A method for evaluating the characteristics of wood damage evolution based on active acoustic emission

Abstract

The present invention discloses a method for evaluating the damage evolution characteristics of wood based on active acoustic emission, which relates to the technical field of physical analysis and includes the following steps: The wood to be detected is uniformly loaded through a four-point loading device, and an acoustic emission sensor is used to collect the acoustic emission waveforms at each detection point to obtain the acoustic emission parameters of the wood to be detected during the damage stage, and the damage characteristic index of the wood to be detected is analyzed. From this, the damage modes of the wood to be detected are classified, and the RA-AF correlation analysis method is used to quantitatively analyze the damage modes of the wood to be detected to obtain the damage forms of the wood to be detected during the loading process, and the damage evolution process of the wood to be detected is analyzed and transmitted to the cloud information warehouse for storage. By adopting the active acoustic emission method, the present invention generates acoustic emission events inside the material through an externally applied load, so that it can more flexibly adapt to different environments and requirements, and the acoustic emission data is processed by using a clustering method to reduce human intervention and improve the objectivity and consistency of classification.

Description

Technical Field

[0001] The present invention relates to the technical field of physical analysis, and specifically to a method for evaluating the evolution characteristics of wood damage based on active acoustic emission. Background Art

[0002] As an important engineering material, wood is widely used in fields such as construction, furniture, and packaging. However, during use, wood is often affected by various external factors, resulting in damage and performance degradation. Traditional wood damage assessment methods often have limitations and cannot accurately and real-time reflect the evolution process of wood damage. Acoustic emission technology can collect transient elastic waves generated during internal material failure to obtain damage information during the material failure process in real-time and accurately.

[0003] The prior art, such as the invention patent with the publication number: CN110196283A, is a non-destructive testing method for wood structure damage based on instantaneous frequency, including: (1) taking structural lumber as the research object, installing corresponding sensors to establish an acoustic emission acquisition system for wood bending damage, and obtaining the acoustic emission signals during the wood damage process; (2) filtering and wavelet decomposition of the collected acoustic emission signals to achieve preprocessing of the original signals. (3) Performing EMD decomposition on the wavelet-reconstructed signals to obtain the acoustic emission waveforms for Hilbert transform. (4) Determining the characteristic frequencies of different acoustic emission events based on the frequency domain characteristics of the acoustic emission reconstructed waveforms. (5) Counting the number of different types of acoustic emission events through instantaneous frequency and calculating the corresponding event occurrence density. Finally, using the acoustic emission event occurrence density and its change to evaluate the stress state during the wood damage process. This method can perform real-time dynamic damage monitoring and identification of internal damage in wood structure buildings.

[0004] The prior art, such as the invention patent with the bulletin number: CN104634878B, is a wood damage monitoring method based on acoustic emission technology, including: placing acoustic emission sensors on the surface of the wood at the stress concentration site to collect weak acoustic emission signals generated by wood damage; amplifying, filtering, and analog-to-digital converting the collected acoustic emission signals and transmitting them to the FIFO inside the FPGA control module for caching, and then transmitting the data to the upper computer through a wireless transmission module; the upper computer performs wavelet denoising and reconstruction on the collected acoustic emission signals, locates the position of the wood damage acoustic emission source using a linear positioning method, and monitors the position of the wood damage acoustic emission source in real-time through a human-computer interaction interface designed by LABVIEW; the upper computer uses a wavelet packet analysis system to extract energy characteristic values from the collected acoustic emission signals, constructs a corresponding training sample set, constructs a neural network, and predicts and analyzes the change trend of the wood stress damage acoustic emission signals through cumulative energy, so as to infer the wood damage position.

[0005] As can be seen from the above solution, the current acoustic emission parameter analysis method only analyzes by selecting a certain characteristic parameter, and cannot comprehensively obtain the law of wood damage evolution. Moreover, there is a lack of quantitative analysis of the damage modes at each stage of wood. Summary of the Invention

[0006] Aiming at the deficiencies of the prior art, the present invention provides a method for evaluating the characteristics of wood damage evolution based on active acoustic emission, which can effectively solve the problems involved in the above background technology.

