Power equipment vibration defect identification method and system

Abstract

The invention relates to the technical field of power equipment operation and maintenance, and discloses a power equipment vibration defect identification method and system. Dynamically adjusting an acoustic vortex field orbital angular momentum mode corresponding to the target power equipment through a preset programmable metasurface array, and obtaining a resonance response signal of the equipment; carrying out compressed state laser irradiation and balanced zero beat detection on the resonance response signal to obtain a quantum correlation factor time sequence; reconstructing a resonance response signal phase space based on the quantum correlation factor time sequence, and calculating to obtain a correlation dimension and a maximum Lyapunov index; and performing defect identification based on the correlation dimension, the maximum Lyapunov index and a preset threshold to obtain defect identification data of the power equipment. According to the method, dynamic sound field excitation and quantum detection are combined, the vibration defects of the power equipment are recognized, and the technical problem that the recognition precision is low due to the fact that a sensor is high in measuring point dependence, low in environmental noise anti-interference capacity and insufficient in fault feature extraction in a traditional method is solved.

Description

Technical Field

[0001] This invention relates to the field of power equipment operation and maintenance technology, and in particular to a method and system for identifying vibration defects in power equipment. Background Technology

[0002] In recent years, in the field of power equipment operation and maintenance, accurate identification of vibration defects has been crucial to ensuring stable equipment operation. Traditional methods mainly rely on mechanical vibration-based detection methods, which use accelerometers to collect equipment vibration signals and perform time-domain or frequency-domain analysis to determine the existence and type of defects; or they employ acoustic-based methods to indirectly identify vibration states by analyzing equipment operating noise.

[0003] However, existing methods still have significant limitations in terms of accurate identification. On the one hand, the sensor installation position in mechanical vibration methods is sensitive, easily leading to signal distortion and misjudgment. Meanwhile, the vibration of complex power equipment exhibits non-stationary and nonlinear characteristics, making it difficult for traditional time-frequency analysis methods to effectively extract its essential features, and early, subtle defects are easily masked by noise. On the other hand, acoustic methods are easily affected by background noise, making feature extraction difficult and resulting in low identification accuracy. Furthermore, existing methods generally lack in-depth exploration of the dynamic characteristics of vibration signals, making it difficult to comprehensively analyze vibration signals from chaotic and nonlinear perspectives. Therefore, they cannot achieve more accurate and detailed identification of vibration defects in power equipment, failing to meet the high requirements of modern power equipment for operational stability and reliability. Summary of the Invention

[0004] This invention provides a method and system for identifying vibration defects in power equipment, which solves the technical problems of low identification accuracy caused by the strong dependence of sensor measurement points, weak anti-interference ability of environmental noise, and insufficient extraction of fault features in existing power equipment vibration defect identification methods.

[0005] The first aspect of this invention provides a method for identifying vibration defects in power equipment, comprising:

[0006] The resonance response signal of the target power equipment is obtained by dynamically adjusting the orbital angular momentum mode of the acoustic vortex field corresponding to the target power equipment according to a preset programmable metasurface array.

[0007] The resonant response signal was subjected to compressed state laser irradiation and equilibrium zero-beat detection to obtain the time series of quantum correlation factors;

[0008] The phase space of the resonance response signal is reconstructed based on the time series of the quantum correlation factor, and the chaotic characteristic quantities are calculated to obtain the correlation dimension and the maximum Lyapunov exponent.

[0009] Based on the correlation dimension, the maximum Lyapunov exponent, and the preset threshold, defect identification is performed to obtain defect identification data corresponding to the target power equipment.

[0010] Optionally, the step of dynamically adjusting the orbital angular momentum mode of the acoustic vortex field corresponding to the target power device according to a preset programmable metasurface array to obtain the resonant response signal of the target power device includes:

[0011] The basic programmable metasurface array architecture is obtained by periodically arranging and optimizing the structural parameters of each metasurface unit in the pre-programmable metasurface array and designing impedance matching.

[0012] Based on the aforementioned basic programmable metasurface array architecture, a dynamic phase modulation algorithm is loaded, and the reflection phase of each metasurface unit is controlled by time-domain coding to obtain the target acoustic vortex field generator.

[0013] The resonant frequency and quality factor of the metasurface unit are optimized by the acoustic vortex field orbital angular momentum mode of the target acoustic vortex field generator, respectively, to generate a dynamic control array.

[0014] The topology of the metasurface units of the preset programmable metasurface array is adjusted using the real-time feedback data corresponding to the dynamic control array to obtain the target programmable metasurface array.

[0015] The acoustic vortex field generated by the target programmable metasurface array is applied to the target power device, and the vibration signal generated by the target power device is used as the resonance response signal of the target power device.

[0016] Optionally, the step of periodically arranging and optimizing the structural parameters of each metasurface unit in the preset programmable metasurface array and designing impedance matching to obtain the basic programmable metasurface array architecture includes:

[0017] An electromagnetic simulation model is constructed using the structural parameters of the metasurface units in a pre-programmable metasurface array.

[0018] The electromagnetic response characteristics of each metasurface unit within a preset frequency range are calculated using the electromagnetic simulation model to obtain initial electromagnetic characteristic data.

[0019] Based on the initial electromagnetic characteristic data, with the optimization objective of balancing beam directivity and bandwidth, the spacing and arrangement period of the metasurface units are calculated using the particle swarm optimization algorithm to construct a periodic arrangement scheme;

[0020] Based on a periodic arrangement scheme, an adjustable load element is introduced at the interface of the metasurface unit, and the load impedance of the adjustable load element is adjusted by the Smith chart method to obtain an impedance matching structure.

[0021] The impedance matching structure is used to simulate the scattering parameters within a preset frequency band, and the scattering parameters are used to perform frequency domain integration calculations to generate an effective bandwidth.

[0022] When the effective bandwidth is greater than a preset bandwidth threshold, the impedance matching structure is used as the basic programmable metasurface array architecture.

[0023] Optionally, the step of loading a dynamic phase modulation algorithm based on the basic programmable metasurface array architecture and controlling the reflection phase of each metasurface unit through time-domain coding to obtain the target acoustic vortex field generator includes:

[0024] Based on the aforementioned basic programmable metasurface array, a time-domain-frequency-domain joint control model is constructed.

[0025] A time-domain encoded sequence generator is constructed using the time-domain-frequency domain joint control model and the reflection phase function corresponding to each metasurface unit.

[0026] The time-domain encoded sequence generator uses a hybrid coding strategy of binary phase shift keying and pulse width modulation to encode the signal and generate a time-domain control signal.

[0027] The time-domain control signal is loaded into the tunable element of the metasurface unit in real time through a field-programmable gate array, and the equivalent circuit parameters of each metasurface unit are dynamically adjusted within a preset control range to construct an initial acoustic vortex field generator.

[0028] The initial acoustic vortex field generator is modeled for far-field radiation, and the sound pressure distribution under different time-domain coding modes is calculated using the finite element method based on the generated far-field radiation model to obtain sound pressure distribution data.

[0029] Based on the sound pressure distribution data, the actual phase modulation error corresponding to the acoustic vortex field generator is obtained by fitting using the least squares method.

[0030] The actual phase modulation error is corrected through a closed-loop feedback mechanism to obtain the root mean square error of the phase.

[0031] When the root mean square error of the phase is less than a preset error threshold, the initial acoustic vortex field generator at the current moment is used as the target acoustic vortex field generator.

[0032] Optionally, the step of subjecting the resonant response signal to squeezed-state laser irradiation and equilibrium zero-beat detection to obtain the quantum correlation factor time series includes:

[0033] The resonant response signal is irradiated with a compressed laser to generate scattered light;

[0034] The scattered light is subjected to balanced zero-beat detection to generate a detection signal time series;

[0035] Quantum correlation analysis was performed on the time series of the detected signal to obtain the time series of quantum correlation factors.

[0036] Optionally, the step of reconstructing the phase space of the resonance response signal based on the time series of the quantum correlation factor, calculating the chaotic characteristic quantities, and obtaining the correlation dimension and the maximum Lyapunov exponent includes:

[0037] Phase space reconstruction is performed using the time series of the quantum correlation factor to obtain the reconstructed phase space trajectory;

[0038] The reconstructed phase space trajectory is substituted into the correlation integral function to calculate the correlation dimension, thus obtaining the correlation dimension.

[0039] Based on the divergence data of adjacent orbits in the reconstructed phase space trajectory, the Wolf direct method is used to calculate the Lyapunov exponent, and the maximum Lyapunov exponent is obtained.

[0040] Optionally, the step of using the quantum correlation factor time series to perform phase space reconstruction processing to obtain the reconstructed phase space trajectory includes:

[0041] The mutual information method is used to calculate the mutual information between the time series of the quantum correlation factor and the time series of the preset delayed quantum correlation factor;

[0042] The delay time when the mutual information first reaches its minimum value is used as the optimal parameter;

[0043] Based on the optimal parameters, the proportion of pseudo-neighbors of the state vector in the low-dimensional embedding space is determined by the pseudo-nearest neighbor method.

[0044] When the proportion of pseudo-neighbors is lower than a preset pseudo-neighbor proportion threshold, the embedding dimension corresponding to the pseudo-neighbor proportion is taken as the minimum embedding dimension.

[0045] Based on the minimum embedding dimension, the time series of the quantum correlation factor is converted into a multidimensional state vector;

[0046] The phase space trajectory is reconstructed based on the multidimensional state vector using time delay embedding technology, resulting in a reconstructed phase space trajectory.

