Damping ratio identification method, device and storage medium based on topological signal processing

Abstract

The present invention discloses a damping ratio identification method, device and storage medium based on topological signal processing, the method comprising: obtaining a detection signal collected by a sensor set on a target structure; extracting each order response of the target structure by using an automatic frequency domain decomposition method; performing autocorrelation processing on each order response to obtain each corresponding free decay curve; setting an adjustment horizontal line, using the adjustment horizontal line to process the extreme points on each free decay curve to obtain a persistence pair and a persistence graph; calculating the persistence length corresponding to the persistence pair and the cutoff equation of the persistence graph, using the cutoff equation and the corresponding persistence length to filter the signal persistence pair in the persistence pair; after eliminating the signal persistence pair corresponding to the maximum persistence length, using the remaining persistence length to calculate the damping ratio of the free decay curve and the damping ratio corresponding to the remaining free decay curves. The limitations of traditional damping ratio identification technology are overcome and the damping ratio identification accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of structure monitoring, and in particular to a damping ratio identification method, device and storage medium based on topological signal processing. Background Art

[0002] Damping refers to the phenomenon caused by the dissipation of mechanical energy in a dynamic system. The most obvious manifestation is the attenuation of the amplitude during free vibration. For structures such as long-span bridges and super-high-rise buildings, the damping ratio can play a role in reducing vibration and noise. However, the size of the damping ratio is affected by many factors, and changes in the damping ratio may mean that the structure will suffer a vibration disaster.

[0003] Therefore, establishing an effective bridge structure damping ratio monitoring system and improving the accuracy of damping ratio identification are of great significance for the design, construction and operation of structures such as long-span bridges and super-high-rise buildings.

[0004] However, in the prior art, the damping ratio identification algorithm has the following problems:

[0005] (1) Inaccurate system model assumptions: Identification algorithms are usually based on some simplified system models, while actual systems are often more complex. If the system model assumptions are inaccurate, the identification results of the damping ratio will have large errors.

[0006] (2) Measurement noise interference: The identification algorithm is usually based on the measured response data. However, there is noise interference in the actual measurement, such as the error of the measuring instrument or the environmental noise. These noises will affect the accuracy of the measurement data, resulting in inaccurate identification results of the damping ratio.

[0007] Therefore, it is urgent to propose a damping ratio identification algorithm that can solve the above problems. Summary of the invention

[0008] The purpose of the present invention is to provide a damping ratio identification method, device and storage medium based on topological signal processing to solve one or more technical problems existing in the prior art and at least provide a beneficial choice or create conditions.

[0009] The solution of the present invention to solve the technical problem is to provide a damping ratio identification method, device and storage medium based on topological signal processing.

[0010] According to an embodiment of the first aspect of the present invention, a damping ratio identification method based on topological signal processing is provided, comprising:

[0011] S100, determining a target structure, acquiring a detection signal collected by a sensor disposed on the target structure, wherein the detection signal includes a relationship between a detection signal value and a time variation;

[0012] S200, performing low-pass filtering on the detection signal, and extracting each order response of the target structure by using an automatic frequency domain decomposition method;

[0013] S300, performing autocorrelation processing on the responses of each order to obtain corresponding free decay curves, and selecting a free decay curve from the free decay curves;

[0014] S400, setting an adjustment horizontal line, using the adjustment horizontal line to process the extreme value points on the free attenuation curve, forming Group persistence pairs and obtain persistence graphs;

[0015] S500, calculating the The duration length corresponding to the group duration pair and the cutoff equation of the duration graph are used to screen and retain the Signal continuation pairs in group continuation pairs;

[0016] S600, after eliminating the signal duration pairs corresponding to the duration length of the maximum value, the damping ratio corresponding to the free decay curve is calculated using the remaining duration length;

[0017] S700 , repeating S400 to S600 to calculate the damping ratios corresponding to the remaining free decay curves.

[0018] Furthermore, the extraction process of each order response of the target structure specifically includes:

[0019] According to the detection signal, the cross power spectrum density of the detection signal response is calculated to obtain a three-dimensional matrix and frequency series ;

[0020] Decompose the three-dimensional matrix using SVD , and we get the left singular vector , right singular vectors and singular values ;

[0021] Smoothing the frequency sequence Then, obtain the local extreme point as the frequency peak , according to the smoothed frequency series , frequency peak and singular values , and the various order responses of the target structure are calculated.

