A method and device for risk assessment of composite material structure under multiple impacts

Abstract

The present application belongs to the field of structural damage detection, and specifically discloses a method and device for risk assessment of composite material structures under multiple impacts, the method comprising: obtaining a probability imaging map of the impact location; thresholding the probability imaging map to obtain a thresholded probability imaging of the impact location; obtaining the total signal energy and energy feature vector; inputting the total signal energy, energy feature vector and position feature vector into a trained BP neural network to obtain the predicted value of the impact energy corresponding to each position point; thresholding the predicted value of the impact energy at each position point to obtain a thresholded impact energy distribution; repeatedly obtaining the risk matrix of multiple impacts, accumulating the risk matrix of each impact, and obtaining the risk assessment result of the composite material structure after multiple impacts. The present application can improve the positioning accuracy and reliability, thereby more accurately assessing the risk situation of the structure.

Description

Technical Field

[0001] The present application belongs to the field of structural damage detection, and more specifically, to a method and device for risk assessment of composite material structures under multiple impacts. Background Art

[0002] For example, structures such as aircraft may be subjected to multiple impacts during a testing cycle. How to evaluate the impact of multiple impacts on the structure is of great significance to the determination of structural integrity, structural damage degree and damage mode.

[0003] The existing methods for evaluating the impact degree usually use the coordinates of the real impact position as input, but in real-time impact monitoring, the impact position estimated by the positioning algorithm will have a certain error, and using the position information with errors to estimate the impact degree will lead to wrong judgments. In addition, the factors that affect the impact response signal of composite materials are complex, and the existing positioning algorithms are difficult to accurately judge such complex influences, which affects the decision-making of the repair and maintenance location and timing of the device structure and the service life, which may cause safety problems or cause excessive maintenance costs.

[0004] Therefore, the existing impact monitoring system has inaccurate assessment of the impact degree due to errors in the positioning algorithm and the inability to accurately judge the complex influence of the composite material impact response signal, which may cause safety problems or excessive maintenance costs. Summary of the invention

[0005] In view of the defects of the prior art, the purpose of this application is to provide a method and device for risk assessment of composite material structures under multiple impacts, aiming to solve the defects of poor accuracy and reliability in impact degree assessment caused by errors in positioning algorithms and the complex influence of inability to accurately judge the impact response signals of composite materials.

[0006] To achieve the above objectives, the present application provides a method and device for risk assessment of composite material structures under multiple impacts.

[0007] The present application provides a risk assessment method for composite material structures under multiple impacts, comprising:

[0008] Obtaining the arrival time difference of the composite material structure at each pair of sensors under this impact, inputting the arrival time difference into a Gaussian regression model, and obtaining a probability imaging diagram of the impact location;

[0009] Performing threshold processing on the probability imaging map, determining the position feature vector and positioning probability corresponding to each position point according to the probability imaging map and a preset positioning probability threshold, and obtaining a thresholded impact position probability imaging;

[0010] Acquire the response signal of each sensor under this impact, calculate the total signal energy of the response signal and extract the energy feature to obtain the energy feature vector;

[0011] The total signal energy, energy feature vector and position feature vector are input into the trained BP neural network to obtain the predicted value of the impact energy corresponding to each position point;

[0012] Performing threshold processing on the predicted value of the impact energy at each position point, and determining the thresholded impact energy distribution according to the predicted value of the impact energy and a preset impact energy threshold;

[0013] The risk matrix of the current impact is obtained based on the impact position probability imaging and the impact energy distribution, and the risk matrices of multiple impacts are repeatedly obtained. The risk matrices of each impact are accumulated to obtain the risk assessment results of the composite material structure after multiple impacts.

[0014] In some embodiments, the step of determining the position feature vector and the positioning probability corresponding to each position point according to the probability imaging map and a preset positioning probability threshold comprises:

[0015] Traversing each position point in the imaging matrix of the probability imaging map, and comparing the positioning probability of each position point with a positioning probability threshold;

[0016] If the positioning probability of the position point n is greater than the positioning probability threshold, the positioning probability of the position point n is retained, and the position vector of the position point n is determined; wherein n is a positive integer;

[0017] If the positioning probability of the position point n is less than the positioning probability threshold, the positioning probability of the position point n is set to 0.

[0018] In some embodiments, the thresholding of the predicted value of the impact energy at each position point and determining the thresholded impact energy distribution according to the predicted value of the impact energy and a preset impact energy threshold value include:

[0019] Compare the predicted value m with the impact energy threshold; wherein the predicted value m is the predicted value corresponding to the position point n;

[0020] If the predicted value m is greater than the impact energy threshold, the impact energy of the composite material structure at the position point n is retained;

[0021] If the predicted value m is less than the impact energy threshold, the impact energy of the composite material structure at the position point n is set to 0, and the thresholded impact energy distribution of each position point of the composite material structure is obtained.