[0047] Optionally, the step of identifying defects based on the correlation dimension, the maximum Lyapunov exponent, and a preset threshold to obtain defect identification data corresponding to the target power equipment includes:

[0048] When the correlation dimension is greater than a preset dimension threshold and the maximum Lyapunov exponent is greater than a preset exponent threshold, the defect identification data corresponding to the target power equipment is winding deformation;

[0049] When the correlation dimension bifurcation amount corresponding to the correlation dimension is greater than the preset bifurcation amount threshold and the mutation amount corresponding to the maximum Lyapunov exponent is greater than the preset mutation amount threshold, the defect identification data corresponding to the target power equipment is insulation damage.

[0050] A second aspect of the present invention provides a vibration defect identification system for power equipment, comprising:

[0051] An adaptive vortex field control module is used to dynamically adjust the orbital angular momentum mode of the acoustic vortex field corresponding to the target power equipment according to a preset programmable metasurface array, so as to obtain the resonance response signal of the target power equipment.

[0052] A quantum precision measurement module is used to irradiate the resonant response signal with compressed state laser and detect equilibrium zero beats to obtain a time series of quantum correlation factors.

[0053] The nonlinear dynamics feature extraction module is used to reconstruct the phase space of the resonance response signal based on the time series of the quantum correlation factor, and to calculate the chaotic feature quantity to obtain the correlation dimension and the maximum Lyapunov exponent;

[0054] The defect identification decision module is used to identify defects based on the correlation dimension, the maximum Lyapunov exponent, and a preset threshold, and obtain defect identification data corresponding to the target power equipment.

[0055] A third aspect of the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the power equipment vibration defect identification method as described in any of the preceding claims.

[0056] As can be seen from the above technical solutions, the present invention has the following advantages:

[0057] This invention first utilizes a pre-programmable metasurface array (PPA) designed with periodic arrangement optimization and impedance matching. Through a dynamic phase modulation algorithm, the acoustic vortex field is controlled in real time to effectively acquire a resonant response signal containing rich frequency components, overcoming the early defect omission problems caused by the strong dependence on measurement points and bandwidth limitations of traditional sensors. Next, compressed-state laser irradiation and balanced zero-beat detection techniques are employed to extract the quantum correlation factor time series. This technique has sensitivity exceeding the classical detection limit and significantly improves the ability to resist environmental noise interference. Based on this, the phase space of the resonant response signal is reconstructed based on the quantum correlation factor time series, and chaotic characteristic quantities are calculated to obtain the correlation dimension and the maximum Lyapunov exponent. These characteristic quantities can profoundly reveal the essential dynamic characteristics of the vibration system, solving the technical bottlenecks of traditional time-frequency analysis methods in comprehensively characterizing complex vibration modes and insufficiently extracting fault features. Finally, based on the intelligent comparison of the correlation dimension and the maximum Lyapunov exponent with preset thresholds, accurate identification of defects such as winding deformation and insulation damage is achieved, overcoming the technical dilemma of low identification accuracy caused by sensor limitations, noise interference, and insufficient feature extraction in traditional methods. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is a flowchart illustrating the steps of a method for identifying vibration defects in power equipment according to Embodiment 1 of the present invention.

[0060] Figure 2 This is a flowchart illustrating the steps of a method for identifying vibration defects in power equipment according to Embodiment 2 of the present invention.

[0061] Figure 3 This is a structural block diagram of a power equipment vibration defect identification system provided in Embodiment 3 of the present invention;

[0062] Figure 4 This is a structural block diagram of a computer device provided in Embodiment 4 of the present invention. Detailed Implementation

[0063] This invention provides a method and system for identifying vibration defects in power equipment, which addresses the technical problem of low identification accuracy caused by existing methods for identifying vibration defects in power equipment due to strong dependence on sensor measurement points, weak anti-interference ability against environmental noise, and insufficient extraction of fault features.

[0064] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0065] Please see Figure 1 , Figure 1 The flowchart illustrates the steps of a method for identifying vibration defects in power equipment according to Embodiment 1 of the present invention.

[0066] This invention provides a method for identifying vibration defects in power equipment, comprising:

[0067] Step 101: Dynamically adjust the orbital angular momentum mode of the acoustic vortex field corresponding to the target power equipment according to the preset programmable metasurface array to obtain the resonance response signal of the target power equipment.

[0068] Preset programmable metasurface arrays refer to programmable metasurface arrays that have undergone periodic arrangement optimization and impedance matching design. Acoustic vortex field orbital angular momentum mode refers to the orbital angular momentum (OAM) state carried by the vortex structure with a helical wavefront formed during sound wave propagation; its essence is the physical characteristic of sound wave energy rotating around the propagation axis.

[0069] In this embodiment of the invention, a basic programmable metasurface array architecture with wideband response characteristics is obtained by periodically arranging and impedance matching the metasurface units of a preset programmable metasurface array based on their structural parameters. A dynamic phase modulation algorithm is then applied to this basic programmable metasurface array architecture, and the reflection phase of each unit is controlled through time-domain coding to obtain a reconfigurable acoustic vortex field generator. Based on the orbital angular momentum mode of the acoustic vortex field of the reconfigurable acoustic vortex field generator, the resonant frequency and quality factor of the metasurface units of the programmable metasurface array are optimized to obtain a dynamically controlled array with mode orthogonality. Based on the real-time feedback data from the dynamically controlled array with mode orthogonality, the topology of the metasurface units of the programmable metasurface array is adjusted to obtain the improved programmable metasurface array, i.e., the target programmable metasurface array. The acoustic vortex field generated by the target programmable metasurface array is applied to a target power device, and the vibration signal generated by the target power device is used as the resonant response signal of the target power device.

[0070] Through a series of operations including periodic arrangement optimization, impedance matching design, dynamic phase modulation, and resonant frequency and quality factor optimization using a pre-programmable metasurface array, it possesses wideband response characteristics and mode orthogonality. It can dynamically and precisely adjust the orbital angular momentum mode of the acoustic vortex field according to different vibration states and requirements of power equipment. This flexible adjustment method enables the acoustic vortex field to interact more effectively with the target power equipment, thereby stimulating the equipment to generate clearer and more accurate resonant response signals. This effectively overcomes the defect identification errors caused by incomplete or inaccurate signal acquisition in traditional methods, laying a solid foundation for subsequent accurate analysis of equipment vibration.

[0071] Step 102: Irradiate the resonant response signal with compressed laser and detect the equilibrium zero beat to obtain the time series of quantum correlation factors.

[0072] In this embodiment of the invention, compressed-state laser irradiation is performed based on the resonant response signal to obtain scattered light. Equilibrium zero-beat detection is then performed on the scattered light to obtain a detection signal time series. Quantum correlation analysis is then performed on the detection signal time series to obtain a quantum correlation factor time series.

[0073] This invention employs compressed-state laser irradiation of the resonant response signal. Compressed-state lasers possess unique quantum properties, reducing light field noise and improving the signal-to-noise ratio, thus highlighting the vibration-related information carried in the scattered light. Furthermore, balanced zero-beat detection technology can further precisely measure the quantum state information of the scattered light, effectively suppressing classical noise interference during the detection process. By combining these two technologies, the time series of quantum correlation factors of the scattered light can be extracted from the resonant response signal. This series contains deep-level quantum correlation information of the equipment vibration signal, and compared to signals obtained by traditional methods, it can more sensitively reflect subtle changes in equipment vibration, providing richer and more reliable evidence for accurately identifying equipment vibration defects.

[0074] Step 103: Reconstruct the phase space of the resonance response signal based on the time series of the quantum correlation factor, and calculate the chaotic characteristic quantities to obtain the correlation dimension and the maximum Lyapunov exponent.

[0075] In this embodiment of the invention, phase space reconstruction is performed based on the time series of quantum correlation factors to obtain the reconstructed phase space trajectory. The correlation dimension is then calculated based on the reconstructed phase space trajectory using the correlation integral function to obtain the correlation dimension. Finally, the maximum Lyapunov exponent is obtained by calculating the Lyapunov exponent based on the reconstructed phase space trajectory using the Wolf direct method.

[0076] This invention reconstructs the phase space of resonant response signals based on quantum correlation factor time series, transforming one-dimensional time series data into high-dimensional phase space trajectories, thus more comprehensively revealing the dynamic characteristics of equipment vibration systems. By calculating the correlation dimension, the complexity and fractal characteristics of the system in phase space can be quantitatively described. The magnitude of the correlation dimension reflects the dimension of the system's attractors, helping to determine whether abnormal nonlinear behavior exists in equipment vibration. The maximum Lyapunov exponent measures the system's sensitivity to initial conditions, i.e., the strength of its chaos. When the maximum Lyapunov exponent is greater than zero, it indicates that the system possesses chaotic characteristics, and its value reflects the degree of chaos. These two chaotic characteristic quantities characterize the dynamic essence of equipment vibration systems from different perspectives, providing key quantitative indicators for accurately identifying equipment vibration defects. Compared to traditional methods that rely solely on simple time-domain or frequency-domain feature analysis, this approach can reveal the intrinsic laws of equipment vibration more deeply and accurately.

[0077] Step 104: Based on the correlation dimension, the maximum Lyapunov exponent, and the preset threshold, defect identification is performed to obtain the defect identification data corresponding to the target power equipment.

