Crack detection method, electronic equipment and computer readable storage medium

Abstract

The invention is suitable for the technical field of structural health detection, and provides a crack detection method, electronic equipment and a computer readable storage medium. The method comprises the following steps: determining a slope of a strain value change rate corresponding to each sampling position according to a strain value of a target structure at each sampling position acquired by distributed optical fiber sensing; determining a strain observation sequence corresponding to each sampling position according to each slope; determining a target observation sequence according to the strain distribution characteristics corresponding to all the strain observation sequences; determining the target position of the crack based on each strain value in the target observation sequence; determining a target integral interval according to the target position, performing progressive integral operation taking an interval midpoint as an interval segmentation line on a continuous strain function in the target integral interval by adopting a self-adaptive Simpson algorithm, and determining the sum of a plurality of obtained target integral values as the width of the crack, automatic detection and width calculation of the crack can be realized, and the detection efficiency and accuracy of the crack are improved.

Description

Technical Field

[0001] This application belongs to the technical field of structural health monitoring, and particularly relates to a crack detection method, an electronic device, and a computer-readable storage medium. Background Art

[0002] During the service life of structures such as bridges, road surfaces, or tunnels, cracks will inevitably occur due to factors such as loads, environmental erosion, or material aging. Cracks are the initial signs of the decline in the durability of structures and even insufficient load-bearing capacity. The continuous expansion of micro-cracks will weaken the integrity and stability of structures, and may even cause safety accidents in severe cases. Therefore, timely and accurate detection of cracks in structures can reduce potential safety risks.

[0003] One traditional crack detection mainly relies on manual regular inspections and visual observations, which is not only inefficient but also unable to achieve real-time monitoring of cracks. Another traditional crack detection method is to locate the crack position by performing threshold analysis and simple gradient calculations on the strain data of the structure to be detected. However, this method is sensitive to environmental noise, resulting in low accuracy of crack detection. Summary of the Invention

[0004] In view of this, the embodiments of this application provide a crack detection method, an electronic device, and a computer-readable storage medium to solve the technical problems that the existing crack detection methods have low crack detection efficiency and accuracy and are unable to achieve real-time monitoring of cracks.

[0005] In a first aspect, the embodiments of this application provide a crack detection method, including: Determining the slope of the strain value change rate corresponding to each sampling position according to the strain values of the target structure collected by distributed optical fiber sensing at each sampling position; Determining the strain observation sequence corresponding to each sampling position according to the slopes of the strain value change rates; the strain observation sequence includes the strain values corresponding to a continuous plurality of sampling positions; Determining the strain distribution characteristics corresponding to each strain observation sequence, and determining the target observation sequence from the strain observation sequences according to the strain distribution characteristics corresponding to all the strain observation sequences; Based on the strain values in the target observation sequence, determining the target position where the crack is located; Determining a target integration interval according to the target position, and using the adaptive Simpson algorithm to perform progressive integration operations on the continuous strain function within the target integration interval with the midpoint of the interval as the interval dividing line, and determining the sum of the multiple target integral values obtained by integration as the width of the crack; the continuous strain function is constructed according to the target observation sequence.

[0006] In an alternative implementation of the first aspect, determining the slope of the strain value change rate corresponding to each sampling position according to the strain values of the target structure collected by distributed optical fiber sensing includes: For the i th sampling position, according to the strain values of the target structure at the i th sampling position, the i +1th sampling position, and the i +2th sampling position, calculate the second derivative of the strain value corresponding to the i th sampling position, and determine the second derivative as the slope of the strain value change rate corresponding to the i th sampling position; 1 ≤ i ≤ n - 2, n is the number of sampling positions, i is an integer.

[0007] In an alternative implementation of the first aspect, determining the strain observation sequence corresponding to each sampling position according to the slopes of the strain value change rates includes: For the 1st sampling position, determine the first strain value sequence composed of the strain values of the continuous first number of sampling positions including the 1st sampling position as the strain observation sequence corresponding to the 1st sampling position; For the j th sampling position, when the slope of the strain value change rate corresponding to the j th sampling position is negative and the slope of the strain value change rate corresponding to the j - 1th sampling position is positive, determine the second strain value sequence composed of the strain values of the continuous second number of sampling positions including the j th sampling position as the strain observation sequence corresponding to the j th sampling position; For the j th sampling position, when the difference between the slope corresponding to the j th sampling position and the slope corresponding to the j - th sampling position is less than the first preset threshold, determine the third strain value sequence composed of the strain values of the continuous third number of sampling positions including the j th sampling position as the strain observation sequence corresponding to the j th sampling position; Wherein, 2 ≤ j ≤ n , n is the number of sampling positions, jis an integer; the second quantity is greater than the first quantity, and the third quantity is less than the first quantity.

[0008] In an alternative implementation of the first aspect, determining a target observation sequence from the strain observation sequences according to the strain distribution characteristics corresponding to all the strain observation sequences includes: Performing dimensionality reduction processing on the strain distribution characteristics corresponding to each of the strain observation sequences to obtain low-dimensional strain data corresponding to each of the strain observation sequences; Performing clustering processing on all the low-dimensional strain data, and determining the strain observation sequences under a target category in the clustering result as the target observation sequence; the target category refers to the category containing crack information.

[0009] In an alternative implementation of the first aspect, determining a target position where a crack is located based on the strain values in the target observation sequence includes: Calculating a first-order backward difference sequence of the target observation sequence; Processing the first-order backward difference sequence by using a unit step function to obtain a binary sequence corresponding to the first-order backward difference sequence; each element in the binary sequence corresponds to each element in the first-order backward difference sequence, and the value of each element in the binary sequence is 0 or 1; Shifting all elements in the binary sequence one bit to the right to obtain a right-shifted sequence corresponding to the binary sequence, and performing an exclusive OR operation on the binary sequence and the right-shifted sequence to obtain a target sequence; Determining the sampling positions corresponding to the elements with a value of 1 in the target sequence as the target position.

[0010] In an alternative implementation of the first aspect, determining a target integration interval according to the target position includes: Performing smoothing processing on the strain values in the target observation sequence by using a moving average algorithm to obtain a smoothed strain value sequence corresponding to the target observation sequence; Constructing a continuous strain function corresponding to the smoothed strain value sequence; Selecting a target integration interval containing the target position from the sampling position interval corresponding to the continuous strain function.

[0011] In an alternative implementation of the first aspect, using an adaptive Simpson algorithm to perform progressive integration operation on the continuous strain function within the target integration interval with the midpoint of the interval as the interval dividing line includes: Performing integration operation on the continuous strain function within the target integration interval by using an adaptive Simpson algorithm to obtain an integration value corresponding to the target integration interval; Taking the midpoint of the target integration interval as the interval dividing line, the target integration interval is divided into two sub-intervals, and the adaptive Simpson's algorithm is used to perform integration operations on the continuous strain function in the two sub-intervals respectively to obtain the integration values corresponding to the two sub-intervals respectively; When the first difference between the integration value corresponding to the target integration interval and the sum of the integration values corresponding to the two sub-intervals is less than the second preset threshold, stop re-dividing the sub-intervals, and determine the integration values corresponding to the two sub-intervals respectively as the target integration value; When the first difference is greater than or equal to the second preset threshold, taking the midpoint of each sub-interval as the interval dividing line, re-divide each sub-interval respectively, and use the adaptive Simpson's algorithm to calculate the integration values corresponding to the sub-intervals obtained by re-division until the first difference between the sum of the integration values corresponding to every two sub-intervals obtained by re-division and the integration value corresponding to the corresponding parent interval is less than the second preset threshold, stop the interval re-division, and determine the integration values corresponding to the sub-intervals whose corresponding first difference is less than the second preset threshold as the target integration value.

[0012] In a second aspect, an embodiment of the present application provides an electronic device, including: A first determination unit, configured to determine the slope of the strain value change rate corresponding to each sampling position according to the strain values of the target structure at each sampling position collected by distributed optical fiber sensing; A second determination unit, configured to determine the strain observation sequence corresponding to each sampling position according to the slope of the strain value change rate of each; the strain observation sequence includes the strain values corresponding to a continuous plurality of sampling positions; A third determination unit, configured to determine the strain distribution characteristics corresponding to each strain observation sequence, and determine a target observation sequence from the strain observation sequences according to the strain distribution characteristics corresponding to all the strain observation sequences; A fourth determination unit, configured to determine the target position where the crack is located based on the strain values in the target observation sequence; A fifth determination unit, configured to determine a target integration interval according to the target position, and use the adaptive Simpson's algorithm to perform a progressive integration operation on the continuous strain function in the target integration interval with the interval midpoint as the interval dividing line, and determine the sum of the multiple target integration values obtained by integration as the width of the crack; the continuous strain function is constructed according to the target observation sequence.