[0022] In some embodiments, the energy feature vector extraction process includes:

[0023] The signal energy of the response signal is acquired, and the response signal is decomposed in multiple layers using a wavelet function to obtain the energy feature vector.

[0024] In some embodiments, the obtaining the risk matrix of the current impact based on the impact position probability imaging and the impact energy distribution includes:

[0025] The sensitivity coefficient is used to determine the sensitivity of the composite structure to the impact at position n;

[0026] The sensitivity is combined with the impact position probability imaging and the impact energy distribution to determine the risk value at the target position, and the risk matrix of this impact is obtained based on the risk value.

[0027] In some embodiments, obtaining the arrival time difference of the composite material structure at each pair of sensors under the impact includes:

[0028] Acquire the acoustic emission signals of this impact on different sensors, and perform real-time impact monitoring based on the acoustic emission signals;

[0029] When an impact signal is detected, a stress wave associated with the impact is determined, and an arrival time difference of the stress wave at each pair of sensors is obtained.

[0030] In some embodiments, it also includes:

[0031] Performing risk assessment on the risk assessment result, if the risk assessment criteria are met, continuously performing real-time impact monitoring based on the acoustic emission signal, if the risk assessment criteria are not met, performing inspection and maintenance on the composite material structure.

[0032] In a second aspect, the present application also provides a composite material structure risk assessment device under multiple impacts, comprising:

[0033] A probability imaging module is used to obtain the arrival time difference of the composite material structure on each pair of sensors under this impact, and input the arrival time difference into a Gaussian regression model to obtain a probability imaging diagram of the impact location;

[0034] A first thresholding module is used to perform thresholding processing on the probability imaging map, determine the position feature vector and positioning probability corresponding to each position point according to the probability imaging map and a preset positioning probability threshold, and obtain a thresholded impact position probability imaging;

[0035] A feature extraction module is used to obtain the response signal of each sensor under this impact, calculate the total signal energy of the response signal and extract the energy feature to obtain an energy feature vector;

[0036] A prediction value acquisition module is used to input the signal total energy, energy feature vector and position feature vector into the trained BP neural network to obtain the prediction value of the impact energy corresponding to each position point;

[0037] A second thresholding module, used to perform thresholding processing on the predicted value of the impact energy of each position point, and determine the thresholded impact energy distribution according to the predicted value of the impact energy and a preset impact energy threshold;

[0038] The risk assessment module is used to obtain the risk matrix of the current impact based on the impact position probability imaging and the impact energy distribution, repeatedly obtain the risk matrix of multiple impacts, accumulate the risk matrix of each impact, and obtain the risk assessment result of the composite material structure after being subjected to multiple impacts.

[0039] In a third aspect, the present application provides an electronic device comprising: at least one memory for storing programs; and at least one processor for executing the programs stored in the memory. When the program stored in the memory is executed, the processor is used to execute the method described in the first aspect or any possible implementation of the first aspect.

[0040] In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method described in the first aspect or any possible implementation of the first aspect.

[0041] In a fifth aspect, the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the method described in the first aspect or any possible implementation of the first aspect.

[0042] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0043] In general, the above technical solutions conceived by this application have the following beneficial effects compared with the prior art:

[0044] (1) This application uses sensors to obtain arrival time differences and Gaussian regression models to locate the impact position with high precision. Through thresholding of the probability imaging map, the location of the impact and its probability distribution can be clearly determined, thereby improving positioning accuracy and reliability, thereby more accurately assessing the risk status of the structure.

[0045] (2) Combining the position eigenvector, energy eigenvector and the trained BP neural network, inputting a variety of characteristic information into the model helps to comprehensively analyze the response of the structure under multiple impacts and improve the comprehensiveness, accuracy and reliability of the assessment.

[0046] (3) By obtaining and accumulating the risk matrix of multiple impacts, the present application can obtain the comprehensive risk assessment results of the composite material structure after being subjected to multiple impacts. Considering the cumulative response of the composite material structure under multiple impact conditions helps to more comprehensively evaluate the durability and safety of the structure, realize real-time monitoring and tracking of the risks of the composite material structure, discover potential problems in a timely manner, and help prevent accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is one of the flow diagrams of a composite material structure risk assessment method under multiple impacts provided in an embodiment of the present application;

[0048] Figure 2 This is the second flow chart of a method for risk assessment of composite material structure under multiple impacts provided in an embodiment of the present application;

[0049] Figure 3 This is a third flow chart of a method for risk assessment of composite material structure under multiple impacts provided in an embodiment of the present application;

[0050] Figure 4 This is a fourth flow chart of a method for risk assessment of composite material structure under multiple impacts provided in an embodiment of the present application;

[0051] Figure 5 is a schematic diagram of the locations of multiple impacts in an example of a method for risk assessment of a composite material structure under multiple impacts provided in an embodiment of the present application;

[0052] Figure 6 It is a structural risk accumulation diagram when both the positioning threshold and the impact energy threshold are 0 in an example of a composite material structure risk assessment method under multiple impacts according to an embodiment of the present application;

[0053] Figure 7 It is a risk accumulation diagram when the positioning threshold is 0.98 and the impact energy threshold is 40 mJ in an example of a composite material structure risk assessment method under multiple impacts according to an embodiment of the present application;

[0054] Figure 8 This is a schematic structural diagram of a composite material structure risk assessment device under multiple impacts according to an embodiment of the present application;

[0055] Fig. 9 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0057] The term "and / or" in this article is 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 at the same time, and B exists alone. The symbol " / " in this article indicates that the associated objects are in an or relationship, for example, A / B means A or B.