[0078] Preset thresholds refer to the critical values ​​that are set in advance for each data point based on actual needs, including preset dimensionality thresholds. Preset index threshold Preset bifurcation threshold and preset mutation threshold .

[0079] In this embodiment of the invention, the identification result of vibration defects in power equipment is obtained based on the correlation dimension and the maximum Lyapunov exponent. Specifically: when the correlation dimension is greater than a preset dimension threshold and the maximum Lyapunov exponent is greater than a preset exponent threshold, the defect identification data corresponding to the target power equipment is winding deformation. When the correlation dimension bifurcation amount corresponding to the correlation dimension is greater than a preset bifurcation amount threshold and the mutation amount corresponding to the maximum Lyapunov exponent is greater than a preset mutation amount threshold, the defect identification data corresponding to the target power equipment is insulation damage.

[0080] This invention utilizes the previously calculated correlation dimension and maximum Lyapunov exponent, comparing them with preset thresholds to establish a scientific and accurate set of criteria for identifying vibration defects in power equipment. Different threshold settings correspond to different defect types. For example, a correlation dimension and maximum Lyapunov exponent exceeding specific values ​​can be identified as winding deformation, while bifurcation and abrupt changes exceeding other thresholds can be identified as insulation damage. This identification method based on chaotic features fully considers the nonlinear and chaotic characteristics of the equipment vibration system, accurately distinguishing different types of vibration defects. It avoids misjudgments and omissions caused by insufficient feature extraction or unscientific identification criteria in traditional methods, greatly improving the accuracy and reliability of power equipment vibration defect identification and providing strong technical support for the maintenance and repair of power equipment.

[0081] In this embodiment of the invention, the resonance response signal is obtained by dynamically adjusting the orbital angular momentum mode of the acoustic vortex field using an improved programmable metasurface array. Then, the time series of quantum correlation factors is obtained by combining compressed state laser irradiation and equilibrium zero-beat detection. Subsequently, the phase space is reconstructed to calculate the correlation dimension of chaotic characteristic quantities and the maximum Lyapunov exponent for defect identification. Compared with traditional methods, this method can deeply explore the deep dynamic characteristics of vibration signals of power equipment and effectively overcome the shortcomings of traditional methods, such as susceptibility to environmental noise interference, difficulty in capturing early weak defect signals, and incomplete characterization of complex vibration modes.

[0082] Please see Figure 2 , Figure 2 This is a flowchart illustrating the steps of a method for identifying vibration defects in power equipment according to Embodiment 2 of the present invention.

[0083] This invention provides a method for identifying vibration defects in power equipment, comprising:

[0084] Step 201: Periodically arrange and optimize the impedance matching of each metasurface unit in the preset programmable metasurface array using structural parameters to obtain the basic programmable metasurface array architecture.

[0085] Further, step 201 may include the following sub-steps:

[0086] S11. An electromagnetic simulation model is constructed using the structural parameters of the metasurface units in a pre-programmable metasurface array.

[0087] In this embodiment of the invention, an electromagnetic simulation model is established based on the structural parameters of the metasurface units of a preset programmable metasurface array. By establishing this model, the behavior of the metasurface units in an electromagnetic environment can be accurately simulated, avoiding the high costs and complex processes of actual fabrication and testing. Through this model, various structural parameters of the metasurface units can be simulated and analyzed in a virtual environment, quickly understanding their electromagnetic properties and providing theoretical basis and directional guidance for subsequent optimization design. This significantly shortens the R&D cycle, reduces R&D costs, and improves the accuracy and reliability of the design, ensuring that subsequent optimization work can be based on a scientifically sound model.

[0088] S12. Calculate the electromagnetic response characteristics of each metasurface unit within a preset frequency range using an electromagnetic simulation model to obtain initial electromagnetic characteristic data.

[0089] In this embodiment of the invention, the reflection / transmission coefficients and phase response data of the metasurface unit within a preset frequency range are calculated based on an electromagnetic simulation model, and initial electromagnetic characteristic data are obtained by solving Maxwell's equations. By calculating the reflection / transmission coefficients and phase response data, a deeper understanding of the electromagnetic response characteristics of the metasurface unit at different frequencies can be obtained. These initial electromagnetic characteristic data are the key basis for subsequent optimization, reflecting the behavior of the metasurface unit under the action of electromagnetic waves. Through the analysis of these data, the advantages and disadvantages of the metasurface unit within a specific frequency range can be identified, providing specific reference indicators for further optimization of its performance and helping to adjust design parameters in a targeted manner to achieve more ideal electromagnetic characteristics.

[0090] S13. Based on the initial electromagnetic characteristic data, with the optimization objective of balancing beam directivity and bandwidth, the spacing and arrangement period of the metasurface units are calculated using the particle swarm optimization algorithm to construct a periodic arrangement scheme.

[0091] In this embodiment of the invention, based on initial electromagnetic characteristic data, and with the optimization objective of balancing beam directivity and bandwidth, the spacing and arrangement period of the metasurface units are calculated using a particle swarm optimization (PSO) algorithm to obtain a periodic arrangement scheme. Specifically, by establishing an objective function and iteratively updating particle positions to evaluate directivity and bandwidth, the optimal unit spacing and arrangement period are ultimately output. The PSO algorithm efficiently searches for the optimal solution. By using this algorithm to calculate the spacing and arrangement period of the metasurface units with the optimization objective of balancing beam directivity and bandwidth, an arrangement scheme that maximizes bandwidth while meeting beam directivity requirements can be found. This optimization method overcomes the blindness and inefficiency of traditional manual parameter adjustments, enabling the rapid and accurate identification of the optimal periodic arrangement scheme, improving the efficiency and scientific rigor of the design, and achieving a better balance in the electromagnetic performance of the metasurface array.

[0092] S14. Based on the periodic arrangement scheme, an adjustable load element is introduced at the interface of the metasurface unit, and the load impedance of the adjustable load element is adjusted by the Smith chart method to obtain an impedance matching structure.

[0093] In this embodiment of the invention, adjustable capacitors / inductors, or adjustable load elements, are introduced at the interface of the metasurface units according to a periodic arrangement scheme. The load impedance is adjusted using the Smith chart method to achieve conjugate matching between the metasurface units and free space, resulting in an impedance matching structure. It should be noted that free space refers to a vacuum or air medium with a fixed characteristic impedance of approximately 377 ohms. Specifically, the impedance matching structure refers to a circuit network designed for each metasurface unit, composed of adjustable capacitors / inductors. Its function is to adjust the impedance of this network to achieve conjugate matching between the overall impedance of the metasurface unit and the free space impedance (377 ohms), thereby maximizing energy transfer and eliminating reflections. By introducing adjustable capacitors / inductors and using the Smith chart method for impedance matching, the energy transfer efficiency between the metasurface units and free space can be effectively improved. Achieving conjugate matching by adjusting the load impedance minimizes signal reflection, allowing more electromagnetic energy to smoothly enter the metasurface units, thereby improving the array's response sensitivity and signal strength. This step is crucial for ensuring the stable and efficient operation of the metasurface array in practical applications, contributing to improved overall array performance and reliability.

[0094] S15. The scattering parameters within the preset frequency band are simulated using an impedance matching structure, and the effective bandwidth is generated by frequency domain integration using the scattering parameters.

[0095] In this embodiment of the invention, the preset frequency band refers to the 2-18 GHz band. Scattering parameters within the 2-18 GHz band are simulated using an impedance matching structure, and the effective bandwidth is calculated through frequency domain integration. The scattering parameters within the preset frequency band are simulated using an impedance matching structure, and the frequency range below -10 dB is calculated. A fractional bandwidth formula is then used to quantify the effective bandwidth. It is worth noting that the fractional bandwidth formula is:

[0096] ;

[0097] in, For effective bandwidth; This is the high-frequency cutoff frequency; This is the low-frequency cutoff frequency; The center frequency.

[0098] S16. When the effective bandwidth is greater than the preset bandwidth threshold, the impedance matching structure is used as the basic programmable metasurface array architecture.

[0099] In this embodiment of the invention, the preset bandwidth threshold is 8 GHz. When the -10 dB bandwidth exceeds 8 GHz, the impedance matching structure is used as the basic programmable metasurface array architecture, resulting in a basic programmable metasurface array architecture with wide bandwidth response characteristics. Simulating scattering parameters and calculating the effective bandwidth within the 2-18 GHz frequency band allows for a comprehensive evaluation of the impedance matching structure's performance at different frequencies. The setting of a -10 dB bandwidth exceeding 8 GHz ensures that the metasurface array has a sufficiently wide bandwidth response. When this condition is met, it indicates that the array can effectively process signals over a wide frequency range, better adapting to the vibration signal detection needs of different frequencies in practical applications. This provides a solid hardware foundation for accurately acquiring the resonance response signal of the target power equipment, contributing to improved accuracy and reliability in identifying vibration defects in power equipment.

[0100] Step 202: Load the dynamic phase modulation algorithm based on the basic programmable metasurface array architecture, and control the reflection phase of each metasurface unit through time-domain coding to obtain the target acoustic vortex field generator.

[0101] Furthermore, step 202 may include the following sub-steps:

[0102] S21. Based on the basic programmable metasurface array frame, a time-domain-frequency domain joint control model is constructed.

