Self-driven wave synchronous monitoring device and method integrating swinging-rolling movement

Abstract

The invention discloses a self-driven wave synchronous monitoring device and method integrating swinging-rolling motion, and the device comprises a pendulum bob main body with a hollow structure, and the interior of the hollow structure of the pendulum bob main body is fixedly provided with a roller-shaped friction nanometer generator. The pendulum bob main body is connected with the disc-shaped friction nanometer generator through a transmission system, and the drum-shaped friction nanometer generator and the disc-shaped friction nanometer generator synchronously respond to wave changes in different directions under the driving of the pendulum bob main body. The pendulum bob structure responds to wave excitation in different directions, through a double-structure joint sensing and multi-modal data fusion algorithm, the high-sensitivity sensing capability for multi-dimensional wave information is achieved, meanwhile, battery-free energy acquisition and signal output are achieved through the triboelectric effect, the advantages of being high in efficiency, compact in structure, high in adaptability and the like are achieved, and the application range is wide. The monitoring requirements for long time efficiency, low power consumption and high reliability in complex environments such as mariculture and offshore early warning are met.

Description

Technical Field

[0001] The present invention relates to the technical field of ocean monitoring, and in particular to a self-driven wave synchronous monitoring device and method integrating rocking and rolling motion. Background Art

[0002] With the rapid development of the marine economy, marine environmental monitoring is playing an increasingly important role in a variety of fields, including marine aquaculture, shipping, meteorological warning, and nearshore disaster prevention. Wave height, wave frequency, and wave propagation direction, as core parameters that characterize the dynamic state of the ocean, are of great significance for determining the energy distribution on the sea surface, sea state levels, precursors to extreme weather, and the stress conditions of aquaculture facilities. In the field of marine aquaculture, drastic changes in waves are often closely related to catastrophic weather such as storms and typhoons. Real-time understanding of the changes in the three elements of waves can help to provide early risk warnings and prevention, reducing economic losses such as damage to aquaculture facilities and the death of organisms. In addition, these parameters also have direct guiding significance for the safety of water operations, marine engineering construction, and ecological and environmental protection.

[0003] However, traditional wave monitoring devices generally rely on batteries or external power supplies. In extreme environments such as deep sea and offshore, they face bottlenecks such as unstable power supply, high maintenance costs, and short equipment lifespans, which seriously restrict the long-term stable operation of the monitoring network. Deep-sea buoys or fixed monitoring stations often experience data interruptions due to battery exhaustion, while cable power supply systems present complex deployment and susceptibility to corrosion. Existing sensors mostly use a split design, which makes it difficult to simultaneously capture multi-dimensional wave parameters with high precision. This results in low data fusion efficiency and insufficient sensitivity to low-frequency waves, limiting their application in complex marine environments. Summary of the Invention

[0004] In order to solve the above technical problems, this application proposes the following technical solutions:

[0005] In the first aspect, an embodiment of the present application provides a self-driven wave synchronization monitoring device with integrated rocking-rolling motion, comprising: a pendulum body with a hollow structure, a drum-shaped friction nanogenerator fixedly arranged in the hollow structure of the pendulum body, the pendulum body being connected to the disc-shaped friction nanogenerator through a transmission system, and the drum-shaped friction nanogenerator and the disc-shaped friction nanogenerator synchronously respond to wave changes in different directions under the drive of the pendulum body.

[0006] In one possible implementation, the drum-shaped friction nanogenerator includes: a rotating shaft, the two ends of the rotating shaft are respectively connected to a first fixed terminal and a second fixed terminal, the first fixed terminal and the second fixed terminal are fixed at the two ends of the hollow structure of the pendulum body, a drum rotor is provided on the rotating shaft, the drum rotor is rigidly connected to the impeller, the drum rotor adopts a hollow fan blade structure, and the cross-section presents a number of fan blades distributed at equal intervals, the outer surface of the drum rotor is provided with a drum stator shell, the inner surface of the drum stator shell is provided with contactless sensing electrodes at intervals, the outer surface of the drum rotor is provided with a polymer film, and there is a small air gap between the polymer film and the sensing electrode.

[0007] In one possible implementation, the drum stator shell is a hollow drum, the first end of the drum stator shell is fixedly connected to the first end of the first sealing cover, the second end of the first sealing cover is fixedly connected to the second fixed terminal, and the second sealing cover is arranged on the rotating shaft between the drum rotor and the impeller.

[0008] In one possible implementation, the transmission system includes an input shaft, on which a first bevel gear and a second bevel gear are fixedly provided, the first bevel gear and the second bevel gear are meshed and connected with a third bevel gear provided on a connecting rod, the first end of the connecting rod is fixedly connected to the third bevel gear, the second end of the connecting rod is provided with a first spur gear, the first spur gear is meshed and connected with a second spur gear provided on the connecting shaft, and the second spur gear is connected to the disc-shaped friction nanogenerator.