[0058] The terms "first" and "second" in the specification and claims herein are used to distinguish different objects rather than to describe a specific order of the objects. For example, a first response message and a second response message are used to distinguish different response messages rather than to describe a specific order of the response messages.

[0059] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0060] In the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more than two. For example, multiple processing units refer to two or more processing units, etc.; multiple elements refer to two or more elements, etc.

[0061] Next, the technical solutions provided in the embodiments of the present application are introduced.

[0062] Reference Figure 1 The present application provides a risk assessment method for composite material structures under multiple impacts, comprising:

[0063] S101. Obtaining the arrival time difference of the composite material structure at each pair of sensors under this impact, inputting the arrival time difference into a Gaussian regression model, and obtaining a probability imaging diagram of the impact location;

[0064] Specifically, a plurality of acoustic emission sensors are installed on the surface or inside of the composite material structure in this embodiment to facilitate accurate location of the acoustic emission event. When the composite material structure is impacted, an acoustic emission event will be triggered, and the sensor network will record information such as the arrival time and possible signal strength.

[0065] Furthermore, the arrival time difference is obtained as follows:

[0066] Acquire the acoustic emission signals of this impact on different sensors, and perform real-time impact monitoring based on the acoustic emission signals; when the impact signal is detected, determine the stress wave related to this impact, and obtain the arrival time difference of the stress wave at each pair of sensors.

[0067] By calculating the arrival time difference of the acoustic emission signals between each pair of sensors, the approximate location information of the acoustic emission source within the structure can be provided.

[0068] In some optional implementations, the acquired arrival time difference data needs to be preprocessed, including noise filtering, outlier processing, etc., to ensure the accuracy and reliability of the data.

[0069] The pre-processed arrival time difference data is input into the Gaussian regression model to obtain the probability imaging map. In this embodiment, the Gaussian regression model is used to locate the acoustic emission event, and the location of the acoustic emission source is estimated based on the time difference between different sensor pairs. The Gaussian regression model will generate a probability distribution of the possible location of the acoustic emission source and display it through a two-dimensional or three-dimensional probability imaging map, showing the possible location and relative probability of the acoustic emission source.

[0070] S102. Thresholding the probability image, determining the position feature vector and positioning probability corresponding to each position point according to the probability image and a preset positioning probability threshold, and obtaining a thresholded impact position probability image;

[0071] Specifically, before thresholding, this embodiment needs to set a positioning probability threshold. The threshold is determined by the specific application requirements and the characteristics of the structure. The positioning probability of each position point in the generated probability imaging map is compared with the preset positioning probability threshold. If the positioning probability of a certain position point is greater than or equal to the set threshold, it is considered that this position point has a sufficiently high probability of being one of the locations where the impact occurs. If the positioning probability is less than the threshold, this position point is excluded from the location where the impact may occur.

[0072] Specifically, the position feature vector and the positioning probability corresponding to each position point are determined according to the probability imaging map and the preset positioning probability threshold, including:

[0073] Traversing each position point in the imaging matrix of the probability imaging map, and comparing the positioning probability of each position point with a positioning probability threshold;

[0074] If the positioning probability of the position point n is greater than the positioning probability threshold, the positioning probability of the position point n is retained, and the position vector of the position point n is determined; wherein n is a positive integer;

[0075] If the positioning probability of the position point n is less than the positioning probability threshold, the positioning probability of the position point n is set to 0.

[0076] After thresholding, the generated thresholded impact position probability image reflects the result that each position point on the structure is judged as the position where the impact may occur.

[0077] Through thresholding processing, the location affected by the impact on the composite structure can be accurately determined, providing important input data for subsequent energy feature extraction and risk assessment. By combining the advantages of probabilistic inference and threshold processing, the accuracy and reliability of impact location identification are effectively improved.

[0078] S103. Obtain the response signal of each sensor under this impact, calculate the total signal energy of the response signal and extract the energy feature to obtain an energy feature vector;

[0079] Specifically, the signal energy of the response signal is obtained, and the response signal is decomposed by multi-layers using a wavelet function to obtain the energy feature vector. In this embodiment, the response signal is decomposed by 8 layers using the db5 wavelet to extract the energy feature.

[0080] The response signal is usually measured by a sensor to record the response of the composite structure under an impact event. The response signal can be a change in physical quantities such as displacement, acceleration or pressure. Furthermore, the response signal collected by each sensor is analyzed in the time domain to calculate the total energy of each signal. The response signal is multi-layered decomposed by wavelet analysis to obtain the energy feature vector.