[0103] In this embodiment of the invention, a time-domain-frequency domain joint control model is constructed based on a basic programmable metasurface array, and the reflection phase function of each metasurface unit is defined. By establishing the time-domain-frequency domain joint control model, the characteristics of the signal in both time and frequency dimensions can be comprehensively considered, making the control of the metasurface array more comprehensive and precise. This model breaks through the limitations of traditional single-dimensional control and provides a theoretical basis for the subsequent generation and dynamic adjustment of more complex acoustic vortex fields. Defining the reflection phase function of each metasurface unit clarifies the phase response law of each unit under specific time and frequency domain conditions, providing a mathematical basis for accurately controlling the reflection phase of each unit, making subsequent phase control work systematic, and laying the foundation for achieving high-precision acoustic vortex field control.

[0104] It should be noted that the reflection phase function is expressed as:

[0105] ;

[0106] in, This represents the reflection phase function value of the nth metasurface unit at time t, and v represents the frequency of the acoustic vortex field orbital angular momentum mode. This represents the initial phase bias of the nth metasurface unit.

[0107] S22. A time-domain encoded sequence generator is constructed by using a time-domain-frequency domain joint control model and the reflection phase function corresponding to each metasurface unit.

[0108] In this embodiment of the invention, a time-domain coded sequence generator is designed based on the time-domain-frequency domain joint control model and the reflection phase function of each metasurface unit. Designing the time-domain coded sequence generator to transform the time-domain-frequency domain joint control model and the reflection phase function into a practically operable time-domain control signal is a key step in achieving dynamic control.

[0109] S23. The signal is encoded by a time-domain encoded sequence generator using a hybrid coding strategy of binary phase offset keying and pulse width modulation to generate a time-domain control signal.

[0110] In this embodiment of the invention, a hybrid coding strategy of binary phase-shift keying (BPSK) and pulse width modulation (PWM) is employed to generate a time-domain control signal containing orbital angular momentum mode information. The specific process of this hybrid coding strategy is as follows: firstly, BPSK is used to assign 0 or 1 to each time unit ("0" or "1"). A fixed reflection phase is used to carry the mode state information of OAM; then, the duration of each "0" or "1" state within a cycle is dynamically adjusted by PWM (duty cycle) to precisely control the energy intensity and timing of each OAM mode in the time domain. It should be noted that the orbital angular momentum (OAM) mode information specifically refers to the characteristic of the electromagnetic wavefront phase being distributed in a spiral shape. Its core parameter is the topological charge number l (an integer), and the value of l determines the tightness of the wavefront spiral and the magnitude of the angular momentum carried by each photon.

[0111] By employing a hybrid coding strategy combining binary phase shift keying (BPS) and pulse width modulation (PWM), the advantages of both coding methods are fully utilized. BPS provides discrete phase changes, ensuring signal stability and reliability; PWM enables continuous phase control, enhancing signal flexibility. The time-domain control signal generated by this hybrid coding strategy effectively carries orbital angular momentum mode information, providing precise control commands for the subsequent generation of acoustic vortex fields with specific characteristics, thus improving the programmability and functionality of the acoustic vortex field generator.

[0112] S24. The time-domain control signal is loaded into the tunable element of the metasurface unit in real time through a field-programmable gate array, and the equivalent circuit parameters of each metasurface unit are dynamically adjusted within a preset control range to construct the initial acoustic vortex field generator.

[0113] In this embodiment of the invention, a time-domain control signal is loaded in real time onto the tunable elements of the metasurface unit via a field-programmable gate array (FPGA). By dynamically adjusting the equivalent circuit parameters of each unit of the metasurface, the reflection phase within the range of 0 to 2π is continuously controlled. Specifically, the dynamic adjustment process involves outputting a sequence of specific voltage / digital signals through the FPGA to control the bias voltage of the varactor diode or PIN diode integrated in each unit, changing its junction capacitance or impedance, thereby perturbing the unit's resonant characteristics and achieving continuous and dynamic control of its reflection phase. Reflection phase data is then obtained, and an initial acoustic vortex field generator is constructed using this data. The FPGA enables rapid loading of the time-domain control signal onto the tunable elements of the metasurface unit, achieving real-time phase control. By dynamically adjusting the equivalent circuit parameters of each unit, the reflection phase can be continuously and precisely controlled within the range of 0 to 2π. This high-precision phase control capability allows the acoustic vortex field generator to change the characteristics of the acoustic vortex field in real time according to different needs, such as orbital angular momentum mode and beam shape. This real-time, precise phase control provides a strong guarantee for the subsequent generation of high-quality, diverse acoustic vortex fields, which helps to improve the efficiency and accuracy of the interaction between the acoustic vortex field and the target power equipment.

[0114] S25. Perform far-field radiation modeling on the initial acoustic vortex field generator, and use the finite element method to calculate the sound pressure distribution under different time-domain coding modes using the generated far-field radiation model to obtain sound pressure distribution data.

[0115] In this embodiment of the invention, an initial acoustic vortex field generator far-field radiation model is constructed based on the reflection phase within the range of 0~2π. The sound pressure distribution under different time-domain coding modes is calculated using the finite element method, yielding the calculation results of the far-field radiation model, i.e., the sound pressure distribution data. By constructing the far-field radiation model, the propagation and distribution characteristics of the acoustic vortex field in the far field can be intuitively described, providing a theoretical framework for the analysis and optimization of the acoustic vortex field. The finite element method, as a numerical calculation method, has high computational accuracy and adaptability, and can accurately calculate the sound pressure distribution of the acoustic vortex field under different time-domain coding modes. The calculation results of the far-field radiation model provide a deeper understanding of the performance of the acoustic vortex field under different conditions, such as beam directivity and energy distribution. These results provide specific reference for subsequent optimization and adjustment of the acoustic vortex field generator, helping to further improve the quality and performance of the acoustic vortex field and better meet the needs of vibration detection in power equipment.

[0116] S26. Based on the sound pressure distribution data, the actual phase modulation error corresponding to the acoustic vortex field generator is obtained by fitting using the least squares method.

[0117] In this embodiment of the invention, the actual phase modulation error is obtained by fitting the calculation results of the far-field radiation model using the least squares method. By fitting the actual phase modulation error using the least squares method, the deviation between the actual phase and the theoretical phase can be accurately quantified, providing precise data support for subsequent error correction.

[0118] S27. The time-domain coding sequence is corrected through a closed-loop feedback mechanism based on the actual phase modulation error to obtain the root mean square error of the phase.

[0119] In this embodiment of the invention, the time-domain encoding sequence is corrected through a closed-loop feedback mechanism to obtain the root mean square error of the phase. The closed-loop feedback mechanism can monitor the phase modulation error in real time and automatically adjust the time-domain encoding sequence according to the error, achieving dynamic optimization of phase modulation. This closed-loop feedback mechanism greatly improves the accuracy and stability of phase modulation, ensuring that the acoustic vortex field generator can generate an acoustic vortex field that meets expectations.

[0120] S28. When the root mean square error of the phase is less than the preset error threshold, the initial acoustic vortex field generator at the current moment is used as the target acoustic vortex field generator.

[0121] In this embodiment of the invention, the preset error threshold refers to a critical value corresponding to the pre-set root mean square error of the phase. When the root mean square error of the phase is lower than the preset error threshold, a reconfigurable acoustic vortex field generator is determined. That is, by comparing the actual acoustic field phase distribution generated by the metasurface with the helical phase distribution of the ideal vortex field theory (… A global comparison is performed. When the root mean square error (RMSE) of the deviation between the two is lower than a preset error threshold, the initial acoustic vortex field generator at the current moment is determined from the "Measurement Results" and taken as the target acoustic vortex field generator. When the phase RMSE is lower than the preset error threshold, it indicates that the phase control accuracy of the acoustic vortex field generator has reached a high level, enabling it to work stably and reliably. This confirms the reconfigurable acoustic vortex field generator, providing a strong guarantee for the subsequent accurate acquisition of the resonance response signal of the target power equipment.

[0122] Step 203: Optimize the resonant frequency and quality factor of the metasurface unit by using the orbital angular momentum mode of the acoustic vortex field of the target acoustic vortex field generator to generate a dynamic control array.

[0123] In this embodiment of the invention, a dynamically modulated array is generated by optimizing the resonant frequency and quality factor of the metasurface unit through the orbital angular momentum mode of the acoustic vortex field of the target acoustic vortex field generator. Optimizing the resonant frequency aims to lay the foundation for dynamic control, because the resonant frequency determines the center point of the phase response of the metasurface unit. The phase change exerted by the unit on the wave is most drastic near its resonant frequency, enabling control from 0 to... The phase transition of the array, achieved by tuning the resonant frequency (e.g., using piezoelectric materials or tunable structures), allows the cells to operate dynamically at different points on the phase curve, thus obtaining the desired reflection phase. The reconfigurability of the array essentially depends on whether the resonant frequency of each cell can be precisely and independently controlled by an external signal such as voltage. Simultaneously, optimizing the quality factor ensures mode orthogonality. The quality factor determines the slope of the phase response and the operating bandwidth. High-Q cells have sharp resonant peaks, and their phase changes dramatically with frequency, contributing to more extreme and precise phase shifts, thus generating high-purity, high-order OAM modes. However, they are extremely sensitive to manufacturing errors and control precision, and have a narrow operating bandwidth. Low-Q cells, on the other hand, have a smooth phase response and a wider operating bandwidth; they are less sensitive to errors and more robust, but their phase tuning range may be insufficient, making it difficult to achieve complete... Phase overlay leads to a decrease in mode purity and crosstalk between different modes, resulting in poor orthogonality. Therefore, the fundamental purpose of optimization is to find an optimal Q-value trade-off for a specific operating frequency bandwidth and target OAM mode group, so that the unit response is both accurate and robust while ensuring a sufficient phase control range. Ultimately, the coordinated control of the resonant frequency generates a high-performance dynamic control array.

