Thin-walled structure impact location method and system based on dense array and intelligent optimization strategy

Abstract

The application discloses a thin-wall structure impact positioning method and system based on a dense array and an intelligent optimization strategy, which comprises the following steps: acquiring an impact stress wave signal and extracting an impact stress narrowband wave signal; calculating the propagation group velocity and the wave number of the impact stress narrowband wave signal; setting initial parameters of an intelligent optimization algorithm; obtaining a fitness function of the intelligent optimization strategy according to a spatial beam focusing algorithm; formulating a population screening mechanism of the direction fitness function; constructing the population screening mechanism of the intelligent optimization strategy, and calculating individuals that are less than the half-power beam width in the angle with the optimal particle and the reference array element in the dense array and whose coordinates are within the monitoring area range at each iteration; iteratively searching the retained individuals that meet the conditions; and predicting the impact source coordinates by using a DBSCAN density clustering algorithm and a Gaussian kernel density estimation. The method and system disclosed by the application can realize accurate positioning at a low sampling rate, and there is no blind area.

Description

Technical Field

[0001] This invention belongs to the field of structural health monitoring, specifically relating to an impact location method and system for thin-walled structures based on dense arrays and intelligent optimization strategies. Background Technology

[0002] Aircraft structures inevitably suffer from various forms of external impact loads during service, such as bird strikes, hail, and runway debris. Repeated impacts can cause structural damage or even failure. Most aircraft structures are thin-walled, and the response signal generated by impact propagates primarily as an elastic wave, characterized by long propagation distance and low attenuation. Piezoelectric sensors, with their wide frequency response range and fast dynamic response speed, are commonly used to acquire impact response signals. With increasing industrial demands for lightweight sensor networks and precise lead wire alignment, as well as the limitations imposed by complex structures on sensor placement areas, dense sensor arrays have become a trend. Therefore, developing impact location algorithms based on dense sensor arrays is of great significance for impact monitoring research on aircraft structures.

[0003] Currently, impact location algorithms based on dense arrays still face several challenges, primarily including: First, the sampling rate of the data acquisition equipment is directly related to the resolution of the acquired signal. A lower sampling rate leads to poor signal resolution, resulting in weak spatial characteristic differences in the signals received by each element in the dense array. A single dense array can only estimate the relative direction of the impact source, making it difficult to estimate its accurate location, leading to blind spots in the monitoring area. Therefore, conventional dense array location methods typically require a high sampling rate (usually in MHz) for the data acquisition equipment, significantly increasing the operation and maintenance costs of the monitoring system. Balancing blind spots and system equipment costs to achieve comprehensive, blind-spot-free monitoring even at lower sampling rates is one of the problems to be solved in impact location technology. Second, classic dense array location algorithms typically map the phase, wavenumber, and other characteristic information of the impact signal from the spatial spectrum to all discrete spatial grids on the structure to assess the probability of an impact. For large real-world structures, this grid-by-grid scanning approach is extremely time-consuming, making it difficult to meet real-time online monitoring requirements. Furthermore, the accuracy of location and imaging depends on the number and size of the spatial grids. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of the aforementioned problems and to provide an improved method and system for impact positioning of thin-walled structures.

[0005] Therefore, some embodiments of this application provide a method for impact localization of thin-walled structures based on dense arrays and intelligent optimization strategies. The method includes the following steps: deploying at least two dense sensor arrays on the monitoring area of ​​the monitored structure to acquire impact stress wave signals from each sensor array; extracting narrowband impact stress wave signals at a specific center frequency using Morlet wavelet transform, Shannon wavelet transform, or short-time Fourier transform, and calculating the propagation group velocity and wave number of the narrowband impact stress wave signals; setting initial parameters for the intelligent optimization algorithm; obtaining the fitness function of the intelligent optimization strategy based on a spatial beam focusing algorithm; and constructing a directional fitness function, including calculating the phase difference between different array elements in the two dense arrays, constructing the direction vector of each array element, and deriving the dense array output power function P with the individual population position as the independent variable based on the spatial beam focusing algorithm. SBF ; and constructing the path fitness function, including calculating the time difference of arrival between array elements at corresponding positions in at least two dense arrays, and obtaining the imaging function P on the path according to the hyperbolic Gaussian distribution imaging algorithm. HL The following steps are taken: Develop a population selection mechanism for the stated orientation fitness function; derive the half-power beamwidth characteristics of the dense array based on its specific element distribution shape, element spacing, and element quantity. Construct a population selection mechanism using an intelligent optimization strategy, calculating individuals whose angle with the optimal particle and reference elements in the dense array is less than the half-power beamwidth and whose coordinates are within the monitoring area at each iteration. Retain these individuals and release the rest that do not meet the conditions. Iterate through the search for retained individuals that meet the conditions; use intelligent optimization algorithms to iteratively find the retained individuals with the largest fitness function value that meet the selection conditions; and use the DBSCAN density clustering algorithm and Gaussian kernel density estimation to predict the impact source coordinates.