[0081] It should be noted that wavelet analysis is a signal analysis tool that provides local information in the time-frequency domain. It reveals the implicit characteristics of the signal by decomposing the signal into components of different frequencies.

[0082] S104. Inputting the total signal energy, energy feature vector and position feature vector into the trained BP neural network to obtain the predicted value of the impact energy corresponding to each position point;

[0083] Specifically, the position feature vector corresponding to each position point is obtained based on the impact position probability image generated in the previous step. The position feature vector is usually a mathematical representation of the position, which can be a two-dimensional coordinate or other forms of encoding.

[0084] The total signal energy, energy feature vector and position feature vector are used as input to the pre-trained BP neural network for feature recognition to obtain the predicted value of impact energy.

[0085] BP neural network is a commonly used back propagation neural network, which is suitable for processing the mapping relationship from input vector to output (i.e. impact energy prediction). It is usually composed of input layer, hidden layer and output layer, and performs information processing and learning through weight adjustment and activation function calculation between multiple layers of neurons.

[0086] In the BP neural network, the neurons in the output layer will generate the impact energy prediction value for each location point. The prediction value is obtained through optimization during network learning and training, and can accurately reflect the degree of impact or the distribution of impact energy at each location point on the structure.

[0087] S105. Performing threshold processing on the predicted value of the impact energy at each position point, and determining the thresholded impact energy distribution according to the predicted value of the impact energy and a preset impact energy threshold;

[0088] Before threshold processing, an impact energy threshold needs to be set. The impact energy threshold can be determined based on the safety requirements of the structure, the level classification of impact intensity or other engineering requirements.

[0089] For the impact energy prediction value of each position point, the following processing is performed:

[0090] If the impact energy prediction value is greater than or equal to the preset impact energy threshold, the threshold impact energy distribution of the position point is set to a certain high value (indicating that an impact has occurred).

[0091] If the impact energy prediction value is less than the preset impact energy threshold, the thresholded impact energy distribution of the position point is set to a low value (indicating that the impact does not reach the threshold) or zero.

[0092] The details are as follows:

[0093] Compare the predicted value m with the impact energy threshold; wherein the predicted value m is the predicted value corresponding to the position point n;

[0094] If the predicted value m is greater than the impact energy threshold, the impact energy of the composite material structure at the position point n is retained;

[0095] If the predicted value m is less than the impact energy threshold, the impact energy of the composite material structure at the position point n is set to 0, and the thresholded impact energy distribution of each position point of the composite material structure is obtained.

[0096] The impact energy distribution after thresholding will be presented as a discrete image or data set, where different values ​​represent the impact energy at different locations or whether the preset impact energy threshold is reached. The thresholded impact energy distribution can be used for decision support, structural health monitoring or post-event analysis, and further structural inspection, maintenance actions or prediction of structural performance can be taken based on these data.

[0097] S106. Obtain the risk matrix of the current impact based on the impact position probability imaging and impact energy distribution, repeatedly obtain the risk matrices of multiple impacts, accumulate the risk matrices of each impact, and obtain the risk assessment results of the composite material structure after multiple impacts.

[0098] Specifically, this embodiment generates a single impact risk matrix for each impact event, and the risk matrix combines the position probability imaging and the impact energy distribution data to evaluate the risk level of each location point.

[0099] For each location point in the composite structure that may be affected by the impact, the corresponding risk assessment result is determined. The risk assessment result includes various risk indicators, such as impact energy, structural fragility, type and degree of possible damage, etc., which are used to determine the risk value of each location point.

[0100] Reference Figure 2 After multiple impact events occur, the single impact risk matrix generated each time is accumulated to obtain the cumulative risk assessment results of the composite material structure after multiple impacts. By analyzing the accumulated risk matrix, comprehensive evaluation and analysis can be performed to obtain the overall strength and damage accumulation of the composite material structure after multiple impacts.

[0101] This application uses sensors to obtain arrival time difference and Gaussian regression model to locate the impact position with high precision. Through threshold processing of probability imaging graph, the location of the impact and its probability distribution can be clearly determined, improving positioning accuracy and reliability, so as to more accurately assess the risk of the structure. In addition, by combining position feature vectors, energy feature vectors and trained BP neural networks, multiple feature information is input into the model, which helps to comprehensively analyze the response of the structure under multiple impacts and improve the comprehensiveness, accuracy and reliability of the assessment.

[0102] In some optional embodiments, referring to Figure 3 and Figure 4 , Figure 3 and Figure 4 They are respectively a schematic diagram of the flow chart of multiple impacts and a schematic diagram of the flow chart of single impact in the risk assessment method of composite material structure under multiple impacts provided in the embodiment of the present application.