[0022] Furthermore, the process of obtaining the persistence graph specifically includes:

[0023] Setting the adjustment level ,make ,in, is a constant;

[0024] make Continuously increasing, with the free decay curve The local minimum on Intersection, adjust the horizontal line With free decay curve Constitute the intersection area ;

[0025] When the local minimum The corresponding intersection area Will be incorporated The corresponding intersection area When , the local maximum Paired local minima , forming a continuous pair ,in, , ;

[0026] Free decay curve The local maximum on and local minima Pair them one by one to form Group continuous pairs and obtain continuous graph .

[0027] Further, the calculation The persistence length corresponding to the group persistence pair and the cutoff equation of the persistence graph specifically include:

[0028] The cutoff equation of the persistence diagram is:

[0029]

[0030] ,in, is the distribution parameter of the noise, is the probability constant;

[0031] According to the calculation formula of continuous length:

[0032]

[0033] ,calculate The duration length corresponding to the group duration pair .

[0034] Furthermore, the process of continuously screening and retaining the signal specifically includes:

[0035] The corresponding duration With cutoff equation 's values ​​for comparison;

[0036] If there is a persistence length Greater than cutoff equation The value of the duration The corresponding continuous pair is a signal continuous pair and is retained.

[0037] Furthermore, the damping ratio calculation process of the free decay curve specifically includes:

[0038] After removing the signal duration pair corresponding to the maximum duration length, obtain the maximum duration length of the remaining signal duration pairs. 0.3299 times the persistence length ;

[0039] Get the duration The corresponding cycle , using the damping ratio calculation formula:

[0040]

[0041] ,in, is the optimal cycle.

[0042] Further, the smoothed frequency sequence , frequency peak and singular values , the calculated responses of the target structure are as follows:

[0043] According to the peak frequency , set the low frequency value and high frequency values ;

[0044] Get the low frequency value and high frequency values The corresponding horizontal axis in the spectrum diagram and ,according to , and singular values Perform calculations,

[0045]

[0046]

[0047] ,in, for and The median of the difference, express to The number of frequencies between

[0048] calculate , and obtain the various order responses of the target structure.

[0049] Furthermore, the process of obtaining each free attenuation curve specifically includes:

[0050] The inverse Fourier transform of the cross power spectrum density is calculated by Fourier transform to obtain the cross correlation function, thereby obtaining the respective free attenuation curves.

[0051] According to an embodiment of the second aspect of the present invention, there is provided an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a damping ratio identification method based on topological signal processing as an embodiment of the first aspect of the present invention.

[0052] According to an embodiment of the third aspect of the present invention, a storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, a damping ratio identification method based on topological signal processing as in an embodiment of the first aspect of the present invention is implemented.

[0053] The beneficial effects of the present invention are: obtaining the detection signal collected by the sensor set on the target structure; calculating the impulse response function and autocorrelation function of the detection signal response, obtaining the free decay curve of each order, and being able to provide a more robust free decay peak point; constructing a persistence pair using the topological signal processing method, and filtering out the true topological characteristics of the normalized detection signal from the measurement noise; calculating the optimal number of cycles, and calculating the damping ratio based on the remaining signal persistence after deleting the maximum value in the signal persistence pair. The present invention has the advantages of high computational efficiency, few parameters that need to be manually determined, and strong robustness. It can overcome the various limitations of traditional damping ratio identification technology and achieve better damping ratio identification accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a schematic flow chart of a damping ratio identification method based on topological signal processing provided by an embodiment of the present invention;

[0055] Figure 2 It is a schematic diagram of processing the extreme value point on the free decay curve by adjusting the horizontal line provided by an embodiment of the present invention;

[0056] Figure 3 is a schematic diagram of a three-degree-of-freedom mass-spring linear structure provided by an embodiment of the present invention;

[0057] Figure 4 is a persistence diagram of an acceleration autocorrelation function provided by an embodiment of the present invention;

[0058] Figure 5 It is a timing diagram of the acceleration autocorrelation function provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and cannot be understood as limiting the present invention.

