Underwater object recognition device based on low-frequency magnetic conductivity frequency dispersion characteristic

Through the underwater object recognition device based on the dispersion characteristics of low-frequency magnetic permeability, the coordinated cooperation of maneuver control, low-frequency magnetic wave excitation and dispersion signal capture modules is adopted to solve the problem of insufficient recognition accuracy and anti-interference ability of traditional underwater detection equipment in complex media environments, and high-precision underwater target exploration and submarine facility inspection are achieved.

CN120507792APending Publication Date: 2025-08-19BEIHANG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202510715736.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-05-30
Publication Date
2025-08-19

AI Technical Summary

Technical Problem

Traditional underwater object detection technology has low target recognition accuracy, weak environmental anti-interference ability and insufficient dynamic adaptability in complex media environments, making it difficult to achieve high-precision underwater target exploration and submarine facility inspection.

Method used

The underwater object recognition device based on the dispersion characteristics of low-frequency magnetic permeability is adopted, including a maneuver control module, a low-frequency magnetic wave excitation module, a dispersion signal capture module and an intelligent analysis module. Through the coordinated cooperation of wide-band magnetic wave dynamic excitation, high-sensitivity capture of dispersion signals and intelligent analysis technology of material electromagnetic fingerprints, combined with rigid anti-interference bracket and multi-dimensional signal processing, high-precision target recognition is achieved.

Benefits of technology

Significantly improve target recognition accuracy and anti-interference ability in complex underwater environments, support high-precision detection of underwater target exploration, submarine facility inspection and submerged substrate archaeological identification, and adapt to dynamic adaptability of different underwater working conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120507792A_ABST
    Figure CN120507792A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of underwater detection, in particular to an underwater object recognition device based on low-frequency magnetic conductivity frequency dispersion characteristics. Comprising a maneuvering control module used for moving underwater, a low-frequency magnetic guided wave excitation module used for identifying different underwater media, a frequency dispersion signal capturing module used for separating interference components, and an intelligent analysis module used for improving underwater identification precision. And the intelligent analysis module is electrically connected with the maneuvering control module, the low-frequency magnetic guided wave excitation module and the frequency dispersion signal acquisition module. According to the invention, through cooperation of the maneuvering control module, the low-frequency magnetic guided wave excitation module, the frequency dispersion signal capturing module and the intelligent analysis module, problems of low target identification precision, weak environment anti-interference capability and insufficient dynamic adaptability of traditional underwater detection equipment in a complex medium environment can be effectively solved; therefore, the method can be effectively applied to high-precision detection scenes such as underwater target exploration, submarine facility inspection and substrate archaeological identification.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of underwater detection technology, and in particular to an underwater object recognition device based on low-frequency magnetic permeability dispersion characteristics. Background Art

[0002] Conventional underwater object detection technology has the following bottlenecks: (1) Conventional magnetic detection devices rely on fixed-frequency excitation (such as DC or a single low frequency) and cannot capture the dispersion characteristics of the material's magnetic permeability as it changes with frequency, resulting in insufficient metal / non-metal target identification capabilities; (2) Sonar detection technology is susceptible to water reverberation, multipath effects, and biological noise interference, and its resolution is significantly reduced in complex terrain or turbid waters; (3) Existing magnetic field measurements mostly use a single sensor, which is difficult to adapt to the needs of dynamic detection in underwater three-dimensional space, and lacks anti-interference acquisition technology for weak abnormal signals; (4) Target recognition algorithms are generally based on static magnetic parameter threshold judgments and cannot extract the intrinsic electromagnetic characteristics of materials from dynamic dispersion responses.

[0003] These issues have led to traditional underwater detection equipment facing challenges such as high misjudgment rates and weak multi-target classification capabilities in scenarios like archaeological exploration and submarine pipeline inspection. Furthermore, the mechanical design of existing magnetic detection devices is limited to fixed platforms or towed carriers, making it difficult to achieve high-precision three-dimensional trajectory control and adaptive detection in complex terrain.