[0103] In some embodiments, the obtaining the risk matrix of the current impact based on the impact position probability imaging and the impact energy distribution includes:

[0104] The sensitivity coefficient is used to determine the sensitivity of the composite structure to the impact at position n;

[0105] The sensitivity is combined with the impact position probability imaging and the impact energy distribution to determine the risk value at the target position, and the risk matrix of this impact is obtained based on the risk value.

[0106] Specifically, this embodiment sets the sensitivity coefficient α (i,j) , used to characterize the sensitivity to shocks at position n(i,j), the risk value R (i,j) It is expressed as:

[0107] R (i,j) =α (i,j) ×P (i,j) ×C (i,j)

[0108] Where P is the probability value of a certain impact occurring at position n(i,h), and C is the estimated impact energy at position n(i,j). A larger α is set for locations that are sensitive to impact. (i,j) value, and when the coefficient α (i,j) When it is set to 1, if the impact positioning result is accurate, that is, the positioning probability value obtained at the true position is 1, then the risk value R represents the impact energy value estimated at the impact position.

[0109] In some embodiments, it also includes:

[0110] Performing risk assessment on the risk assessment result, if the risk assessment criteria are met, continuously performing real-time impact monitoring based on the acoustic emission signal, if the risk assessment criteria are not met, performing inspection and maintenance on the composite material structure.

[0111] Specifically, in this embodiment, it is necessary to predefine the risk assessment criteria and standards in order to determine when real-time impact monitoring or overhaul maintenance should be performed. The assessment criteria are usually based on the requirements of structural safety and the preset risk acceptance criteria.

[0112] Based on real-time monitoring technologies such as acoustic emission signals, monitoring strategies are set to track the impact of the structure in real time, and acoustic emission sensors are used to continuously monitor the sound signals in the structure. The collected acoustic emission signal data is analyzed in real time and compared with the preset risk assessment criteria. If a signal or change that exceeds the safety limit is detected, further risk assessment or maintenance actions are triggered.

[0113] If the real-time monitoring data shows that the structure is still within the safe range, continue to conduct real-time impact monitoring of the acoustic emission signal and regularly review the risk assessment results to ensure that the structure continues to meet safety standards.

[0114] If the monitoring data shows that the condition of the structure has reached or exceeded the limit of the risk assessment criteria, immediate maintenance measures will be taken. This may include repairing, reinforcing or replacing damaged parts to restore the safety and performance of the structure.

[0115] The following is a detailed description of an illustrative example of a composite material structure risk assessment method under multiple impacts provided according to an embodiment of the present invention. It should be understood that this illustrative example is only used to illustrate and illustrate the embodiment of the present invention in an exemplary manner, and does not limit the scope of the present invention in any way.

[0116] In order to demonstrate the proposed method, three different small balls are used to impact the composite sandwich structure respectively. The material properties and signal acquisition equipment are the same as above. A 15g small ball is used to impact the same position three times at different heights. The specific impact positions and impact energies of the five impacts are shown in Figure 2. Figure 5 And as shown in Table 1.

[0117] Table 1: Detailed information of the five impacts

[0118]

[0119] Assuming that there is no weak position in the structure, the α at each position (i,j) All are set to 1. Three impacts are carried out at position A (300, 150), and it is considered that each impact will cause damage to the structure. The cumulative impact energy value at this position is 108.8. The cumulative impact energy values ​​at positions B (50, 350) and C (400, 300) are 73.5. According to the cumulative impact energy value, the risk position of the structure is position.

[0120] The results are analyzed. First, without setting the positioning threshold and impact energy threshold, the risk of the structure after five impacts is evaluated. It can be found that the overall form of the structural risk assessment result is similar to the positioning result. This is because the impact energy value formed by each impact in the experimental setting is less than the positioning probability value. At this time, the positioning result has a greater impact on the risk value calculation.

[0121] To describe in more detail the changes in the risk matrix after each shock, Figure 6The cumulative process of the risk matrix is ​​shown. The ordinate in the figure represents the risk, i.e., the product of the positioning probability value and the predicted impact energy value. From the figure, three peaks can be clearly observed, indicating three areas where there are risks after the impact. During the maintenance process, these areas can be tested to identify potential damage. The prediction results of each impact are shown in Table 22. It can be seen from the table that the error between the predicted position and the actual position of each impact is small. Since the predicted impact position is very close to the actual position, the risk value predicted at this point can represent the actual risk value to a certain extent, and even represent the impact energy size.

[0122] Table 2: Prediction results of a single shock

[0123]

[0124] The main purpose of proposing this method is to identify the most severely affected part of the structure after multiple impacts. Therefore, the results of the three wave peak positions are analyzed here, and the results are shown in Table 3.