[0124] A dynamic control array with mode orthogonality is represented as:

[0125] ;

[0126] in, This represents a dynamically controlled array with mode orthogonality, controlled by the acoustic vortex field orbital angular momentum mode l and the current time t. Let represent the dynamic amplitude coefficient of the i-th metasurface element controlled by the acoustic vortex field orbital angular momentum mode l and the current time t. This represents the total number of metasurface units. Let the response vector of the i-th metasurface unit satisfy the following condition: , This represents the response vector of the j-th metasurface unit. Indicates the orthogonal symbol. Let represent the reflection phase function value of the i-th metasurface unit controlled by the acoustic vortex field orbital angular momentum mode l and the current time t, where e represents the exponential function.

[0127] Step 204: Adjust the topology of the metasurface units of the preset programmable metasurface array using the real-time feedback data corresponding to the dynamic control array to obtain the target programmable metasurface array.

[0128] In this embodiment of the invention, the real-time feedback data specifically refers to the mode purity and intensity distribution of far-field acoustic vortices measured by the microphone array and processed by a mode decomposition algorithm. By comparing the error between the measured values ​​and the target values, the optimization algorithm (such as gradient descent) is driven to adjust the control voltage applied to each metasurface unit by the FPGA (i.e., equivalent to adjusting its topology), thereby achieving closed-loop optimization of array performance. Based on the real-time feedback data of the dynamically controlled array with mode orthogonality, the topology of the metasurface units of the programmable metasurface array is adjusted to obtain the improved programmable metasurface array, i.e., the target programmable metasurface array.

[0129] Step 205: Apply the acoustic vortex field generated by the target programmable metasurface array to the target power equipment, and use the vibration signal generated by the target power equipment as the resonance response signal of the target power equipment.

[0130] In this embodiment of the invention, the previously optimized target programmable metasurface array is used as a precision excitation source to generate an acoustic vortex field carrying a specific orbital angular momentum, which is then precisely applied to the target power equipment. This special acoustic field can perform efficient mode matching with the equipment structure. When the angular momentum mode of the acoustic vortex matches the vibration mode of a potential defect in the equipment, it significantly excites its structural resonance. The vibration signal generated by the equipment is thus synchronously acquired and specifically identified as the resonance response signal of the target power equipment.

[0131] Step 206: Irradiate the resonant response signal with compressed state laser and detect equilibrium zero beats to obtain the time series of quantum correlation factors.

[0132] Furthermore, step 206 may include the following sub-steps:

[0133] S31. The resonant response signal is irradiated with a compressed laser to generate scattered light.

[0134] In this embodiment of the invention, the resonant response signal is subjected to compressed-state laser irradiation to obtain scattered light. By using compressed-state laser to irradiate the resonant response signal, the signal-to-noise ratio of the optical field is improved. When it irradiates the resonant response signal, it can more sensitively perceive subtle changes in the signal, enabling the scattered light to carry more precise information related to the vibration of the target power equipment. This provides a better data foundation for subsequent precise analysis of the equipment's vibration characteristics, helping to detect potential vibration defects in the equipment earlier and more accurately, and ensuring the stable operation of the power system.

[0135] S32. Perform balanced zero-beat detection on the scattered light to generate a time series of detection signals.

[0136] In this embodiment of the invention, a time series of the detection signal is obtained by performing balanced zero-beat detection processing on the scattered light. The specific detection process is as follows: the signal light and the local oscillator light are interfered with at a 50:50 beam splitter, and the two output optical signals are respectively fed into a pair of performance-matched photodiodes for subtraction. Differential amplification is used to eliminate classical intensity noise and amplify quantum noise containing information. Finally, the time series of photocurrent, which is proportional to the orthogonal component of the signal light field, is output as the detection signal time series. By applying balanced zero-beat detection technology to scattered light processing, quantum state information in the scattered light can be effectively extracted. This technology, by interfering the scattered light with the local oscillator light and measuring the interference result using a photodetector, can accurately measure the quantum properties of the scattered light while suppressing classical noise interference during the detection process. The detection signal time series obtained after this processing more clearly reflects the quantum correlation information related to the vibration of power equipment compared to the original scattered light signal. It removes a large amount of irrelevant noise, making the key features in the signal more prominent, providing a reliable basis for subsequent accurate analysis of the quantum correlation characteristics of the equipment, and improving the accuracy of vibration defect identification.

[0137] S33. Perform quantum correlation analysis on the time series of the detection signal to obtain the time series of quantum correlation factors.

[0138] In this embodiment of the invention, based on the Takens embedding theorem, the quantum correlation factor time series is used. Based on itself, by selecting the optimal delay time ( (Commonly obtained using the autocorrelation function method or mutual information method) and embedding dimension ( (Used commonly by the pseudo-nearest neighbor method), constructing Vector points in 3D phase space Connecting all vector points in chronological order yields the reconstructed phase space trajectory, i.e., the quantum correlation factor time series. This quantum correlation factor time series quantifies the degree of correlation between different components at different times in the signal, providing a more comprehensive description of the quantum characteristics of the power equipment vibration system. Compared to traditional analysis methods based solely on time-domain or frequency-domain features, the quantum correlation factor time series provides richer and deeper information, more accurately reflecting the complex dynamic behavior of equipment vibration, and offering a more powerful tool for precisely identifying vibration defects in power equipment.

[0139] Step 207: Reconstruct the phase space of the resonance response signal based on the time series of the quantum correlation factor, and calculate the chaotic characteristic quantities to obtain the correlation dimension and the maximum Lyapunov exponent.

[0140] Furthermore, step 207 may include the following sub-steps:

[0141] S41. Phase space reconstruction is performed using quantum correlation factor time series to obtain the reconstructed phase space trajectory.

[0142] Further, step S41 may include the following sub-steps:

[0143] S411. Calculate the mutual information between the time series of quantum correlation factors and the time series of quantum correlation factors with a preset delay using the mutual information method.

[0144] In this embodiment of the invention, the original quantum correlation factor time series is copied and delayed backward. Each time step generates a corresponding time series with a pre-defined delayed quantum correlation factor. The mutual information method is used to statistically analyze these two series, calculating their mutual information content. The advantage of this method is that it can capture comprehensive statistical correlations between series, including nonlinear dependencies, rather than being limited to linear relationships. By systematically calculating different... The mutual information at a given value can be observed to change with... The curve increases and then decreases, and the first minimum point on this curve is selected as the optimal delay time. This selection means that in this... At this value, the delayed sequence provides new, non-redundant dynamic information about the original sequence to the greatest extent, thus providing a key data foundation and parameter basis for the subsequent accurate reconstruction of the phase space trajectory that can truly reflect the dynamic characteristics of the system.

[0145] S412. The delay time when the mutual information first reaches its minimum value is taken as the optimal parameter.

[0146] In this embodiment of the invention, the delay time at which the mutual information first reaches its minimum value is used as the optimal parameter. Using this delay time as the optimal parameter indicates that the correlation between the two sequences is at its lowest at this point, meaning the sequences have maximum independence at this delay time. Choosing this delay time as the optimal parameter ensures sufficient independence between state vectors during subsequent phase space reconstruction, avoiding information redundancy caused by excessive correlation of time series. The constructed state vectors can more accurately reflect the true state of the system, laying the foundation for accurate reconstruction of the phase space trajectory, thereby improving the accuracy of judging the vibration state of the equipment.

[0147] S413. Based on the optimal parameters, determine the proportion of pseudo-neighbors of the state vector in the low-dimensional embedding space using the pseudo-nearest neighbor method.

[0148] In this embodiment of the invention, based on optimal parameters, the proportion of pseudo-nearest neighbors of the state vector in the low-dimensional embedding space is determined using the pseudo-nearest neighbor method. By calculating the proportion of pseudo-nearest neighbors, the suitability of the current embedding dimension can be evaluated. If the proportion of pseudo-nearest neighbors is too high, it indicates that the current dimension is insufficient to accurately describe the system's state, and the dimension needs to be increased; conversely, if the proportion is low, it indicates that the current dimension may already be sufficient. This step helps determine a suitable embedding dimension, avoiding inaccurate phase space reconstruction due to improper dimension selection, and providing a guarantee for the accurate calculation of chaotic features in the subsequent process.

[0149] It is worth mentioning that the proportion of false neighbors refers to the proportion of points that are "falsely nearby" due to dimensional compression in the low-dimensional embedding space. Specifically, it is calculated as follows: for each point, after finding its nearest neighbor in the low-dimensional space, the ratio of the distance of the neighbor in the high-dimensional space to the distance in the low-dimensional space is calculated. If the ratio exceeds a set threshold (such as 2), it is determined to be a false neighbor. Finally, the number of false neighbors of all points is counted and divided by the total number of neighbors to obtain the proportion.

[0150] S414. When the proportion of pseudo-neighbors is lower than the preset pseudo-neighbor proportion threshold, the embedding dimension corresponding to the pseudo-neighbor proportion is taken as the minimum embedding dimension.