[0004] Therefore, there is an urgent need for an underwater object identification device based on the dispersion characteristics of low-frequency magnetic permeability. Through the collaboration of broadband magnetic wave dynamic excitation, high-sensitivity capture of dispersion signals and intelligent analysis technology of material electromagnetic fingerprints, it can be effectively applied to high-precision detection scenarios such as underwater target exploration, seabed facility inspection and archaeological identification of sunken objects. Summary of the Invention

[0005] The purpose of the present invention is to provide an underwater object identification device based on the dispersion characteristics of low-frequency magnetic permeability to solve the problems existing in the above-mentioned prior art.

[0006] To achieve the above-mentioned objectives, the present invention provides the following solution: an underwater object identification device based on the dispersion characteristics of low-frequency magnetic permeability, comprising a maneuverable control module for moving underwater, a low-frequency magnetic wave excitation module for identifying different underwater media, a dispersion signal capture module for separating interference components, and an intelligent analysis module for improving underwater identification accuracy, wherein the intelligent analysis module is electrically connected to the maneuverable control module, the low-frequency magnetic wave excitation module, and the dispersion signal capture module; the maneuverable control module includes an air cushion, and a propulsion system for mobilizing the movement of the air cushion is provided at the bottom of the air cushion; the low-frequency magnetic wave excitation module includes a motorized control module provided on the air cushion. A square carbon fiber honeycomb array perforated plate, a plurality of low-frequency electromagnetic transmitting units are installed on the carbon fiber honeycomb array perforated plate; the dispersion signal capture module includes an aluminum alloy partition arranged above the carbon fiber honeycomb array perforated plate, and sensor arrays are respectively arranged on both sides of the aluminum alloy partition, and the sensor array includes a mesh support plate, and the mesh support plate is connected to the sensor array support plate through a programmable rotating drum, and a plurality of fluxgate sensors are respectively installed on the mesh support plate and the sensor array support plate; the intelligent analysis module includes a central control module fixedly connected to the top surface of the aluminum alloy partition, and a communication antenna is installed on the top surface of the central control module.

[0007] Preferably, the propulsion system includes an underwater propeller motion control module fixedly connected to the top surface of the air cushion, the underwater propeller motion control module is electrically connected to a plurality of servo motors, the servo motors are fixedly connected to the bottom of the air cushion, and the output shafts of the servo motors are transmission-connected to a titanium alloy propeller.

[0008] Preferably, a backup energy supply unit is provided on one side of the underwater propeller motion control module, and the backup energy supply unit is fixedly connected to the top surface of the air cushion.

[0009] Preferably, the carbon fiber honeycomb array perforated plate is fixedly connected to the top surface of the air cushion through a plurality of carbon fiber pillars, and the underwater propeller motion control module and the backup energy supply unit are arranged between the carbon fiber honeycomb array perforated plate and the air cushion.

[0010] Preferably, a square permanent magnet, piezoelectric ceramics and an inverter amplifier circuit are installed in the low-frequency electromagnetic transmitting unit.

[0011] Preferably, one end of the carbon fiber support column away from the air cushion passes through the carbon fiber honeycomb array perforated plate and is fixedly connected to the aluminum alloy partition.

[0012] Preferably, rectangular cable channels are installed on both sides of the mesh support plate and the sensor array support plate.

[0013] Preferably, the mesh support plate is rotatably connected to a commutator on both sides of one end away from the sensor array support plate, the commutator is connected to a brushless DC motor through a ceramic shaft, and the commutator, the brushless DC motor and the aluminum alloy partition are fixedly connected.

[0014] Preferably, an acoustic distance sensor is installed on one end of the sensor array support plate away from the mesh support plate.

[0015] Preferably, the central control module is provided with a central controller, an intelligent power supply station, a signal processing module, a GPS and an inertial measurement unit.

[0016] The present invention discloses the following technical effects:

[0017] Through the collaborative cooperation of a maneuvering control module, a low-frequency magnetic wave excitation module, a dispersion signal capture module, and an intelligent analysis module, the present invention can effectively solve the problems of low target recognition accuracy, weak environmental interference resistance, and insufficient dynamic adaptability of traditional underwater detection equipment in complex media environments; making the present invention effectively applicable to high-precision detection scenarios such as underwater target exploration, seabed facility inspection, and archaeological identification of sunken objects. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 It is a schematic diagram of the overall structure of the present invention;

[0020] Figure 2 This is a schematic diagram of the central control integration system structure of the present invention;

