each of the radar units includes a DA power amplifier component, a signal acquisition AD component, an edge computing control and filtering component, and an integrated antenna module; the DA power amplifier component is used to convert the baseband digital signal into an analog radio frequency signal and amplify the signal in terms of power; the signal acquisition AD component is used to collect the echo analog signal received by the antenna and convert it into a digital signal after conditioning; the edge computing control and filtering component is used for real-time signal processing, system control, and noise filtering, removing noise and clutter through digital filtering to improve the signal-to-noise ratio of the signal; and completing data processing locally to reduce the computing pressure at the backend and reduce the transmission delay.

[0015] The integrated antenna module is assembled by an internal three-channel antenna transceiver module, a magnetic flux protection head, and a transmitting antenna protection cover. The internal three-channel antenna transceiver module is composed of a radar signal transmitting antenna, a miniature receiving antenna for radar reflected signals, and a miniature laser displacement transceiver sensor arranged at equal intervals longitudinally; the radar signal transmitting antenna, the miniature receiving antenna for radar reflected signals, and the miniature laser displacement transceiver sensor are connected by a shielded cable I.

[0016] The present invention also provides a real-time monitoring device for piping, which includes a self-propelled platform, an opto-acoustic-electric integrated mast installed on the self-propelled platform, and the piping detection radar device described above; the piping detection radar device is fixedly connected to one end of the self-propelled platform through a flange connector, and rollers are installed at the bottom of the piping detection radar device.

[0017] A mast connection seat is fixed on the top wall of the self-propelled platform, and the opto-acoustic-electric integrated mast is fixedly installed on the mast connection seat. An opto-acoustic-electric signal front-end processing module and its three-proof cover are arranged at the bottom of the opto-acoustic-electric integrated mast, which can remove noise and preprocess the monitoring data to meet the data analysis preparation requirements of the backend early warning platform.

[0018] Further, a lateral probe module, a longitudinal probe module, a voiceprint recognition module, and a Beidou satellite guidance device are provided on the opto-acoustic-electric integrated mast.

[0019] The lateral probe module is composed of a visible infrared dual-spectrum probe on the backwater slope side and a visible light probe on the adjacent water side; the visible infrared dual-spectrum probe and the visible light probe are respectively arranged on the lateral sides of the top outer wall of the opto-acoustic-electric integrated mast.

[0020] The longitudinal probe module is composed of a forward video probe and a rearward video probe, which are respectively arranged on both longitudinal sides of the outer side wall of the top of the acoustic-optical integrated mast;

[0021] The probe of the voiceprint recognition module is placed at the top of the acoustic-optical-electrical integrated mast; the Beidou satellite guidance device is installed on the top of the front outer side wall of the three-proof cover to provide a reference for the positioning of pipe bursts.

[0022] The present invention also provides a method for intelligently identifying piping, using the piping real-time monitoring device according to claim 7, the identification method comprising the following steps:

[0023] S1: According to the height of the dam to be detected and the detection depth, adjust the longitudinal extension width of the variable array pipe burst detection radar device;

[0024] S2: Based on the width of the dike top to be detected, adjust the lateral extension width of the variable array pipe burst detection radar device;

[0025] S3: According to the surface flatness of the detection target, adjust the ground clearance of the variable array pipe burst detection radar device;

[0026] S4: According to the depth and accuracy of pipe burst detection and the inspection speed requirements, set the number of pipe burst radar units to be started and determine the spatial layout of the front-end radar units;

[0027] S5: The high-frequency electromagnetic signal transmitting and receiving array of the variable array piping detection radar device is set to adapt to the complex monitoring environment of different soil types and heights. The piping detection radar device can set the shock wave frequency according to the above array matching;

[0028] S6: According to the bottom width and detection range of the target dike, adjust the up and down height and angle of the acoustic-optical-electrical integrated mast to cover the required specific monitoring surface range, so that the monitoring range of the visible infrared dual spectrum probe covers the potential pipe bursting exposed area, so as to obtain visible and infrared image data of the backwater slope and a certain range of extension, and enable the audio probe of the voiceprint recognition module to effectively receive audio signals;

[0029] S7: The data information collected by the variable array pipe burst detection radar device is preprocessed to form a clear ripple data graph. The frame processing of the signal preprocessing is set to a frame length of 20-40ms. The time-frequency analysis can generate a Mel spectrum graph, and then Z-score standardization processing is performed;

[0030] S8: Analyze the data obtained by the variable array type piping detection radar device and preprocessed through a bidirectional LSTM deep model A to capture the front and back associated features; to improve the analysis efficiency, introduce an attention mechanism to focus on the key frequency bands in the analysis, use depthwise separable convolution to assist feature extraction, and use an improved Sigmoid classification at the output layer;

[0031] S9: After the corrugated data map is identified by the deep learning algorithm, the piping identification data inside the dike is generated in real time. The model quickly judges the piping category and penetration degree according to the piping characteristics, and preliminarily judges the current state and potential risks of the piping, and gives the piping risk factor A1;