[0013] In a third aspect, an embodiment of the present application provides another electronic device, including a memory and a computer program stored in the memory and executable on a processor. When the processor executes the computer program, the method described in any optional implementation manner of the first aspect above is implemented.

[0014] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the crack detection method described in any optional implementation manner of the first aspect above is implemented.

[0015] In a fifth aspect, an embodiment of the present application provides a computer program product, and when the computer program product runs on an electronic device, the electronic device is enabled to implement the crack detection method described in any optional implementation manner of the first aspect.

[0016] Implementing the crack detection method, electronic device, computer-readable storage medium, and computer program product provided by the embodiments of the present application has the following beneficial effects: The crack detection method provided by the embodiment of the present application can roughly determine the position interval where the crack is located by determining the slope of the strain value change rate corresponding to each sampling position. On this basis, in order to more accurately locate the crack, all strain values are divided into multiple strain observation sequences according to each slope. Since the strain distribution characteristics corresponding to the strain observation sequences can accurately reflect the strain value distribution in the strain observation sequences, by determining the strain distribution characteristics corresponding to each strain observation sequence and according to the strain distribution characteristics corresponding to all strain observation sequences, the target observation sequence containing crack information can be accurately determined from the strain observation sequences. Furthermore, based on the strain values in the target observation sequence, the target position where the crack is located can be accurately determined, thereby improving the accuracy and efficiency of crack detection. At the same time, by determining the target integration interval where the target position is located and using the adaptive Simpson algorithm to perform progressive integration operations on the continuous strain function within the target integration interval with the midpoint of the interval as the interval division line, the estimation accuracy of the integral value can be improved. Since the crack width is the sum of multiple target integral values obtained by integration, the accuracy of crack width calculation can be improved.

[0017] In addition, since this crack detection method only needs to deploy distributed optical fiber sensors on the target structure to realize automatic monitoring of cracks on the target structure without manual regular inspections, it not only improves the crack monitoring efficiency but also reduces the labor cost. Description of the Drawings

[0018] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 Schematic structural diagram of a crack detection system provided by an embodiment of the present application; Figure 2 Schematic flowchart of a crack detection method provided by an embodiment of the present application; Figure 3 Schematic diagram of the interface of an electronic device provided by an embodiment of the present application; Figure 4 Schematic diagram of the implementation process of S203 in a crack detection method provided by an embodiment of the present application; Figure 5 Schematic structural diagram of a preset encoder provided by an embodiment of the present application; Figure 6 Schematic diagram of the implementation process of S204 in a crack detection method provided by an embodiment of the present application; Figure 7 Schematic diagram of the implementation process of S205 in a crack detection method provided by an embodiment of the present application; Figure 8 Schematic structural diagram of an electronic device provided by an embodiment of the present application; Figure 9 Schematic structural diagram of an electronic device provided by another embodiment of the present application. Detailed implementation manners

[0020] The following embodiments are only used to more clearly illustrate the technical solutions of the present application, so they are only examples and cannot be used to limit the protection scope of the present application.

[0021] In the description of the embodiments of the present application, technical terms such as "include", "comprise", "have" and any variations thereof all mean "including but not limited to", unless otherwise specifically emphasized in other ways. In the description of the embodiments of the present application, unless otherwise specified, the technical term "a plurality of" means two or more than two, and the technical terms "at least one", "one or more" mean one, two or more than two. Technical terms such as "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity, specific order or primary-secondary relationship of the indicated technical features. The technical term "and / or" is merely a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after.

[0022] The embodiments of the present application first provide a crack detection system. Please refer to Figure 1 , which is a schematic structural diagram of a crack detection system provided by the embodiments of the present application. As Figure 1 shown, the crack detection system may include an electronic device 11 and a collection device 12. A communication connection is established between the electronic device 11 and the collection device 12, and this communication connection can be a wireless communication connection or a wired communication connection. The embodiments of the present application do not limit the communication connection method between the electronic device 11 and the collection device 12.

[0023] Exemplarily, the electronic device 11 may include devices such as a mobile phone, a tablet computer, a notebook computer, a desktop computer or a server, etc. The embodiments of the present application do not limit the type of the electronic device 11.

[0024] In some embodiments, the collection device 12 may include a distributed optical fiber sensor 121 and a fiber grating demodulator 122. The fiber grating demodulator 122 can be connected to the distributed optical fiber sensor 121 and the electronic device 11.

[0025] Exemplarily, the distributed optical fiber sensor 121 may include, from the inside to the outside: a fused silica optical fiber core, a fused silica optical fiber cladding, and a protective coating made of multiple layers of polymers.

[0026] Among them, a plurality of equally spaced fiber Bragg gratings (FBGs) can be integrated on the optical fiber core of the distributed optical fiber sensor 121, and the interval between every two adjacent FBGs can be set according to actual needs. For example, in order to make the distributed optical fiber sensor 121 have a high spatial resolution, the interval between every two adjacent FBGs can be set to 5 millimeters (mm) to 10 mm.

[0027] In practical applications, when it is necessary to detect the health status of a target structure (such as whether there are cracks), the distributed fiber optic sensor 121 can be arranged on the surface or inside of the key stress area of the target structure. Exemplarily, the target structure may include concrete structures such as buildings, bridges, tunnels or road surfaces, or may also include structures made of other materials (such as wood structures or ceramic structures, etc.). The specific type of the target structure is not limited in the embodiments of the present application.

[0028] Optionally, when the distributed fiber optic sensor 121 is arranged on the surface of the target structure, epoxy resin or other suitable adhesives can be used to bond the distributed fiber optic sensor 121 to the surface of the target structure, so that the distributed fiber optic sensor 121 is closely attached to the surface of the target structure, thereby effectively transmitting the strain of the target structure, reducing the measurement error of the distributed fiber optic sensor, and improving the accuracy of crack detection.

[0029] It should be noted that different FBGs in the distributed fiber optic sensor 121 can be used to reflect light of different wavelengths. Since the distributed fiber optic sensor 121 is arranged on the surface or inside of the target structure, each FBG in the distributed fiber optic sensor 121 can correspond to a sampling position on the target structure, and is used to collect the strain value of the target structure at this sampling position. Multiple FBGs on the optical fiber core form an FBG array along the optical fiber axis, which can realize multi-point distributed monitoring of the strain value of the target structure.

[0030] Based on this, the fiber Bragg grating demodulator 122 can emit light to the distributed fiber optic sensor 121, receive the light reflected back by the distributed fiber optic sensor 121 (i.e., the reflected light), and analyze the actual wavelength of the reflected light, so as to obtain the strain values of the target structure at each sampling position. Among them, the strain value can be used to represent the relative deformation degree of the target structure.

[0031] Exemplarily, the sampling frequency of the acquisition device 12 can be set according to actual needs. For example, in order to meet the high time resolution requirements of the strain change of the target structure during the crack propagation process, the sampling frequency of the acquisition device 12 can be set to 100 Hertz (Hz) to 1000 Hz.

[0032] Optionally, after the fiber Bragg grating demodulator 122 obtains the strain values of the target structure at each sampling position, it can store the strain values of the target structure at each sampling position in a local server or database for subsequent application in the crack detection of the target structure. Exemplarily, the storage format of the strain values of the target structure at each sampling position can be set according to actual needs, and the embodiments of the present application do not limit it.

[0033] Optionally, when receiving a strain value acquisition instruction from the electronic device 11, the fiber Bragg grating demodulator 122 may also send the strain values of the target structure at each sampling position stored in the local server or database to the electronic device 11, so that the electronic device 11 can determine the target position where the crack is located based on the strain values of the target structure at each sampling position and calculate the width of the crack.

[0034] In some other embodiments, the acquisition device 12 may further include a temperature sensor. The temperature sensor may be connected to the electronic device 11. In practical applications, the temperature sensor may be disposed in the key stress areas of the target structure to collect the actual temperature of the key stress areas of the target structure and send the actual temperature to the electronic device 11, so that the electronic device 11 can subsequently perform temperature compensation on the strain values of the target structure at each sampling position based on the actual temperature, thereby reducing the influence of the temperature change of the target structure on the measurement accuracy of the strain values and improving the accuracy of crack detection.