[0007] To achieve the above object, the present invention is realized through the following technical solutions: The present invention provides a method for evaluating the characteristics of wood damage evolution based on active acoustic emission, including the following steps:

[0008] S1, the wood to be detected is uniformly loaded through a four-point loading device, and at the same time, an acoustic emission sensor is used to collect the acoustic emission waveforms at each detection point, and data processing is performed to obtain the acoustic emission parameters at the wood damage stage to be detected.

[0009] S2, perform clustering processing on the acoustic emission parameters at the wood damage stage to be detected, analyze and obtain the damage characteristic index of the wood to be detected, and thereby classify the damage modes of the wood to be detected.

[0010] S3, use the RA-AF correlation analysis method to quantitatively analyze the damage mode of the wood to be detected, and obtain the damage form of the wood to be detected during the loading process.

[0011] S4, according to the damage form of the wood to be detected during the loading process, analyze the damage evolution process of the wood to be detected and transmit it to the cloud information warehouse for storage.

[0012] As a preferred technical solution, the process of using the acoustic emission sensor to collect the acoustic emission waveforms at each detection point is as follows: Set a test period, and extract and collect the acoustic emission waveforms at each detection point through the acoustic emission sensor during the test period. After the acoustic emission waveforms at each detection point pass through amplifier discharge and filter filtering, they are not cut any further and are directly stored in the computer storage device.

[0013] As a preferred technical solution, the acoustic emission parameters at the wood damage stage to be detected are specifically: According to the acoustic emission waveforms at each detection point, extract the typical characteristic parameters in the acoustic emission waveforms, and obtain the pre-judged damage mode of the wood to be detected during the test period and combine them as the acoustic emission parameters at the wood damage stage to be detected.

[0014] The typical characteristic parameters specifically include the ring count, amplitude, energy, and rise time at each detection point.

[0015] As a preferred technical solution, the clustering process of the acoustic emission parameters in the detection of wood damage stages is as follows: Obtain the acoustic emission parameters in the detection of wood damage stages, randomly select K points from each detection point as clustering centers, mark them as K clustering centers, and statistically analyze the typical feature vectors of the K clustering centers, and statistically analyze the typical feature parameters of each detection point to obtain the typical feature vectors of each detection point.

[0016] Statistically analyze the distances between the typical feature vectors of each detection point and the typical feature vectors of the K clustering centers respectively, and sequentially assign each detection point to the category of the clustering center with the closest distance, thereby statistically analyzing all detection points in the K clustering centers.

[0017] Calculate the average value of all detection points in the K clustering centers, solve for the new K clustering centroids, and compare them with the K clustering centroids obtained in the previous calculation. When the centroids do not change, stop and output the clustering results, and the clustering results include typical feature clustering vectors.

[0018] As a preferred technical solution, the analysis to obtain the damage feature index of the detected wood is as follows: Extract the reference typical feature parameters stored in the cloud database, including reference ring counts, reference amplitudes, reference energies, and reference rise times.

[0019] Statistically analyze the ring counts, amplitudes, energies, and rise times of each detection point, and analyze and process them to obtain the damage feature index of the detected wood.

[0020] The damage feature index of the detected wood is specifically obtained by analyzing and processing the acoustic emission parameters in the detection of wood damage stages, and is used to reflect the damage characteristics of the detected wood and as a numerical classification basis for detecting the wood damage mode.

[0021] As a preferred technical solution, the classification of the detected wood damage mode is as follows: According to the damage feature index of the detected wood, compare it with the damage feature index boundary value stored in the cloud database. If the damage feature index of the detected wood is lower than the damage feature index boundary value, then classify the wood damage mode.

[0022] The classification of the wood damage mode includes statistically analyzing the wood damage mode in the detection test cycle, the typical feature clustering vectors and typical feature parameters under this mode.

[0023] As a preferred technical solution, the RA-AF correlation analysis method is used to distinguish the shear cracks of the detected wood and the tensile cracks of the detected wood.