[0032] S10: The visible infrared dual-spectrum probe collects the visible light image information on the backwater surface in real time, introduces an environment adaptive module to dynamically adjust the suppression parameters, and based on the pre-trained deep learning model B, conducts the identification of the piping outlet morphology, removes the influence of environmental interference through a denoising algorithm, and gives the piping risk factor A2;

[0033] S11: Based on the difference between the piping water temperature and the backwater side environmental temperature, the visible infrared dual-spectrum probe perceives the infrared image within the set range on the backwater surface in real time. At the same time, the pre-trained deep learning model C conducts the identification of the piping infrared morphology, judges the possibility of the piping, and gives the piping risk factor A3;

[0034] S12: The acoustic wave probe of the voiceprint recognition module captures the audio signal in real time. The noise suppression uses a noise reduction deep learning method and conducts adversarial training in combination with a noise sample library; in the construction of feature engineering, the feature sequence selects a 50-100 frame sliding window, and conducts data augmentation by adding random background noise, time shift / speed change processing, and dynamically mixing piping features; adds noise perturbation through adversarial training, and optimizes the noise robustness of the analysis method through a multi-task learning and time-domain-frequency-domain double verification mechanism; conducts identification and judgment through a noise removal algorithm in combination with the pre-trained deep learning voiceprint recognition model D, and gives the piping risk factor A4;

[0035] S13: According to the pre-trained model, combined with the hydrological and geomorphic morphology around the dike, give the environmental risk factor A5;

[0036] S14: Based on the set fusion algorithm, conduct a weighted calculation and analysis of the above five risk factors to obtain the risk value of the occurrence of piping; combined with spatial positioning and correction technology, obtain the piping risk probability at the spatial positioning point of the dike, and automatically broadcast a warning message for the points where the risk probability exceeds the set value.

[0037] Furthermore, in the step S14, the weighted calculation formula for the risk value is:

[0038]

[0039] where k is the risk value, and A i is the risk factor, and ω i is the weight coefficient corresponding to the risk factor.

[0040] Furthermore, in the step S14, when it is in the daytime with good visual conditions:

[0041] The weight coefficients corresponding to each risk factor are respectively: weight A1(ω1)=0.253, weight A2(ω2)=0.382, weight A3(ω3)=0.185, weight A4(ω4)=0.136, weight A5(ω5)=0.044;

[0042] When it is in the night time:

[0043] The weight coefficients corresponding to each risk factor are respectively: weight A1(ω1)=0.286, weight A2(ω2)=0.019, weight A3(ω3)=0.425, weight A4(ω4)=169, weight A5(ω5)=0.101.

[0044] Advantages of the present invention:

[0045] 1. The variable array type piping detection radar device of the present invention realizes the coverage adaptation for different dam widths, different soil qualities and complex terrains through the longitudinally and laterally multi - adjustable array layout, and solves the contradiction that it is difficult for traditional fixed - array radars to balance the monitoring range and accuracy;

[0046] 2. Each radar unit of the variable array type piping detection radar device of the present invention is built - in with edge - computing control and filtering components, which can complete signal noise reduction and data compression in real - time, shortening the time delay of traditional back - end centralized processing from the second level to the millisecond level, meeting the real - time early - warning requirements; based on the dynamic weight allocation of deep learning, it realizes the adaptive adjustment of the monitoring strategy and avoids the response lag caused by manual intervention.

[0047] 3. The piping real - time monitoring device of the present invention is equipped with a self - moving platform and a Beidou satellite positioning system. Combining with the offline map interpolation algorithm, it can still maintain a positioning accuracy of <5 cm in areas without satellite signals, solving the problems of high missed - detection rate and fuzzy positioning in traditional manual inspections.

[0048] 4. The piping real - time monitoring device of the present invention adopts a modular design and architecture. The front - end radar unit of the piping detection radar device is miniaturized and arranged in an array, and the piping audio detector is placed on the top of the self - moving platform, and an omnidirectional video probe is set. It can collect the panoramic information of sound, light and electricity in the complex field in real - time, can monitor the multi - physical intuitive data related to piping to the greatest extent, and based on the data fusion algorithm, completes the identification and comparison of piping, significantly improving the identification speed and accuracy of piping under complex working conditions.

[0049] 5. The real-time monitoring device for pipe bursts of the present invention integrates radar electromagnetic signals, acoustic wave signals, infrared thermal imaging and visual images to construct a multimodal monitoring system, thereby solving the defect that a single sensor is susceptible to environmental interference and improving recognition reliability under complex working conditions.

[0050] 6. The present invention combines variable array mechanical design, multimodal sensor fusion and dynamic weight deep learning to construct a closed-loop monitoring system with environmental self-adaptation, multi-source signal perception and intelligent decision-making, thereby realizing efficient and high-precision monitoring and early warning of pipe bursts; and achieving comprehensive breakthroughs in core indicators such as positioning accuracy, recognition accuracy, and environmental adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of the overall structure of a real-time monitoring device for piping in the present invention;

[0052] Figure 2 It is a schematic diagram of a partial structure of a longitudinal telescopic bracket in the present invention;

[0053] Figure 3 It is a structural schematic diagram of the variable array piping detection radar device in the present invention;

[0054] Figure 4 It is a schematic diagram of the structure of the radar unit in the present invention;

[0055] Figure 5 A schematic diagram of a horizontal or vertical test grouping of a piping detection radar unit in the present invention;

[0056] Figure 6 It is a schematic diagram of the structure of the mast used to embody the integration of sound, light and electricity in the present invention;

[0057] Figure 7 This is a schematic diagram of the overall control logic of modular monitoring information fusion and early warning based on deep learning of the present invention;

[0058] Figure 8 This is a schematic diagram of the LSTM recognition process of the pipe surge soundprint feature of the present invention;

[0059] Figure 9 A schematic diagram of a visible light and infrared light recognition algorithm for piping according to the present invention;

[0060] Figure 10 It is a schematic diagram of the comprehensive prediction algorithm of piping probability of the present invention.