[0021] Figure 3 This is a schematic structural diagram of the low-frequency electromagnetic transmitting unit of the present invention;

[0022] Among them, 1. Communication antenna; 2. Mesh support plate; 3. Commutator; 4. Brushless DC motor; 5. Aluminum alloy partition; 6. Low-frequency electromagnetic transmitting unit; 61. Square permanent magnet; 62. Piezoelectric ceramics; 63. Inverter amplifier circuit; 7. Carbon fiber honeycomb array perforated plate; 8. Underwater propeller motion control module; 9. Servo motor; 10. Ceramic shaft; 11. Backup energy supply unit; 12. Titanium alloy propeller; 13. Water grass cutter; 14. Air cushion; 15. Acoustic distance sensor; 16. Fluxgate sensor; 17. Sensor array support plate; 18. Programmable drum; 19. Rectangular cable channel; 20. Carbon fiber pillar; 21. Central control module; 211. Central controller; 212. Intelligent power supply station; 213. Signal processing module; 214. GPS; 215. Inertial measurement unit. DETAILED DESCRIPTION

[0023] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0024] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0025] Reference Figure 1-Figure 3 The present invention provides an underwater object recognition device based on the dispersion characteristics of low-frequency magnetic permeability, comprising a maneuvering control module for moving underwater, a low-frequency magnetic wave excitation module for identifying different underwater media, a dispersion signal capture module for separating interference components, and an intelligent analysis module for improving underwater recognition accuracy. The intelligent analysis module is electrically connected to the maneuvering control module, the low-frequency magnetic wave excitation module, and the dispersion signal capture module; the maneuvering control module comprises an air cushion 14, the bottom of which is provided with a propulsion system for mobilizing the air cushion 14; the low-frequency magnetic wave excitation module comprises a carbon fiber honeycomb array perforated plate arranged above the air cushion 14 7. A plurality of low-frequency electromagnetic transmitting units 6 are installed on the carbon fiber honeycomb array perforated plate 7; the dispersion signal capture module includes an aluminum alloy partition plate 5 arranged above the carbon fiber honeycomb array perforated plate 7, and sensor arrays are respectively arranged on both sides of the aluminum alloy partition plate 5, and the sensor array includes a mesh support plate 2, and the mesh support plate 2 is connected to the sensor array support plate 17 by a programmable rotating drum 18, and a plurality of fluxgate sensors 16 are respectively installed on the mesh support plate 2 and the sensor array support plate 17; the intelligent analysis module includes a central control module 21 fixedly connected to the top surface of the aluminum alloy partition plate 5, and a communication antenna 1 is installed on the top surface of the central control module 21.

[0026] Through the collaborative cooperation of a maneuvering control module, a low-frequency magnetic wave excitation module, a dispersion signal capture module, and an intelligent analysis module, the present invention can effectively solve the problems of low target recognition accuracy, weak environmental interference resistance, and insufficient dynamic adaptability of traditional underwater detection equipment in complex media environments; making the present invention effectively applicable to high-precision detection scenarios such as underwater target exploration, seabed facility inspection, and archaeological identification of sunken objects.

[0027] To further optimize the solution, the propulsion system includes an underwater propeller motion control module 8 fixedly connected to the top surface of the air cushion 14. The underwater propeller motion control module 8 is electrically connected to multiple servo motors 9. The servo motors 9 are fixedly connected to the bottom of the air cushion 14. The output shaft of the servo motor 9 is connected to the titanium alloy propeller 12.

[0028] The titanium alloy propeller 12 is driven by the servo motor 9 to realize propulsion vector control.

[0029] In order to prevent entanglements such as water plants from affecting the servo motor 9 in driving the titanium alloy propeller 12, a water plant cutter 13 is provided between the servo motor 9 and the titanium alloy propeller 12, and the water plant cutter 13 can be used to clean up entanglements such as water plants.

[0030] As a further optimization solution, a backup energy supply unit 11 is provided on one side of the underwater propeller motion control module 8, and the backup energy supply unit 11 is fixedly connected to the top surface of the air cushion 14. The backup energy supply unit 11 supplies power to the central control module 21 first.