[0024] As a preferred technical solution, the process of obtaining the damage form of the tested wood during the loading process is as follows: Define the ratio of the rising time of each detection point of the tested wood to the amplitude of each detection point of the tested wood as the acoustic emission parameter RA, and define the ratio of the ring count of each detection point of the tested wood to the set detection period duration as the acoustic emission parameter AF.

[0025] According to the acoustic emission parameter RA and the acoustic emission parameter AF, a quantitative analysis of the damage mode of the tested wood is carried out.

[0026] When the tested wood has the characteristics of high RA and low AF, the damage form of the tested wood is shear-type crack; when the tested wood has the characteristics of low RA and high AF, the damage form of the tested wood is tensile-type crack.

[0027] As a preferred technical solution, the damage evolution process of the tested wood specifically includes: an initial stage, a linear damage accumulation stage, a non-linear damage accumulation stage, and a failure stage.

[0028] In the initial stage, specifically, some microscopic changes will occur inside the tested wood, such as the generation of microcracks or small-scale deformation on the surface.

[0029] In the linear damage accumulation stage, specifically, with the action of an external load, the tested wood will enter the linear damage accumulation stage, and the damage usually increases linearly with the increase of the load.

[0030] In the non-linear damage accumulation stage, specifically, when the tested wood is under a higher load, it will enter the non-linear damage accumulation stage.

[0031] In the failure stage, specifically, it is the failure form or failure morphology presented by the tested wood when it bears an external load. When the tested wood is subjected to an excessive external load, it will enter the failure stage.

[0032] As a preferred technical solution, the damage characteristic index of the tested wood is specifically calculated by the following formula:

[0033]

[0034] Where A is the damage characteristic index of the tested wood, α i is the ring count of the i-th detection point of the tested wood, α0 is the reference ring count, β i is the amplitude of the i-th detection point of the tested wood, β0 is the reference amplitude, γ i is the energy of the i-th detection point of the tested wood, γ0 is the reference energy, δ iTo detect the rise time of the i-th detection point of the wood, δ0 is the reference rise time, x1 is the preset ringing count weight, x2 is the preset amplitude weight, x3 is the preset energy weight, x4 is the preset rise time weight, i is the number of each detection point of the wood to be detected, i = 1, 2,..., n, n is the number of detection points of the wood to be detected, and e is the natural constant.

[0035] Compared with the prior art, the embodiments of the present invention have at least the following beneficial effects:

[0036] (1) By providing a method for evaluating the damage evolution characteristics of wood based on active acoustic emission, the method of clustering acoustic emission characteristic parameters can more accurately classify the wood damage patterns. At the same time, based on the propagation principle of acoustic emission signals in wood, we can deeply understand the propagation process and characteristics of acoustic emission signals inside the wood, quantitatively analyze the evolution law of wood damage, and thus more clearly grasp the development process and trend of wood damage.

[0037] (2) The present invention adopts the active acoustic emission method to generate acoustic emission events inside the material by applying an external load, so as to be able to more flexibly adapt to different environments and requirements.

[0038] (3) By processing the acoustic emission data using the clustering method, the present invention does not require prior knowledge and subjective judgment, reduces human intervention, and improves the objectivity and consistency of classification.

[0039] (4) The present invention adopts the RA-AF correlation analysis method to quantify the wood damage patterns detected at different stages, which is beneficial to analyzing the damage evolution law of the wood to be detected under the action of loading.

[0040] Of course, it is not necessary for any product implementing the present invention to achieve all the above-mentioned advantages simultaneously. Description of the Drawings

[0041] Figure 1 is a schematic flow chart of the method of the present invention;

[0042] Figure 2 is a schematic flow chart of the clustering analysis related to the embodiments of the present invention;

[0043] Figure 3 is a schematic diagram of the principle of acoustic emission signal generation related to the embodiments of the present invention;

[0044] Figure 4 is a schematic diagram of the acoustic emission waveform data related to the embodiments of the present invention;

[0045] Figure 5 is a structural diagram of the active acoustic emission system related to the embodiments of the present invention;

[0046] Figure 6 This is the classification diagram of tensile cracks and shear cracks involved in the embodiments of the present invention. Specific embodiments

[0047] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0048] Please refer to Figure 1 As shown, the embodiments of the present invention provide a method for evaluating the characteristics of wood damage evolution based on active acoustic emission, including the following steps: uniformly loading the wood to be detected through a four-point loading device, and simultaneously using an acoustic emission sensor to collect the acoustic emission waveforms of each detection point, and performing data processing to obtain the acoustic emission parameters at the wood damage stage.