[0061] In the figure: 1. Self-propelled platform; 2. Mast connection seat; 3. Tri-proof cover; 4. Acoustic, optical and electrical integrated mast; 41. Flange seat; 42. Protective cover; 43. Shielded cable II; 5. Transverse probe module; 51. Visible and infrared dual-spectrum probe; 52. Visible light probe; 6. Voiceprint recognition module; 7. Longitudinal probe module; 71. Forward video probe; 72. Rearward video probe; 8. Satellite guidance device; 9. 5G transmitting antenna; 10. Pipe burst detection radar device; 11. Radar unit; 111. Magnetic flux protection head; 112. Radar signal transmitting antenna; 113. Miniature radar reflection signal receiving antenna; 114. Laser displacement transceiver sensor; 115. Signal transmission optical cable; 150. Transmitting antenna protective cover; 151. DA power amplifier component; 152. Signal acquisition AD component; 153. Edge computing control and filtering component; 154. Shielded cable I; 155. Sealing gasket; 156. Connecting screw; 101. Radar bracket; 102. Flange connector; 103. Roller; 105. Transverse telescopic bracket; 106. Transverse guide rail; 107. Transverse linear bearing; 108. Transverse limit buckle; 116. Longitudinal telescopic bracket; 117. Longitudinal guide rail; 118. Longitudinal linear bearing; 119. Longitudinal limit buckle; 120. Bracket bottom plate; 12. Data storage module; 13. Integrated computing power integrated cabinet; 14. Protective base. Detailed implementation mode

[0062] Now, the present invention will be further described in detail with reference to the accompanying drawings. These drawings are all simplified schematic diagrams, only illustrating the basic structure of the present invention in a schematic manner, so they only show the components related to the present invention.

[0063] The present invention discloses an active variable array type pipe burst detection radar device.

[0064] Refer to Figures 1 to 3 , an active variable array type pipe burst detection radar device includes a radar bracket 101 and a plurality of radar units 11 arranged on the radar bracket 101; the radar bracket 101 includes a bracket bottom plate 120, a transverse telescopic bracket 105 and a longitudinal telescopic bracket 116, the longitudinal telescopic bracket 116 is installed on the bracket bottom plate 120, the longitudinal telescopic bracket 116 includes a longitudinal guide rail 117 and a longitudinal linear bearing 118 slidably connected to the longitudinal guide rail 117, and a longitudinal limit buckle 119 is arranged on the longitudinal linear bearing 118, and the longitudinal linear bearing 118 is relatively fixed to the longitudinal guide rail 117 through the longitudinal limit buckle 119.

[0065] Refer to Figure 2, The horizontal telescopic bracket 105 is installed on the longitudinal linear bearing 118 of the longitudinal telescopic bracket 116. The horizontal telescopic bracket 105 includes a horizontal guide rail 106 and a horizontal linear bearing 107 slidably connected to the horizontal guide rail 106. A horizontal limit buckle 108 is provided on the horizontal linear bearing 107, and the horizontal linear bearing 107 is relatively fixed to the horizontal guide rail 106 through the horizontal limit buckle 108.

[0066] Refer to Figure 2 , Each radar unit 11 is installed on the horizontal linear bearing 107. The radar units 11 are slidably adjusted longitudinally through the position of the longitudinal telescopic bracket 116 in the longitudinal direction and are fixed and limited by the longitudinal limit buckle 119. In this embodiment, the radar units 11 are three-stage adjustable longitudinally between unit spacings of 0.4 m, 0.6 m, and 0.8 m. The radar units 11 are slidably adjusted transversely through the position of the horizontal telescopic bracket 105 in the transverse direction and are fixed and limited by the horizontal limit buckle 108. In this embodiment, the radar units 11 are three-stage adjustable transversely between unit spacings of 0.5 m, 0.75 m, and 1 m. The longitudinal limit buckle 119 and the horizontal limit buckle 108 in this embodiment are locked and fixed by locking bolts.

