Typical damage monitoring device and method for load test flight

Abstract

The invention belongs to the technical field of aircraft flight tests, and relates to a typical damage monitoring device and method for load test flight. The typical damage monitoring device comprises a loading actuator cylinder (1), an environmental box (2), a fixed constraint (4), a test loading piece (5) and a damage feature (6), the test loading piece (5) is an aluminum alloy box section, the upper skin of the box section is divided into three pieces, and the upper skins at the two ends of the box section are non-detachable skins and are connected with front and rear beam flanges and rib flanges through rivets; the upper skin of the middle section can be independently detached, a bolt and supporting plate nut connecting mode is adopted, and the piezoelectric sensor network array is arranged in the test process, and damage and damage characteristics of the upper skin structure are monitored; the middle section of the test loading piece (5) is arranged in the environment box (2); the loading actuating cylinder (1) acts on the free end of the test loading piece (5); and the other end of the test loading piece (5) is fixed through the fixing constraint (4).

Description

Technical Field

[0001] The invention belongs to the technical field of aircraft flight test, and relates to a typical damage monitoring device and method for load test flight. Background Art

[0002] Damage to aircraft key structures has a significant impact on flight safety. Monitoring the damage before it reaches a critical size through reasonable methods can, on the one hand, ensure flight safety; on the other hand, it can avoid unnecessary disassembly and maintenance of the airframe structure, improve the efficiency of airframe structure maintenance, and ensure aircraft attendance rate. Therefore, research on aircraft structure damage monitoring methods is of great significance.

[0003] The current aircraft structural damage monitoring methods are mainly divided into two categories: passive monitoring and active monitoring. Passive monitoring focuses on the measurement of various flight parameters and evaluates the structural health through these parameters. Passive monitoring of structural damage is widely used, but it cannot directly detect whether the structure is damaged. It is an indirect monitoring method. Active structural damage monitoring can directly detect existing and potential structural damage, thereby evaluating the health of the structure. The main monitoring technical means include Lamb waves and strain. This technology contains many specific types and is currently under research and exploration. In actual use, there are problems such as poor reliability and high false alarm rate. Therefore, it has not yet been widely used in practice, especially in the current real-time and online damage monitoring technology of aircraft structures in real flight environments. The development is relatively slow. Summary of the invention

[0004] Purpose of the invention: To propose a typical damage monitoring device and method for load test flight. The method adopts a piezoelectric active Lamb wave structure monitoring method and a strain distribution damage monitoring method to build a dual damage monitoring system, thereby improving the reliability of the damage monitoring system. Then, the structural damage of the detection object is classified and simulated, and a damage monitoring environment is constructed to simulate aircraft structural damage monitoring under real flight conditions, thereby achieving accurate damage identification to improve the alarm success rate, and laying a foundation for the subsequent installation and application of the damage monitoring method proposed in the present invention.

[0005] The technical solution of the present invention:

[0006] A typical damage monitoring device for load test flight includes: a loading actuator 1, an environmental chamber 2, a fixed constraint 4, a test load 5, and a damage feature 6, wherein:

[0007] The test loader 5 is an aluminum alloy box section, and the upper skin of the box section is divided into three parts, of which the upper skins at both ends are non-detachable skins, connected to the front and rear beam flanges and rib flanges by rivets; the upper skin of the middle section can be removed separately, and is connected by bolts + support plate nuts, which is used to arrange a piezoelectric sensor network array during the test to monitor the damage of the upper skin structure, and the damage feature 6 is prepared on the upper skin of the middle section; the middle section of the test loader 5 is set in the environmental chamber 2; the loading actuator 1 acts on the free end of the test loader 5; the other end of the test loader 5 is fixed by a fixed constraint 4; the environmental chamber 2 is used to realize temperature field simulation, which is used to keep the structural temperature field to a specific value and to change the structural temperature field in a step-by-step manner; the deformation field simulation is realized by distributed loading through the loading actuator 1, which is used to keep the structural loading load at a specific value and for step-by-step loading; the complex time-varying environment simulation is realized by distributed loading through the loading actuator 1 and the joint loading of the environmental chamber 2.