[0009] In one possible implementation, the disc-shaped friction nanogenerator includes: a disc stator and a disc rotor with a grid structure, the disc rotor uses a fur-like triboelectric material and is attached to the disc rotor base, the disc rotor is connected to the second spur gear through the connecting shaft, a disc-shaped polymer contact layer is provided between the disc stator and the disc rotor, an induction electrode is provided on the surface of the disc stator, and the polymer contact layer is attached to the surface of the induction electrode.

[0010] In a possible implementation, a waterproof shell is further included. A fixing mechanism is provided on the waterproof shell. The material of the waterproof shell includes: glass fiber reinforced composite material, closed-cell foamed polypropylene and foamed polyurethane material.

[0011] In a second aspect, an embodiment of the present application provides a self-driven wave synchronous monitoring method integrating rocking and rolling motion, comprising:

[0012] Acquire a first channel electrical signal generated in the transverse direction by the disk-shaped triboelectric nanogenerator and a second channel electrical signal generated in the longitudinal direction by the roller-shaped triboelectric nanogenerator during wave propagation;

[0013] After full-bridge rectification of the first channel electrical signal and the second channel electrical signal, decoupling and feature extraction are performed on the channel electrical signals to obtain wave height, wave frequency and wave direction information;

[0014] The decoupled wave frequency, wave height, and wave direction information are used as training samples and input into the constructed physics-deep learning hybrid model for training.

[0015] After the training is completed, the wave parameters are predicted by the physical-deep learning model, and the predicted parameter values are displayed in real time through the host computer interface.

[0016] In one possible implementation, after full-bridge rectification of the first channel electrical signal and the second channel electrical signal, decoupling and feature extraction of the channel electrical signals are performed to obtain wave height, wave frequency, and wave direction information, including:

[0017] After denoising the rectified first channel electrical signal and the second channel electrical signal using an adaptive filtering algorithm, the rectified first channel electrical signal is subjected to a fast Fourier transform, and the main frequency is extracted as the wave frequency. The calculation formula is:

[0018] f=FFT(V(t))

[0019] Where V(t) is the rectified voltage signal, and f is the wave frequency;

[0020] Performing multi-resolution analysis on the rectified second channel signal using wavelet transform to extract features at different scales, dividing the rectified second channel electrical signal into different frequency bands through wavelet packet decomposition and analyzing the wave height using a short-time energy envelope demodulation algorithm;

[0021] Combining the dual-channel phase difference, the wave direction information is extracted, and the calculation formula for the relationship between wave height and wave direction is:

[0022]

[0023] Among them, W i (t) is the sub-signal of wavelet transform, E i (t) is the wave height feature extracted by energy envelope demodulation, and H(t) is the wave direction.

[0024] In one possible implementation, the decoupled wave frequency, wave height, and wave direction information are used as training samples and input into a constructed physics-deep learning hybrid model for training, including:

[0025] The wave frequency, wave height and wave direction information obtained by decoupling are combined with the motion response model constructed based on wave theory as physical priors;

[0026] A deep learning model combining convolutional neural networks and gated recurrent unit models is used for training. At the same time, the Transformer model is introduced to capture the timing relationship under different wave states, and the network training process is dynamically adjusted through an adaptive optimization algorithm.

[0027] In one possible implementation, the calculation formula for predicting wave parameters using a physics-deep learning model is:

[0028]

[0029] in, represents the predicted wave parameters, X t is the input signal feature, f transformer 、f CNN and f GRU are the output functions of different models for input signals, W i is the corresponding weight.

[0030] Compared with the prior art, the present invention has the following advantages:

[0031] This application innovatively integrates the sensing capabilities of drum-shaped and disk-shaped friction nanogenerators. Through the principle of friction nano-power generation, it can respond to three-dimensional wave disturbances, greatly reducing the frequency of offshore maintenance and having long-term deployment capabilities. Through structural design and signal processing strategies, the integrated voltage characteristics for wave height identification, the spectrum main frequency characteristics for wave frequency identification, and the voltage phase difference and ratio for wave direction identification are extracted respectively, realizing independent modeling and identification of the three parameters of wave height, frequency, and direction, avoiding the traditional coupling interference problem. The modular and lightweight design allows it to be flexibly fixed in various scenarios such as surface platforms and aquaculture facilities. The solid-solid contact mode and optimized electrode structure design enable stable output in high-humidity, high-salt, and highly corrosive marine environments, maintaining passive self-driven characteristics, and completing intelligent monitoring without the need for additional power supply. It can be widely used in marine ranch management, offshore wind farm operation and maintenance, waterway safety monitoring, disaster prevention and early warning systems, and other fields, with important economic and social value. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a schematic diagram of the overall structure of a self-driven wave synchronization monitoring device with integrated rocking and rolling motion provided by an embodiment of the present application;