[0049] In this embodiment, the detection technology adopted is the active acoustic emission method. The active acoustic emission technology is an advanced detection means that actively applies an excitation to the object to be measured, generates elastic waves, and collects, analyzes, and processes these elastic waves to obtain information about the characteristics of wood damage evolution. This technology can monitor the internal damage of wood in real time and dynamically, providing strong support for accurately evaluating the evolution process of wood damage.

[0050] In a specific embodiment, the principle of generating acoustic emission signals is as follows: when the internal stress of a material exceeds the yield limit strength under the action of external conditions, resulting in plastic deformation or micro-changes such as crack initiation and propagation, transient stress waves are generated accompanied by energy release. Acoustic emission is a common physical phenomenon. Due to the existence of defects or non-uniform distribution of microstructures in local areas inside the material, when the material is subjected to external loads, stress concentration will occur at local positions, causing the internal defects of the material to yield and deform. The energy of the material always transitions from an unstable high-energy state to a low-energy state, attempting to make the energy of the material tend to the lowest energy state. At this time, the unstable energy concentrated inside the material is released in a certain form, thus generating acoustic emission signals.

[0051] For easy understanding, the principle of generating acoustic emission signals in this embodiment is shown in Figure 3 as shown.

[0052] It should be understood that when a material or component generates elastic waves under the action of an external excitation, the elastic waves are transmitted to the surface of the detection object, causing mechanical vibrations on the surface of the object. The displacement change on the surface of the object caused by this mechanical vibration can be converted into an identifiable and analyzable signal through a sensor and output as a waveform.

[0053] In this embodiment, the process of collecting the acoustic emission waveforms at each detection point using an acoustic emission sensor is as follows: Set a test period, and extract and collect the acoustic emission waveforms at each detection point through the acoustic emission sensor during the test period. After the acoustic emission waveforms at each detection point are discharged by an amplifier and filtered by a filter, they are directly stored in a computer storage device without further cutting.

[0054] For ease of understanding, an example of the waveform data stored in the computer in this embodiment is as Figure 4 shown.

[0055] It should be understood that the detection points specifically detect each part of the wood to be detected, and the detected parts of the wood to be detected are marked as detection points.

[0056] In this embodiment, the active acoustic emission method is adopted, and acoustic emission events are generated inside the material by applying an external load, so that different environments and requirements can be more flexibly adapted.

[0057] Please refer to Figure 5 shown. The structure of the active acoustic emission system involved in this embodiment includes a sensor, a preamplifier, an acoustic emission analyzer, and a computer.

[0058] In this embodiment, the process of obtaining the acoustic emission parameters at the damage stage of the wood to be detected is as follows: According to the acoustic emission waveforms at each detection point, typical characteristic parameters in the acoustic emission waveforms are extracted, and the pre-judged damage mode of the wood to be detected during the test period is obtained and jointly used as the acoustic emission parameters at the damage stage of the wood to be detected.

[0059] Specifically, the typical characteristic parameters specifically include the ring count, amplitude, energy, and rise time at each detection point.

[0060] In this embodiment, the ring count is specifically the number of oscillations of the signal exceeding the threshold, the amplitude is specifically the maximum peak value of a single-frame signal, with the unit of decibel, the energy is specifically the envelope area of the oscillation within the duration of a single-frame signal, which can reflect the relative energy and intensity of the event, and the rise time is specifically the time difference between the moment corresponding to the maximum peak value and the moment when the signal first exceeds the threshold.

[0061] Cluster processing is performed on the acoustic emission parameters at the damage stage of the wood to be detected, and the damage characteristic index of the wood to be detected is analyzed, thereby classifying the damage mode of the wood to be detected.

[0062] In a specific embodiment, the cluster analysis is a multivariate statistical method for studying the classification of samples or indicators. The so-called "class" is a set of similar elements, and "classification" is to assign an observation object to a certain class.