[0006] In some embodiments, the selectable intelligent optimization algorithms include, but are not limited to, gray wolf, bat, and particle swarm optimization algorithms.

[0007] In some embodiments, methods such as Morlet wavelet transform, Shannon wavelet transform, or short-time Fourier transform are used to extract the impact stress narrowband wave signal at a specific center frequency, and the propagation group velocity and wave number of the impact stress narrowband wave signal are calculated.

[0008] In some embodiments, constructing the orientation fitness function includes calculating the phase difference between different elements in any two dense arrays, constructing the orientation vector of each element, deriving the dense array output power function with the individual position of the population as the independent variable based on the spatial beam focusing algorithm, and using it as the orientation fitness function of the optimization algorithm.

[0009] In some embodiments, constructing the path fitness function includes calculating the time difference of arrival between array elements at corresponding positions in at least two dense arrays, obtaining the imaging function on the path according to the hyperbolic Gaussian distribution imaging algorithm, and using it as the path fitness function.

[0010] In some embodiments, individuals that do not meet the conditions are released.

[0011] In some embodiments, the retained individuals obtained by searching the direction fitness function are defined as direction individuals, and the individuals obtained by searching the path fitness function are defined as path individuals.

[0012] In some embodiments, the DBSCAN density clustering algorithm is used to cluster individuals in all directions and along the path. The approximate location of the impact source can be determined by the distribution area of ​​all individuals retained by the clustering. Then, the probability density distribution of the individual position coordinates within the clustered area is calculated using the Gaussian kernel density estimation method. The maximum probability value corresponds to the coordinates of the predicted impact location.

[0013] This application also proposes a corresponding system.

[0014] This invention combines dense arrays to develop an online impact location recognition technology that achieves blind-zone-free, high-accuracy, and real-time identification at low sampling rates. The beneficial technical effects include:

[0015] First, the method and system proposed in this invention can achieve accurate positioning at low sampling rates and have no blind spots.

[0016] Secondly, this invention overcomes the shortcomings of traditional methods that require scanning all spatial grids one by one, greatly shortening the positioning time and improving the real-time performance of impact monitoring.

[0017] Third, this invention uses the beamwidth characteristics of the dense array as a constraint condition for the intelligent optimization algorithm to screen individuals, taking into account the ambiguity and uncertainty brought about by the beamwidth characteristics of the dense array itself to the positioning results.

[0018] Fourth, this invention designs a multi-channel synchronous data acquisition module, in which each channel of a single module acquires data synchronously, and each module is responsible for acquiring signals from a dense array, with no interference between modules. Compared to traditional data acquisition systems, this module is more miniaturized and lightweight. Furthermore, it eliminates the need for all data acquisition channels across the monitoring area to remain synchronized, improving the real-time performance of the method. Attached Figure Description

[0019] Figure 1A , Figure 1B and Figure 1CA flowchart of the thin-walled structure impact positioning method based on dense array and intelligent optimization strategy according to the present invention;

[0020] Figure 2 This is a flowchart of the DBSCAN density clustering algorithm in the method according to the present invention;

[0021] Figure 3 This is a schematic diagram of the monitored structure, the passive monitoring system, and the arrangement of its sensor clusters in an embodiment of the present invention;

[0022] Figure 4 The array pattern of a sensor cluster with M = 6 elements;

[0023] Figure 5 The half-power beamwidth of a sensor cluster with M = 6 array elements;

[0024] Figure 6 Time-domain signal diagrams of impact stress waves acquired by each sensor;

[0025] Figure 7 for Figure 6 The time-domain signal of the impact stress wave is obtained by Fourier transform, which is then used to obtain the frequency-domain signal of the impact stress wave.

[0026] Figure 8 Distribution diagrams of retained individuals in the 2nd, 3rd, 4th and 5th generations according to the method of the present invention.