[0067] Refer to Figure 3 and Figure 5 , In this embodiment, there are 16 radar units 11, which are arranged in a regular 4×4 matrix distribution; a shielding cover is provided outside the radar units 11 to improve the signal-to-noise ratio. The 16 units are connected by a signal transmission optical cable 115; the radar units 11 are adapted to different soil types and heights and other complex dam detection environments by setting longitudinal and transverse transmitting and receiving arrays; they are expanded in multiple stages longitudinally and transversely to adapt to different dam widths and improve the monitoring range and monitoring accuracy. The 16 radar units 11 are divided into four groups, A1, A2, A3, and A4, longitudinally. The same frequency can be used between the four groups, or A1 and A3 rows can be combined, and A2 and A4 rows can be combined, so as to synchronously transmit and collect two groups of signals with different frequency characteristics. The collected signals are filtered and processed through digital filtering technology and then grouped for analysis; they can also be divided into four groups, B1, B2, B3, and B4, transversely. The same frequency can be used between the four groups, or B1 and B3 rows can be combined, and B2 and B4 rows can be combined, so as to synchronously transmit and collect two groups of signals with different frequency characteristics. The collected signals are filtered and processed through digital filtering technology and then grouped for analysis. Each group of radar units 11 selects 2 - 4 groups to turn on according to the complexity of the actual working environment and the usage time. The piping detection radar device 10 can match and set the shock wave frequency according to the above array, and the variable frequency range is 20 KHz to 600 MHz, which is grouped and controlled by the main control operation panel.

[0068] Refer to Figure 4, each radar unit 11 includes a DA power amplifier component 151, a signal acquisition AD component 152, an edge computing control and filtering component 153, and an integrated antenna module;

[0069] Among them, the DA power amplifier component 151 is used to convert the baseband digital signal into an analog radio frequency signal and amplify the signal in terms of power; the signal acquisition AD component 152 is used to collect the echo analog signal received by the antenna and convert it into a digital signal after conditioning. The edge computing control and filtering component 153 is used for real-time signal processing, system control, and noise filtering, removing noise and clutter through digital filtering to improve the signal-to-noise ratio of the signal; and completing data processing locally to reduce the computing pressure at the backend and reduce the transmission delay. There is a sealant pad 155 between the edge computing control and filtering component and the integrated antenna module for waterproofing and tightly connecting through connecting screws 156.

[0070] The integrated antenna module is assembled by an internal three-channel antenna transceiver module, a magnetic flux protection head 111, and a transmitting antenna protection cover 150. The internal three-channel antenna transceiver module is composed of a radar signal transmitting antenna 112, a radar reflected signal micro receiving antenna 113, and a micro laser displacement transceiver sensor 114 arranged at equal intervals longitudinally. Both the radar signal transmitting antenna 112 and the radar reflected signal micro receiving antenna 113 are made of gallium oxide. The measurement range of the micro laser displacement transceiver sensor 114 is 1 m, and the accuracy is 0.1 mm. The radar signal transmitting antenna 112, the radar reflected signal micro receiving antenna 113, and the micro laser displacement transceiver sensor 114 are connected through a shielded cable Ⅰ154.

[0071] Refer to Figure 1 , the present invention also discloses a real-time monitoring device for piping, including a self-propelled platform 1, an optoelectronic and acoustic integrated mast 4 installed on the self-propelled platform 1, and the above-mentioned piping detection radar device 10. The piping detection radar device 10 is fixedly connected to one end of the self-propelled platform 1 through a flange connector 102. In order to avoid damage caused by the contact between the piping detection radar device 10 and the ground due to uneven road surfaces during the movement of the self-propelled platform 1, a roller 103 is installed at the bottom of the piping detection radar device 10. This roller 103 is a silent roller 103 with an adjustable ground clearance to adapt to different road surface conditions.

[0072] The data obtained by the piping detection radar device 10 is transmitted to the data storage module 12 through a signal transmission optical cable 115, and then transmitted to the integrated computing power integrated cabinet 13 for data processing and analysis. The data storage module 12 and the integrated computing power integrated cabinet 13 are vibration-isolatedly connected to the inside of the self-propelled platform 1 through a protection base 14, and the power supply is provided by the self-propelled platform 1.

[0073] Refer to Figure 1 and Figure 6The top wall of the self-propelled platform 1 is fixed with a mast connection seat 2, and the acoustic, optical and electrical integrated mast 4 is fixedly mounted on the mast connection seat 2. The bottom of the acoustic, optical and electrical integrated mast 4 is provided with an acoustic, optical and electrical signal front-end processing module and its three-proof cover 3, which can remove noise and pre-process the monitoring data to meet the data analysis preparation requirements of the back-end early warning platform. The acoustic, optical and electrical integrated mast 4 is connected to the top wall of the three-proof cover 3 through a flange base. A protective cover 42 is provided on the top of the acoustic, optical and electrical integrated mast 4 to protect its precision sensors and electrical circuits in complex and extreme environments.

[0074] Reference Figure 1 and Figure 6 The height and angle of the acoustic-optical-electrical integrated mast 4 are adjustable to ensure effective coverage of the required detection range. The acoustic-optical-electrical integrated mast 4 is provided with a transverse probe module 5, a longitudinal probe module 7, a voiceprint recognition module 6 and a Beidou satellite guidance device 8. Among them, the transverse probe module 5 is composed of a visible infrared dual-spectrum probe 51 on the back slope side and a visible light probe 52 on the water side. The visible infrared dual-spectrum probe 51 and the visible light probe 52 are respectively arranged on the transverse sides of the outer wall of the top of the acoustic-optical-electrical integrated mast 4. Considering that infrared recognition and resolution on the water side is almost impossible, no infrared probe is set on this side. The longitudinal probe module 7 is composed of a forward video probe 71 and a rearward video probe 72. The forward video probe 71 and the rearward video probe 72 are respectively arranged on the longitudinal sides of the outer wall of the top of the acoustic-optical-electrical integrated mast 4, which is convenient for providing auxiliary monitoring of the surrounding environment and spatial obstacle avoidance information for the self-propelled platform 1, preventing the platform from entering the water and dangerous areas by mistake, and combining the vehicle's own automatic driving function to further improve the platform's own safety and self-defense level.