[0125] Table 3: Prediction results when both the positioning threshold and the impact energy threshold are 0

[0126]

[0127] At peak 1, the predicted position coordinates are (249,111), and the positioning error is about 10mm. This peak is caused by 3 impacts with a total impact energy of 108.8mJ, and the risk value is 108.8. The risk value calculated at the predicted position is 139.2, with an error of about 27.9%. The predicted position coordinates of peak 2 are (0,300), which is consistent with the actual position, but the risk value calculated at this point is 113.4, with an error as high as 54.0%. The predicted position coordinates of peak 3 are (351,252), with a positioning error of about 2mm. The risk value calculated at the predicted position is 100.8, with an error of 37.1%. The reason for the large error in the risk value is that the predicted value at this point is obtained by accumulating the prediction results of multiple impacts, such as Figure 6 Although the accumulated risk at this point is less after being impacted at other locations, when the number of impacts increases, the error in the predicted risk value will increase.

[0128] From the prediction results, we can find that the risk value of peak 1 is the largest, and the risk value of peak 3 is the smallest, which means that the peak 1 area is the part of the structure with the highest risk after the impact, that is, the part most affected by the impact, which is consistent with the experimental settings. Therefore, the prediction results can still be used to qualitatively identify the most severely affected part of the structure.

[0129] From the previous analysis, it can be found that the reason for the large error in the risk value in the prediction result is the existence of the positioning probability value. A probability value is still assigned to the part where no impact occurs. Although the probability value is very small, when the impact energy is very large, the product of the probability and the impact energy at this point will also become very large. Therefore, it is necessary to set a positioning threshold to filter the positions with smaller probabilities. Another advantage of setting a positioning threshold is that it can improve the calculation efficiency. When the positioning threshold is not set, it is not suitable for real-time monitoring of large-scale structural impacts because the impact energy of each position needs to be predicted, which increases the calculation time. When setting the threshold, if the setting is too small, the calculation efficiency is still low. When the threshold is set too large, it may cause the real impact position to not exist within the prediction range. When the threshold is set to 0.98, that is, the upper 0.02 quantile of each impact probability result, the positioning probability value of the real impact position is between the upper 0.02 quantile and the upper limit. Therefore, the positioning threshold is set to 0.98 here.

[0130] In the study of impact damage of composite materials, it was found that when the impact energy is lower than a certain threshold, the impact causes elastic deformation of the structure and does not cause internal damage to the structure. In this application, it is assumed that when the impact energy is lower than 40mJ, the damage caused by the impact is negligible. The impact energy threshold is set to 40mJ. When the predicted impact energy is lower than the threshold, the result is not used for the final accumulation. After setting the two thresholds, the analysis results are as follows: Figure 7 shown.

[0131] As can be seen from the figure, there are still three peaks. Peak 1 (246, 108) is least affected by the shock, with a calculated risk of 42.7 and a prediction error of 3.6%. Peak 2 (0, 303) has a calculated risk of 74.6 and a prediction error of 1.5%. Peak 3 (351, 252) has a calculated risk of 75.1 and a prediction error of 2.2% at this position.

[0132] Due to the setting of the impact energy threshold, the two impacts with impact energy lower than 40mJ are not accumulated in the results. At this time, the risk position of the structure is at peak 3. By setting the impact energy threshold, it is possible to avoid the situation where a certain part is subjected to multiple low-energy impacts and the cumulative risk value is too large, resulting in the wrong judgment of the most seriously affected part of the structure. It is worth noting that the risk values ​​of the three peaks are consistent with the results of impact 3, impact 4, and impact 5 in Table 2. This is because the setting of the positioning threshold makes it possible to avoid accumulating risks at low-probability positions when impacts are applied to the three positions. In this case, the risk matrix can not only qualitatively characterize the most seriously affected parts of the structure, but also quantitatively describe the impact energy received at each position. In addition, the calculation efficiency is significantly improved compared to when the positioning threshold is not set. The calculation time is reduced from 790.7s to 16.4s, an increase of 97.9%. Obviously, the improved calculation efficiency is related to the setting of the positioning threshold.

[0133] This application is suitable for real-time impact monitoring of large structures. In most cases, there is a certain relationship between impact energy and damage. The greater the energy, the greater the damage that may be caused. The prediction results can largely help detection and maintenance personnel to carry out their work in a planned and targeted manner.

[0134] Reference Figure 8 The present application also provides a composite material structure risk assessment device under multiple impacts, comprising:

[0135] The probability imaging module 810 is used to obtain the arrival time difference of the composite material structure on each pair of sensors under the impact, and input the arrival time difference into the Gaussian regression model to obtain the probability imaging diagram of the impact location;

[0136] A first thresholding module 820 is used to perform thresholding processing on the probability imaging map, determine the position feature vector and positioning probability corresponding to each position point according to the probability imaging map and a preset positioning probability threshold, and obtain a thresholded impact position probability imaging;

[0137] The feature extraction module 830 is used to obtain the response signal of each sensor under the current impact, calculate the total signal energy of the response signal and extract the energy feature to obtain the energy feature vector;

[0138] A prediction value acquisition module 840 is used to input the signal total energy, energy feature vector and position feature vector into a trained BP neural network to obtain a prediction value of the impact energy corresponding to each position point;

[0139] A second thresholding module 850 is used to perform thresholding processing on the predicted value of the impact energy of each position point, and determine the thresholded impact energy distribution according to the predicted value of the impact energy and a preset impact energy threshold;

[0140] The risk assessment module 860 is used to obtain the risk matrix of the current impact based on the impact position probability imaging and the impact energy distribution, repeatedly obtain the risk matrices of multiple impacts, accumulate the risk matrix of each impact, and obtain the risk assessment result of the composite material structure after multiple impacts.