[0151] In this embodiment of the invention, the minimum embedding dimension refers to the smallest dimension value that makes the proportion of pseudo-neighbors in the low-dimensional embedding space lower than a preset threshold. The method is as follows: starting from a low dimension, the embedding dimension is gradually increased. For each dimension, the proportion of pseudo-neighbors for all points is calculated (by comparing the ratio of neighbor distances between high-dimensional and low-dimensional spaces to whether it exceeds a threshold). When the proportion first falls below the threshold, that dimension is the minimum embedding dimension. The preset pseudo-neighbor proportion threshold is used to determine whether the current embedding dimension meets the requirements. When the pseudo-neighbor proportion is lower than this threshold, it indicates that the number of pseudo-neighbors between state vectors is within an acceptable range at the current dimension, meaning that this dimension can better preserve the dynamic information of the system and avoid information loss. The minimum embedding dimension determined at this point can minimize computational load and improve the efficiency of subsequent analysis while ensuring the accuracy of phase space reconstruction, providing a suitable embedding space for accurately calculating the correlation dimension and chaotic features such as the maximum Lyapunov exponent.

[0152] S415. Based on the minimum embedding dimension, the time series of quantum correlation factors is converted into a multidimensional state vector.

[0153] In this embodiment of the invention, the quantum correlation factor time series is converted into a multidimensional state vector according to the minimum embedding dimension, wherein each multidimensional state vector is composed of quantum correlation factor values ​​at consecutive time delays. It is worth noting that the quantum correlation factor values ​​are obtained through quantum experimental measurements (such as quantum state tomography, correlation function detection) or numerical simulations of quantum system evolution (such as calculating quantum mutual information, entanglement entropy, and other correlation quantities). Specifically, this is achieved by collecting quantum system correlation characteristic data at consecutive time points to form a time series, and then extracting the values ​​at consecutive time delays from it to construct the multidimensional state vector.

[0154] Converting the quantum correlation factor time series into multidimensional state vectors transforms one-dimensional time series data into points in a high-dimensional space, enabling a more comprehensive description of the system's state. Each multidimensional state vector consists of quantum correlation factor values ​​at consecutive time delays. This construction method fully utilizes the information in the time series, allowing the state vectors to encompass the dynamic changes of the system at different times. This approach more accurately reflects the system's dynamic behavior, provides a richer data foundation for subsequent phase space reconstruction, facilitates more precise calculation of chaotic characteristic quantities, and improves the accuracy of identifying equipment vibration defects.

[0155] S416. The phase space trajectory is reconstructed based on the multidimensional state vector by using time delay embedding technology to obtain the reconstructed phase space trajectory.

[0156] In this embodiment of the invention, reconstructing the phase space trajectory using time-delay embedding technology requires first determining the optimal delay time and the minimum embedding dimension *m*, then constructing state vectors, and finally plotting the trajectory formed by these vectors in m-dimensional space, ensuring that the attractor topology of the original dynamic system is preserved. By utilizing multi-dimensional state vectors, this technique can recover the dynamic behavior of the system in high-dimensional space, revealing the complex structure originally hidden in a one-dimensional time series. The reconstructed phase space trajectory can more intuitively reflect the state evolution process of the system, facilitating the analysis of the system's attractors, bifurcation, and other characteristics. This step provides an intuitive geometric object for subsequent calculations of chaotic features such as the correlation dimension and the maximum Lyapunov exponent, helping to gain a deeper understanding of the dynamic essence of equipment vibration systems and improving the accuracy and reliability of vibration defect identification.

[0157] S42. Substitute the reconstructed phase space trajectory into the correlation integral function to calculate the correlation dimension, and obtain the correlation dimension.

[0158] In this embodiment of the invention, the correlation dimension is calculated based on the reconstructed phase space trajectory using a correlation integral function. Statistical analysis of the distances between point pairs in the reconstructed phase space trajectory using the correlation integral function accurately quantifies the complexity of the attractor in a chaotic system. Different power equipment vibration systems exhibit significant differences in attractor structure under normal and fault conditions, and the correlation dimension can effectively distinguish these states. For example, the attractor structure of a normal equipment vibration system is relatively regular, resulting in a smaller correlation dimension; however, when a fault exists, such as wear or loosening of components, the dynamic characteristics of the vibration system change, the attractor structure becomes more complex, and the correlation dimension increases. This provides a quantitative basis for accurately judging the vibration state of equipment and helps in the early detection of potential faults.

[0159] Reconstructing the phase space extends a one-dimensional time series to a higher-dimensional space. Analysis based on the correlation integral function can fully uncover the hidden nonlinear information within the time series. Power equipment vibration systems often exhibit complex nonlinear dynamic characteristics that are difficult to accurately describe using traditional linear analysis methods. The correlation dimension, as a nonlinear index, can comprehensively reflect the structural characteristics of the system in a higher-dimensional space, providing strong support for a deeper understanding of the intrinsic mechanisms of equipment vibration systems and helping to grasp the fundamental causes and development patterns of equipment vibration defects.

[0160] S43. Based on the divergence data of adjacent orbits in the reconstructed phase space trajectory, the Wolf direct method is used to calculate the Lyapunov exponent, and the maximum Lyapunov exponent is obtained.

[0161] In this embodiment of the invention, based on the Wolf direct method, the phase space trajectory is reconstructed (the optimal embedding dimension and delay time are determined by time delay embedding technology), and the exponential separation rate of the initial vector and its nearest neighbor over time is tracked in the phase space. Finally, by linearly fitting the natural logarithm of the separation distance and the slope of the evolution time, the dynamic behavior of the system is accurately quantified, and the maximum Lyapunov exponent is obtained as the core indicator of the chaos criterion.

[0162] The maximum Lyapunov exponent is a key indicator for measuring the sensitivity of a chaotic system to initial conditions. The Wolf direct method calculates this exponent by tracking the divergence of adjacent trajectories in the reconstructed phase space trajectory, accurately reflecting the amplification capability of a power equipment vibration system to small initial disturbances. If the maximum Lyapunov exponent is greater than zero, it indicates that the system exhibits chaotic characteristics and is highly sensitive to initial conditions; small initial differences can lead to significant differences in the system's future state. This helps determine whether the equipment vibration system is in a chaotic state, providing important evidence for fault diagnosis.

[0163] The long-term behavioral trend of a power equipment vibration system can be predicted based on the sign and magnitude of the maximum Lyapunov exponent. When the maximum Lyapunov exponent is positive and large, the system trajectory diverges rapidly, the vibration behavior becomes unpredictable, the equipment may be in an unstable state, and the risk of failure increases. Conversely, if the maximum Lyapunov exponent is negative, the system trajectory converges, and the vibration behavior is relatively stable. By monitoring changes in the maximum Lyapunov exponent, abnormal changes in the equipment vibration state can be detected in advance, allowing for timely measures to prevent failures and ensure the safe operation of power equipment.

[0164] The magnitude of the maximum Lyapunov exponent directly reflects the degree of chaos in a power equipment vibration system. A larger value indicates stronger chaos and more complex, disordered vibration behavior; a smaller value indicates weaker chaos and more regular vibration behavior. This provides a concise and effective indicator for assessing the complexity of equipment vibration systems, enabling engineers to quickly understand the equipment's vibration state, formulate appropriate maintenance and repair strategies, and improve the reliability and stability of equipment operation.

[0165] Step 208: Based on the correlation dimension, the maximum Lyapunov exponent, and the preset threshold, defect identification is performed to obtain the defect identification data corresponding to the target power equipment.

[0166] Furthermore, step 208 may include the following sub-steps:

[0167] S51. When the correlation dimension is greater than the preset dimension threshold and the maximum Lyapunov exponent is greater than the preset exponent threshold, the defect identification data corresponding to the target power equipment is winding deformation.

[0168] S52. When the number of bifurcations in the correlation dimension corresponding to the correlation dimension is greater than the preset bifurcation threshold and the number of mutations corresponding to the maximum Lyapunov exponent is greater than the preset mutation threshold, the defect identification data corresponding to the target power equipment is insulation damage.

[0169] In this embodiment of the invention, if the correlation dimension Preset dimensionality threshold And the maximum Lyapunov index Preset index threshold If so, it is determined to be winding deformation;

[0170] If the correlation dimension bifurcation Greater than the preset bifurcation threshold And the largest Lyapunov index mutation Greater than the preset mutation threshold If so, it is determined to be insulation damage.

[0171] The correlation dimension reflects the complexity of the attractor in a chaotic system, while the maximum Lyapunov exponent measures the system's sensitivity to initial conditions. Using only one of these indicators for identification may lead to misjudgment due to incomplete information. For example, relying solely on the correlation dimension may make it difficult to distinguish between normal and fault states with similar complex attractor structures; conversely, relying solely on the maximum Lyapunov exponent may fail to accurately capture some early faults with subtle chaotic characteristics. Combining both allows for a comprehensive assessment of the vibration state of power equipment from two different dimensions: attractor structure and system dynamic sensitivity. This significantly improves the accuracy of identification results and reduces misdiagnosis and missed diagnosis.

[0172] In this embodiment of the invention, the present invention utilizes an improved programmable metasurface array to dynamically adjust the orbital angular momentum mode of the acoustic vortex field to obtain the resonance response signal. Then, it combines compressed state laser irradiation and equilibrium zero-beat detection to obtain the time series of quantum correlation factors. Subsequently, it reconstructs the phase space to calculate the correlation dimension of chaotic characteristic quantities and the maximum Lyapunov exponent for defect identification. Compared with traditional methods, it can deeply explore the deep dynamic characteristics of vibration signals of power equipment and effectively overcome the shortcomings of traditional methods, such as susceptibility to environmental noise interference, difficulty in capturing early weak defect signals, and incomplete characterization of complex vibration modes.