[0075] Reference Figure 1 and Figure 6 The probe of the voiceprint recognition module 6 is placed at the top of the acoustic-optical integrated mast 4 and waterproofed. The voiceprint recognition probe can identify the specific fluctuating audio frequency information of micro-pipe surges with high resolution. After training, the module is sensitive to the sound wave signals in the relevant bands of pipe surges, and after denoising, it can identify the soundprint characteristics of pipe surges with high definition.

[0076] Reference Figure 1 and Figure 6 The Beidou satellite guidance device 8 is installed on the top of the front outer wall of the three-proof cover 3. Through the whip antenna, it can perform high-precision real-time positioning in the field with satellite signals, providing a reference for the positioning of pipe bursts. In areas with weak or no satellite signals, the system automatically determines the position based on its own offline map and travel speed and performs interpolation evaluation based on the signal segment.

[0077] After the signals detected by the above-mentioned probes are preprocessed by the acoustic-optic-electric signal front-end processing module, they are transmitted to the data storage module 12 through the shielded cable II 43, and further transmitted to the integrated computing power integrated cabinet 13 for algorithm prediction. A 5G transmitting antenna 9 is provided at the rear end of the top wall of the three-proof cover 3 for sending real-time warning signals to the flood control headquarters or the command center; the monitoring warning signals and pre-judgment results generated by the intelligent pipe burst identification can be intuitively displayed in the vehicle-mounted integrated display system and can be quickly transmitted to the regional warning center through the 5G transmitting antenna 9.

[0078] The present invention also discloses a method for intelligent pipe burst identification.

[0079] Referring to Figures 7 to 10 , a method for intelligent pipe burst identification uses the above-mentioned pipe burst real-time monitoring device, including the following steps:

[0080] S1: According to the height of the dike to be detected and the detection depth, adjust the longitudinal extension width of the variable array type pipe burst detection radar device 10;

[0081] S2: Based on the top width of the dike to be detected, adjust the lateral extension width of the variable array type pipe burst detection radar device;

[0082] S3: According to the surface flatness of the detection target dike, adjust the ground clearance of the variable array type pipe burst detection radar device 10;

[0083] S4: According to the depth and accuracy of pipe burst detection and the requirements of the inspection speed, set the number of groups of the radar units 11 to be turned on, and determine the spatial layout of the front-end radar units 11;

[0084] S5: Set the high-frequency electromagnetic signal transmitting and receiving formation of the variable array type pipe burst detection radar device 10 to adapt to complex monitoring environments such as different soil qualities and heights. The pipe burst detection radar device 10 can match and set the shock wave frequency according to the above formation, and the variable frequency range is 20KHz to 600MHz. The monitoring depth is inversely proportional to the frequency selection. The deeper the depth to be monitored, the lower frequency range is selected for detection;

[0085] S6: According to the bottom width and detection range of the detection target dike, adjust the up and down height and angle of the acoustic-optic-electric integrated mast 4 to cover the specific monitored surface range required, so that the monitoring range of the visible infrared dual-spectrum probe 51 covers the potential pipe burst outcrop area to obtain visible and infrared image data of the backwater slope and a certain range of extension, and enable the audio probe of the voiceprint recognition module 6 to effectively receive audio signals;

[0086] S7: The data information collected by the variable array type piping detection radar device 10 forms a clear corrugation data graph after preprocessing. The frame processing of the signal preprocessing is set to a frame length of 20 - 40 ms, and the time-frequency analysis can generate a Mel spectrogram, followed by Z-score normalization processing;

[0087] S8: As shown in Figure 7 and Figure 9 , perform bidirectional LSTM deep model analysis A on the data obtained by the variable array type piping detection radar device 10 and preprocessed to capture the front and back correlation features; to improve the analysis efficiency, introduce an attention mechanism to focus on key frequency bands, use depthwise separable convolution to assist feature extraction, and use an improved Sigmoid classification at the output layer;

[0088] S9: After the corrugation data graph is identified by the deep learning algorithm, the piping identification data in the dike body is generated in real time. The model quickly judges the piping category and penetration degree according to the piping characteristics, and preliminarily judges the current state and potential risks of the piping, and gives the piping risk factor A1;

[0089] S10: As shown in Figure 7 and Figure 9 , the visible infrared dual-spectrum probe 51 collects the visible light image information of the backwater surface in real time, introduces an environment adaptive module, dynamically adjusts the suppression parameters, and performs the identification of the piping outlet morphology based on the pre-trained deep learning model B, and removes the influence of environmental interference through the denoising algorithm, and gives the piping risk factor A2;