[0035] It should be noted that the specific processes of the electronic device 11 for temperature compensation of strain values, determining the target position where the crack is located, and calculating the width of the crack can refer to the relevant descriptions in the subsequent method embodiments, and will not be elaborated here.

[0036] The embodiment of the present application also provides a crack detection method. The execution subject of this crack detection method may be the electronic device 11 in the above crack detection system. Exemplarily, please refer to Figure 2 for a schematic flowchart of a crack detection method provided by the embodiment of the present application. As Figure 2 shown, in some embodiments, this crack detection method may include S201 - S205, which are described in detail as follows: S201, determine the slope of the strain value change rate corresponding to each sampling position according to the strain values of the target structure at each sampling position collected by distributed optical fiber sensing.

[0037] Among them, the slope of the strain value change rate corresponding to each sampling position can be used to represent the change speed of the strain value change rate at each sampling position.

[0038] In an optional implementation manner, when receiving a crack detection instruction, the electronic device may obtain the strain values of the target structure at each sampling position collected by the distributed sensor from the fiber Bragg grating demodulator.

[0039] Among them, the crack detection instruction may be manually triggered by the user. Exemplarily, please refer to Figure 3 for a schematic diagram of the interface of an electronic device provided by the embodiment of the present application. As Figure 3As shown in the figure, a crack detection control 31 can be configured on the electronic device. The user can issue a crack detection instruction to the electronic device by clicking on the crack detection control 31. Based on this, when the electronic device detects that the crack detection control 31 is clicked, it can determine that it has received the crack detection instruction and obtain the strain values of the target structure at each sampling position collected by the distributed sensor from the fiber Bragg grating demodulator.

[0040] In another alternative implementation, the electronic device can automatically obtain the strain values of the target structure at each sampling position collected by the distributed sensor from the fiber Bragg grating demodulator every first time period. The first time period can be set according to actual needs, and the embodiments of the present application do not limit it.

[0041] Optionally, in the case where the crack detection system does not include a temperature sensor, in order to reduce the computational complexity, the electronic device can directly determine the slope of the strain value change rate corresponding to each sampling position based on the strain values of the target structure at each sampling position. In order to reduce the strain value fluctuations caused by environmental changes, the electronic device can perform denoising processing on the strain values of the target structure at each sampling position and determine the slope of the strain value change rate corresponding to each sampling position based on the denoised strain values of each sampling position.

[0042] Optionally, in the case where the crack detection system includes a temperature sensor, the electronic device can further perform temperature compensation on the strain values of each sampling position after denoising processing according to the actual temperature of the key stress area of the target structure collected by the temperature sensor, and determine the slope of the strain value change rate corresponding to each sampling position based on the temperature-compensated strain values of each sampling position. This can reduce the influence of temperature changes on the measurement accuracy of strain values and improve the accuracy of crack detection.

[0043] In a specific implementation, the electronic device can determine the slope of the strain value change rate corresponding to each sampling position through step 1.1, which is described in detail as follows: Step 1.1, for the i th sampling position, according to the strain values of the target structure at the i th sampling position, the i +1th sampling position, and the i +2th sampling position, calculate the second derivative of the strain value corresponding to the i th sampling position, and determine the second derivative as the slope of the strain value change rate corresponding to the i th sampling position.

[0044] Among them, 1 ≤ i ≤ n -2, n is the number of sampling positions, i is an integer.

[0045] In an alternative implementation, the electronic device can specifically calculate the second derivative of the strain value corresponding to the i th sampling position through the following formula (1): Gradient’ ( i ) = Strain ( i + 2) - 2 Strain ( i + 1) + Strain ( i )] / Δ x 2 ; Formula (1) where Gradient’ ( i ) is the second derivative of the strain value corresponding to the i th sampling position, Strain ( i ) is the strain value of the target structure at the i th sampling position, Strain ( i + 1) is the strain value of the target structure at the i + 1th sampling position, Strain ( i + 2) is the strain value of the target structure at the i + 2th sampling position, and Δ x is the interval between every two adjacent sampling positions.

[0046] The second derivative of the strain value corresponding to each sampling position can be used to measure the curvature (i.e., the degree of bending) of the strain curve at that sampling position. Among them, the strain curve can be a continuous curve obtained by connecting the strain values of the target structure at each sampling position in spatial order, which can reflect the strain distribution of the target structure.

[0047] It can be understood that when a crack occurs in the target structure, the material at the crack will break and separate, resulting in relative displacement of the structures on both sides of the crack. Since the distributed fiber optic sensor is closely attached to the target structure, the corresponding part of the outermost protective coating of the distributed fiber optic sensor will be forced to stretch to adapt to the opening of the crack. This stretching will cause the corresponding part of the protective coating to bear a very high strain, and the protective coating will transmit this high strain to the FBG at the corresponding position. In this way, not only can the FBG at the corresponding position sense this high strain, but also the probability of the distributed fiber optic sensor being torn is greatly reduced, making the distributed fiber optic sensor have a long service life. It should be noted that the above high strain will cause a peak in the strain curve, that is, the peak region in the strain curve usually corresponds to the crack position of the target structure.

[0048] It can also be understood that, due to the large curvature of the strain curve near the peak, the second derivative of the strain value near the peak is usually negative, and the strain value reaches the minimum (the largest absolute value) at the peak; since the strain curve is close to a straight line in the gently changing region, the second derivative of the strain value in this region is usually close to 0. Based on this, the location of the crack can be located according to the sign change of the second derivative of the strain value corresponding to each sampling position.

[0049] S202. Determine the strain observation sequence corresponding to each sampling position according to the slope of the change rate of each strain value; the strain observation sequence includes the strain values corresponding to a continuous plurality of sampling positions.

[0050] To improve the accuracy of crack detection, when locating the target position where the crack is located, multiple strain observation sequences can be determined according to the sampling positions. Each strain observation sequence can be composed of the strain values corresponding to a continuous plurality of sampling positions. For example, the strain observation sequence corresponding to each sampling position can be determined according to the slope of the change rate of the strain value corresponding to each sampling position, so as to obtain n a plurality of strain observation sequences.

[0051] In an optional implementation manner, the electronic device can determine the strain observation sequence corresponding to each sampling position through steps 2.1 to 2.3, which are described in detail as follows: Step 2.1. For the first sampling position, determine the first strain value sequence composed of the strain values corresponding to the continuous first number of sampling positions including the first sampling position as the strain observation sequence corresponding to the first sampling position.

[0052] Step 2.2. For the j th sampling position, when the slope of the change rate of the strain value corresponding to the j th sampling position is negative and the slope of the change rate of the strain value corresponding to the j -1th sampling position is positive, determine the second strain value sequence composed of the strain values corresponding to the continuous second number of sampling positions including the j th sampling position as the strain observation sequence corresponding to the j th sampling position.

[0053] Step 2.3. For the j th sampling position, when the difference between the slope corresponding to the j th sampling position and the slope corresponding to the j -th sampling position is less than the first preset threshold, determine the third strain value sequence composed of the strain values corresponding to the continuous third number of sampling positions including the j th sampling position as the strain observation sequence corresponding to the j th sampling position.

[0054] where 2 ≤ j ≤ n , n is the number of sampling positions, j is an integer.

[0055] The second quantity may be greater than the first quantity, and the third quantity may be less than the first quantity. For example, the first quantity may be 60, the second quantity may be 90, and the third quantity may be 30. That is, the first strain value sequence may be composed of 60 strain values, the second strain value sequence may be composed of 90 strain values, and the third strain value sequence may be composed of 30 strain values.

[0056] In the embodiments of the present application, by using a smaller strain observation sequence for fast data processing in a region where the strain value changes gently, and using a larger strain observation sequence for stable and comprehensive data analysis in a region where the strain value changes violently (i.e., a region containing crack information), the size of the strain observation sequence can be matched with the local complexity of the data, which can not only accurately detect the location of the crack, but also reduce the computational complexity and save computational resources.

[0057] S203. Determine the strain distribution characteristics corresponding to each strain observation sequence, and determine the target observation sequence from the strain observation sequences according to the strain distribution characteristics corresponding to all the strain observation sequences.