[0141] In some embodiments, the step of determining the position feature vector and the positioning probability corresponding to each position point according to the probability imaging map and a preset positioning probability threshold comprises:

[0142] Traversing each position point in the imaging matrix of the probability imaging map, and comparing the positioning probability of each position point with a positioning probability threshold;

[0143] If the positioning probability of the position point n is greater than the positioning probability threshold, the positioning probability of the position point n is retained, and the position vector of the position point n is determined; wherein n is a positive integer;

[0144] If the positioning probability of the position point n is less than the positioning probability threshold, the positioning probability of the position point n is set to 0.

[0145] In some embodiments, the thresholding of the predicted value of the impact energy at each position point and determining the thresholded impact energy distribution according to the predicted value of the impact energy and a preset impact energy threshold value include:

[0146] Compare the predicted value m with the impact energy threshold; wherein the predicted value m is the predicted value corresponding to the position point n;

[0147] If the predicted value m is greater than the impact energy threshold, the impact energy of the composite material structure at the position point n is retained;

[0148] If the predicted value m is less than the impact energy threshold, the impact energy of the composite material structure at the position point n is set to 0, and the thresholded impact energy distribution of each position point of the composite material structure is obtained.

[0149] In some embodiments, the energy feature vector extraction process includes:

[0150] The signal energy of the response signal is acquired, and the response signal is decomposed in multiple layers using a wavelet function to obtain the energy feature vector.

[0151] In some embodiments, the obtaining the risk matrix of the current impact based on the impact position probability imaging and the impact energy distribution includes:

[0152] The sensitivity coefficient is used to determine the sensitivity of the composite structure to the impact at position n;

[0153] The sensitivity is combined with the impact position probability imaging and the impact energy distribution to determine the risk value at the target position, and the risk matrix of this impact is obtained based on the risk value.

[0154] In some embodiments, obtaining the arrival time difference of the composite material structure at each pair of sensors under the impact includes:

[0155] In some embodiments, a risk assessment module is further included, the risk assessment module being configured to:

[0156] Performing risk assessment on the risk assessment result, if the risk assessment criteria are met, continuously performing real-time impact monitoring based on the acoustic emission signal, if the risk assessment criteria are not met, performing inspection and maintenance on the composite material structure.

[0157] It can be understood that the detailed functional implementation of each of the above-mentioned units / modules can be found in the introduction of the aforementioned method embodiment, and will not be repeated here.

[0158] It should be understood that the above-mentioned device is used to execute the method in the above-mentioned embodiment. The implementation principle and technical effect of the corresponding program module in the device are similar to those described in the above-mentioned method. The working process of the device can refer to the corresponding process in the above-mentioned method, which will not be repeated here.

[0159] Reference Fig. 9 Based on the method in the above embodiment, the embodiment of the present application provides an electronic device, which may include: a processor (processor) 910, a communication interface (Communications Interface) 920, a memory (memory) 930 and a communication bus 940, wherein the processor 910, the communication interface 920, and the memory 930 communicate with each other through the communication bus 940. The processor 910 can call the logic instructions in the memory 930 to execute the method in the above embodiment.

[0160] In addition, the logic instructions in the above-mentioned memory 930 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application.

[0161] Based on the method in the above embodiment, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method in the above embodiment.

[0162] Based on the method in the above embodiment, an embodiment of the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the method in the above embodiment.

[0163] It is understandable that the processor in the embodiment of the present application can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor can be a microprocessor or any conventional processor.

[0164] The method steps in the embodiments of the present application can be implemented by hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, mobile hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC.

[0165] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions may be transmitted from a website site, a computer, a server or a data center to another website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.

[0166] It should be understood that the various numerical numbers involved in the embodiments of the present application are only used for the convenience of description and are not used to limit the scope of the embodiments of the present application.