[0033] Figure 2A front view of a self-propelled wave synchronization monitoring device with integrated rocking and rolling motion provided by an embodiment of the present application;

[0034] Figure 3 A schematic diagram of the structure of a roller-shaped triboelectric nanogenerator provided in an embodiment of the present application;

[0035] Figure 4 A front view of a roller-shaped triboelectric nanogenerator provided in an embodiment of the present application;

[0036] Figure 5 A schematic structural diagram of a drum stator housing provided in an embodiment of the present application;

[0037] Figure 6 A diagram showing the working principle of the roller-shaped triboelectric nanogenerator provided in an embodiment of the present application;

[0038] Figure 7 A schematic diagram of the structure of the transmission system provided in an embodiment of the present application;

[0039] Figure 8 A schematic diagram of the structure of a disc-shaped triboelectric nanogenerator provided in an embodiment of the present application;

[0040] Figure 9 A schematic diagram of a disc rotor in a disc-shaped triboelectric nanogenerator provided in an embodiment of the present application;

[0041] Figure 10 Schematic diagram of a disc stator in a disc-shaped triboelectric nanogenerator provided in an embodiment of the present application;

[0042] Figure 11 A diagram showing the working principle of a disc-shaped triboelectric nanogenerator provided in an embodiment of the present application;

[0043] Figure 12 A flow chart of a self-driven wave synchronous monitoring method with integrated rocking and rolling motion provided in an embodiment of the present application.

[0044] in, Figure 1-9 The symbols in the figure are: 1-pendulum body, 2-drum-shaped friction nanogenerator, 3-disk-shaped friction nanogenerator, 4-transmission system, 5-waterproof housing, 6-first fixed terminal, 7-second fixed terminal, 8-rotating shaft, 9-drum rotor, 10-impeller, 11-drum stator housing, 12-first sealing cover, 13-second sealing cover, 14-input shaft, 15-first bevel gear, 16-second bevel gear, 17-connecting rod, 18-third bevel gear, 19-first spur gear, 20-connecting shaft, 21-second spur gear, 22-disc stator, 23-disc rotor, 24-disc rotor base, 25-disc-shaped polymer contact layer, 26-sensing electrode. DETAILED DESCRIPTION

[0045] The present solution will be described below with reference to the accompanying drawings and specific implementation methods.

[0046] Figure 1 This is a schematic diagram of the overall structure of a self-driven wave synchronous monitoring device with integrated rocking and rolling motion provided by an embodiment of the present application. The entire device can be fixedly deployed on the coast, floating platform or offshore structure to sense multiple parameters such as wave height, wave frequency and wave direction. Figure 1 and Figure 2 In this embodiment, a self-driven wave synchronization monitoring device with integrated rocking and rolling motion comprises a pendulum body 1 with a hollow structure. A drum-shaped triboelectric nanogenerator 2 is fixedly mounted within the hollow structure of the pendulum body 1. The pendulum body 1 is connected to a disc-shaped triboelectric nanogenerator 3 via a transmission system 4. Driven by the pendulum body 1, the drum-shaped triboelectric nanogenerator 2 and the disc-shaped triboelectric nanogenerator 3 synchronously respond to wave variations in different directions. In this embodiment, the pendulum body 1 includes a pendulum bracket and a pendulum. The pendulum has overall dimensions of 950 mm × 500 mm. The hollow structure within the pendulum is a rectangular parallelepiped with dimensions of 300 mm × 150 mm. The pendulum body 1 is constructed from materials including glass fiber reinforced plastic (GFRP), closed-cell expanded polypropylene (EPP), and expanded polyurethane (PU). In this embodiment, the closed-cell expanded polypropylene (EPP) is selected for its buoyancy, lightness, corrosion resistance, and structural strength. The self-propelled wave-synchronized monitoring device with integrated rocking and rolling motion in this application also includes a waterproof housing 5, which is equipped with a fixing mechanism that allows it to be fixedly deployed on the coast, floating platform, or offshore structure. The waterproof housing 5 is made of materials including glass fiber reinforced plastic (GFRP), closed-cell expanded polypropylene (EPP), and expanded polyurethane (PU). In this embodiment, the waterproof housing 5 is made of expanded polyurethane.