[0058] where the number of strain distribution characteristics may be multiple. Different strain distribution characteristics can be used to reflect the strain value distribution in the strain observation sequence from different angles.

[0059] The target observation sequence may refer to a strain observation sequence containing crack information.

[0060] The strain distribution characteristics may include: basic statistical characteristics, slope distribution characteristics of the change rate, strain difference characteristics, distribution fitting characteristics, peak characteristics, direction characteristics, tail characteristics, and comprehensive calculation characteristics, etc.

[0061] The basic statistical characteristics can be used to represent the central tendency, dispersion degree, strain situation at the starting sampling position, and strain situation at the ending sampling position of the strain value distribution in the strain observation sequence. Exemplarily, the basic statistical characteristics may include: the maximum strain value in the strain observation sequence, the standard deviation of the strain value, the first quartile of the strain value, the third quartile of the strain value, the strain starting value, and the strain ending value.

[0062] Among them, the first quartile of the strain values in the strain observation sequence can be the strain value at the first 1 / 4 position after sorting all the strain values in the strain observation sequence in ascending order. Similarly, the third quartile of the strain values in the strain observation sequence can be the strain value at the first 3 / 4 position after sorting all the strain values in the strain observation sequence in ascending order. The starting strain value in the strain observation sequence can refer to the strain value corresponding to the first sampling position in the strain observation sequence. The ending strain value in the strain observation sequence can refer to the strain value corresponding to the last sampling position in the strain observation sequence.

[0063] The slope distribution feature can be used to represent the change trend of the strain value change rate in the strain observation sequence. Exemplarily, the slope distribution feature can include: the sampling position corresponding to the maximum strain value in the strain observation sequence, the slope of the strain value change rate before the maximum strain value (referred to as the forward slope for short), and the slope of the strain value change rate after the maximum strain value (referred to as the backward slope for short).

[0064] Optionally, the electronic device can perform linear regression fitting on the strain values corresponding to several sampling positions before the maximum strain value in the strain observation sequence, and determine the slope of the first curve obtained by fitting as the forward slope of the maximum strain value. Optionally, the electronic device can also perform linear regression fitting on the strain values corresponding to several sampling positions after the maximum strain value in the strain observation sequence, and determine the slope of the second curve obtained by fitting as the backward slope of the maximum strain value.

[0065] The strain difference feature can be used to represent the difference situation of the strain values in the strain observation sequence. Exemplarily, the strain difference feature can include: the difference between the maximum strain value and the minimum strain value in the strain observation sequence, the difference between the cumulative relative strain values before and after the maximum strain value, and the difference between the starting strain value and the ending strain value.

[0066] Among them, the difference between the cumulative relative strain values before and after the maximum strain value in the strain observation sequence can refer to the difference between the sum of the strain values corresponding to all sampling positions before the sampling position corresponding to the maximum strain value and the sum of the strain values corresponding to all sampling positions after the sampling position corresponding to the maximum strain value.

[0067] The distribution fitting feature can be used to represent the distribution characteristics of the strain values in the strain observation sequence, and can provide a basis for crack identification. Exemplarily, the distribution fitting feature can include: the difference between the true distribution of the strain values in the strain observation sequence and the fitted normal distribution.

[0068] Optionally, the electronic device can calculate the difference between the true distribution of the strain values in the strain observation sequence and the fitted normal distribution through the following steps 3.1 to 3.5: Step 3.1, calculate the average value of all strain values in the strain observation sequence s 1 and the standard deviation δ1.

[0069] Step 3.2, construct a normal distribution with the average value s 1 as the mean and the standard deviation δ1 as the standard deviation ( s 1, δ1 2 ).

[0070] Step 3.3, calculate the first empirical cumulative distribution function corresponding to the strain observation sequence ECDF 1( x ).

[0071] Step 3.4, calculate the second empirical cumulative distribution function corresponding to the normal distribution ( s 1, δ1 2 ). ECDF 2( x ).

[0072] Step 3.5, according to the formula D = max x | ECDF 1( x ) - ECDF 2( x )|, calculate the difference measure ECDF 1( x ) and the second empirical cumulative distribution function ECDF 2( x ), and determine this difference measure D as the difference between the true distribution of the strain values in the strain observation sequence and the fitted normal distribution. D

[0073] Exemplarily, the peak feature may include the number of peaks in the strain observation sequence. Since cracks will cause peaks in the strain distribution, the peak feature can be used to assist in identifying the location of the cracks.

[0074] The direction feature can be used to evaluate the symmetry of the strain values in the strain observation sequence to judge the direction of the strain distribution. Exemplarily, the direction feature may include the skewness coefficient. Optionally, the skewness coefficient may be, for example, the Fisher-Pearson skewness coefficient. It should be noted that the specific calculation method of the Fisher-Pearson skewness coefficient can refer to the relevant descriptions in the prior art and will not be elaborated here.

[0075] The tail feature can be used to represent the end feature of the strain distribution corresponding to the strain observation sequence. Exemplarily, the tail feature may include the kurtosis coefficient corresponding to the strain observation sequence. It should be noted that the specific calculation method of the kurtosis coefficient can refer to the relevant descriptions in the prior art and will not be elaborated here.​

[0076] Exemplarily, the comprehensive calculation features may include: the sum of all strain values in the strain observation sequence, the test statistic of the null hypothesis test corresponding to the strain observation sequence, the p-value corresponding to the test statistic, and the full width at half maximum of the strain value. Among them, the sum of all strain values in the strain observation sequence can reflect the cumulative situation of the strain values as a whole. The test statistic of the null hypothesis test corresponding to the strain observation sequence and the p-value corresponding to the test statistic are used to test whether the strain values in the strain observation sequence follow a normal distribution to assist in judging the distribution characteristics of the strain values. The full width at half maximum of the strain value can be used to represent the width characteristics of the peak.

[0077] It should be noted that the specific calculation methods for the sum of all strain values in the strain observation sequence, the test statistic of the null hypothesis test corresponding to the strain observation sequence, the p-value corresponding to the test statistic, and the full width at half maximum of the strain value can refer to the relevant descriptions in the prior art and will not be elaborated here.

[0078] In a specific implementation manner, the electronic device can determine the target observation sequence from the strain observation sequence through S2031 to S2032 as shown in Figure 4 and is described in detail as follows: S2031, perform dimensionality reduction processing on the strain distribution characteristics corresponding to each strain observation sequence to obtain the low-dimensional strain data corresponding to each strain observation sequence.

[0079] Optionally, the electronic device can input the strain distribution characteristics corresponding to each strain observation sequence into a preset encoder for processing to obtain the low-dimensional strain data corresponding to each strain observation sequence.

[0080] Exemplarily, please refer to Figure 5 for a schematic structural diagram of a preset encoder provided in an embodiment of the present application. As shown in Figure 5 , the preset encoder may include a plurality of cascaded feature extraction networks 51. Each feature extraction network 51 may include a fully connected layer and an activation layer connected to the output end of the fully connected layer.

[0081] Among them, the fully connected layer can be used to map the high-dimensional strain to low-dimensional strain data and send the low-dimensional strain data to the activation layer. The activation layer can be used to perform a non-linear transformation on the received low-dimensional strain data using a preset activation function. It should be noted that the data output by the activation layer of the last feature extraction network 51 is the low-dimensional strain data corresponding to the strain observation sequence.

[0082] Exemplarily, the preset activation function can be the hyperbolic tangent function or other types of activation functions, and the present application embodiment does not limit the type of the preset activation function.

[0083] In the embodiment of the present application, by setting a plurality of cascaded feature extraction networks 51 in a preset encoder, the input multiple strain distribution features can be gradually compressed into low-dimensional strain data, so as to realize the dimensionality reduction processing and feature extraction of the strain distribution features, and reduce the computational complexity of subsequent steps.

[0084] S2032, perform clustering processing on all the low-dimensional strain data, and determine the strain observation sequences under the target category in the clustering result as the target observation sequences.

[0085] Exemplarily, the electronic device can adopt the K-means clustering algorithm to perform clustering processing on the low-dimensional strain data corresponding to all the strain observation sequences, so as to cluster the low-dimensional strain data into K categories. Exemplarily, K can be 2, and the 2 categories can be the target category containing crack information and the non-target category not containing crack information respectively. Among them, the similarity between the low-dimensional strain data in the same category is relatively high, and the similarity between the low-dimensional strain data in different categories is relatively low.