[0027] Figure 9A A schematic diagram for all retained individuals.

[0028] Figure 9B This is a schematic diagram illustrating the process of clustering all retained individuals using the DBSCAN density clustering algorithm.

[0029] Figure 9C A schematic diagram showing the approximate locations of the regions where all individuals are preserved through clustering.

[0030] Figure 9D This is the probability density distribution of the impact location calculated using the Gaussian kernel density estimation method. Detailed Implementation

[0031] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0032] like Figure 1A , Figure 1B and Figure 1CAs shown, a thin-walled structure impact location method based on dense array and intelligent optimization strategy can be divided into five parts according to the functions implemented, mainly including: preprocessing of impact stress wave signal and algorithm parameters, obtaining the fitness function of intelligent optimization strategy according to spatial beam focusing algorithm, formulating population screening mechanism for directional fitness function, iteratively searching for retained individuals that meet the conditions, and predicting the coordinates of impact source using density clustering algorithm and Gaussian kernel density estimation.

[0033] Each part shall be implemented in accordance with the following steps:

[0034] Step S1: Acquisition of impact stress wave signal and preprocessing of algorithm parameters.

[0035] N dense sensor arrays (N≥2) are deployed on the monitoring area of ​​the structure being monitored, and the impact stress wave signals of each array element are acquired using data acquisition equipment. The shape of the dense sensor array can be circular, rectangular, etc., and the number can be customized according to requirements. The array element PZT in the nth dense sensor array is used to receive the impact stress wave signal excited by the impact source. i n The position coordinates are as follows Where I is the total number of array elements in the nth dense sensor array. The radius of the circular array is denoted as R. s The nth dense sensor array element PZT i n The received shock stress wave signal was recorded as follows Based on the frequency distribution information of the impact stress wave signal collected by each array element, a specific center frequency f is extracted. c Narrowband wave signal

[0036] The specific center frequency can be determined through the following steps: First, perform a Fourier transform to obtain the signal's spectrum; the extracted center frequency needs to be within the spectral range. For continuous wavelets, higher frequency signals have higher time resolution; for Lamb wave signals, higher frequency signals have more complex mode structures. The specific center frequency can be selected based on a comprehensive consideration of different scenarios.

[0037] Then, the group velocity v of the shock stress wave propagating at this specific frequency is calculated using methods such as Morlet wavelet transform, Shannon wavelet transform, or short-time Fourier transform. g In the imaging, the wave number k and wavelength λ of the shock stress wave are calculated based on formula (1).

[0038]

[0039] Step S2, Intelligent Algorithm Initialization. The intelligent optimization algorithm in the method of this invention can use optimization algorithms such as gray wolf, bat, and particle swarm optimization. Here, the gray wolf optimization algorithm is used as an example. Initialize the parameters set based on the principle of the gray wolf optimization algorithm: g is used to control the maximum number of generations g before the algorithm terminates. m Initialize the population size M, individual orientation control parameter A, and individual distance control parameter C. The two-dimensional spatial vector δ∈(x,y) is used to locate the impact zone of a large thin-walled structure. Formula (2) is used to determine the wolf pack p... g Initialization, when the monitoring area is in x∈[x] min ,x max ]mm、y∈[y min ,y max When mm, the upper and lower limits of the spatial vector of the monitoring area are δ respectively. U =(x max ,y max ) and δ L =(x min ,y min ). ζ r This represents the spatial resolution of individuals during the search process, typically taken as ζ. r = (1,1). At this point, the population is initialized with a 1mm spatial interval x∈[x min ,x max ]mm、y∈[y min ,y max M individuals are randomly distributed within mm. It should be understood that although this step is labeled S2 for ease of description, the intelligent algorithm initialization only needs to be performed before the actual execution of the intelligent algorithm.

[0040]

[0041] Step S3: Obtain the fitness function of the intelligent optimization strategy based on the spatial beam focusing algorithm.

[0042] Based on the individual locations distributed within the monitoring area (x pg ,y pg Based on formula (3), the relative distance r between the individual and each element in the nth dense array is calculated. i (x pg ,y pg ).

[0043]

[0044] The turning vector of each element in the nth dense array can be calculated by formula (4), r i It is the relative distance of an individual from each element in the nth dense array. iIt is the turning vector of the array elements, referring to the direction from the center of the array to the signal source. It is usually represented by spatial coordinates and reflects the angle or direction when the signal arrives at the array.