[0173] This invention optimizes the periodic arrangement and impedance matching of metasurface units in a programmable metasurface array, establishes an electromagnetic simulation model, and applies particle swarm optimization and the Smith chart method to obtain a basic programmable metasurface array architecture with wideband response characteristics. This architecture can capture richer frequency components in the vibration signals of power equipment, especially low-energy, high-frequency signals generated by early weak defects, effectively overcoming the problem of traditional methods missing early defects due to bandwidth limitations.

[0174] This invention utilizes a dynamic phase modulation algorithm based on a fundamental programmable metasurface array architecture. It establishes a joint time-domain and frequency-domain control model, defines a reflection phase function, and designs a time-domain encoded sequence generator. A hybrid encoding strategy is employed to generate a time-domain control signal containing orbital angular momentum mode information. This signal is then applied in real-time to the tunable elements of the metasurface unit via a field-programmable gate array (FPGA), enabling continuous control of the reflection phase within the range of 0 to 2π. Simultaneously, by constructing a far-field radiation model, calculating the sound pressure distribution, fitting the phase modulation error, and using a closed-loop feedback mechanism to correct the time-domain encoded sequence, the high-precision reconfigurability of the acoustic vortex field generator is ensured. This allows for real-time adjustment of the acoustic vortex field based on the actual operating state and vibration characteristics of the power equipment, thereby more accurately acquiring the resonance response signal of the target power equipment. This effectively reduces external interference and signal distortion, enhancing the reliability of signal acquisition.

[0175] This invention utilizes compressed-state laser irradiation and equilibrium zero-beat detection techniques to obtain the time series of quantum correlation factors of scattered light, thereby reconstructing the phase space of the resonant response signal and calculating chaotic characteristic quantities, namely the correlation dimension and the maximum Lyapunov exponent. Phase space reconstruction is achieved by processing the quantum correlation factor time series using methods such as mutual information and pseudo-nearest neighbor methods. The correlation dimension and the maximum Lyapunov exponent are then calculated based on the correlation integral function and the Wolf direct method, respectively. Finally, by comparing the correlation dimension and the maximum Lyapunov exponent with preset thresholds, the type of vibration defect in power equipment, such as winding deformation or insulation damage, can be accurately determined. Compared with traditional methods that rely solely on simple time-domain or frequency-domain feature analysis, this invention provides a more comprehensive and accurate reflection of the equipment's vibration state.

[0176] Please see Figure 3 , Figure 3 This is a structural block diagram of a power equipment vibration defect identification system provided in Embodiment 3 of the present invention.

[0177] This invention provides a vibration defect identification system for power equipment, comprising:

[0178] The adaptive vortex field control module 301 is used to dynamically adjust the orbital angular momentum mode of the acoustic vortex field corresponding to the target power equipment according to the preset programmable metasurface array, so as to obtain the resonance response signal of the target power equipment.

[0179] The quantum precision measurement module 302 is used to irradiate the resonant response signal with compressed state laser and detect the equilibrium zero beat to obtain the time series of quantum correlation factors.

[0180] The nonlinear dynamics feature extraction module 303 is used to reconstruct the phase space of the resonance response signal based on the time series of quantum correlation factors, and to calculate the chaotic feature quantities to obtain the correlation dimension and the maximum Lyapunov exponent.

[0181] The defect identification decision module 304 is used to identify defects based on the correlation dimension, the maximum Lyapunov exponent and a preset threshold, and obtain defect identification data corresponding to the target power equipment.

[0182] Furthermore, the adaptive vortex field control module 301 can perform the following steps:

[0183] The basic programmable metasurface array architecture is obtained by periodically arranging and optimizing the structural parameters of each metasurface unit in the pre-programmable metasurface array and designing impedance matching.

[0184] A target acoustic vortex field generator is obtained by loading a dynamic phase modulation algorithm on a basic programmable metasurface array architecture and controlling the reflection phase of each metasurface unit through time-domain coding.

[0185] The resonant frequency and quality factor of the metasurface unit are optimized by using the orbital angular momentum mode of the acoustic vortex field of the target acoustic vortex field generator to generate a dynamically adjustable array.

[0186] The topology of the metasurface units of the preset programmable metasurface array is adjusted by using real-time feedback data corresponding to the dynamic control array to obtain the target programmable metasurface array.

[0187] The acoustic vortex field generated by the target programmable metasurface array is applied to the target power equipment, and the vibration signal generated by the target power equipment is used as the resonance response signal of the target power equipment.

[0188] Furthermore, the adaptive vortex field control module 301 can also perform the following steps:

[0189] An electromagnetic simulation model is constructed using the structural parameters of the metasurface units in a pre-programmable metasurface array.

[0190] The electromagnetic response characteristics of each metasurface unit within a preset frequency range are calculated using an electromagnetic simulation model to obtain initial electromagnetic characteristic data.

[0191] Based on the initial electromagnetic characteristic data, with the optimization objective of balancing beam directivity and bandwidth, the spacing and arrangement period of the metasurface units are calculated using the particle swarm optimization algorithm to construct a periodic arrangement scheme;

[0192] Based on a periodic arrangement scheme, an adjustable load element is introduced at the interface of the metasurface unit, and the load impedance of the adjustable load element is adjusted by the Smith chart method to obtain an impedance matching structure.

[0193] The scattering parameters within a preset frequency band are simulated using an impedance matching structure, and the effective bandwidth is generated by frequency domain integration using the scattering parameters.

[0194] When the effective bandwidth is greater than the preset bandwidth threshold, the impedance matching structure is used as the basic programmable metasurface array architecture.

[0195] Furthermore, the adaptive vortex field control module 301 can also perform the following steps:

[0196] A time-domain-frequency-domain joint control model is constructed based on a basic programmable metasurface array architecture.

[0197] A time-domain coded sequence generator is constructed using a time-domain-frequency domain joint control model and the reflection phase function corresponding to each metasurface unit;

[0198] The signal is encoded using a hybrid coding strategy of binary phase shift keying and pulse width modulation through a time-domain encoded sequence generator to generate a time-domain control signal.

[0199] The time-domain control signal is loaded into the tunable element of the metasurface unit in real time through a field-programmable gate array, and the equivalent circuit parameters of each metasurface unit are dynamically adjusted within a preset control range to construct an initial acoustic vortex field generator.

[0200] The initial acoustic vortex field generator is modeled for far-field radiation, and the sound pressure distribution under different time-domain coding modes is calculated using the finite element method based on the generated far-field radiation model to obtain sound pressure distribution data.

[0201] Based on the sound pressure distribution data, the actual phase modulation error corresponding to the acoustic vortex field generator is obtained by fitting the least squares method.

[0202] The root mean square error of the phase is obtained by correcting the time-domain encoded sequence according to the actual phase modulation error through a closed-loop feedback mechanism.

[0203] When the root mean square error of the phase is less than the preset error threshold, the initial acoustic vortex field generator at the current moment is used as the target acoustic vortex field generator.

[0204] Furthermore, the quantum precision measurement module 302 can perform the following steps:

[0205] The resonant response signal is irradiated with a compressed laser to generate scattered light;

[0206] The scattered light is subjected to balanced zero-beat detection to generate a time series of detection signals;

[0207] Quantum correlation analysis was performed on the time series of the detection signal to obtain the time series of quantum correlation factors.

[0208] Furthermore, the nonlinear dynamics feature extraction module 303 can perform the following steps:

[0209] Phase space reconstruction was performed using quantum correlation factor time series to obtain the reconstructed phase space trajectory;

[0210] The reconstructed phase space trajectory is substituted into the correlation integral function to calculate the correlation dimension, thus obtaining the correlation dimension.

[0211] Based on the divergence data of adjacent orbits in the reconstructed phase space trajectory, the Wolf direct method is used to calculate the Lyapunov exponent, and the maximum Lyapunov exponent is obtained.

[0212] Furthermore, the nonlinear dynamics feature extraction module 303 can also perform the following steps:

[0213] The mutual information method is used to calculate the mutual information between the time series of quantum correlation factors and the time series of quantum correlation factors with a preset delay.

[0214] The delay time when the mutual information first reaches its minimum value is used as the optimal parameter;

[0215] Based on the optimal parameters, the proportion of pseudo-neighbors of the state vector in the low-dimensional embedding space is determined by the pseudo-nearest neighbor method.

[0216] When the proportion of false neighbor points is lower than the preset threshold for the proportion of false neighbor points, the embedding dimension corresponding to the proportion of false neighbor points is taken as the minimum embedding dimension.

[0217] Based on the minimum embedding dimension, the time series of quantum correlation factors are transformed into multidimensional state vectors;

[0218] The phase space trajectory is reconstructed by using time delay embedding technology based on multidimensional state vectors.

[0219] Furthermore, the defect identification decision module 304 can perform the following steps:

[0220] When the correlation dimension is greater than the preset dimension threshold and the maximum Lyapunov exponent is greater than the preset exponent threshold, the defect identification data corresponding to the target power equipment is winding deformation;

[0221] When the number of bifurcations in the correlation dimension corresponding to the correlation dimension is greater than the preset bifurcation threshold and the number of mutations corresponding to the maximum Lyapunov exponent is greater than the preset mutation threshold, the defect identification data corresponding to the target power equipment is insulation damage.