[0047] See also Figure 3 and Figure 4 In this embodiment, the drum-shaped triboelectric nanogenerator 2 includes a rotating shaft 8, the ends of which are respectively connected to a first fixed terminal 6 and a second fixed terminal 7, which are fixed to the two ends of the hollow structure of the pendulum body 1. A drum rotor 9 is provided on the rotating shaft 8, and the drum rotor 9 is rigidly connected to the impeller 10. The impeller 10 drives the drum rotor 9 to rotate through wave energy. The drum rotor 9 adopts a hollow fan blade structure, and the cross-section is a plurality of fan blades distributed at equal intervals. The number of fan blades can be increased according to specific needs. The outer surface of the drum rotor 9 is provided with a drum stator shell 11. The inner surface of the drum stator shell 11 is provided with non-contact sensing electrodes at intervals. The outer surface of the drum rotor 9 is provided with a polymer film, and there is a small air gap between the polymer film and the sensing electrodes. Figure 5In this embodiment, the drum stator housing 11 is a hollow drum. The first end of the drum stator housing is fixedly connected to the first end of the first sealing cover 12. The second end of the first sealing cover 12 is fixedly connected to the second fixed terminal 7. The second sealing cover 13 is disposed on the rotating shaft 8 between the drum rotor 9 and the impeller 10. Because the pendulum reciprocates in one direction, the drum-shaped triboelectric nanogenerator 2 is used for sensing perpendicular to the direction of the pendulum's motion. Since the pendulum reciprocates in one direction, the impeller 10 within the hollow structure of the pendulum can receive wave responses in other directions, converting the wave force received into rotational motion of the rotating shaft 8, driving the drum rotor 9 to rotate within the drum stator housing 11, thereby achieving relative sliding friction and charge transfer.

[0048] In this embodiment, each surface of the drum stator housing 11 is provided with at least two rectangular electrodes. Multiple sets of symmetrical electrodes can also be provided, and the number of electrodes can be adjusted as needed. In this embodiment, the overall dimensions of the drum-shaped triboelectric nanogenerator 2 are 300 mm × 120 mm. The outer diameter of the drum rotor 9 is 115 mm, and the outer diameter of the drum stator housing 11 is 120 mm, with a thickness of 2 mm. The drum rotor 9 is 180 mm long, with eight blades, corresponding to eight friction layers (polymer films). The impeller 10 is 50 mm long, with eight stator sensing electrodes. The fixed terminal dimensions are 45 × 45 × 20 mm, and the rotating shaft 8 has a diameter of 15 mm.

[0049] The polymer film materials include: PTFE, PE, PP, PET, PDMS and PVC. The materials for the drum rotor 9, the drum stator housing 11, and the fixed terminals include: acrylic, carbon fiber board, polycarbonate (PC), glass fiber reinforced plastic (GFRP), ABS plastic and polyetheretherketone (PEEK). The materials for the sensing electrodes include: copper, aluminum, gold and conductive semiconductor materials. In this embodiment, the polymer film material is polytetrafluoroethylene (PTFE) with a thickness of 80 μm. The materials for the drum rotor 9, the drum stator housing 11, and the fixed terminals are acrylic. The material for the sensing electrodes is commercial conductive copper foil with a thickness of 0.2 mm.

[0050] The working principle of the roller-shaped friction nanogenerator 2 is as follows Figure 6As shown, in this embodiment, the surface charge density of the polymer film is further enhanced by pre-rubbing it with the sensing electrodes. The charged polymer film then induces electrons from one sensing electrode to flow along the load circuit to the other sensing electrode through electrostatic induction, generating opposite net charges on the two sensing electrodes. Note that the induced charge density on the sensing electrode, regardless of charge sign, is lower than that on the polymer film in the presence of an air gap. Upon application of external excitation, the drum rotor naturally swings left or right. When the polymer film layer swings rightward, electrons are driven through the external circuit from the right electrode to the left electrode, generating an induced current from the left electrode to the right electrode. For a symmetrical structure, the current reaches a peak. As the swinging polymer film passes across the sensing electrodes, the opposing charge accumulation trends on each sensing electrode cause electrons to flow backward, generating opposite current pulses. Swinging across the two sensing electrodes completes a complete electron flow cycle. When atmospheric drag, bearing drag, and electrostatic forces are neglected during the swinging motion, the output current is a quasi-sinusoidal signal. Due to the symmetrical and centrosymmetrical positions of the polymer film and the sensing electrodes, the waveform of the electrical signal is the same whether the movement is clockwise or counterclockwise.