[0063] In this embodiment, the clustering process of the acoustic emission parameters in the wood damage detection stage is as follows: Obtain the acoustic emission parameters in the wood damage detection stage. Randomly select K points among the detection points as the clustering centers, denoted as K clustering centers, and statistically analyze the typical feature vectors of the K clustering centers, and statistically analyze the typical feature parameters of each detection point to obtain the typical feature vectors of each detection point.

[0064] Statistically analyze the distances between the typical feature vectors of each detection point and the typical feature vectors of the K clustering centers, and sequentially assign each detection point to the category of the nearest clustering center, thereby statistically analyzing all the detection points in the K clustering centers.

[0065] Calculate the average value of all the detection points in the K clustering centers to obtain a new K clustering centroid, and compare it with the K clustering centroid obtained in the previous calculation. When the centroid does not change, stop and output the clustering result, and the clustering result includes the typical feature clustering vector.

[0066] It should be understood that the reasons for the change of the centroid mainly include the uncertainty of the initial clustering and the randomness of the data point distribution.

[0067] The uncertainty of the initial clustering specifically refers to that the initially selected clustering centers may not be optimal, and as the iteration progresses, the clustering centers will gradually adjust to more appropriate positions.

[0068] The randomness of the data point distribution specifically refers to that the distribution of sample data points in different clusters may be relatively random and complex, which will affect the position of the centroid.

[0069] In this embodiment, by processing the acoustic emission data using the clustering method, prior knowledge and subjective judgment are not required, reducing human intervention and improving the objectivity and consistency of classification.

[0070] In this embodiment, the analysis to obtain the damage feature index of the detected wood is as follows: Extract the reference typical feature parameters stored in the cloud database, including the reference ring count, reference amplitude, reference energy, and reference rise time.

[0071] It should be noted that the cloud database is used to store reference index data, and the reference index data includes reference typical feature parameters and damage feature index boundary values.

[0072] The reference typical feature parameters include the reference ring count, reference amplitude, reference energy, and reference rise time.

[0073] It should be noted that the reference index data is collected and verified by relevant personnel and obtained after review, and can be used as the reference data for wood damage determination after verification.

[0074] Count the ringing count, amplitude, energy, and rise time at each detection point, and analyze and process them to obtain the damage characteristic index of the detected wood.

[0075] The damage characteristic index of the detected wood is specifically obtained by analyzing and processing the acoustic emission parameters in the damage stage of the detected wood, which is used to reflect the damage characteristics of the detected wood and is used as the numerical classification basis for the damage mode of the detected wood.

[0076] In this embodiment, the damage characteristic index of the detected wood can be obtained not only by performing principal component analysis on the detected wood, but also by calculation. The specific formula is as follows:

[0077]

[0078] where A is the damage characteristic index of the detected wood, α i is the ringing count at the i-th detection point of the detected wood, α0 is the reference ringing count, β i is the amplitude at the i-th detection point of the detected wood, β0 is the reference amplitude, γ i is the energy at the i-th detection point of the detected wood, γ0 is the reference energy, δ i is the rise time at the i-th detection point of the detected wood, δ0 is the reference rise time, x1 is the preset weight value of the ringing count, x2 is the preset weight value of the amplitude, x3 is the preset weight value of the energy, x4 is the preset weight value of the rise time, i is the number of each detection point of the detected wood, i = 1, 2,..., n, n is the number of detection points of the detected wood, and e is the natural constant.

[0079] It should be noted that the preset weight value of the ringing count needs to be determined according to factors such as the damage type and material microstructure of the detected wood, and its value range is between 0 and 0.5;

[0080] The preset weight value of the amplitude needs to consider the strength, fracture toughness, etc. of the detected wood, and its value range is between 0.2 and 0.4;

[0081] The preset weight value of the energy needs to consider the energy absorption characteristics, damage degree, etc. of the detected wood, and its value range is between 0.2 and 0.5;

[0082] The preset weight value of the rise time needs to be determined according to the dynamic response, fracture process, etc. of the detected wood, and its value range is between 0.1 and 0.4.