[0060] It should be noted that although the functional modules are divided in the system diagram, in some cases, the steps shown or described may be performed in a different order than the module division in the system or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0061] In the description of the present invention, it should be noted that, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in the present invention in combination with the specific content of the technical solution.

[0062] Reference Figure 1 and Figure 2 According to an embodiment of the first aspect of the present invention, a damping ratio identification method based on topological signal processing comprises the following steps:

[0063] S100, determining a target structure, acquiring a detection signal collected by a sensor disposed on the target structure, wherein the detection signal includes a relationship between a detection signal value and a time variation.

[0064] In this embodiment, the structures to be analyzed are determined, and the target structures may include bridges, houses, large machinery, etc. A collection sensor is set on the target structure to obtain a detection signal, and the detection signal includes a corresponding relationship between the detection signal and changes over time.

[0065] Typically, an acceleration sensor is installed on the target structure to collect the acceleration time series to be analyzed.

[0066] S200, low-pass filtering is performed on the detection signal, and the various-order responses of the target structure are extracted using an automatic frequency domain decomposition method.

[0067] In this embodiment, the detection signal obtained in S100 is low-pass filtered to eliminate the influence of high-frequency noise, and the automatic frequency domain decomposition method is used to extract each order response, and the extracted each order response is regarded as a single degree of freedom system.

[0068] S300, performing autocorrelation processing on each order response to obtain corresponding free decay curves, and selecting a free decay curve from the free decay curves.

[0069] In this embodiment, autocorrelation processing is performed on each order response in turn, that is, the inverse Fourier transform of the cross power spectrum density is calculated using fast Fourier transform, thereby obtaining a cross correlation function, and correspondingly obtaining each free attenuation curve corresponding to each order response.

[0070] Select the free decay curve. The free decay curve can be selected arbitrarily or from large to small or from small to large in order. For example: The response of each order of the target structure is , then the free decay curve is , the free decay curve can be selected as Processing, or select the free decay curve as Processing, or select the free decay curve as to be processed.

[0071] S400, set the adjustment level line, use the adjustment level line to process the extreme points on the free decay curve, and form Group persistence pairs and obtain persistence graphs.

[0072] In this embodiment, a horizontal line is set, and the horizontal line is adjustable. The horizontal line is adjusted to process all extreme points on the selected free decay curve, and a continuous pair is constructed according to the extreme points. There are several extreme points on the free decay curve, so as to obtain Group continuous pair, according to Group persistence pairs and design a persistence diagram, which includes The group of continuous pairs is the set of all continuous pairs.

[0073] S500, calculation The duration length corresponding to the group duration pair and the cutoff equation of the duration graph are used to screen and retain the corresponding duration lengths. Signal continuation pair in group continuation pair.

[0074] In this embodiment, using the data obtained in S400 The group duration is calculated to obtain the corresponding duration length, that is, Calculate the cutoff equation of the persistence diagram obtained in S400, and use the cutoff equation to The duration length is used to filter and distinguish, from Several groups of noise continuous pairs and several groups of signal continuous pairs are separated from the group of continuous pairs, and several groups of signal continuous pairs are retained. Separation of continuous pairs Group noise continuous pair and The group signal continues to

[0075] Among them, several groups of noise persistence pairs can be eliminated or stored in a memory for subsequent processing.

[0076] S600, after eliminating the signal duration pairs corresponding to the duration length of the maximum value, the damping ratio of the free decay curve is calculated using the remaining signal duration pairs;

[0077] In this embodiment, the maximum value of the duration length corresponding to the signal duration pair is found, the signal duration pair corresponding to the duration length is eliminated, and the damping ratio of the selected free decay curve is obtained by calculating the duration length corresponding to the remaining signal duration pairs. That is, the signal duration corresponding to the maximum value of the duration length is eliminated, and the damping ratio of the selected free decay curve is obtained. The group signal continues to duration, using The group signal continues to The damping ratio of the selected free decay curve is calculated based on the persistence length.

[0078] S700 , repeating S400 to S600 to calculate the damping ratios corresponding to the remaining free decay curves.