[0051] See also Figure 7 In this embodiment, the transmission system 4 includes an input shaft 14, on which are fixedly mounted a first bevel gear 15 and a second bevel gear 16. These first and second bevel gears 15, 16 mesh with a third bevel gear 18 mounted on a connecting rod 17. The first end of the connecting rod 17 is fixedly connected to the third bevel gear 18. A first spur gear 19 is mounted on the second end of the connecting rod 17. This first spur gear 19 meshes with a second spur gear 21 mounted on a connecting shaft 20. The second spur gear 21 is connected to the disc-shaped triboelectric nanogenerator 3. When the pendulum rotates with the waves, the gears rotate. The disc rotor 23 rotates with the transmission gears, causing friction between the copper layer of the disc-shaped triboelectric nanogenerator 3 and the disc-shaped polymer contact layer 25 on the disc stator 22. Electrostatic induction generates a voltage at the induction electrodes on the back of the contact layer. The disc-shaped polymer contact layer 25 effectively suppresses direct air breakdown between the metal electrodes, thereby generating a higher charge density.

[0052] See also Figure 8In this embodiment, the disc-shaped triboelectric nanogenerator 3 includes a disc stator 22 and a disc rotor 23 with a grid structure. The disc rotor 23 is made of a fur-like triboelectric material and is attached to a disc rotor base 24 to improve friction performance and flexible contact durability. The disc rotor 23 is connected to the second spur gear 21 via a connecting shaft 20. A disc-shaped polymer contact layer 25 is provided between the disc stator 22 and the disc rotor 23. A sensing electrode 26 is provided on the surface of the disc stator 22. The disc-shaped polymer contact layer 25 is attached to the surface of the sensing electrode 26 to prevent air breakdown and increase charge density. Figure 9 and Figure 10 In this embodiment, the number of disc rotors 23 and disc stators 22 is expandable, and there is no limit to the number. The number can be set as needed. The 3D printed shell is made of acrylic, and the seams are sealed with marine-grade epoxy resin. Fur-based triboelectric materials include: rabbit hair, wool, mink hair, and artificial fur. The materials of the disc-shaped polymer contact layer 25 include: PTFE, PE, PP, PET, PDMS, and PVC. The materials of the disc rotor 23 and disc stator 22 include: acrylic, carbon fiber plate, polycarbonate (PC), glass fiber reinforced plastic (GFRP), ABS plastic, and polyetheretherketone (PEEK). In this embodiment, the fur-based triboelectric material is rabbit hair with a length of 5mm. While maintaining high electrical performance, it creates soft contact, reduces friction resistance and torque, and enhances the durability of the device. The material of the disc-shaped polymer contact layer 25 is PET with a thickness of 2mm, and the material of the disc rotor 23 and disc stator 22 is carbon fiber with a thickness of 5mm.

[0053] The working principle of the disc-shaped triboelectric nanogenerator 3 is as follows Figure 11 As shown, the basic mechanism in this embodiment is the coupling of triboelectric charging and electrostatic induction. In the original state, after a period of rotation, the fur and polymer contact layer are respectively charged positively and negatively due to the difference in electronegativity. However, the charge density on the fur surface is twice that of the polymer contact layer. In addition, the left and right sensing electrodes below the polymer contact layer acquire equal and opposite charges. Due to the opposite charges generated by electrostatic induction, when the hair slides to the right, electrons flow from the left electrode to the right electrode to balance the potential difference. When the hair completely overlaps the right polymer contact layer, all the electrons are transferred to the right electrode. A potential difference is then generated between the two electrodes, and a reverse current flows through the external circuit until the final state is reached.

[0054] Both the disc-shaped triboelectric nanogenerator 3 and the roller-shaped triboelectric nanogenerator 2 utilize a solid-solid friction mechanism. They feature a simple structure, stable output, strong adaptability, and are suitable for long-term, high-efficiency energy harvesting in marine and lake environments. The friction between the metal electrodes of the disc-shaped triboelectric nanogenerator 3 and the roller-shaped triboelectric nanogenerator 2 generates a triboelectric signal representing angle information. The phase difference in the friction signal is used to distinguish wave direction. When the pendulum and impeller 10 are driven by wave energy, they rotate clockwise and counterclockwise, generating an AC signal with a phase difference between the two rotations.