[0031] In a further optimized solution, the carbon fiber honeycomb array perforated plate 7 is fixedly connected to the top surface of the air cushion 14 through a plurality of carbon fiber pillars 20, and the underwater propeller motion control module 8 and the backup energy supply unit 11 are arranged between the carbon fiber honeycomb array perforated plate 7 and the air cushion 14. This enables the carbon fiber honeycomb array perforated plate 7 to be stably installed above the air cushion 14, and enables the underwater propeller motion control module 8 and the backup energy supply unit 11 to be effectively installed on the air cushion 14 below the carbon fiber honeycomb array perforated plate 7.

[0032] According to a further optimized solution, a square permanent magnet 61 , a piezoelectric ceramic 62 and an inverter amplifier circuit 63 are installed in the low-frequency electromagnetic transmitting unit 6 .

[0033] The low-frequency electromagnetic transmitting unit 6 is polygonal. The present invention adopts a hexagonal prism-shaped low-frequency electromagnetic transmitting unit 6. The piezoelectric ceramic 62 generates 1-100 Hz mechanical vibration through the inverse piezoelectric effect, drives the square permanent magnet 61 to generate a swept-frequency magnetic signal, and multiple low-frequency electromagnetic transmitting units 6 achieve magnetic field intensity vector superposition through phase synchronization technology, and directionally focus toward the target direction.

[0034] The present invention uses 45 low-frequency electromagnetic transmitting units 6 through calculation and phase synchronization technology to achieve magnetic field intensity vector superposition and directionally focus toward the target direction.

[0035] In a further optimized solution, one end of the carbon fiber support 20 away from the air cushion 14 passes through the carbon fiber honeycomb array perforated plate 7 and is fixedly connected to the aluminum alloy partition 5, so that the aluminum alloy partition 5 can be stably placed above the carbon fiber honeycomb array perforated plate 7.

[0036] In a further optimized solution, rectangular cable channels 19 are respectively installed on both sides of the mesh support plate 2 and the sensor array support plate 17. The rectangular cable channels 19 facilitate signal transmission.

[0037] To further optimize the solution, the mesh support plate 2 is rotatably connected to both sides of one end away from the sensor array support plate 17 with a commutator 3, the commutator 3 is connected to a brushless DC motor 4 through a ceramic shaft 10, and the commutator 3, the brushless DC motor 4 and the aluminum alloy partition 5 are fixedly connected.

[0038] The brushless DC motor 4 drives the ceramic shaft 10 to rotate, the ceramic shaft 10 drives the commutator 3 to work, the commutator 3 drives the mesh support plate 2 to rotate, and the programmable drum 18 drives the sensor array support plate 17 to expand and contract.

[0039] A glass fiber reinforced vinyl ester resin mesh support structure is used between the mesh support plate 2 and the sensor array support plate 17 , and a plurality of fluxgate sensors 16 are installed in the glass fiber reinforced vinyl ester resin mesh support structure.

[0040] In a further optimized solution, an acoustic distance sensor 15 is installed at one end of the sensor array support plate 17 away from the mesh support plate 2. That is, a programmable rotating drum 18 is installed between the sensor array support plate 17 and the mesh support plate 2; the acoustic distance sensor 15 is used to monitor obstacles in real time.

[0041] To further optimize the solution, the central control module 21 is provided with a central controller 211 , an intelligent power supply station 212 , a signal processing module 213 , a GPS 214 and an inertial measurement unit 215 .

[0042] The inertial measurement unit 215 of the central control module 21 is used to collect attitude data in real time. The collected attitude data is calculated by the central controller 211 and used to adjust the speed difference of the titanium alloy propeller 12. Combined with the positioning of GPS214, it can realize underwater three-dimensional motion trajectory control.

[0043] The present invention breaks through the deformation error limitations of traditional flexible structures and adopts a rigid anti-interference bracket and multi-dimensional signal processing technology: the low-frequency magnetic wave excitation module is based on the collaborative emission of multi-stage oscillators in a honeycomb array, and realizes the directional radiation enhancement of low-frequency electromagnetic waves through phase synchronization control; the dispersion signal capture module is combined with a mesh magnetic sensor array fixed by a rigid anti-interference bracket, and uses time domain correlation analysis technology to improve dynamic signal stability; the intelligent analysis module realizes high-resolution identification of the dispersion characteristics of material magnetic permeability through joint feature extraction in the time-frequency domain.