[0090] S11: As shown in Figure 7 and Figure 9 , based on the difference between the piping water temperature and the backwater side environmental temperature, the visible infrared dual-spectrum probe 51 perceives the infrared image within the set range of the backwater surface in real time, and at the same time performs the identification of the piping infrared morphology through the pre-trained deep learning model C, judges the possibility of the piping, and gives the piping risk factor A3;

[0091] S12: As shown in Figure 8 , the acoustic wave probe of the voiceprint recognition module 6 captures the audio signal in real time. The noise suppression uses the deep learning method for noise reduction and performs adversarial training in combination with the noise sample library; in the construction of the feature engineering, the feature sequence selects a 50 - 100 frame sliding window, and performs data enhancement by adding random background noise, time shift / speed change processing and dynamically mixing piping features; add noise perturbation through adversarial training, and optimize the noise robustness of the analysis method through multi-task learning and time-domain - frequency-domain double verification mechanism; perform identification and judgment through the noise removal algorithm in combination with the pre-trained deep learning voiceprint recognition model D, and give the piping risk factor A4;

[0092] S13: According to the pre-trained model, combined with the hydrological and geomorphic features around the dam (the probability of piping increases significantly for specific geomorphic features), further give the environmental risk factor A5. The environmental risk factor levels and values at the current stage of the present invention are as follows in the table:

[0093]

[0094] S14: As Figure 7 , based on the set fusion algorithm, perform weighted calculation and analysis on the above five risk factors to obtain the risk value k of piping occurrence. The weighted calculation formula is:

[0095] A i is the risk factor, and ω i is the weight coefficient corresponding to the risk factor.

[0096] In the above analysis, when in the daytime with good visibility conditions:

[0097] The weight coefficients corresponding to each risk factor are respectively: weight A1(ω1) = 0.253, weight A2(ω2) = 0.382, weight A3(ω3) = 0.185, weight A4(ω4) = 0.136, weight A5(ω5) = 0.044;

[0098] When in the night time:

[0099] The weight coefficients corresponding to each risk factor are respectively: weight A1(ω1) = 0.286, weight A2(ω2) = 0.019, weight A3(ω3) = 0.425, weight A4(ω4) = 169, weight A5(ω5) = 0.101.

[0100] Combined with spatial positioning and correction technology, obtain the piping risk probability P of the dam spatial positioning point, and automatically broadcast warning information for the points with a probability greater than 75% (adjustable).

[0101] Through the above steps, the spatial and temporal positions of piping can be self-detected, real-time, and efficiently, and early warnings can be given. Through the automatic parallel comparison of multiple monitoring data, the accuracy of the monitoring effect can be further improved, the efficiency can be significantly improved, and the original work intensity can be reduced.

[0102] The present invention can autonomously patrol and locate the spatial position of piping and further give early warnings. The accuracy and timeliness are qualitatively improved compared with the previous manual inspections, and it can greatly save the funds and personnel input for dike inspections, and significantly reduce the difficult workload of dike inspections with a large area and a large quantity;

[0103] Through modular design and architecture, the front-end radar unit 11 of the piping detection radar device 10 is miniaturized and arranged in an array, and the piping audio detector is placed on the top of the self-propelled platform. An omnidirectional video probe is set up to be able to collect the panoramic information of sound, light, and electricity monitoring at the complex field site in real time, and can monitor the most multi-physical intuitive data related to piping to the greatest extent. Based on the data fusion algorithm, the identification and comparison of piping are completed, significantly improving the identification speed and accuracy of piping under complex working conditions;

[0104] Based on the detected piping field data, based on the deep learning algorithm, signal preprocessing is carried out, and adversarial training is combined with the noise sample library to identify and suppress noise. Based on feature engineering, a time series feature sequence is constructed. Through the data enhancement strategy and LSTM model design, combined with noise robustness optimization, high-quality single-signal discrimination results can be obtained;

[0105] Through the fusion analysis of multiple signals, the evaluation of the spatial occurrence state and connectivity of piping is significantly improved, and the positioning error is less than 5 cm. Through active confrontation between deep feature learning and environmental noise, high-precision piping identification under complex working conditions can be achieved. Experiments show that the accuracy rate can reach more than 90% when the signal-to-noise ratio ≥ 5 dB.