[0045]

[0046] The signal received by each element in the nth dense array can be represented in the following form:

[0047] F n (t)=A n (x pg ,y pg V n (t)+N n (t) (5)

[0048] in,

[0049]

[0050] The output power of the nth dense array is constructed based on a space beam focusing algorithm based on beamforming theory. Use it as the directional fitness function of the optimization algorithm:

[0051]

[0052] The normalized root mean square magnitude method described in formulas (8) and (9) is used to calculate the PZT of the array element in the nth dense array. i n and the array element PZT in the m-th dense array i m The normalized root mean square sequence.

[0053]

[0054] Where V represents the length of the impact stress wave signal, S(t) w ) represents the signal at t w The signal amplitude at time , Indicates 0 <t<t v The average amplitude of the signal over a given time period.

[0055]

[0056] Then, select an appropriate normalized root mean square threshold κ, and measure the PZT of the array elements in the nth dense array. i n and the array element PZT in the m-th dense array i m Time difference of arrival Where m, n = 1, 2, ..., N.

[0057] Based on the group velocity of the wave, the array element PZT in the nth dense array i n and the array element PZT in the m-th dense array i m The coordinates are used to calculate the theoretical time difference of arrival.

[0058]

[0059] Finally, the imaging function along this path is obtained using the hyperbolic Gaussian distribution imaging algorithm, or hyperbolic imaging algorithm for short. As a path fitness function:

[0060]

[0061] Step S4: Establish a population selection mechanism based on the orientation fitness function.

[0062] Based on formulas (12) and (13), the pattern function E of a uniform circular array containing 8 elements is derived. n (φ):

[0063]

[0064] Extract the angle value φ corresponding to the 3dB beamwidth. 3dB Define individual constraint parameters

[0065]

[0066] Individuals are selected in each iteration to satisfy the retention constraint. Individuals that meet the criteria are retained, while those that do not are removed.

[0067] Step S5: Iteratively search for individuals that meet the conditions for retention, including:

[0068] Step S51: Based on the orientation fitness function obtained in the above steps, calculate the orientation fitness of all gray wolf individuals in the current dense array, save the top three wolves with the highest fitness (α, β, and δ), and update the current gray wolf individual positions. Select the gray wolf individuals to retain based on their orientation. Update A and C, and calculate the orientation fitness value and position of the retained gray wolf individuals. Continue until the number of iterations reaches g. m Furthermore, the search yields the set of orientation-preserving individuals obtained under all dense arrays.

[0069] Step S52: Based on the path fitness function obtained in the previous steps, calculate the path fitness of all gray wolf individuals in the current dense array, save the top three wolves with the highest fitness (α, β, and δ), and update the current position of the gray wolf individuals. Update A and C, and calculate and save the orientation fitness value and position of the gray wolf individuals. Continue until the number of iterations reaches g. m Furthermore, the search yields a path-preserving set of all dense array pairs.

[0070] Step S6: Predict the coordinates of the impact source using density clustering algorithm and Gaussian kernel density estimation.

[0071] Including through fusion and Obtain all retained individuals within the monitoring area, and employ methods such as... Figure 2 The DBSCAN density clustering algorithm shown clusters individuals from all directions and along the path. By retaining the distribution area of ​​all individuals through clustering, the approximate location of the impact source can be determined (step S61). Then, using methods such as... Figure 2 The Gaussian kernel density estimation method shown calculates the probability density distribution of individuals within the cluster region. The location with the highest probability is the predicted impact location, step S62.

[0072] In another embodiment, this application also provides a thin-walled structure impact positioning system based on dense arrays and intelligent optimization strategies. The system includes at least two dense sensor arrays deployed in the monitoring area of ​​the monitored structure to acquire impact stress wave signals from each array element; a multi-channel data synchronous acquisition module to acquire the impact stress wave signals; and a memory and a processor. The memory stores computer instructions, and the processor executes the following method steps after executing the computer instructions: extracting impact stress narrowband wave signals from the impact stress wave signals; algorithm parameter preprocessing, including calculating the propagation group velocity and wavenumber of the impact stress narrowband wave signals; setting initial parameters for the intelligent optimization algorithm; and obtaining the fitness function of the intelligent optimization strategy based on a spatial beam focusing algorithm. This includes constructing a directional fitness function and a path fitness function; establishing a population selection mechanism for the directional fitness function; deriving the half-power beamwidth characteristics of the dense array based on the element distribution shape, element spacing, and number of elements of each dense array; constructing a population selection mechanism for an intelligent optimization strategy, calculating individuals whose angle with the optimal particle and the reference element in the dense array is less than the half-power beamwidth and whose coordinates are within the monitoring area at each iteration, and retaining the calculated individuals; iteratively searching for the retained individuals that meet the conditions, including using an intelligent optimization algorithm to iteratively find the retained individuals with the largest fitness function value that meet the selection conditions; and using the DBSCAN density clustering algorithm and Gaussian kernel density estimation to predict the coordinates of the impact source.