[0079] In this embodiment, S400 to S600 are looped until the damping ratios corresponding to the remaining free decay curves are calculated. For example, the free decay curve calculated through S400 to S600 is: The corresponding damping ratio , and the free decay curve is calculated again from S400 to S600: The corresponding damping ratio .

[0080] Through S100 to S300, the autocorrelation function and the impulse response function of the detection signal response collected by the sensor are calculated to obtain each order response and the subsequent free decay curve, thereby reducing the parameters of the target structure that need to be manually determined, and solving the problem of inaccurate damping ratio identification of the single degree of freedom system model;

[0081] Then, through S400 to S500, the topological signal processing method is used to construct The real topological features of the normalized detection signal are separated from the measurement noise by forming a continuous pair, thereby solving the problem of measurement noise interference such as measurement instrument error or environmental noise and reducing recognition errors;

[0082] Through S600 to S700, the optimal number of cycles is calculated according to the error propagation analysis theory, and the damping ratio of each order response is calculated according to the residual signal persistence pair after elimination, so as to achieve better damping ratio identification accuracy and improve calculation efficiency.

[0083] In some embodiments of the present invention, in S200, extracting each order response of the target structure using the automatic frequency domain decomposition method specifically includes the following steps:

[0084] S210, using the detection signal in S100, calculating the cross power spectrum density of the detection signal response, obtaining a three-dimensional matrix And the frequency sequence .

[0085] In this embodiment, initialization The zero matrix of and , the cross power spectral density of the detection signal response is calculated using the cpsd frequency response function in MATLAB, and the initialized zero matrix is ​​used to store the three-dimensional matrix calculated by the cpsd frequency response function The cross power spectrum density of the detection signal response is calculated by the cpsd frequency response function in MATLAB, and the frequency sequence is obtained. .

[0086] S220, decompose the three-dimensional matrix obtained in S210 using SVD , get the left singular vector , right singular vectors And the singular values ;

[0087] In this embodiment, in order to simplify the data and remove noise and redundant data, the three-dimensional matrix in S210 is Perform singular value decomposition (SVD decomposition) and store the decomposed left singular vectors, right singular vectors and singular values ​​in matrices respectively , as well as Singular value decomposition is also a dimensionality reduction method that can convert a three-dimensional matrix Mapped to two-dimensional space, reducing the amount of later calculations. Then the left singular vector Can be used for row number compression, right singular vector It can be used to compress the number of columns, i.e. the feature dimension.

[0088] S230, for the frequency sequence in S210 Perform smoothing to obtain the smoothed frequency sequence The local extreme point in the , according to the peak frequency , singular values And the smoothed frequency series Perform calculations to obtain responses of various orders.

[0089] In this embodiment, the frequency sequence in S210 is smoothed using Perform smoothing and find local extreme points as frequency peaks , where in the frequency sequence There are several local extreme points in the frequency peak There are also several. According to several frequency peaks , singular value matrix And the smoothed frequency series Calculations are performed to obtain the various order responses of the target structure, and each order response obtained is regarded as a single degree of freedom system.

[0090] Through S210 to S230, the detection collected by the sensor is processed by pulse frequency response function, the various order responses of the target structure are extracted, the parameters of the target structure that need to be manually determined are reduced, the influence of high-frequency noise is eliminated, SVD decomposition and local extreme points are used to remove noise and redundant data, and the problem of inaccurate damping ratio identification of the single-degree-of-freedom system model is solved.

[0091] In some embodiments of the present invention, in S230, the process of obtaining each order response specifically includes the following steps:

[0092] S231, for frequency peak , define the low frequency value and high frequency values .

[0093] S232, get low frequency value and high frequency values The corresponding horizontal axes in the spectrum diagram are set as and ,according to , And the singular values , calculated

[0094]

[0095]

[0096] ,in, for and The median of the differences, express to The number of frequencies between

[0097] S233, calculation , and obtain the various order responses of the target structure.

[0098] In this embodiment, in the frequency sequence There are several local extreme points in the frequency peak There are also several, according to the frequency sequence The spectrum diagram is available.

[0099] Using frequency peaks Set the low frequency value accordingly and high frequency values , , .