[0083] In this embodiment, the classification of the damage mode of the detected wood is as follows: According to the damage characteristic index of the detected wood, compare it with the defined value of the damage characteristic index stored in the cloud database. If the damage characteristic index of the detected wood is lower than the defined value of the damage characteristic index, then classify the damage mode of the wood.

[0084] The classification of the wood damage patterns includes detecting the wood damage patterns during the statistical test cycle, the typical feature clustering vectors under this pattern, and the typical feature parameters.

[0085] It should be understood that the typical feature clustering vectors refer to the typical feature vectors of the detected wood obtained after clustering analysis.

[0086] Use the RA-AF correlation analysis method to quantitatively analyze the detected wood damage patterns, and obtain the damage forms of the detected wood during the loading process.

[0087] In this embodiment, the RA-AF correlation analysis method is used to distinguish the shear cracks of the detected wood and the tensile cracks of the detected wood.

[0088] In this embodiment, the specific process of obtaining the damage forms of the detected wood during the loading process is as follows: Define the ratio of the rising time of each detection point of the detected wood to the amplitude of each detection point of the detected wood as the acoustic emission parameter RA, and define the ratio of the ring count of each detection point of the detected wood to the set detection cycle duration as the acoustic emission parameter AF.

[0089] According to the acoustic emission parameter RA and the acoustic emission parameter AF, quantitatively analyze the detected wood damage patterns.

[0090] When the detected wood has the characteristics of high RA and low AF, the damage form of the detected wood is shear cracks; when the detected wood has the characteristics of low RA and high AF, the damage form of the detected wood is tensile cracks.

[0091] In this embodiment, by adopting the RA-AF correlation analysis method, the damage patterns of the detected wood at different stages can be quantified, which is beneficial to studying the damage evolution law of the detected wood under the loading effect.

[0092] It should be noted that in a specific embodiment, the acoustic emission signal wave usually propagates in the form of a wave packet superimposed by longitudinal waves (compression waves) and transverse waves (shear waves). The propagation speed of longitudinal waves is faster than that of transverse waves, and the main energy arrival time is earlier than that of transverse waves. Therefore, the RA value of longitudinal waves is lower. Generally, it is considered that tensile cracks mainly propagate in the form of longitudinal waves, and the lateral crack expansion caused by them has the characteristics of high frequency and short time (high AF), while shear cracks mainly propagate in the form of transverse waves and have the characteristics of low frequency and long time (low AF). Therefore, it is considered that signals with high RA and low AF correspond to shear cracks, while signals with low RA and high AF correspond to tensile cracks.

[0093] According to the damage forms of the detected wood during the loading process, analyze the damage evolution process of the detected wood and transmit it to the cloud information warehouse for storage.

[0094] It should be understood that the detection of the wood damage evolution process generally refers to the change rules of the internal microstructure, mechanical properties or physical characteristics of wood during the process of being stressed or damaged, including the entire process and rules of detecting the deformation, cracking, fracture or other forms of damage of wood under different loading conditions.

[0095] In a specific embodiment, the detection of the wood damage evolution process specifically includes: an initial stage, a linear damage accumulation stage, a non-linear damage accumulation stage, and a failure stage.

[0096] In the initial stage, specifically, some microscopic changes will occur inside the wood, such as the generation of microcracks or small-scale deformation on the surface.

[0097] In the linear damage accumulation stage, specifically, with the action of an external load, the wood to be detected will enter the linear damage accumulation stage, and the damage usually increases linearly with the increase of the load.

[0098] In a specific embodiment, the main modes of damage in the linear damage accumulation stage include the propagation of microcracks, the increase of local deformation, etc.

[0099] In the non-linear damage accumulation stage, specifically, when the wood to be detected is under a higher load, it will enter the non-linear damage accumulation stage.

[0100] In a specific embodiment, the damage in the non-linear damage accumulation stage will show the characteristics of non-linear growth, and some complex damage forms may appear, such as the acceleration of crack propagation, local plastic deformation, etc.