[0086] In practical applications, the specific types of the above 2 categories can be determined according to the test statistic of the null hypothesis test corresponding to the strain observation sequences in each category and the p-value corresponding to the test statistic. For example, when both the test statistic of the null hypothesis test corresponding to the strain observation sequences in a certain category and the p-value corresponding to the test statistic are close to 0, the electronic device can determine that this category is the non-target category not containing crack information, and determine the other category as the target category containing crack information.

[0087] It should be noted that the target category can include one or more strain observation sequences.

[0088] S204, based on each strain value in the target observation sequence, determine the target position where the crack is located.

[0089] Among them, the target position where the crack is located can be the sampling positions corresponding to one or more strain values in the target observation sequence, that is, the number of the target positions where the crack is located can be one or more.

[0090] Optionally, S204 can be implemented by S2041~S2044 as shown in Figure 6 The details are as follows: S2041, calculate the first-order backward difference sequence of the target observation sequence.

[0091] Among them, the first-order backward difference sequence can be used to represent the differences between the 2nd to the m th strain values in the target observation sequence and their previous strain values respectively. m is the number of strain values included in the target observation sequence.

[0092] Optionally, the electronic device may shift all the strain values in the target observation sequence one position to the left to obtain a left-shifted sequence corresponding to the target observation sequence, and calculate the difference values (i.e., the differences) between each strain value in the target observation sequence and each strain value in the left-shifted sequence, and determine the sequence composed of all the difference values as the first-order backward difference sequence of the target observation sequence.

[0093] Exemplarily, assume that the target observation sequence is [2, 3, 5, 6, 10, 6, 5, 3, 2]. Then, the left-shifted sequence obtained by shifting all the strain values in the target observation sequence one position to the left can be [3, 5, 6, 10, 6, 5, 3, 2, X1], where X1 is data out of range and can be ignored when calculating the first-order backward difference sequence. Based on this, the first-order backward difference sequence can be [2 - 3, 3 - 5, 5 - 6, 6 - 10, 10 - 6, 6 - 5, 5 - 3, 3 - 2], that is, [-1, -2, -1, -4, 4, 1, 2, 1]. It can be seen that -4 and 4 in the first-order backward difference sequence are relatively prominent compared to other values, indicating that the target structure may have cracks at the positions corresponding to -4 and 4.

[0094] S2042. Process the first-order backward difference sequence using the unit step function to obtain a binary sequence corresponding to the first-order backward difference sequence.

[0095] The purpose of the electronic device to process the first-order backward difference sequence using the unit step function is to remove the negative part of the first-order backward difference sequence and only retain the positive part that may have mutations.

[0096] Optionally, the electronic device may assign 0 to all negative values in the first-order backward difference sequence and assign 1 to all positive values, so as to obtain a binary sequence corresponding to the first-order backward difference sequence, that is, each element in the binary sequence corresponds to each element in the first-order backward difference sequence, and the value of each element in the binary sequence is 0 or 1.

[0097] Exemplarily, assume that the first-order backward difference sequence is [-1, -2, -1, -4, 4, 1, 2, 1]. Then, the binary sequence obtained by processing the first-order backward difference sequence using the unit step function can be [0, 0, 0, 0, 1, 1, 1, 1].

[0098] S2043. Shift all elements in the binary sequence one position to the right to obtain a right-shifted sequence corresponding to the binary sequence, and perform an exclusive OR operation on the binary sequence and the right-shifted sequence to obtain the target sequence.

[0099] Exemplarily, assume that the binary sequence is [0, 0, 0, 0, 1, 1, 1, 1]. After shifting all elements in the binary sequence one position to the right, the shifted sequence obtained can be [X2, 0, 0, 0, 0, 1, 1, 1], where X2 is an unknown value and can be ignored or replaced with 0. Based on this, the target sequence obtained after performing an exclusive OR operation on the binary sequence and the shifted sequence can be [0, 0, 0, 0, 1, 0, 0, 0].

[0100] S2044, determine the acquisition positions corresponding to the elements with a value of 1 in the target sequence as the target positions.

[0101] Exemplarily, assume that the 1 in the target sequence [0, 0, 0, 0, 1, 0, 0, 0] corresponds to the 5th sampling position. Then it can be shown that there is a crack at the 5th sampling position, that is, the target position where the crack is located is the 5th sampling position.

[0102] S205, determine the target integration interval according to the target position, and use the adaptive Simpson's algorithm to perform a progressive integration operation on the continuous strain function within the target integration interval with the midpoint of the interval as the interval division line, and determine the sum of multiple target integral values obtained by the integration as the width of the crack.

[0103] Among them, the target integration interval refers to a continuous position interval that includes the target position where the crack is located.

[0104] The continuous strain function can be constructed according to the target observation sequence.

[0105] Optionally, in S205, determining the target integration interval according to the target position and using the adaptive Simpson's algorithm to perform a progressive integration operation on the continuous strain function within the target integration interval with the midpoint of the interval as the interval division line may include S2051 to S2057 as shown in Figure 7 and are detailed as follows: S2051, use the moving average algorithm to smooth each strain value in the target observation sequence to obtain a smoothed strain value sequence corresponding to the target observation sequence.

[0106] Since the strain values within the target integration interval (i.e., the crack area) usually have large fluctuations, the moving average algorithm can be used to preprocess the strain values within the target integration interval to reduce noise interference. For example, the electronic device can use a moving window of size h to perform a moving average process on the strain values in the target observation sequence to obtain a smoothed strain value sequence corresponding to the target observation sequence. Among them, h can be set according to actual needs. For example h can be 3.

[0107] Exemplarily, the value of each element in the smoothed strain value sequence can be the mean of all strain values within the moving window corresponding to the corresponding sampling position in the target observation sequence. That is y q =( x q-1 + x q + x q+1 ) / 3, where y q is the value of the q -th element in the smoothed strain value sequence, x q-1 is the strain value corresponding to the q -1-th sampling position in the target observation sequence, x q is the strain value corresponding to the q -th sampling position in the target observation sequence, x q+1 is the strain value corresponding to the q +1-th sampling position in the target observation sequence, 1 ≤ q ≤ m .

[0108] It should be noted that for the sampling positions corresponding to the boundary strain values in the target observation sequence, the missing strain values in the moving window corresponding to the sampling position can be determined by mirror extension (i.e., mirror symmetry supplementation). For example, for the sampling position corresponding to the first strain value in the target observation sequence, the strain value x 0 in its corresponding moving window can be determined according to the second strain value x 2 in the target observation sequence. For the sampling position corresponding to the m -th strain value in the target observation sequence, the strain value x m+1 in its corresponding moving window can be determined according to the m -1-th strain value x m-1 in the target observation sequence.

[0109] S2052. Construct a continuous strain function corresponding to the smoothed strain value sequence.

[0110] It can be understood that since the adaptive Simpson algorithm repeatedly calculates the integral value of the function in a continuous interval, it is necessary to construct a continuous strain function corresponding to the smoothed strain value sequence to meet the calculation requirements of the adaptive Simpson algorithm.

[0111] Exemplarily, the electronic device may use the cubic spline interpolation method to convert each discrete strain value in the smoothed strain value sequence into a continuous strain function. Specifically, the electronic device may construct a continuous strain function through the following steps 4.1 to 4.3, which are described in detail as follows: Step 4.1, construct m -1 cubic polynomials according to the smoothed strain value sequence.

[0112] Wherein, m is the number of strain values included in the smoothed strain value sequence.

[0113] m The -1 cubic polynomials can be respectively the cubic polynomials corresponding to the sampling position intervals between every two adjacent strain values in the smoothed strain value sequence.

[0114] Exemplarily, the electronic device may take each strain value in the smoothed strain value sequence as a corresponding ordinate value (i.e., y 1, y 2, ……, y m ), and take the sampling position corresponding to each strain value in the smoothed strain value sequence as a corresponding abscissa value (i.e., x 1, x 2, ……, x m ), so as to obtain m discrete coordinate points: ([[]] x 1, y 1), ([[]] x 2, y 2), ……, ([[]] x m , y m ). The electronic device may construct m -1 cubic polynomials m using the cubic spline interpolation method based on these S d ( x ). Among them, each polynomial S d ( x ) corresponds to a coordinate point interval x d , x d+1 , 1 ≤ d ≤ m -1. Exemplarily, the electronic device may define each cubic polynomial S d ( x ) as: S d( x ) = a d + b d ( x - x d ) + c d ( x - x d ) 2 + d d ( x - x d ) 3 . Among them, a d , b d , c d and d d are the coefficients of the cubic polynomial S d ( x ).