[0100] according to Get the corresponding horizontal coordinate in the spectrum diagram and according to Get the corresponding horizontal coordinate in the spectrum diagram ,use , And the singular values ​​obtained in S220 , calculated

[0101]

[0102]

[0103] , according to several frequency peaks The corresponding low frequency value and high frequency values Calculate the difference between two adjacent items, get several corresponding differences, and select the median from the several differences , and using the singular values calculate ;

[0104]

[0105] ,in, express to The number of frequencies between .

[0106] calculate , corresponding to the various order responses of the target structure , where the calculated and Both can be sets.

[0107] Reference Figure 2 In some embodiments of the present invention, in S400, The process of obtaining the group persistence pairs and persistence graphs specifically includes the following steps:

[0108] S410, set an adjustment horizontal line ,make ,in, is a constant.

[0109] In this embodiment, let the adjustment level be , . Represented as a free decay curve The minimum value in is expressed as a very small constant. That is, the initial adjustment level Specific free decay curve The minimum value in , there is no free decay curve There are intersection points.

[0110] S420, let Continue to increase and adjust the horizontal line Free decay curve The local minimum on Intersect, thereby enclosing the corresponding intersection area .

[0111] In this embodiment, adjusting ,let Continue to increase, so that the horizontal line can be adjusted Towards The positive half axis moves in During the enlargement process, adjust the horizontal line Free decay curve The local minima on intersect, and the local minima are defined as , adjust the horizontal line With free decay curve The enclosed area is defined as the intersection area .

[0112] S430, when , ,exist During the growth process, Will be incorporated ,form The corresponding new intersection area ,when Will be incorporated hour, , then the local maximum and local minimum Pair up to form a continuous pair .

[0113] In this embodiment, During the enlargement process, adjust the horizontal line Free decay curve The local maxima on intersect. And the larger local minimum The corresponding intersection area will merge into smaller local minima The corresponding intersection area , that is, the intersection area Will intersect area Connected, intersecting areas disappear, forming a larger new intersection area .when Will be incorporated , the intersection area When it disappears, adjust the horizontal line to ,but is a local minimum Paired local maxima, local maxima and local minimum Pair up to form a continuous pair .

[0114] For example: Known , then the corresponding intersection area ,when During the growth process, the intersection area Intersecting Area Connected, intersecting areas Disappear and merge into the intersection area , forming a new intersection area , then adjust the horizontal line to ,Will and Pair up to form a continuous pair , the remaining corresponding intersection area ,along with Increase, intersection area Intersecting Area Connected, intersecting areas Disappear and merge into the intersection area , forming a new intersection area , then adjust the horizontal line to ,Will and Pair up to form a continuous pair , the remaining corresponding intersection area .

[0115] S440, using S430, for local maximum and local minima Pair them one by one to form Group continuous pair, according to Group persistence pairs to get persistence graph .

[0116] In this embodiment, as Increase, adjust the horizontal line Towards The positive half axis of the axis moves, using the pairing rule of S430 to make the free decay curve The local maximum on One by one and local minimum Pairing, thus forming Group continuous pair, according to The continuous design of the group obtains the continuous graph , that is, get the set of all continuous pairs.

[0117] Through S410 to S440, the adjustment horizontal line is set, and the topological signal processing method is used to construct The real topological features of the normalized detection signal are separated from the measurement noise by a continuous pairing, thereby solving the problem of measurement noise interference such as measurement instrument error or environmental noise and reducing recognition errors.

[0118] In some embodiments of the present invention, S500 specifically includes the following steps:

[0119] S510, calculate the cutoff equation of the persistence graph:

[0120]

[0121] ,in, is the distribution parameter of the noise, is a probability constant.

[0122] In this embodiment, is a probability constant, where Take 0.1%, is the number of continuous pairs.

[0123] S520, according to the calculation formula of the duration:

[0124]

[0125] ,calculate The duration length corresponding to the group duration pair .

[0126] In this embodiment, the duration length corresponding to each group of duration pairs is calculated. ,get Duration .

[0127] S530, compared with S520 Duration With cutoff equation The value of .