[0167] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A risk assessment method for composite material structures under multiple impacts, characterized in that: include: Obtaining the arrival time difference of the composite material structure at each pair of sensors under this impact, inputting the arrival time difference into a Gaussian regression model, and obtaining a probability imaging diagram of the impact location; Performing threshold processing on the probability imaging map, determining the position feature vector and positioning probability corresponding to each position point according to the probability imaging map and a preset positioning probability threshold, and obtaining a thresholded impact position probability imaging; Acquire the response signal of each sensor under this impact, calculate the total signal energy of the response signal and extract the energy feature to obtain the energy feature vector; The total signal energy, energy feature vector and position feature vector are input into the trained BP neural network to obtain the predicted value of the impact energy corresponding to each position point; Performing threshold processing on the predicted value of the impact energy at each position point, and determining the thresholded impact energy distribution according to the predicted value of the impact energy and a preset impact energy threshold; Based on the impact position probability imaging and the impact energy distribution, the risk matrix of the current impact is obtained, and the risk matrix of multiple impacts is repeatedly obtained, and the risk matrix of each impact is accumulated to obtain the risk assessment result of the composite material structure after multiple impacts; The step of obtaining the risk matrix of the current impact based on the impact position probability imaging and the impact energy distribution includes: The sensitivity coefficient is used to determine the sensitivity of the composite structure to the impact at position n; The sensitivity is combined with the impact position probability imaging and the impact energy distribution to determine the risk value at the target position, and the risk matrix of this impact is obtained based on the risk value.

2. The risk assessment method for composite material structures under multiple impacts according to claim 1, characterized in that: The determining of the position feature vector and the positioning probability corresponding to each position point according to the probability imaging map and the preset positioning probability threshold comprises: Traversing each position point in the imaging matrix of the probability imaging map, and comparing the positioning probability of each position point with a positioning probability threshold; If the positioning probability of the position point n is greater than the positioning probability threshold, the positioning probability of the position point n is retained, and the position vector of the position point n is determined; wherein n is a positive integer; If the positioning probability of the position point n is less than the positioning probability threshold, the positioning probability of the position point n is set to 0.

3. The risk assessment method for composite material structures under multiple impacts according to claim 1, characterized in that: The step of performing threshold processing on the predicted value of the impact energy at each position point and determining the thresholded impact energy distribution according to the predicted value of the impact energy and a preset impact energy threshold comprises: Compare the predicted value m with the impact energy threshold; wherein the predicted value m is the predicted value corresponding to the position point n; If the predicted value m is greater than the impact energy threshold, the impact energy of the composite material structure at the position point n is retained; If the predicted value m is less than the impact energy threshold, the impact energy of the composite material structure at the position point n is set to 0, and the thresholded impact energy distribution of each position point of the composite material structure is obtained.

4. The risk assessment method for composite material structures under multiple impacts according to claim 1, characterized in that: The energy feature vector extraction process includes: The signal energy of the response signal is acquired, and the response signal is decomposed in multiple layers using a wavelet function to obtain the energy feature vector.

5. The risk assessment method for composite material structures under multiple impacts according to claim 1, characterized in that: The step of obtaining the arrival time difference of the composite material structure at each pair of sensors under the impact comprises: Acquire the acoustic emission signals of this impact on different sensors, and perform real-time impact monitoring based on the acoustic emission signals; When an impact signal is detected, a stress wave associated with the impact is determined, and an arrival time difference of the stress wave at each pair of sensors is obtained.

6. The risk assessment method for composite material structures under multiple impacts according to claim 5, characterized in that: Also includes: Performing risk assessment on the risk assessment result, if the risk assessment criteria are met, continuously performing real-time impact monitoring based on the acoustic emission signal, if the risk assessment criteria are not met, performing inspection and maintenance on the composite material structure.

7. A composite material structure risk assessment device under multiple impacts, characterized in that: include: A probability imaging module is used to obtain the arrival time difference of the composite material structure on each pair of sensors under this impact, and input the arrival time difference into a Gaussian regression model to obtain a probability imaging diagram of the impact location; A first thresholding module is used to perform thresholding processing on the probability imaging map, determine the position feature vector and positioning probability corresponding to each position point according to the probability imaging map and a preset positioning probability threshold, and obtain a thresholded impact position probability imaging; A feature extraction module is used to obtain the response signal of each sensor under this impact, calculate the total signal energy of the response signal and extract the energy feature to obtain an energy feature vector; A prediction value acquisition module is used to input the signal total energy, energy feature vector and position feature vector into the trained BP neural network to obtain the prediction value of the impact energy corresponding to each position point; A second thresholding module, used to perform thresholding processing on the predicted value of the impact energy of each position point, and determine the thresholded impact energy distribution according to the predicted value of the impact energy and a preset impact energy threshold; A risk assessment module, configured to obtain a risk matrix of the current impact based on the impact position probability imaging and the impact energy distribution, repeatedly obtain risk matrices of multiple impacts, accumulate the risk matrices of each impact, and obtain a risk assessment result of the composite material structure after multiple impacts; The step of obtaining the risk matrix of the current impact based on the impact position probability imaging and the impact energy distribution includes: The sensitivity coefficient is used to determine the sensitivity of the composite structure to the impact at position n; The sensitivity is combined with the impact position probability imaging and the impact energy distribution to determine the risk value at the target position, and the risk matrix of this impact is obtained based on the risk value.

8. An electronic device, characterized in that: include: at least one memory for storing a computer program; At least one processor is used to execute the program stored in the memory. When the program stored in the memory is executed, the processor is used to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program runs on a processor, the processor is caused to execute the method according to any one of claims 1 to 6.

Images