[0073] like Figure 3 The diagram shows a specific embodiment of the method and system of this application, illustrating the monitored thin-walled plate, sensor cluster, and multi-channel data synchronous acquisition module. The monitored structure measures 900mm × 1250mm × 3mm. A sensor cluster with a radius of 13mm is composed of piezoelectric sensors with a diameter of 8mm and a thickness of 0.5mm. This sensor cluster refers to the aforementioned dense sensor array, such as sensor cluster 1 and sensor cluster 2.

[0074] like Figure 4 The image shows the array pattern when the sensor cluster 1 is pointed at 90°. Further, it can be determined based on... Figure 5 The half-power beamwidth was calculated to be 58°. A Cartesian coordinate system was established with the lower left corner of the structure under test as the origin. The coordinates of the piezoelectric sensor and the impact source were recorded. An impact event was triggered using a drop hammer device at the coordinates (382mm, 830mm) on the structure. Each dense sensor array was connected to a multi-channel data synchronous acquisition module, for example... Figure 3 The sampling rate of the system in modules 1 and 2 shown is set to 25kHz. The multi-channel data synchronous acquisition module can also work with existing data storage modules, power management modules, and signal conditioning modules to acquire the required signals.

[0075] like Figure 6 The figure shows a time-domain plot of a typical impact stress wave signal received by sensor 1. Fourier transform is then used to obtain... Figure 7 The spectrum diagram shown indicates that the main frequency components of the signal are concentrated below 3kHz. Following step S1, Morlet wavelet transform is used to extract the impact stress narrowband wave signal with a center frequency of 1kHz, and the group velocity is calculated to be 113m / s and the wave number to be 8.813m. -1 The wavelength is 0.113m. The intelligent optimization algorithm is initialized, with a maximum number of generations g. m =5. Initialize the population size M = 3000. Following steps S2 and S3, obtain the fitness function of the intelligent optimization strategy based on the spatial beam focusing algorithm, and formulate a population selection mechanism based on the directional fitness function.

[0076] like Figure 8 As shown, the set of individuals with preserved orientation is obtained according to step S4. and path-preserving individual set According to step S5, the following is obtained: Figure 9A All the retained individuals shown were clustered using the DBSCAN density clustering algorithm for individuals in all directions and along paths, as follows: Figure 9B As shown, the approximate location of the impact source can be determined by the distribution area of ​​all individuals preserved through clustering. Figure 9CAs shown, the probability density distribution of individuals within the cluster region is finally calculated using the Gaussian kernel density estimation method, as follows: Figure 9D As shown, the predicted impact location coordinates are (386mm, 833mm).

[0077] It should be understood that the processor in the embodiments of this application can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0078] It should also be understood that the memory in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), which serves as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous interconnected dynamic random access memory (SLDRAM), and direct memory bus random access memory (DR RAM).

[0079] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0080] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0081] In this application, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or multiple items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0082] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0083] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0084] Those skilled in the art will 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.

[0085] 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; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0086] 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.

[0087] In addition, the functional units in the various embodiments of this application 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.

[0088] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion 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 described in the various embodiments of this application. 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.