[0128] S540, greater than the cutoff equation The duration of the value The corresponding continuous pair is the signal continuous pair, and the signal continuous pair is retained.

[0129] In this embodiment, the S520 obtained Duration and calculate the cutoff equation Determine if it is greater than the cutoff equation The value of the duration , then the duration The corresponding continuous pair is the signal continuous pair. Keep the signal continuous pair and get Group signal duration pairs and their corresponding duration lengths.

[0130] Where, less than or equal to the cutoff equation The value of the duration , the duration can be The corresponding continuous pair is considered as the noise continuous pair, and we get Group noise duration pairs and their corresponding duration lengths;

[0131] Through S510 to S540, the persistence length of each group of persistence pairs is calculated, and the cutoff equation of the persistence graph is calculated. The noise persistence pairs are screened out through the cutoff equation, and the signal persistence pairs are retained. The true topological characteristics of the normalized detection signal are separated from the measurement noise, thereby solving the problem of measurement noise interference such as measurement instrument error or environmental noise, and reducing recognition errors.

[0132] In some embodiments of the present invention, in S600, the following steps are specifically included:

[0133] S610, after eliminating the signal continuation pair corresponding to the maximum duration length, obtain the maximum duration length of the remaining signal continuation pairs 0.3299 times the persistence length .

[0134] In this embodiment, the maximum duration is selected from the signal pairs retained in S500. , according to the maximum duration , determine the corresponding signal continuation pairs, and remove them to obtain the remaining signal continuation pairs and their corresponding continuation lengths , that is, Group signal duration pairs and their corresponding duration lengths.

[0135] Then determine the corresponding duration from the remaining signal continuation pairs The maximum value of , according to the duration of this time The maximum value of , select 0.3299 times its duration ;

[0136] That is to say, from The group signal is continuously aligned to determine the corresponding maximum continuous length ,exist The group signal is continuously aligned and the duration is selected , .

[0137] S620, according to the duration , determine its corresponding period , using the damping ratio calculation formula:

[0138]

[0139] ,in, is the optimal cycle.

[0140] In this embodiment, according to the duration , determine its corresponding period ,cycle is the optimal cycle, through the cycle , duration as well as Maximum value of the group signal continuous alignment duration , using the damping ratio calculation formula:

[0141]

[0142] , calculate the damping ratio corresponding to the free decay curve selected in S300 .

[0143] Reference Figures 3 to 5 In some embodiments of the present invention, Figure 3 The three-degree-of-freedom mass-spring linear structure shown in the figure is used as an example to identify the damping ratio. Assume that the mass of the structure is , the stiffness is , proportional damping is adopted, the damping ratio of each order is 0.035, and Gaussian white noise is used as excitation.

[0144] Determine the target structure, set up three acceleration sensors on the target structure, and use the Newmark method to calculate the acceleration of each detection point. According to the detection signal, the structural acceleration response has an integration time interval of 0.01s and a total duration of 3600s.

[0145] Initialize size Zero matrix, determined from the detection signal and ,in, , the cross power spectral density of the detection signal response is calculated using the cpsd frequency response function in MATLAB, and the initialized zero matrix is ​​used to store the three-dimensional matrix calculated by the cpsd frequency response function The cross power spectrum density of the detection signal response is calculated by the cpsd frequency response function in MATLAB, and the frequency sequence is obtained. .

[0146] Decomposing a 3D matrix using SVD , the decomposed left singular vectors, right singular vectors and singular values ​​are stored in the matrix , as well as middle.

[0147] Use smoothness to adjust the frequency series Perform smoothing and find local extreme points as frequency peaks , for the peak frequency ,definition and ;

[0148] according to Get the corresponding horizontal coordinate in the spectrum diagram and according to Get the corresponding horizontal coordinate in the spectrum diagram ,but

[0149]

[0150]

[0151] , is the median of the difference between two consecutive terms in the frequency vector. ,in, express to The number of frequencies between . Calculate , and obtain the various order responses of the target structure , and all the responses obtained are regarded as single-degree-of-freedom systems.

[0152] Sequentially Autocorrelation processing is performed, that is, the inverse Fourier transform of the cross spectral density is calculated using the fast Fourier transform to obtain the cross correlation function, and three free decay curves are obtained. .