[0055] In this embodiment, the drum-shaped triboelectric nanogenerator (TGN) 2 responds to transverse wave disturbances, with its output voltage signal correlated with transverse wave amplitude and direction. The disk-shaped TGN 3 responds to longitudinal wave excitation, with its output electrical signal reflecting wave height and longitudinal disturbance characteristics. Through an innovative structural integration of the drum-shaped and disk-shaped TGNs, their orthogonal arrangement enables the acquisition of wave signals from different directions, collaboratively sensing information such as wave height, frequency, and direction. The two systems work together to achieve self-energy collection and signal acquisition through solid-solid contact friction without the need for an external power source, improving the system's stability and environmental adaptability. By employing a signal feature decoupling algorithm and a hybrid physics-deep learning model, wave frequency monitoring is achieved through single-channel data acquisition. A pendulum drives the disk-shaped TGN 3 to detect wave direction and height. Data fusion and algorithms from the two TGNs extract wave height, direction, and amplitude from the original dual-channel electrical signals, enabling real-time, low-power, and high-precision monitoring of key wave dynamics. This device has a compact structure, lightweight and corrosion-resistant materials, and excellent underwater sealing performance. It is suitable for long-term deployment in complex marine environments and provides a high-performance, low-power, and easy-to-deploy new solution for in-situ sensing of multi-parameter waves.

[0056] Corresponding to the self-driven wave synchronous monitoring device with integrated rocking and rolling motion provided in the above embodiment, the present application also provides an embodiment of a self-driven wave synchronous monitoring method with integrated rocking and rolling motion.

[0057] See also Figure 12 In this embodiment, a self-driven wave synchronous monitoring method integrating rocking and rolling motion includes:

[0058] S101, obtaining a first channel electrical signal generated by the disk-shaped triboelectric nanogenerator in the transverse direction and a second channel electrical signal generated by the roller-shaped triboelectric nanogenerator in the longitudinal direction during wave propagation.

[0059] In this embodiment, during wave propagation, the pendulum structure is driven to produce two motion modes: lateral swing and longitudinal tumbling of the drum. The pendulum's up-and-down swing drives the transmission system, which in turn drives the disc-shaped triboelectric nanogenerator's disc rotor, generating a periodic first-channel electrical signal. The longitudinal tumbling of the drum-shaped triboelectric nanogenerator under wave disturbances stimulates rotational contact between the friction pairs within the drum, generating a second-channel electrical signal.

[0060] S102 , performing full-bridge rectification on the first channel electrical signal and the second channel electrical signal, decoupling and extracting features of the channel electrical signals to obtain wave height, wave frequency and wave direction information.

[0061] In this embodiment, the first and second channel electrical signals are first converted into pulsed DC signals via two full-bridge rectifier circuits. The rectified signals are then fed into a data processing system via a voltage acquisition module, ensuring the complete restoration of the frequency characteristics of the wave excitation. The rectified dual-channel triboelectric nanogenerator signals undergo decoupling and feature extraction. To accurately extract wave frequency, height, and direction, a signal processing method combining an adaptive filtering algorithm, a fast Fourier transform, and a wavelet transform is employed.

[0062] First, an adaptive filtering algorithm is used to denoise the rectified first-channel electrical signal and the second-channel electrical signal to ensure signal stability and reliability. In complex ocean environments, wave signals are often interfered with by noise. Adaptive filtering can adjust the filter parameters in real time and automatically adapt to environmental changes. Since the rectified first-channel electrical signal has obvious periodicity, a fast Fourier transform is performed on the rectified first-channel electrical signal to extract the main frequency as the wave frequency. The calculation formula is:

[0063] f=FFT(V(t))

[0064] Where V(t) is the rectified voltage signal, and f is the wave frequency;

[0065] In order to extract the wave height and direction characteristics, the wavelet transform is used to perform multi-resolution analysis on the rectified second channel signal to extract features at different scales. The rectified second channel electrical signal is divided into different frequency bands through wavelet packet decomposition and the wave height is analyzed using the short-time energy envelope demodulation algorithm. The wave direction information is extracted by combining the phase difference of the two friction nanogenerator signals. The calculation formula for the relationship between wave height and wave direction is:

[0066]

[0067] Among them, W i (t) is the sub-signal of wavelet transform, E i (t) is the wave height feature extracted by energy envelope demodulation, and H(t) is the wave direction.

[0068] S103: The decoupled wave frequency, wave height, and wave direction information are used as training samples and input into the constructed physics-deep learning hybrid model for training.

[0069] In this embodiment, the wave frequency, wave height and wave direction information obtained by decoupling are used as training samples and input into the physical-deep learning model. In order to improve the accuracy and adaptability of the model, these features are first combined with the motion response model constructed based on wave theory as a physical prior to ensure that the model has a stronger physical basis. A deep learning framework combining a convolutional neural network (CNN) and a gated recurrent unit (GRU) model is used for training. CNN is used to extract spatial features in the time-frequency domain from wave signals, and GRU is used to process the temporal dependencies of wave data. The output of the model is the predicted value of multiple parameters (wave frequency, wave height, wave direction) of the wave. Based on the CNN and GRU models, the Transformer model is introduced. By inputting the time series signals of wave frequency, wave height and wave direction into the Transformer model, the temporal relationship under different wave states can be more effectively captured, thereby improving the prediction accuracy. In addition, an adaptive optimization algorithm is introduced to dynamically adjust the training process of the network, so that the model can maintain good adaptability under different ocean conditions.