[0089] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application. It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

Claims

1. A method for impact localization of thin-walled structures based on dense arrays and intelligent optimization strategies, characterized in that: The method includes: At least two dense sensor arrays are deployed in the monitoring area of ​​the monitored structure to obtain the impact stress wave signal of each array element and extract the impact stress narrowband wave signal. Algorithm parameter preprocessing includes calculating the propagation group velocity and wave number of the impact stress narrowband wave signal; Set the initial parameters for the intelligent optimization algorithm; The fitness function of the intelligent optimization strategy is obtained based on the spatial beam focusing algorithm, including the construction of the orientation fitness function and the construction of the path fitness function; A population selection mechanism based on the orientation fitness function is established, including deriving the half-power beamwidth characteristics of the dense array based on the element distribution shape, element spacing, and number of elements of each dense array; a population selection mechanism based on an intelligent optimization strategy is constructed, calculating individuals whose angle with the optimal particle and the reference element in the dense array is less than the half-power beamwidth and whose coordinates are within the monitoring area at each iteration, and retaining the calculated individuals; Iteratively searching for the individuals to be retained that meet the conditions includes using an intelligent optimization algorithm to iteratively find the individuals with the largest fitness function value that meet the selection conditions. The DBSCAN density clustering algorithm and Gaussian kernel density estimation are used to predict the coordinates of the impact source.

2. The thin-walled structure impact positioning method based on dense array and intelligent optimization strategy according to claim 1, characterized in that: The intelligent optimization algorithms include, but are not limited to, gray wolf, bat, and particle swarm optimization algorithms.

3. The thin-walled structure impact positioning method based on dense array and intelligent optimization strategy according to claim 1, characterized in that: The impact stress narrowband wave signal at a specific center frequency is extracted using Morlet wavelet transform, Shannon wavelet transform, or short-time Fourier transform, and the propagation group velocity and wave number of the impact stress narrowband wave signal are calculated.

4. The thin-walled structure impact positioning method based on dense array and intelligent optimization strategy according to claim 1, characterized in that: The construction of the orientation fitness function includes calculating the phase difference between different elements in any two dense arrays, constructing the orientation vector of each element, deriving the dense array output power function with the individual position of the population as the independent variable based on the spatial beam focusing algorithm, and using it as the orientation fitness function of the optimization algorithm.

5. The thin-walled structure impact positioning method based on dense array and intelligent optimization strategy according to claim 1, characterized in that: The construction of the path fitness function includes calculating the time difference of arrival between array elements at corresponding positions in at least two dense arrays, obtaining the imaging function on the path according to the hyperbolic Gaussian distribution imaging algorithm, and using it as the path fitness function.

6. The thin-walled structure impact positioning method based on dense array and intelligent optimization strategy according to claim 1, characterized in that: Individuals that do not meet the conditions will be released.

7. The thin-walled structure impact positioning method based on dense array and intelligent optimization strategy according to claim 1, characterized in that: The individuals retained by searching the directional fitness function are defined as directional individuals, and the individuals obtained by searching the path fitness function are defined as path individuals.

8. The thin-walled structure impact positioning method based on dense array and intelligent optimization strategy according to claim 1, characterized in that: The DBSCAN density clustering algorithm is used to cluster individuals in all directions and along the path. The approximate location of the impact source is determined by the distribution area of ​​all individuals retained by the clustering. Then, the probability density distribution of the individual position coordinates within the clustering area is calculated using the Gaussian kernel density estimation method. The coordinates of the predicted impact location are the maximum probability values.

9. A thin-walled structure impact positioning system based on dense array and intelligent optimization strategy, characterized in that: This includes at least two dense sensor arrays deployed in the monitoring area of ​​the structure being monitored to obtain the impact stress wave signal of each array element; A multi-channel data synchronous acquisition module acquires the impact stress wave signal; It also includes a memory and a processor, wherein the memory stores computer instructions, and the processor executes the computer instructions to perform the following method steps: Extract the narrowband wave signal of impact stress from the impact stress wave signal; Algorithm parameter preprocessing includes calculating the propagation group velocity and wave number of the impact stress narrowband wave signal; Set the initial parameters for the intelligent optimization algorithm; The fitness function of the intelligent optimization strategy is obtained based on the spatial beam focusing algorithm, including the construction of the orientation fitness function and the construction of the path fitness function; A population selection mechanism based on the orientation fitness function is established, including deriving the half-power beamwidth characteristics of the dense array based on the element distribution shape, element spacing, and number of elements of each dense array; a population selection mechanism based on an intelligent optimization strategy is constructed, calculating individuals whose angle with the optimal particle and the reference element in the dense array is less than the half-power beamwidth and whose coordinates are within the monitoring area at each iteration, and retaining the calculated individuals; Iteratively searching for the individuals to be retained that meet the criteria includes using an intelligent optimization algorithm to iteratively find the individuals with the largest fitness function value that also meet the selection criteria; and The DBSCAN density clustering algorithm and Gaussian kernel density estimation are used to predict the coordinates of the impact source.

Images