[0101] In the failure stage, specifically, it refers to the failure form or failure mode presented by the wood when it is under an external load. When the wood to be detected is subjected to an excessive external load, it will enter the failure stage.

[0102] In a specific embodiment, the main modes of damage in the failure stage include fracture, rupture or severe plastic deformation, resulting in the wood to be detected being unable to continue to bear the load.

[0103] In this embodiment, the failure stage detection wood modes include tensile failure and shear failure. The tensile failure is a failure form that may occur when the wood is subjected to a tensile force, and may include the tearing or tensile deformation of the wood. The shear failure is a failure form that may occur when the wood is subjected to a shear force, including the shear deformation or fracture of the wood.

[0104] Please refer to Figure 6As shown, the horizontal axis represents the acoustic emission signal RA, where RA is the ratio of the rise time to the amplitude, and the vertical axis represents the acoustic emission signal AF, where AF is the ratio of the ring count to the set detection period duration. A dashed line with a slope of k is used as the dividing line. Signals with high AF and low RA above the dividing line correspond to tensile cracks, and signals with low AF and high RA below the dividing line correspond to shear cracks. In this embodiment, the value of k is defined as 50.

[0105] Please refer to Figure 2 As shown, the specific process of the clustering analysis involved in the embodiment of the present invention is as follows: Input the number of clusters and initialize. Assign the data of each detection point to the nearest class, then recalculate the center of each class, and determine whether the result converges.

[0106] If the clustering result converges, output the clustering result. If the clustering result does not converge, continue to calculate the center of each class.

[0107] It should be understood that the convergence of the clustering result means that the cluster centers before and after calculation do not change.

[0108] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or elements inherent to such process, method, article or device.

[0109] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principle and practical application of the present invention, so that those skilled in the relevant technical field can understand and utilize the present invention well. As long as it does not deviate from the structure of the present invention or exceed the scope defined by the present invention, it should fall within the protection scope of the present invention.

Claims

1. An evaluation method for the characteristics of wood damage evolution based on active acoustic emission, characterized in that: It includes the following steps: S1. Apply a uniform load to the tested wood through a four-point loading device. Meanwhile, use an acoustic emission sensor to collect the acoustic emission waveforms at each detection point, and perform data processing to obtain the acoustic emission parameters at the damage stage of the tested wood; S2. Perform clustering processing on the acoustic emission parameters at the damage stage of the tested wood, analyze and obtain the damage characteristic index of the tested wood, and thereby classify the damage modes of the tested wood; S3. Use the RA-AF correlation analysis method to quantitatively analyze the damage modes of the tested wood, and obtain the damage forms of the tested wood during the loading process; S4. According to the damage forms of the tested wood during the loading process, analyze the damage evolution process of the tested wood and transmit it to the cloud information warehouse for storage; The specific analysis process for obtaining the damage characteristic index of the tested wood is as follows: Extract the reference typical characteristic parameters stored in the cloud database, including reference ring counts, reference amplitudes, reference energies, and reference rise times; Count the ring counts, amplitudes, energies, and rise times of each detection point, and analyze and process them to obtain the damage characteristic index of the tested wood; The damage characteristic index of the tested wood is specifically obtained by analyzing and processing the acoustic emission parameters at the damage stage of the tested wood, which is used to reflect the damage characteristics of the tested wood and is used as the numerical classification basis for the damage modes of the tested wood; The specific process for classifying the damage modes of the tested wood is as follows: According to the damage characteristic index of the tested wood, compare it with the defined value of the damage characteristic index stored in the cloud database. If the damage characteristic index of the tested wood is lower than the defined value of the damage characteristic index, then classify the damage mode of the wood; The classification of the damage mode of the wood includes counting the damage mode of the tested wood in the test cycle, the typical characteristic clustering vector under this mode, and the typical characteristic parameters.

2. The method for evaluating the wood damage evolution characteristics based on active acoustic emission according to claim 1, wherein: The specific process for using the acoustic emission sensor to collect the acoustic emission waveforms at each detection point is as follows: Set the test cycle, and extract and collect the acoustic emission waveforms at each detection point through the acoustic emission sensor during the test cycle. After the acoustic emission waveforms at each detection point pass through amplifier discharge and filter filtering, they are not cut any further and are directly stored in the computer storage device.