[0153] From three free decay curves Select , let a horizontal line be , first let , Represented as a free decay curve The minimum value in is represented as a very small constant. Next, adjust ,let Continue to increase and adjust the horizontal line Towards The positive half axis moves in During the enlargement process, adjust the horizontal line Each time a local minimum or local maximum is encountered, it is defined as the point Or click , adjust the horizontal line With free decay curve The enclosed area is defined as the intersection area The connection between two regions follows: As the value increases, the larger the local minimum The corresponding intersection area will merge into smaller local minima The corresponding intersection area , the intersection area Will intersect area Connected, intersecting areas disappear, forming a larger new intersection area .when Will be incorporated , the intersection area When it disappears, adjust the horizontal line to , the local maximum and local minimum Pair up to form a continuous pair According to the above rules, the free decay curve All extreme points are processed to obtain a persistence graph , refer to Figure 4 , that is, A collection of persistent pairs.

[0154] Calculate the cutoff equation for the persistence graph , And calculate the duration of each pair of persistence The persistence pair corresponding to the persistence length greater than 0.268 is the signal persistence pair. ( Figure 5 The persistence pairs corresponding to the persistence length less than 0.268 are noise persistence pairs. ( Figure 5 The signal is retained and processed in the next step.

[0155] Among the signal continuation pairs, after deleting the signal continuation pair with the longest continuation length, find the maximum continuation length among the remaining signal continuation pairs 0.3299 times the persistence length , , the corresponding period is the optimal period number 5, according to the damping ratio calculation formula:

[0156]

[0157] The first-order response is calculated The corresponding free decay curve The corresponding damping ratio .

[0158] Repeat the above steps to continue calculating the remaining free decay curves , , the damping ratio is , .

[0159] Compared with the prior art, the damping ratio of the three-degree-of-freedom mass-spring linear structure is identified by automatic frequency domain decomposition method, and the corresponding modal damping ratio is obtained. , , .

[0160] By comparing the damping ratio obtained by the identification method of the present invention with the damping ratio obtained based on the automatic frequency domain decomposition method, it can be seen that the maximum relative error of the automatic frequency domain decomposition method is 14.286%, and the average error is 9.524%; the maximum error of the identification method of the present invention is 5.714%, and the average error is 2.857%.

[0161] The damping ratio accuracy of the present invention is better than that obtained by the automatic frequency domain decomposition method. Therefore, the damping ratio identification method of the present invention can more accurately describe the damping ratio in structural health monitoring.

[0162] According to an embodiment of the second aspect of the present invention, an electronic device includes:

[0163] A memory for computer programs; a processor for executing the computer programs stored in the memory. When the processor executes the program stored in the memory, the processor is used to execute a damping ratio identification method based on topological signal processing in an embodiment of the first aspect of the present invention.

[0164] The processor and the memory may be connected via a bus or other means.

[0165] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs, such as a damping ratio identification method based on topological signal processing described in an embodiment of the present invention. The processor implements a damping ratio identification method based on topological signal processing in an embodiment of the first aspect of the present invention by running the non-transient software program and instructions stored in the memory.

[0166] The memory may include a program storage area and a parameter storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the parameter storage area may store a damping ratio identification method based on topological signal processing described above. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0167] According to an embodiment of the third aspect of the present invention, a storage medium is characterized in that it includes: a computer program is stored, and the computer program is used to be executed by a processor such as a damping ratio identification method based on topological signal processing according to the first aspect of the present invention.

[0168] The non-transient software program and instructions required to implement the above-mentioned terminal selection method are stored in the memory, and when executed by one or more processors, they are executed in the first damping ratio identification method based on topological signal processing of the present invention.

[0169] It will be appreciated by those skilled in the art that all or some of the steps and systems in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, parameter structures, program modules, or other parameters). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media generally include computer readable instructions, parameter structures, program modules or other parameters in a modulated parameter signal such as a carrier wave or other transmission mechanism, and may include any information delivery media.