[0070] In order to build a highly reliable physical-deep learning hybrid model, it is necessary to build an experimental platform in a controllable wave environment to collect and verify the signal responses under various wave conditions. The test site includes an artificial controllable wave tank, a data acquisition platform, and a self-driven wave synchronous monitoring device. In this embodiment, the specific experimental process of simulating wave height includes: building an integrated self-driven wave synchronous monitoring device in an artificial wave tank, setting multiple periodic waves with fixed wave heights such as 10cm, 15cm, 20cm, and 25cm, collecting the vertical direction voltage output signal V(t) at a sampling frequency of 500Hz and performing numerical integration for each wave cycle; extracting the voltage peak and integral value as the input of the wave height regression model and establishing a multivariate linear regression model, training the wave height prediction model, inputting the new signal after the training is completed, and outputting the corresponding wave height estimation value.

[0071] The specific experimental process of simulating wave frequency includes: building an integrated self-driven wave synchronization monitoring device in an artificial wave tank; setting wave conditions with different wave frequencies (0.5Hz~2.0Hz), collecting dual-channel voltage signals at a sampling frequency of 1000Hz, and performing short-time Fourier transform on the signal to obtain a power spectrum diagram, extracting the maximum amplitude frequency as the main frequency, and mapping it to the wave period T=1 / f; comparing the frequency changes over time and constructing a wave frequency monitoring curve.

[0072] The specific experimental process of simulating wave direction includes: building an integrated self-driven wave synchronization monitoring device in an artificial wave tank, designing wave direction angles of 0° to 180° for incident waves in different directions, and synchronously collecting the output signals of two friction nanogenerators with a sampling frequency of 1000 Hz; performing Hilbert transform, extracting the phase difference and constructing the direction vector, comparing the phase difference with the direction angle, constructing a direction angle decoupling model, training the angle mapping relationship, and outputting the real-time wave incident direction.

[0073] S104, after the training is completed, the wave parameters are predicted through the physics-deep learning model, and the predicted parameter values are displayed in real time through the host computer interface.

[0074] In this embodiment, the prediction of wave frequency, wave height, and wave direction parameters can be expressed as follows based on a fusion model of multi-layer perceptron (MLP) combined with time series learning:

[0075]

[0076] in, represents the predicted wave parameters, X t is the input signal feature, f transformer 、f CNN and f GRU are the output functions of different models for input signals, W i The system outputs monitoring results such as wave frequency (Hz), wave height (m), and propagation direction. Through the host computer interface, real-time plotting of wave time series data, direction indication, and early warning judgment are performed, providing real-time environmental perception support for water platforms and aircraft.

[0077] In the embodiments of this application, "multiple" refers to two or more. "And / or" describes the relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can mean that A exists alone, A and B exist simultaneously, or B exists alone. A and B can be singular or plural. The character " / " generally indicates that the associated objects are in an "or" relationship.

[0078] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0079] The above description is merely a specific embodiment of the present application. Any person skilled in the art may easily conceive of variations or substitutions within the technical scope disclosed in this application, and such variations or substitutions shall be within the scope of protection of this application. The scope of protection of this application shall be subject to the scope of protection of the claims.

Claims

1. A self-driven wave synchronization monitoring device with integrated rocking and rolling motion, characterized in that: include: A pendulum body with a hollow structure is provided, in which a drum-shaped friction nanogenerator is fixedly provided. The pendulum body is connected to the disc-shaped friction nanogenerator through a transmission system. Driven by the pendulum body, the drum-shaped friction nanogenerator and the disc-shaped friction nanogenerator synchronously respond to wave changes in different directions.

2. The self-propelled wave synchronous monitoring device with integrated rocking and rolling motion according to claim 1, characterized in that: The drum-shaped friction nanogenerator includes: a rotating shaft, the two ends of which are respectively connected to a first fixed terminal and a second fixed terminal, the first fixed terminal and the second fixed terminal are fixed at the two ends of the hollow structure of the pendulum body, a drum rotor is provided on the rotating shaft, the drum rotor is rigidly connected to the impeller, the drum rotor adopts a hollow fan blade structure, and the cross-section presents a plurality of fan blades distributed at equal intervals, the outer surface of the drum rotor is provided with a drum stator shell, the inner surface of the drum stator shell is provided with contactless sensing electrodes at intervals, the outer surface of the drum rotor is provided with a polymer film, and there is a small air gap between the polymer film and the sensing electrode.