3. The method for evaluating the characteristics of wood damage evolution based on active acoustic emission according to claim 1, wherein: The specific content for obtaining the acoustic emission parameters at the damage stage of the tested wood is: According to the acoustic emission waveforms at each detection point, extract the typical characteristic parameters in the acoustic emission waveforms, and obtain the pre-judged damage mode of the tested wood in the test cycle in combination as the acoustic emission parameters at the damage stage of the tested wood; The typical characteristic parameters specifically include the ring counts, amplitudes, energies, and rise times of each detection point.

4. The method for evaluating the characteristics of wood damage evolution based on active acoustic emission according to claim 1, wherein: The specific process for performing clustering processing on the acoustic emission parameters at the damage stage of the tested wood is as follows: Obtain the acoustic emission parameters at the damage stage of the tested wood. Randomly select K points among each detection point as the clustering centers, marked as K clustering centers. Count the typical characteristic vectors of the K clustering centers, and count the typical characteristic parameters of each detection point and convert them into the typical characteristic vectors of each detection point; Respectively count the distances between the typical characteristic vectors of each detection point and the typical characteristic vectors of the K clustering centers, and sequentially assign each detection point to the category of the clustering center with the closest distance, thereby counting all the detection points in the K clustering centers; The average value of all detection points in the K cluster center is calculated to solve the new K cluster centroid, and compared with the K cluster centroid calculated last time. When the centroid does not change, stop and output the clustering result, which includes the typical feature clustering vector.

5. The method for evaluating the characteristics of wood damage evolution based on active acoustic emission according to claim 1, wherein: The RA-AF correlation analysis method is used to distinguish between detecting shear cracks in wood and detecting tension cracks in wood.

6. The method for evaluating the characteristics of wood damage evolution based on active acoustic emission according to claim 5, characterized in that: The damage form of the detected wood during the loading process is obtained, and the specific process is as follows: The ratio of the rise time of each detection point of the detected wood to the amplitude of each detection point of the detected wood is defined as the acoustic emission parameter RA, and the ratio of the ring count of each detection point of the detected wood to the set detection cycle length is defined as the acoustic emission parameter AF; According to the acoustic emission parameters RA and AF, the damage pattern of the detected wood is quantitatively analyzed; When the detected wood has high RA and low AF characteristics, the damage form of the detected wood is shear crack; when the detected wood has low RA and high AF characteristics, the damage form of the detected wood is tension crack.

7. The method for evaluating the characteristics of wood damage evolution based on active acoustic emission according to claim 6, wherein: The detection of the wood damage evolution process specifically includes: an initial stage, a linear damage accumulation stage, a nonlinear damage accumulation stage and a failure stage; The initial stage specifically involves detecting some microscopic changes inside the wood; The linear damage accumulation stage is specifically that with the action of external load, the tested wood will enter the linear damage accumulation stage, and the damage will increase linearly with the increase of load; The nonlinear damage accumulation stage is specifically when the tested wood is subjected to a higher load, it will enter the nonlinear damage accumulation stage; The failure stage refers specifically to the damage form or failure state of the test wood when it is subjected to an external load. When the test wood is subjected to an excessively large external load, it will enter the failure stage.

8. The method for evaluating the characteristics of wood damage evolution based on active acoustic emission according to claim 1, characterized in that: The specific formula for the damage characteristic index of the detected wood is as follows: Among them, A is the damage characteristic index of the detected wood, and α i is the ring count of the i-th detection point of the detected wood, α0 is the reference ring count, β i is the amplitude of the i-th detection point of the detected wood, β0 is the reference amplitude, γ i is the energy of the i-th detection point of the detected wood, γ0 is the reference energy, δ i is the rise time of the i-th detection point of the detected wood, δ0 is the reference rise time, x1 is the preset ring count weight, x2 is the preset amplitude weight, x3 is the preset energy weight, x4 is the preset rise time weight, i is the number of each detection point of the detected wood, i = 1, 2,..., n, n is the number of detection points of the detected wood, and e is the natural constant.

Images