[0170] The preferred embodiments of the present invention are described in detail above, but the invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A damping ratio identification method based on topological signal processing, characterized in that: include: S100, determining a target structure, acquiring a detection signal collected by a sensor disposed on the target structure, wherein the detection signal includes a relationship between a detection signal value and a time variation; S200, performing low-pass filtering on the detection signal, and extracting each order response of the target structure by using an automatic frequency domain decomposition method; S300, performing autocorrelation processing on the responses of each order to obtain corresponding free decay curves, and selecting a free decay curve from the free decay curves; S400, setting an adjustment horizontal line, using the adjustment horizontal line to process the extreme value points on the free attenuation curve, forming Group persistence pairs and obtain persistence graphs; S500, calculating the The duration length corresponding to the group duration pair and the cutoff equation of the duration graph are used to screen and retain the Signal continuation pairs in group continuation pairs; S600, after eliminating the signal duration pairs corresponding to the duration length of the maximum value, the damping ratio corresponding to the free decay curve is calculated using the remaining duration length; S700 , repeating S400 to S600 to calculate the damping ratios corresponding to the remaining free decay curves.

2. According to claim 1, a damping ratio identification method based on topological signal processing is characterized in that: The extraction process of each order response of the target structure specifically includes: According to the detection signal, the cross power spectrum density of the detection signal response is calculated to obtain a three-dimensional matrix and frequency series ; Decompose the three-dimensional matrix using SVD , and we get the left singular vector , right singular vectors and singular values ; Smoothing the frequency sequence Then, obtain the local extreme point as the frequency peak , according to the smoothed frequency series , frequency peak and singular values , and the various order responses of the target structure are calculated.

3. The damping ratio identification method based on topological signal processing according to claim 1 is characterized in that: The process of obtaining the persistence graph specifically includes: Setting the adjustment level ,make ,in, is a constant; make Continuously increasing, with the free decay curve The local minimum on Intersection, adjust the horizontal line With free decay curve Constructing the intersection area ; When the local minimum The corresponding intersection area Will be incorporated The corresponding intersection area When , the local maximum Paired local minima , forming a continuous pair ,in, , ; Free decay curve The local maximum on and local minima Pair them one by one to form Group continuous pairs and obtain continuous graph .

4. The damping ratio identification method based on topological signal processing according to claim 3 is characterized in that: The calculation The persistence length corresponding to the group persistence pair and the cutoff equation of the persistence graph specifically include: The cutoff equation of the persistence diagram is: ， in, is the distribution parameter of the noise, is the probability constant; According to the calculation formula of continuous length: ， calculate The duration length corresponding to the group duration pair .

5. The damping ratio identification method based on topological signal processing according to claim 4 is characterized in that: The process of continuously screening and retaining the signal specifically includes: The corresponding duration With cutoff equation 's values ​​for comparison; If there is a persistence length Greater than cutoff equation The value of the duration The corresponding continuous pair is a signal continuous pair and is retained.

6. The damping ratio identification method based on topological signal processing according to claim 1 is characterized in that: The damping ratio calculation process corresponding to the free decay curve specifically includes: After removing the signal duration pair corresponding to the maximum duration length, obtain the maximum duration length of the remaining signal duration pairs. 0.3299 times the persistence length ; Get the duration The corresponding cycle , using the damping ratio calculation formula: ， in, is the optimal cycle.

7. The damping ratio identification method based on topological signal processing according to claim 2 is characterized in that: According to the smoothed frequency sequence , frequency peak and singular values , the calculated responses of the target structure are as follows: According to the peak frequency , set the low frequency value and high frequency values ; Get the low frequency value and high frequency values The corresponding horizontal axis in the spectrum diagram and ,according to , and singular values Perform calculations, ， in, for and The median of the difference, express to The number of frequencies between calculate , and obtain the various order responses of the target structure.

8. The damping ratio identification method based on topological signal processing according to claim 1 is characterized in that: The process of obtaining each free attenuation curve specifically includes: The inverse Fourier transform of the cross power spectrum density is calculated by Fourier transform to obtain the cross correlation function, thereby obtaining the respective free attenuation curves.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: The processor executes the computer program to implement the damping ratio identification method based on topological signal processing according to any one of claims 1 to 8.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, a damping ratio identification method based on topological signal processing according to any one of claims 1 to 8 is implemented.

Images