[0115] Step 4.2, Solve the coefficients of each cubic polynomial to obtain the complete function expression corresponding to each cubic polynomial.

[0116] For each coordinate point interval x d , x d+1 , the function value x = x d at the starting position (i.e., S d ( x d ) is y d . To make the curve of the finally obtained continuous strain function continuous and smooth, the electronic device can make the function values, first-order derivatives, and second-order derivatives equal at the junction points of every two adjacent coordinate point intervals (i.e., x = x d+1 ), that is, S d ( x d+1 ) = S d+1 ( x d+1 ), S d ’ (x d+1 )= S d+1 ’ ( x d+1 ),and S d ’ ( x d+1 )= S d+1 ’ ( x d+1 ), and the second-order derivatives of the starting coordinate point and the ending coordinate point of all coordinate points are equal to 0, so that the curve of the continuous strain function can be smoother at the beginning and the end. The electronic device solves the simultaneous equations according to the above conditions and obtains the coefficients of each cubic polynomial by solving the equations.

[0117] Step 4.3, determine the function described by the complete function expression corresponding to all cubic polynomials as a continuous strain function.

[0118] S2053, selecting a target integration interval including the target position from the sampling position interval corresponding to the continuous strain function.

[0119] The target integration interval may be the interval between the starting position and the ending position of the crack.

[0120] S2054, using an adaptive Simpson algorithm to perform an integration operation on the continuous strain function within the target integration interval to obtain an integral value corresponding to the target integration interval.

[0121] For example, assuming that the target integration interval is [ a , b ], the electronic device can use the following adaptive Simpson algorithm to perform integration operation on the continuous strain function within the target integration interval to obtain the integral value corresponding to the target integration interval: ; in, S ( a , b ) is the integral value corresponding to the target integral interval, f is a continuous strain function, ( a + b ) / 2 is the midpoint of the target integration interval.

[0122] S2055. Using the midpoint of the target integral interval as the interval division line, divide the target integral interval into two sub-intervals, and use the adaptive Simpson's algorithm to perform integral operations on the continuous strain functions in the two sub-intervals respectively to obtain the integral values corresponding to the two sub-intervals.

[0123] Exemplarily, assume that the target integral interval is a , b . Then the electronic device can divide the midpoint of the target interval a , b as ( a + b ) / 2 as the interval division line, and divide the target interval a , b into two sub-intervals a , r and r , b , where r= ( a + b ) / 2. The electronic device can use the above-mentioned adaptive Simpson's algorithm to perform integral operations on the continuous strain functions in the sub-intervals a , r and r , b respectively, to obtain the integral value a , r corresponding to the sub-interval S ( a , r ) and the integral value r , b corresponding to the sub-interval S ( r , b ).

[0124] S2056. When the first difference between the integral value corresponding to the target integral interval and the sum of the integral values corresponding to the two sub-intervals is less than the second preset threshold, stop the re-division of the sub-intervals, and determine the integral values corresponding to the two sub-intervals respectively as the target integral values.

[0125] S2057. When the first difference is greater than or equal to the second preset threshold, use the midpoint of each sub-interval as the interval division line to re-divide each sub-interval respectively, and use the adaptive Simpson's algorithm to calculate the integral values corresponding to the re-divided sub-intervals. Stop the interval re-division until the first difference between the sum of the integral values corresponding to every two re-divided sub-intervals and the integral value corresponding to the corresponding parent interval is less than the second preset threshold, and determine the integral values corresponding to the sub-intervals with the corresponding first difference less than the second preset threshold as the target integral values.

[0126] Among them, the second preset threshold can be determined according to actual needs, and the embodiments of the present application do not limit it.

[0127] It should be noted that the above first difference is the absolute value of the difference.

[0128] Optionally, in the above S ( a , b ) and S ( a , r ) + S ( r , b ) when the first difference is less than the second preset threshold, it indicates that the calculation accuracy of the current integral value has reached the preset requirement. Therefore, the electronic device can stop further dividing the sub-interval and determine the integral value a , r corresponding to the sub-interval S ( a , r ) and the integral value r , b corresponding to the sub-interval S ( r , b ) as the target integral value, and determine the sum of the target integral values S ( a , r ) + S ( r , b ) as the crack width.

[0129] Optionally, in the above S ( a , b ) and S ( a , r ) + S ( r , b ) when the first difference is greater than or equal to the second preset threshold, it indicates that the calculation accuracy of the current integral value has not reached the preset requirement. Therefore, the electronic device can respectively perform operations such as further dividing, integrating, and comparing integral differences on the sub-intervals a , r and r , b until the first difference between the sum of the integral values corresponding to each two sub-intervals obtained by further division and the integral value corresponding to the corresponding parent interval is less than the second preset threshold, then stop further dividing the interval, and determine the integral values corresponding to the sub-intervals with the first difference less than the second preset threshold as the target integral values.

[0130] Exemplarily, for the sub-interval a , r , the electronic device can perform the following operations: Calculate the midpoint of the sub-interval a , r r 1 = ( a + r ) / 2, and according to the midpoint a , r of the sub-interval r 1, divide the sub-interval into two sub-intervals a , r 1] and r 1, r ; Use the above adaptive Simpson's algorithm to perform integral operations on the continuous strain function within the sub-intervals a , r 1] and r , r to obtain the integral value a , r 1] corresponding to the sub-interval S ( a , r 1) and the integral value r 1, r corresponding to the sub-interval S ( r , r ); When the sum of the integral value corresponding to the sub-interval a , r 1] and the integral value corresponding to the sub-interval r , r S ( a , r 1) + S ( r , r ) and the integral value corresponding to the corresponding parent interval a , r has a first difference less than the second preset threshold, it indicates that the calculation accuracy of the current integral value has reached the preset requirement. Therefore, the electronic device can stop re-dividing the sub-intervals a , r 1] and r , r .

[0131] Exemplarily, for the sub-interval r , b , the electronic device can perform the following operations: Calculate the midpoint of the sub-interval r , b ​r 2 = ([[]] r + b ) / 2, and according to the midpoint of the sub - interval r , b r 2, the sub - interval is further divided into two sub - intervals r , r 2] and r 2, b ; The above - mentioned adaptive Simpson algorithm is used to perform integral operations on the continuous strain function in the sub - intervals r , r 2] and r 2, b , and the integral values r , r 2] corresponding to the sub - interval are obtained S ( r , r 2) and the integral values r 2, b corresponding to the sub - interval S ( r 2, b ); In the case where the sum of the integral values r , r 2] corresponding to the sub - interval and the integral values r 2, b corresponding to the sub - interval S ( r , r 2) + S ( r 2, b ) and the first difference of the integral value corresponding to the corresponding mother - interval r , b is less than the second preset threshold, it indicates that the calculation accuracy of the current integral value has reached the preset requirement. Therefore, the electronic device can stop further dividing the sub - sub - intervals r , r 2] and r 2, b .

[0132] Based on this, the electronic device can take S ( a , r 1), S ( r 1, r ), S ( r , r 2) and S ( r 2, b ) as the target integral values, and the sum of the target integral valuesS ( a , r 1)+ S ( r 1, r )+ S ( r , r 2)+ S ( r 2, b ) is determined as the crack width.

[0133] By performing progressive integration operations in this way, the accuracy of the integral value estimation can be gradually improved. When all the first differences are less than the second preset threshold, the sum of the integral values corresponding to each sub-interval where the corresponding first difference is less than the second threshold is determined as the crack width, which can improve the accuracy of crack width calculation.