3. The self-propelled wave synchronous monitoring device with integrated rocking and rolling motion according to claim 2, characterized in that: The drum stator shell is a hollow drum, the first end of the drum stator shell is fixedly connected to the first end of the first sealing cover, the second end of the first sealing cover is fixedly connected to the second fixed terminal, and the second sealing cover is arranged on the rotating shaft between the drum rotor and the impeller.

4. The self-propelled wave synchronization monitoring device with integrated rocking and rolling motion according to claim 1, characterized in that: The transmission system includes an input shaft, on which a first bevel gear and a second bevel gear are fixedly provided. The first bevel gear and the second bevel gear are meshed and connected with a third bevel gear provided on a connecting rod. The first end of the connecting rod is fixedly connected to the third bevel gear. The second end of the connecting rod is provided with a first spur gear, which is meshed and connected with a second spur gear provided on the connecting shaft. The second spur gear is connected to the disc-shaped friction nanogenerator.

5. The self-propelled wave synchronization monitoring device with integrated rocking and rolling motion according to claim 4, characterized in that: The disc-shaped friction nanogenerator includes: a disc stator and a disc rotor with a grid structure. The disc rotor uses a fur-like triboelectric material and is attached to the disc rotor base. The disc rotor is connected to the second spur gear through the connecting shaft. A disc-shaped polymer contact layer is provided between the disc stator and the disc rotor. An induction electrode is provided on the surface of the disc stator, and the polymer contact layer is attached to the surface of the induction electrode.

6. The self-propelled wave synchronization monitoring device with integrated rocking and rolling motion according to claim 1, characterized in that: It also includes a waterproof shell, which is provided with a fixing mechanism. The material of the waterproof shell includes: glass fiber reinforced composite material, closed-cell foamed polypropylene and foamed polyurethane material.

7. A self-propelled wave synchronous monitoring method with integrated rocking and rolling motion, using the self-propelled wave synchronous monitoring device with integrated rocking and rolling motion according to any one of claims 1 to 6, characterized in that: include: Acquire a first channel electrical signal generated in the transverse direction by the disk-shaped triboelectric nanogenerator and a second channel electrical signal generated in the longitudinal direction by the roller-shaped triboelectric nanogenerator during wave propagation; After full-bridge rectification of the first channel electrical signal and the second channel electrical signal, decoupling and feature extraction are performed on the channel electrical signals to obtain wave height, wave frequency and wave direction information; The decoupled wave frequency, wave height, and wave direction information are used as training samples and input into the constructed physics-deep learning hybrid model for training. After the training is completed, the wave parameters are predicted by the physical-deep learning model, and the predicted parameter values are displayed in real time through the host computer interface.

8. The self-driven wave synchronous monitoring method with integrated rocking and rolling motion according to claim 7 is characterized in that: After full-bridge rectification of the first channel electrical signal and the second channel electrical signal, decoupling and feature extraction of the channel electrical signals are performed to obtain wave height, wave frequency, and wave direction information, including: After denoising the rectified first channel electrical signal and the second channel electrical signal using an adaptive filtering algorithm, the rectified first channel electrical signal is subjected to a fast Fourier transform, and the main frequency is extracted as the wave frequency. The calculation formula is: f=FFT(V(t)) Where V(t) is the rectified voltage signal, and f is the wave frequency; Performing multi-resolution analysis on the rectified second channel signal using wavelet transform to extract features at different scales, dividing the rectified second channel electrical signal into different frequency bands through wavelet packet decomposition and analyzing the wave height using a short-time energy envelope demodulation algorithm; Combining the dual-channel phase difference, the wave direction information is extracted, and the calculation formula for the relationship between wave height and wave direction is: Among them, W i (t) is the sub-signal of wavelet transform, E i (t) is the wave height feature extracted by energy envelope demodulation, and H(t) is the wave direction.

9. The self-driven wave synchronous monitoring method with integrated rocking and rolling motion according to claim 7, characterized in that: The decoupled wave frequency, wave height, and wave direction information are used as training samples and input into the constructed physics-deep learning hybrid model for training, including: The wave frequency, wave height and wave direction information obtained by decoupling are combined with the motion response model constructed based on wave theory as physical priors; A deep learning model combining convolutional neural networks and gated recurrent unit models is used for training. At the same time, the Transformer model is introduced to capture the timing relationship under different wave states, and the network training process is dynamically adjusted through an adaptive optimization algorithm.

10. The method for synchronous monitoring of self-driven waves of composite motion according to claim 7, characterized in that: The calculation formula for predicting wave parameters using the physics-deep learning model is: in, represents the predicted wave parameters, X t is the input signal feature, f transformer 、f CNN and f GRU are the output functions of different models for input signals, W i is the corresponding weight.

Images