[0222] Please see Figure 4 , Figure 4 This is a structural block diagram of a computer device provided in Embodiment 4 of the present invention.

[0223] An electronic device according to an embodiment of the present invention includes: a memory 401 and a processor 402. The memory 401 stores a computer program. When the computer program is executed by the processor 402, the processor 402 performs the power equipment vibration defect identification method as described in any of the above embodiments.

[0224] Memory 401 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Memory 401 has storage space 403 for program code 413 for performing any of the method steps described above. For example, storage space 403 for program code may include individual program codes 413 for implementing the various steps in the methods described above. This program code may be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, CDs, memory cards, or floppy disks. The program code may be compressed, for example, in a suitable form. When run by a computing processing device, this code causes the computing processing device to perform the various steps in the methods described above. This program code may be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, CDs, memory cards, or floppy disks. The program code may be compressed, for example, in a suitable form. When this code is run by a computing processing device, it causes the device to perform the various steps in the power equipment vibration defect identification method described above.

[0225] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0226] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0227] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0228] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0229] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0230] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A power equipment vibration defect identification method, characterized in that, The method comprises the following steps: According to the preset programmable metasurface array, the acoustic vortex field orbital angular momentum mode corresponding to the target power equipment is dynamically adjusted, and the resonance response signal of the target power equipment is obtained; The resonance response signal is irradiated by a compressed state laser and detected by a balanced homodyne detection, and a quantum correlation factor time sequence is obtained; Based on the quantum correlation factor time sequence, the resonance response signal phase space is reconstructed, and the chaotic characteristic quantity is calculated to obtain the correlation dimension and the maximum Lyapunov exponent; Based on the correlation dimension, the maximum Lyapunov exponent and the preset threshold, defect recognition is carried out to obtain the defect recognition data corresponding to the target power equipment.

2. The power equipment vibration defect identification method according to claim 1, characterized in that, The step of adjusting the acoustic vortex field orbital angular momentum mode corresponding to the target power equipment according to the preset programmable metasurface array to obtain the resonance response signal of the target power equipment comprises the following steps: The structural parameters of each metasurface unit in the preset programmable metasurface array are periodically arranged and optimized and impedance matching design is performed to obtain a basic programmable metasurface array architecture; A dynamic phase modulation algorithm is loaded based on the basic programmable metasurface array architecture, and the reflection phase of each metasurface unit is controlled through time domain coding to obtain a target acoustic vortex field generator; The resonance frequency and quality factor of each metasurface unit are optimized through the acoustic vortex field orbital angular momentum mode of the target acoustic vortex field generator to generate a dynamic control array; The topology structure of the metasurface unit of the preset programmable metasurface array is adjusted by using the real-time feedback data corresponding to the dynamic control array to obtain a target programmable metasurface array; The acoustic vortex field generated by the target programmable metasurface array acts on the target power equipment, and the vibration signal generated by the target power equipment is used as the resonance response signal of the target power equipment.

3. The power equipment vibration defect identification method according to claim 2, characterized in that, The step of periodically arranging and optimizing and impedance matching design of the structural parameters of each metasurface unit in the preset programmable metasurface array to obtain a basic programmable metasurface array architecture comprises the following steps: An electromagnetic simulation model is constructed by using the structural parameters of the metasurface unit in the preset programmable metasurface array; The electromagnetic response characteristics of each metasurface unit in a preset frequency range are calculated through the electromagnetic simulation model to obtain initial electromagnetic characteristic data; Based on the initial electromagnetic characteristic data, the spacing and arrangement period of the metasurface unit are calculated by particle swarm algorithm to construct a periodic arrangement scheme, taking balanced beam directivity and frequency bandwidth as optimization objectives; Based on the periodic arrangement scheme, adjustable load elements are introduced at the interfaces of the metasurface units, and the load impedance of the adjustable load elements is adjusted by Smith chart method to obtain an impedance matching structure; The scattering parameters in a preset frequency band are simulated by the impedance matching structure, and the effective bandwidth is generated by frequency domain integral calculation using the scattering parameters; When the effective bandwidth is greater than a preset bandwidth threshold, the impedance matching structure is used as a basic programmable metasurface array architecture.

4. The power equipment vibration defect identification method according to claim 2 or 3, characterized in that, The step of loading a dynamic phase modulation algorithm based on the basic programmable metasurface array architecture and controlling the reflection phase of each metasurface unit through time domain coding to obtain a target acoustic vortex field generator comprises: Based on the basic programmable metasurface array architecture, a time-frequency domain joint regulation model is constructed; The time-frequency domain joint regulation model and the reflection phase function corresponding to each metasurface unit are used to construct a time domain coding sequence generator; A time domain control signal is generated by using a binary phase shift keying and pulse width modulation hybrid coding strategy for signal coding through the time domain coding sequence generator; The time domain control signal is loaded to the tunable element of the metasurface unit through the field programmable gate array in real time, and the equivalent circuit parameters of each metasurface unit are dynamically adjusted within a preset regulation range to construct an initial acoustic vortex field generator; The initial acoustic vortex field generator is modeled for far-field radiation, and the sound pressure distribution under different time domain coding modes is calculated through the generated far-field radiation model using the finite element method to obtain sound pressure distribution data; Based on the sound pressure distribution data, the actual phase regulation error corresponding to the acoustic vortex field generator is fitted using the least squares method; The phase root mean square error is obtained by correcting the time domain coding sequence through a closed-loop feedback mechanism according to the actual phase regulation error; When the phase root mean square error is less than a preset error threshold, the initial acoustic vortex field generator at the current time is taken as the target acoustic vortex field generator.

5. The power equipment vibration defect identification method of claim 1, wherein The step of compressing the resonance response signal and irradiating it with a squeezed state laser and balanced homodyne detection to obtain a quantum correlation factor time series comprises: The resonance response signal is irradiated with a squeezed state laser to generate scattered light; The scattered light is detected by balanced homodyne detection to generate a detection signal time series; Quantum correlation analysis is performed on the detection signal time series to obtain a quantum correlation factor time series.

6. The power equipment vibration defect identification method of claim 1, wherein The step of reconstructing the resonance response signal phase space based on the quantum correlation factor time series and calculating the chaotic characteristic quantity to obtain the correlation dimension and the maximum Lyapunov exponent comprises: The quantum correlation factor time series is used for phase space reconstruction processing to obtain a reconstructed phase space trajectory; The reconstructed phase space trajectory is substituted into the correlation integral function for correlation dimension calculation processing to obtain the correlation dimension; Based on the divergence data of adjacent orbits in the reconstructed phase space trajectory, the Wolf direct method is used for Lyapunov exponent calculation to obtain the maximum Lyapunov exponent.

7. The power equipment vibration defect identification method according to claim 6, characterized in that, The step of using the quantum correlation factor time series for phase space reconstruction processing to obtain a reconstructed phase space trajectory comprises: The mutual information between the quantum correlation factor time series and a preset delay quantum correlation factor time series is calculated using the mutual information method; The delay time when the mutual information first reaches a minimum value is taken as the optimal parameter; Based on the optimal parameter, the proportion of pseudo-neighbor points of state vectors in a low-dimensional embedding space is determined by the pseudo-nearest neighbor method; When the proportion of pseudo-neighbor points is lower than a preset pseudo-neighbor point proportion threshold, the embedding dimension corresponding to the proportion of pseudo-neighbor points is taken as the minimum embedding dimension; convert the quantum correlation factor time series into a multi-dimensional state vector based on the minimum embedding dimension; reconstruct a phase space trajectory based on the multi-dimensional state vector by a time delay embedding technique to obtain a reconstructed phase space trajectory.

8. The power equipment vibration defect identification method of claim 1, wherein, The step of performing defect identification based on the correlation dimension, the maximum Lyapunov exponent, and a preset threshold to obtain defect identification data corresponding to the target power equipment comprises: When the correlation dimension is greater than a preset dimension threshold and the maximum Lyapunov exponent is greater than a preset exponent threshold, the defect identification data corresponding to the target power equipment is winding deformation. When the correlation dimension corresponds to a correlation dimension bifurcation amount greater than a preset bifurcation amount threshold and the maximum Lyapunov exponent corresponds to a mutation amount greater than a preset mutation amount threshold, the defect identification data corresponding to the target power equipment is insulation damage.

9. A power equipment vibration defect identification system, characterized in that, comprise: an adaptive vortex field regulation module configured to dynamically adjust an acoustic vortex field orbital angular momentum mode corresponding to the target power equipment according to a preset programmable metasurface array to obtain a resonance response signal of the target power equipment; a quantum precision measurement module configured to perform squeezed state laser irradiation and balanced homodyne detection on the resonance response signal to obtain a quantum correlation factor time series; a nonlinear dynamics feature extraction module configured to reconstruct a resonance response signal phase space based on the quantum correlation factor time series and calculate chaotic characteristic quantities to obtain a correlation dimension and a maximum Lyapunov exponent; a defect identification decision module configured to perform defect identification based on the correlation dimension, the maximum Lyapunov exponent, and a preset threshold to obtain defect identification data corresponding to the target power equipment.

10. An electronic device, comprising: comprise a memory and a processor, the memory storing a computer program, and the computer program being executed by the processor to cause the processor to perform the steps of the power equipment vibration defect identification method according to any one of claims 1-8.

Images