[0044] Working process:

[0045] During signal capture, the central controller 211 starts the brushless DC motor 4, which drives the ceramic shaft 10 to rotate, the ceramic shaft 10 drives the commutator 3 to work, and the commutator 3 drives the mesh support plate 2 to rotate; the programmable rotating drum 18 drives the sensor array support plate 17 to unfold, so that the sensor array is immersed underwater; each array node uses a high-sensitivity fluxgate sensor 16 to synchronously collect magnetic induction intensity at a sampling rate of 500Hz, and after the environmental magnetic field noise signal is filtered out by the fully digital phase-locked amplifier algorithm, it is transmitted to the signal processing module 213 through the rectangular cable channel 19; the signal processing module 213 has a built-in frequency dispersion feature extraction algorithm, which first separates the fundamental frequency and harmonic components through fast Fourier transform, and then combines continuous wavelet transform CWT to extract the attenuation coefficient of magnetic permeability with frequency, and finally inputs the feature vector into the pre-trained convolutional neural network for material classification.

[0046] In actual operation, after the present invention receives the task instruction through the communication antenna 1, the intelligent power supply station 212 preferentially activates the backup energy supply unit 11 for underwater propulsion; the acoustic distance sensor 15 monitors obstacles in real time, and when encountering entanglements such as water plants, the water plant cutter 13 will clean them up; after the target is identified, the sensor array is automatically retracted to above the aluminum alloy partition 5, and all process data are uploaded to the shore-based control center through the communication antenna 1, forming a closed-loop detection system.

[0047] The maneuvering control module of the present invention adopts the coordinated drive of the air cushion 14 carrier and the multi-directional propulsion system, and realizes precise control of the underwater three-dimensional space through intelligent navigation and positioning technology; the air cushion 14 carrier is integrally formed of lightweight and high-strength composite materials, and the surface is covered with a corrosion-resistant functional coating, combined with real-time terrain feedback to achieve stable operation and obstacle avoidance control in complex seabed environments; the propulsion system is based on a vector thrust adjustment mechanism, supports omnidirectional maneuverability and attitude stability, and can dynamically adapt to different underwater working conditions.

[0048] The low-frequency magnetic wave excitation module of the present invention is a multi-level magnetic vibrator radiation system based on a honeycomb array. It adopts piezoelectric-permanent magnet coupling driving technology and realizes the directional superposition radiation of broadband swept-frequency electromagnetic waves through phase synchronization control; multiple low-frequency electromagnetic transmitting units 6 form an array unit, adopt a multi-layer composite structure design, use the inverse piezoelectric effect to drive the high-frequency vibration of the permanent magnet, and combine the magnetic circuit focusing optimization to enhance the field strength distribution in the target area, effectively stimulating the magnetic permeability dispersion characteristics of underwater objects.

[0049] The low-frequency magnetic wave excitation module of the present invention adopts a multi-level piezoelectric-permanent magnet composite vibrator array design, and realizes the directional focusing and energy superposition of low-frequency swept electromagnetic waves through honeycomb topology arrangement and phase synchronization control technology; this design effectively stimulates the intrinsic magnetic permeability dispersion response of underwater objects, and combined with the frequency domain characteristic decoupling algorithm, significantly improves the ability to distinguish ferromagnets, non-ferromagnets and non-metallic media; in complex seabed terrain, it can dynamically adjust the excitation parameters according to real-time magnetic field feedback, and enhance the effective field strength coverage of the target area through an adaptive sweep strategy.

[0050] The dispersion signal capture module of the present invention adopts a mesh magnetic sensor array formed by multiple fluxgate sensors 16 and a dynamic signal conditioning system, and realizes high-sensitivity monitoring of three-dimensional magnetic field gradients through a distributed sensing topology layout; the array bracket adopts a rigid anti-interference structure design to ensure the stable pointing of the sensor and signal acquisition accuracy in complex water flow environments; the signal processing system integrates multi-level noise reduction and dynamic compensation algorithms, and extracts target dispersion characteristics from strong background noise through time-frequency domain joint analysis technology, supporting real-time capture and enhanced analysis of weak abnormal signals.