[0106] Inspired by the ideal embodiments of the present invention described above, through the above description, relevant staff can completely make various changes and modifications without departing from the technical idea of this invention. The technical scope of this invention is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. An active variable array type piping detection radar device, characterized in that: It includes a radar bracket (101) and a number of radar units (11) arranged on the radar bracket (101); The radar bracket (101) includes a bracket bottom plate (120), a transverse telescopic bracket (105) and a longitudinal telescopic bracket (116). The longitudinal telescopic bracket (116) is installed on the bracket bottom plate (120). The longitudinal telescopic bracket (116) includes a longitudinal guide rail (117) and a longitudinal linear bearing (118) slidably connected to the longitudinal guide rail (117). A longitudinal limit buckle (119) is provided on the longitudinal linear bearing (118), and the longitudinal linear bearing (118) is relatively fixed to the longitudinal guide rail (117) through the longitudinal limit buckle (119); The transverse telescopic bracket (105) is installed on the longitudinal linear bearing (118) of the longitudinal telescopic bracket (116). The transverse telescopic bracket (105) includes a transverse guide rail (106) and a transverse linear bearing (107) slidably connected to the transverse guide rail (106). A transverse limit buckle (108) is provided on the transverse linear bearing (107), and the transverse linear bearing (107) is relatively fixed to the transverse guide rail (106) through the transverse limit buckle (108); Each of the radar units (11) is installed on the transverse linear bearing (107). The radar units (11) are slidably adjusted longitudinally by the position of the longitudinal telescopic bracket (116) in the longitudinal direction and fixed and limited by the longitudinal limit buckle (119). Then, the radar units (11) are adjusted in multiple levels within the unit spacing in the longitudinal direction; Each of the radar units (11) is slidably adjusted transversely by the position of the transverse telescopic bracket (105) in the transverse direction and fixed and limited by the transverse limit buckle (108). Then, the radar units (11) are adjusted in multiple levels within the unit spacing in the transverse direction.

2. The active variable array type piping detection radar device according to claim 1, wherein: There are 16 radar units (11), which are arranged in a regular 4×4 matrix on the radar bracket (101); the radar units (11) adapt to the complex dam detection environment of different soil qualities and heights by setting different transmitting and receiving arrays longitudinally and transversely; the radar units (11) are expanded in multiple levels longitudinally and transversely to adapt to different dam widths and improve the monitoring range and accuracy.

3. The active variable array type piping detection radar device according to claim 2, wherein: The radar units (11) are divided into four groups, A1, A2, A3, and A4, longitudinally. The four groups of radar units (11) use the same frequency, or A1 and A3 rows are combined, and A2 and A4 rows are combined, so as to synchronously transmit and collect two groups of signals with different frequency characteristics. The collected signals are analyzed by digital filtering technology in groups.

4. The active variable array type piping detection radar device according to claim 2, wherein: The radar units (11) are divided into four groups, B1, B2, B3, and B4, transversely. The four groups of radar units (11) use the same frequency, or B1 and B3 rows are combined, and B2 and B4 rows are combined, so as to synchronously transmit and collect two groups of signals with different frequency characteristics. The collected signals are analyzed by digital filtering technology in groups.

5. The active variable array type piping detection radar device according to claim 2, characterized in that: Each radar unit (11) comprises a DA power amplifier component (151), a signal acquisition AD component (152), an edge computing control and filtering component (153) and an integrated antenna module; the DA power amplifier component (151) is used to convert a baseband digital signal into an analog radio frequency signal and to amplify the power of the signal; the signal acquisition AD component (152) is used to collect the echo analog signal received by the antenna and convert it into a digital signal after conditioning; the edge computing control and filtering component (153) is used for real-time signal processing, system control and noise filtering, and removes noise and clutter through digital filtering to improve the signal-to-noise ratio; and completes data processing locally to reduce back-end computing pressure and transmission delay; The integrated antenna module is assembled from a built-in three-channel antenna transceiver module, a magnetic flux protection head (111) and a transmitting antenna protection cover (150); the built-in three-channel antenna transceiver module is composed of a radar signal transmitting antenna (112), a radar reflection signal micro receiving antenna (113) and a micro laser displacement transceiver sensor (114) arranged at equal intervals in the longitudinal direction; the radar signal transmitting antenna (112), the radar reflection signal micro receiving antenna (113) and the micro laser displacement transceiver sensor (114) are connected by a shielded cable I (154).

6. A real-time monitoring device for piping, characterized in that: The invention comprises a self-propelled platform (1), an acoustic-optical-electrical integrated mast (4) installed on the self-propelled platform (1), and a pipe burst detection radar device (10) as claimed in claim 5; the pipe burst detection radar device (10) is fixedly connected to one end of the self-propelled platform (1) via a flange connector (102), and a roller (103) is installed at the bottom of the pipe burst detection radar device (10); A mast connection seat (2) is fixed to the top wall of the self-propelled platform (1), an acoustic, photoelectric integrated mast (4) is fixedly mounted on the mast connection seat (2), and an acoustic, photoelectric signal front-end processing module and a three-proof cover (3) are arranged at the bottom of the acoustic, photoelectric integrated mast (4), which can remove noise and pre-process monitoring data to meet the data analysis preparation requirements of a back-end early warning platform.

7. The real-time monitoring device for piping according to claim 6, characterized in that: The acoustic-optical-electrical integrated mast (4) is provided with a transverse probe module (5), a longitudinal probe module (7), a voiceprint recognition module (6) and a Beidou satellite guidance device (8); The transverse probe module (5) is composed of a visible infrared dual spectrum probe (51) on the back slope side and a visible light probe (52) on the water side; the visible infrared dual spectrum probe (51) and the visible light probe (52) are respectively arranged on the transverse sides of the outer side wall of the top of the acoustic-optical integrated mast (4); The longitudinal probe module (7) is composed of a forward video probe (71) and a backward video probe (72), and the forward video probe (71) and the backward video probe (72) are respectively arranged on the longitudinal sides of the outer side wall of the top of the acoustic-optical integrated mast (4); The probe of the voiceprint recognition module (6) is placed at the top of the acoustic-optical-electrical integrated mast (4); the Beidou satellite guidance device (8) is installed at the top of the front outer side wall of the three-proof cover (3) to provide a reference for the positioning of the pipe burst.