[0134] As can be seen from the above, in the embodiment of the present application, by determining the slope of the strain value change rate corresponding to each sampling position, the position interval where the crack is located can be roughly determined according to the signs of the slopes; on this basis, in order to more accurately locate the crack, all strain values are divided into multiple strain observation sequences according to the slopes. Since the strain distribution characteristics corresponding to the strain observation sequences can accurately reflect the strain value distribution in the strain observation sequences, by determining the strain distribution characteristics corresponding to each strain observation sequence and according to the strain distribution characteristics corresponding to all strain observation sequences, the target observation sequence containing crack information can be accurately determined from the strain observation sequences. Furthermore, according to the strain values in the target observation sequence, the target position where the crack is located can be accurately determined, thereby improving the accuracy and efficiency of crack detection. At the same time, by determining the target integration interval where the target position is located and using the adaptive Simpson algorithm to perform progressive integration operations on the continuous strain function in the target integration interval with the midpoint of the interval as the interval dividing line, the estimation accuracy of the integral value can be improved; since the crack width is the sum of multiple target integral values obtained by integration, the accuracy of crack width calculation can be improved.

[0135] In addition, since this crack detection method only needs to deploy distributed optical fiber sensors on the target structure to realize automatic monitoring of cracks on the target structure without the need for manual regular inspections, not only the crack monitoring efficiency is improved, but also the labor cost is reduced.

[0136] It can be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean 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 to the implementation process of the embodiments of the present application.

[0137] Based on the crack detection method provided in the above embodiments, an embodiment of an electronic device for implementing the method embodiments above is further given in an embodiment of the present application. Please refer to Figure 8 , which is a schematic structural diagram of an electronic device provided in an embodiment of the present application. For ease of description, only parts related to this embodiment are shown. As Figure 8 shown, the electronic device 80 may include: a first determination unit 801, a second determination unit 802, a third determination unit 803, a fourth determination unit 804, and a fifth determination unit 805. Among them: The first determination unit 801 is configured to determine the slope of the strain value change rate corresponding to each sampling position according to the strain values of the target structure at each sampling position collected by distributed optical fiber sensing.

[0138] The second determination unit 802 is configured to determine the strain observation sequence corresponding to each sampling position according to the slope of each strain value change rate; the strain observation sequence includes the strain values corresponding to a continuous plurality of sampling positions.

[0139] The third determination unit 803 is configured to determine the strain distribution characteristics corresponding to each strain observation sequence, and determine the target observation sequence from the strain observation sequences according to the strain distribution characteristics corresponding to all the strain observation sequences.

[0140] The fourth determination unit 804 is configured to determine the target position where the crack is located based on the strain values in the target observation sequence.

[0141] The fifth determination unit 805 is configured to determine the target integration interval according to the target position, and use the adaptive Simpson algorithm to perform a progressive integration operation on the continuous strain function within the target integration interval with the midpoint of the interval as the interval dividing line, and determine the sum of the multiple target integral values obtained by integration as the width of the crack; the continuous strain function is constructed according to the target observation sequence.

[0142] Optionally, the first determination unit 801 is specifically configured to: For the i th sampling position, according to the strain values of the target structure at the i th sampling position, the i +1th sampling position, and the i +2th sampling position, calculate the second derivative of the strain value corresponding to the i th sampling position, and determine the second derivative as the slope of the strain value change rate corresponding to the i th sampling position; 1≤ i ≤ n -2, n is the number of sampling positions, i is an integer.

[0143] Optionally, the second determination unit 802 is specifically configured to: For the first sampling position, determine the first strain value sequence composed of the strain values corresponding to the consecutive first number of sampling positions including the first sampling position as the strain observation sequence corresponding to the first sampling position; For the j th sampling position, when the slope of the strain value change rate corresponding to the j th sampling position is negative and the slope of the strain value change rate corresponding to the j -1th sampling position is positive, determine the second strain value sequence composed of the strain values corresponding to the consecutive second number of sampling positions including the j th sampling position as the strain observation sequence corresponding to the j th sampling position; For the j th sampling position, when the difference between the slope corresponding to the j th sampling position and the slope corresponding to the j -th sampling position is less than the first preset threshold, determine the third strain value sequence composed of the strain values corresponding to the consecutive third number of sampling positions including the j th sampling position as the strain observation sequence corresponding to the j th sampling position; where 2 ≤ j ≤ n , n is the number of sampling positions, j is an integer; the second number is greater than the first number, and the third number is less than the first number.

[0144] Optionally, the third determination unit 803 is specifically configured to: Perform dimensionality reduction processing on the strain distribution characteristics corresponding to each of the strain observation sequences to obtain the low-dimensional strain data corresponding to each of the strain observation sequences; Perform clustering processing on all the low-dimensional strain data, and determine the strain observation sequences under the target category in the clustering result as the target observation sequences; the target category refers to the category containing crack information.

[0145] Optionally, the fourth determination unit 804 is specifically configured to: Calculate the first-order backward difference sequence of the target observation sequence; The unit step function is used to process the first-order backward difference sequence to obtain a binary sequence corresponding to the first-order backward difference sequence; each element in the binary sequence corresponds to each element in the first-order backward difference sequence, and the value of each element in the binary sequence is 0 or 1; All elements in the binary sequence are shifted one bit to the right to obtain a right-shifted sequence corresponding to the binary sequence, and an exclusive OR operation is performed on the binary sequence and the right-shifted sequence to obtain a target sequence; The sampling positions corresponding to the elements with a value of 1 in the target sequence are determined as the target positions.

[0146] Optionally, the fifth determination unit 805 is specifically configured to: The moving average algorithm is used to smooth each strain value in the target observation sequence to obtain a smoothed strain value sequence corresponding to the target observation sequence; Construct a continuous strain function corresponding to the smoothed strain value sequence; A target integration interval including the target position is selected from the sampling position interval corresponding to the continuous strain function.

[0147] Optionally, the fifth determination unit 805 is further specifically configured to: The adaptive Simpson algorithm is used to perform an integration operation on the continuous strain function within the target integration interval to obtain an integration value corresponding to the target integration interval; Taking the midpoint of the target integration interval as the interval division line, the target integration interval is divided into two sub-intervals, and the adaptive Simpson algorithm is used to perform an integration operation on the continuous strain function within the two sub-intervals respectively to obtain the integration values corresponding to the two sub-intervals respectively; When the first difference between the integration value corresponding to the target integration interval and the sum of the integration values corresponding to the two sub-intervals is less than a second preset threshold, stop re-dividing the sub-intervals, and determine the integration values corresponding to the two sub-intervals respectively as the target integration value; When the first difference is greater than or equal to the second preset threshold, taking the midpoint of each sub-interval as the interval division line, each sub-interval is re-divided respectively, and the adaptive Simpson algorithm is used to calculate the integration values corresponding to the sub-intervals obtained by re-division until the first difference between the sum of the integration values corresponding to every two sub-intervals obtained by re-division and the integration value corresponding to the corresponding parent interval is less than the second preset threshold, stop performing interval re-division, and determine the integration values corresponding to the sub-intervals with the corresponding first difference less than the second preset threshold as the target integration value.

[0148] Those skilled in the art can clearly understand that, for the convenience and conciseness of description, only the above division of each functional unit is used as an example. In actual applications, the above functions can be allocated to different functional units as needed, that is, the internal structure of the electronic device is divided into different functional units to complete all or part of the functions described above. Each functional unit in the embodiment can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit are only for the convenience of mutual distinction and do not limit the protection scope of this application. The specific working processes of each unit in the above electronic device can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0149] Please refer to Figure 9 , Figure 9 which is a schematic structural diagram of an electronic device provided by another embodiment of this application. As Figure 9 shown, the electronic device 9 provided in this embodiment may include: a processor 90, a memory 91, and a computer program 92 stored in the memory 91 and executable on the processor 90, such as the program corresponding to the crack detection method. When the processor 90 executes the computer program 92, the steps in the above crack detection method embodiment are implemented, such as Figure 2 the S201~S205 shown. Or when the processor 90 executes the computer program 92, the functions of each module / unit in the above electronic device embodiment are implemented, such as Figure 8 the functions of the units 801~805 shown.

[0150] Exemplarily, the computer program 92 can be divided into one or more modules / units. One or more modules / units are stored in the memory 91 and executed by the processor 90 to complete this application. One or more modules / units can be a series of computer program instruction segments capable of completing specific functions, and these instruction segments are used to describe the execution process of the computer program 92 in the electronic device 9. For example, the computer program 92 can be divided into a first determination unit, a second determination unit, a third determination unit, a fourth determination unit, and a fifth determination unit. For the specific functions of each unit, please refer to Figure 8 the relevant descriptions in the corresponding embodiments and will not be elaborated here.