[0051] The dispersion signal capture module of the present invention also introduces dynamic environmental noise suppression technology: the mesh magnetic sensor array adopts a rigid anti-interference bracket, and through the collaboration of time domain correlation and frequency domain phase-locked amplification technology, it effectively separates the target dispersion signal from interference components such as geomagnetism and eddy currents; the intelligent analysis module extracts the nonlinear frequency-varying characteristics of the real / imaginary part of the magnetic permeability based on asymmetric wavelet packet decomposition, and combines the convolutional neural network optimized by transfer learning to realize cross-scene migration identification of the material's electromagnetic fingerprint; this technology can maintain high recognition accuracy in turbid waters or strong ocean current environments.

[0052] The present invention breaks through the deformation error limitations of traditional flexible sensing structures through the use of rigid anti-interference brackets and honeycomb array magnetic vibrator collaborative radiation technology. Combined with the time-domain phase-locked integration and transient response capture algorithm, it significantly suppresses water flow disturbances and geomagnetic drift interference, thereby improving the signal-to-noise ratio of weak magnetic permeability dispersion signals in complex media by two orders of magnitude, thereby achieving stable detection of metal / non-metal targets.

[0053] The intelligent analysis module of the present invention constructs a dynamic identification system for electromagnetic fingerprints of materials based on a deep learning framework. It extracts the nonlinear pattern of the permeability dispersion response through asymmetric wavelet packet decomposition and multi-scale feature fusion technology. The algorithm embeds a transfer learning mechanism and combines convolutional neural networks to perform high-dimensional mapping and classification of dispersion features to achieve multi-dimensional intelligent identification of metal / non-metallic materials. The system supports online learning and model iteration, and improves the robustness of target classification in complex scenarios through dynamic database updates.

[0054] The present invention is based on an intelligent analysis framework of time-frequency domain joint feature fusion, adopts dynamic time warping (DTW) to align multi-track detection data, combines improved short-time Fourier transform with cross-scale feature mapping of convolutional neural network, excites the intrinsic characteristics of the material by covering the full frequency band of 1-100Hz, and combines the asymmetric wavelet packet decomposition algorithm to extract the real / imaginary frequency variation characteristics, so that the material classification dimension is expanded to three dimensions.

[0055] The present invention adopts a modular and scalable architecture: the maneuvering control module supports rapid adaptation of different propulsion units to adapt to shallow-water maneuvering operation modes; the low-frequency magnetic wave excitation module adopts a standardized vibrator unit design, which can flexibly expand the array size according to the detection depth requirements; the intelligent analysis module reserves a multi-source data fusion interface to support the collaborative processing of data from auxiliary sensors such as acoustics and optics; this modular design can not only meet the needs of conventional underwater target exploration, but can also be applied to specialized scenarios such as submarine pipeline corrosion detection and shipwreck artifact material identification through functional reorganization, providing an innovative solution for the field of underwater detection that combines high precision and strong adaptability.

[0056] The modular scalable architecture and the integrated rigid-electric-control anti-interference design of the present invention support the on-demand expansion of the magnetic excitation array scale and flexible configuration of the sensor topology, adapting to the shallow-water maneuverable scanning detection mode; the dynamic matching technology of the rigid frame and the multi-directional thruster ensures that the sub-meter trajectory control accuracy is maintained under the fourth-level sea condition, adapting to the operation requirements of complex ocean current environments.

[0057] The present invention has environmental adaptability enhanced by multi-source data fusion and transfer learning. Through the collaborative modeling of the magnetic permeability dispersion feature library and acoustic and optical auxiliary data, it constructs a dynamic identification map of underwater targets, reduces the target false alarm rate in turbid waters or strong biological interference scenarios, and provides innovative equipment for underwater archaeology and resource exploration.

[0058] In the description of the present invention, it should be understood that the terms "longitudinal", "transverse", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" and the like to indicate orientations or positional relationships are based on the orientations or positional relationships shown in the accompanying drawings and are only for the convenience of describing the present invention, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0059] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by persons skilled in the art should fall within the scope of protection defined by the claims of the present invention.