8. An intelligent identification method for piping, characterized in that: Adopt the piping real-time monitoring device described in claim 7 above, and the identification method includes the following steps: S1: According to the height and detection depth of the dike to be detected, adjust the longitudinal extension width of the variable array type piping detection radar device (10); S2: Based on the crest width of the dike to be detected, adjust the lateral extension width of the variable array type piping detection radar device (10); S3: According to the surface flatness of the detection target dike, adjust the ground clearance of the variable array type piping detection radar device (10); S4: According to the detection depth, accuracy of piping detection and the requirement of inspection speed, set the number of powered-on groups of the piping radar unit (11), and determine the spatial layout of the front-end radar unit (11); S5: Set the high-frequency electromagnetic signal transmitting and receiving formation of the variable array type piping detection radar device (10) to adapt to the complex monitoring environment of different soil qualities and heights. The piping detection radar device (10) can match and set the shock wave frequency according to the above formation; S6: According to the bottom width and detection range of the detection target dike, adjust the up and down height and angle of the acousto-optic-electronic integrated mast (4) to cover the specific monitored ground surface range required, so that the monitoring range of the visible infrared dual-spectrum probe (51) covers the potential piping outcrop area, to obtain visible and infrared image data of the backwater slope and a certain range of extension, and enable the audio probe of the voiceprint recognition module (6) to effectively receive audio signals; S7: The data information collected by the variable array type piping detection radar device (10) forms a clear ripple data graph after preprocessing. The frame processing of the signal preprocessing is set to a frame length of 20 - 40 ms, and the time-frequency analysis can generate a Mel spectrogram, and then perform Z-score normalization processing; S8: Perform bidirectional LSTM depth model analysis A on the data obtained by the variable array type piping detection radar device (10) and preprocessed to capture the front and back correlation features; to improve the analysis efficiency, the analysis introduces an attention mechanism to focus on the key frequency bands, uses depthwise separable convolution to assist feature extraction, and uses improved Sigmoid classification in the output layer; S9: After the ripple data graph is identified by the deep learning algorithm, the piping identification data in the dike body is generated in real time. The model quickly judges the piping category and penetration degree according to the piping characteristics, and preliminarily judges the current state and potential risks of the piping, and gives the piping risk factor A1; S10: The visible infrared dual-spectrum probe (51) introduces an environment adaptive module by real-time collecting the visible light image information of the backwater surface, dynamically adjusts the suppression parameters, and performs identification of the piping outlet morphology based on the pre-trained deep learning model B, and removes the influence of environmental interference through a denoising algorithm, and gives the piping risk factor A2; S11: Based on the difference between the piping water temperature and the backwater side environmental temperature, the visible infrared dual-spectrum probe (51) real-time senses the infrared image within a set range of the backwater surface, and at the same time performs piping infrared morphology identification through the pre-trained deep learning model C, judges the possibility of piping, and gives the piping risk factor A3; S12: The acoustic wave probe of the voiceprint recognition module (6) captures the audio signal in real time. The noise suppression adopts the noise reduction deep learning method and conducts adversarial training in combination with the noise sample library. In the construction of feature engineering, a 50-100-frame sliding window is selected for the feature sequence. Data augmentation is carried out by adding random background noise, time shift / speed change processing, and dynamically mixing the characteristics of pipe bursts. Noise perturbation is added through adversarial training, and the noise robustness of the analysis method is optimized through multi-task learning and the time-domain-frequency-domain dual verification mechanism. The recognition and judgment are carried out by removing the noise algorithm in combination with the pre-trained deep learning voiceprint recognition model D, and the pipe burst risk factor A4 is given. S13: According to the pre-trained model, combined with the hydrological and geomorphic forms around the dam, the environmental risk factor A5 is given. S14: Based on the set fusion algorithm, the above five risk factors are weighted and calculated and analyzed to obtain the risk value of the occurrence of pipe bursts. Combined with the space positioning and correction technology, the pipe burst risk probability of the dam space positioning point is obtained, and an automatic warning message is broadcast for the points where the risk probability exceeds the set value.

9. The intelligent identification method for piping according to claim 8, wherein: In the step S14, the weighted calculation formula for the risk value is: Among them, k is the risk value, A i is the risk factor, ω i is the weight coefficient corresponding to the risk factor.

10. A method for intelligent identification of piping, according to claim 9, characterized in that: In the step S14, when it is in the daytime with good visibility: The weight coefficients corresponding to each risk factor are: weight A1(ω1)=0.253, weight A2(ω2)=0.382, weight A3(ω3)=0.185, weight A4(ω4)=0.136, weight A5(ω5)=0.044; When it is in the night time: The weight coefficients corresponding to each risk factor are: weight A1(ω1)=0.286, weight A2(ω2)=0.019, weight A3(ω3)=0.425, weight A4(ω4)=0.169, weight A5(ω5)=0.101.

Images