[0151] Those skilled in the art can understand that Figure 9 this is only an example of the electronic device 9 and does not constitute a limitation on the electronic device 9. It may include more or fewer components than shown in the figure, or combine some components, or different components.

[0152] The processor 90 may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor may also be any conventional processor, etc.

[0153] The memory 91 may be an internal storage unit of the electronic device 9, such as the hard disk or memory of the electronic device 9. The memory 91 may also be an external storage device of the electronic device 9, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, or flash card equipped on the electronic device 9, etc. Further, the memory 91 may also include both the internal storage unit and the external storage device of the electronic device 9. The memory 91 is used to store computer programs and other programs and data required by the electronic device. The memory 91 may also be used to temporarily store data that has been output or is to be output.

[0154] The embodiment of the present application also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, each step in the crack detection method in the above method embodiment is implemented.

[0155] The embodiment of the present application provides a computer program product, and when the computer program product runs on an electronic device, the electronic device is enabled to implement the steps in the above various method embodiments.

[0156] In the above embodiments, the descriptions of the respective embodiments have their own focuses. For parts not detailed or recorded in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0157] It should be noted that unless otherwise specified, all technical terms used in the embodiments of the present application have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs. The technical terms used in the embodiments of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0158] As used in the description of the embodiments of the present application, the term "embodiment" means that the specific features, structures, or characteristics described in connection with the embodiment may be included in at least one embodiment of the present application. The phrase appearing in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein may be combined with other embodiments.

[0159] Those of ordinary skill in the art will realize that the units and algorithm steps of the examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled professionals may use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0160] The above-described embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements 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 application, and should all be included in the protection scope of the present application.

Claims

1. A crack detection method, characterized in that, Including: Determine the slope of the strain value change rate corresponding to each sampling position according to the strain values of the target structure collected by distributed optical fiber sensing at each sampling position; Determine the strain observation sequence corresponding to each sampling position according to the slopes of the strain value change rates; The strain observation sequence includes the strain values corresponding to a continuous plurality of sampling positions; Determine the strain distribution characteristics corresponding to each strain observation sequence, and determine the target observation sequence from the strain observation sequences according to the strain distribution characteristics corresponding to all the strain observation sequences; Based on the strain values in the target observation sequence, determine the target position where the crack is located; Determine the target integration interval according to the target position, and use the adaptive Simpson algorithm to perform a progressive integration operation on the continuous strain function within the target integration interval with the midpoint of the interval as the interval dividing line, and determine the sum of the multiple target integral values obtained by the integration as the width of the crack; The continuous strain function is constructed according to the target observation sequence.

2. The method according to claim 1, characterized in that, Determine the slope of the strain value change rate corresponding to each sampling position according to the strain values of the target structure collected by distributed optical fiber sensing, including: For the i th sampling position, according to the strain values at the i th sampling position, the i +1 th sampling position, and the i +2 th sampling position, calculate the second derivative of the strain value corresponding to the i th sampling position, and determine the second derivative as the slope of the strain value change rate corresponding to the i th sampling position; 1 ≤ i ≤ n - 2, n is the number of sampling positions, i is an integer.

3. The method according to claim 1, characterized in that, Determine the strain observation sequence corresponding to each sampling position according to the slopes of the strain value change rates, including: For the first sampling position, determine the first strain value sequence composed of the strain values corresponding to the continuous first number of sampling positions including the first sampling position as the strain observation sequence corresponding to the first sampling position; For the j th sampling position, when the slope of the strain value change rate corresponding to the j th sampling position is negative, and the slope of the strain value change rate corresponding to the j -1th sampling position is positive, the second strain value sequence composed of the strain values corresponding to the consecutive second number of sampling positions including the j th sampling position is determined as the strain observation sequence corresponding to the j th sampling position; For the j th sampling position, when the difference between the slope corresponding to the j th sampling position and the slope corresponding to the j -th sampling position is less than the first preset threshold, a third strain value sequence composed of strain values corresponding to a continuous third number of sampling positions including the j th sampling position is determined as the strain observation sequence corresponding to the j th sampling position; where 2 ≤ j ≤ n , n is the number of sampling positions, j is an integer; the second quantity is greater than the first quantity, and the third quantity is less than the first quantity.

4. The method according to claim 1, characterized in that, Determine the target observation sequence from the strain observation sequences according to the strain distribution characteristics corresponding to all the strain observation sequences, including: Perform dimensionality reduction processing on the strain distribution characteristics corresponding to each strain observation sequence to obtain the low-dimensional strain data corresponding to each strain observation sequence; Perform clustering processing on all the low-dimensional strain data, and determine the strain observation sequences under the target category in the clustering result as the target observation sequence; the target category refers to the category containing crack information.

5. The method according to claim 1, characterized in that, Based on the strain values in the target observation sequence, determine the target position where the crack is located, including: Calculate the first-order backward difference sequence of the target observation sequence; Process the first-order backward difference sequence using the unit step function to obtain the binary sequence corresponding to the first-order backward difference sequence; each element in the binary sequence corresponds to each element in the first-order backward difference sequence, and the value of each element in the binary sequence is 0 or 1; Shift all the elements in the binary sequence one bit to the right to obtain the right-shifted sequence corresponding to the binary sequence, and perform an exclusive OR operation on the binary sequence and the right-shifted sequence to obtain the target sequence; Determine the sampling positions corresponding to the elements with a value of 1 in the target sequence as the target position.

6. The method according to any one of claims 1 - 5, characterized in that, Determine the target integration interval according to the target position, including: Perform smoothing processing on the strain values in the target observation sequence using the moving average algorithm to obtain the smoothed strain value sequence corresponding to the target observation sequence; Construct the continuous strain function corresponding to the smoothed strain value sequence; Select a target integration interval that contains the target position from the sampling position intervals corresponding to the continuous strain function.

7. The method according to any one of claims 1 - 5, characterized in that, Using the adaptive Simpson's algorithm, perform a progressive integration operation on the continuous strain function within the target integration interval with the midpoint of the interval as the interval dividing line, including: Using the adaptive Simpson's algorithm to perform an integration operation on the continuous strain function within the target integration interval to obtain the integral value corresponding to the target integration interval; Taking the midpoint of the target integration interval as the interval dividing line, divide the target integration interval into two sub-intervals, and use the adaptive Simpson's algorithm to perform integration operations on the continuous strain function within the two sub-intervals respectively to obtain the integral values corresponding to the two sub-intervals respectively; When the first difference between the integral value corresponding to the target integration interval and the sum of the integral values corresponding to the two sub-intervals is less than the second preset threshold, stop re-dividing the sub-intervals, and determine the integral values corresponding to the two sub-intervals respectively as the target integral value; When the first difference is greater than or equal to the second preset threshold, taking the midpoint of each sub-interval as the interval dividing line, re-divide each sub-interval respectively, and use the adaptive Simpson's algorithm to calculate the integral values corresponding to the sub-intervals obtained by re-division until the first difference between the sum of the integral values corresponding to each two sub-intervals obtained by re-division and the integral value corresponding to the corresponding parent interval is less than the second preset threshold, stop performing interval re-division, and determine the integral values corresponding to the sub-intervals whose corresponding first difference is less than the second preset threshold as the target integral value.

8. An electronic device, characterized in that, Including: A first determination unit for determining the slope of the strain value change rate corresponding to each sampling position according to the strain values of the target structure collected by distributed optical fiber sensing; A second determination unit for determining the strain observation sequence corresponding to each sampling position according to the slopes of the strain value change rates; The strain observation sequence includes the strain values corresponding to a continuous plurality of sampling positions; A third determination unit for determining the strain distribution characteristics corresponding to each strain observation sequence, and determining a target observation sequence from the strain observation sequences according to the strain distribution characteristics corresponding to all the strain observation sequences; A fourth determination unit for determining the target position where the crack is located based on the strain values in the target observation sequence; A fifth determination unit for determining a target integration interval according to the target position, and using the adaptive Simpson's algorithm to perform a progressive integration operation on the continuous strain function within the target integration interval with the midpoint of the interval as the interval dividing line, and determining the sum of the multiple target integral values obtained by integration as the width of the crack; the continuous strain function is constructed according to the target observation sequence.

9. An electronic device, characterized in that, Including a memory and a computer program stored in the memory and executable on a processor, and when the processor executes the computer program, the method described in any one of claims 1-7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, the crack detection method described in any one of claims 1-7 is implemented.

Images