Claims

1. An underwater object recognition device based on the low-frequency permeability dispersion characteristics, characterized by: The system comprises a maneuvering control module for moving underwater, a low-frequency magnetic guide wave excitation module for identifying different underwater media, a dispersion signal capture module for separating interference components, and an intelligent analysis module for improving underwater identification accuracy. The intelligent analysis module is electrically connected to the maneuvering control module, the low-frequency magnetic guide wave excitation module, and the dispersion signal capture module. The maneuvering control module includes an air cushion (14), and a propulsion system for mobilizing the air cushion (14) is provided at the bottom of the air cushion (14); The low-frequency magnetic guided wave excitation module comprises a carbon fiber honeycomb array perforated plate (7) arranged above the air cushion (14), and a plurality of low-frequency electromagnetic emission units (6) are mounted on the carbon fiber honeycomb array perforated plate (7); The dispersion signal capture module comprises an aluminum alloy partition plate (5) arranged above the carbon fiber honeycomb array perforated plate (7), and sensor arrays are respectively arranged on both sides of the aluminum alloy partition plate (5), and the sensor array comprises a mesh support plate (2), and the mesh support plate (2) is rotatably connected to a sensor array support plate (17) via a programmable rotating drum (18), and a plurality of fluxgate sensors (16) are respectively installed on the mesh support plate (2) and the sensor array support plate (17); The intelligent analysis module comprises a central control module (21) fixedly connected to the top surface of the aluminum alloy partition (5), and a communication antenna (1) is installed on the top surface of the central control module (21).

2. The underwater object identification device based on low-frequency magnetic permeability dispersion characteristics according to claim 1, characterized in that: The propulsion system comprises an underwater propeller motion control module (8) fixedly connected to the top surface of the air cushion (14), the underwater propeller motion control module (8) being electrically connected to a plurality of servo motors (9), the servo motors (9) being fixedly connected to the bottom of the air cushion (14), and the output shafts of the servo motors (9) being transmission-connected to the titanium alloy propellers (12).

3. The underwater object identification device based on low-frequency magnetic permeability dispersion characteristics according to claim 2, characterized in that: A backup energy supply unit (11) is provided on one side of the underwater propeller motion control module (8), and the backup energy supply unit (11) is fixedly connected to the top surface of the air cushion (14).

4. The underwater object identification device based on low-frequency magnetic permeability dispersion characteristics according to claim 3, characterized in that: The carbon fiber honeycomb array perforated plate (7) is fixedly connected to the top surface of the air cushion (14) via a plurality of carbon fiber pillars (20); the underwater propeller motion control module (8) and the backup energy supply unit (11) are arranged between the carbon fiber honeycomb array perforated plate (7) and the air cushion (14).

5. The underwater object identification device based on low-frequency magnetic permeability dispersion characteristics according to claim 1, characterized in that: A square permanent magnet (61), piezoelectric ceramics (62) and an inverter amplifier circuit (63) are installed in the low-frequency electromagnetic emission unit (6).

6. The underwater object identification device based on low-frequency magnetic permeability dispersion characteristics according to claim 4, characterized in that: One end of the carbon fiber support (20) away from the air cushion (14) passes through the carbon fiber honeycomb array perforated plate (7) and is fixedly connected to the aluminum alloy partition (5).

7. The underwater object identification device based on low-frequency magnetic permeability dispersion characteristics according to claim 1, characterized in that: Rectangular cable channels (19) are respectively installed on both sides of the mesh support plate (2) and the sensor array support plate (17).

8. The underwater object identification device based on low-frequency magnetic permeability dispersion characteristics according to claim 1, characterized in that: Commutators (3) are rotatably connected to both sides of one end of the mesh support plate (2) away from the sensor array support plate (17); the commutator (3) is connected to a brushless DC motor (4) via a ceramic rotating shaft (10); and the commutator (3), the brushless DC motor (4) and the aluminum alloy partition (5) are fixedly connected.

9. The underwater object identification device based on low-frequency magnetic permeability dispersion characteristics according to claim 1, characterized in that: An acoustic distance sensor (15) is installed on one end of the sensor array support plate (17) away from the mesh support plate (2).

10. The underwater object identification device based on low-frequency magnetic permeability dispersion characteristics according to claim 1, characterized in that: The central control module (21) is provided with a central controller (211), an intelligent power supply station (212), a signal processing module (213), a GPS (214) and an inertial measurement unit (215).