[0008] A typical damage monitoring method for load test flight includes:

[0009] Step 1: Arrange a piezoelectric sensor network in the middle section of the test loading member 5 and apply temperature and / or load, wherein the piezoelectric sensor network is in an array type;

[0010] Step 2: Based on the piezoelectric sensor network data of the healthy structural state, a baseline probability model of each excitation-sensing path in the healthy state of the structure is established through the Gaussian mixture model damage identification algorithm;

[0011] Step 3: construct damage feature 6 on the test loading component 5: damage feature 6 includes three types of damage, namely bolt withdrawal, prefabricated cracks and prefabricated pits, and apply temperature and / or load;

[0012] Step 4: Based on the piezoelectric sensor network data under the structural damage state, a monitoring probability model of each excitation-sensing path under the structural damage state is established through the Gaussian mixture model damage identification algorithm;

[0013] Step 5: Use the probability distribution migration distance calculation method to quantify the migration degree between the monitoring probability model and the benchmark probability model;

[0014] The migration degree is converted into a damage characteristic parameter, and the damage characteristic parameters of all excitation-sensing paths are compared. If there is a damage characteristic parameter greater than a damage alarm threshold, a damage alarm is generated, otherwise no damage alarm is generated;

[0015] Step 6: Using the damage characteristic parameters of each excitation-sensing path, construct the probability ellipse of each excitation-sensing path according to the path probability statistical imaging method; calculate the probability of damage occurrence at any point in the coverage area of ​​the piezoelectric sensor network, and use it as the pixel value of each point for imaging to obtain the damage imaging result.

[0016] Further, in step 1, applying temperature specifically includes:

[0017] The temperature was maintained at room temperature of 26°C, and data of each excitation-sensing path at room temperature was scanned and collected;

[0018] The temperature of the structure was lowered to -40°C, and 6 insulation points were designed between -40°C and 60°C. The data of each excitation-sensing path were scanned and collected at these insulation points. Then the temperature was lowered back to room temperature 26°C to complete a temperature cycle. During the test, the box segment loading load was set to 0kN.

[0019] Further, in step 1, applying temperature specifically includes:

[0020] The rate of temperature change is 0.5℃ / min, and the average time for completing a temperature cycle is 6.5 hours. The structural temperature in the test is measured using a platinum resistance.

[0021] Further, in step 1, applying a load specifically includes:

[0022] The deformation field is constructed by applying loads to two loading points at the free end of the test loading member (5) through two actuators;

[0023] During the test, the ambient temperature was kept at 26°C, and the load cycle was loaded from 0KN to 5KN and finally returned to 0KN to complete a loading cycle.

[0024] During the loading process, four holding points of 25%, 50%, 75%, and 100% are set. All excitation-sensing paths are scanned at each holding point. During the loading process, lamb wave signals are continuously scanned and collected.

[0025] Further, in step 1, applying temperature and load specifically includes:

[0026] The temperature and load are applied according to the following table:

[0027] Load Step Single point loading load cycle / KN Temperature field loading cycle / ℃ 1 0 26 2 0 -40 3 1.25 -20 4 2.5 0 5 3.75 20 6 5 40 7 0 60 8 0 26 .

[0028] A typical damage monitoring method for load test flight includes:

[0029] Step 1: planning a strain monitoring path according to the structural form of the test loading member (5) and the strain distribution after loading, arranging strain sensor measuring points along the strain monitoring path to form a strain sensor network, and applying temperature and / or load;

[0030] Step 2: Based on the strain sensor network data of the healthy structure, the Fréchet distance between each measuring point in the healthy state of the structure is calculated by the strain distribution similarity measurement method, and a cluster analysis is performed to establish a benchmark clustering model of strain distribution of the undamaged structure;

[0031] Step 3: constructing a damage feature (6) on the test loading component (5): the damage feature (6) includes three types of damage, namely bolt pullout, prefabricated cracks, and prefabricated pits, and applying temperature and / or load;

[0032] Step 4: Based on the strain sensor network data of the structural damage state, the Fréchet distance between each measuring point under the structural damage state is calculated by the strain distribution similarity measurement method, and a cluster analysis is performed to establish a monitoring clustering model for the strain distribution of the damaged structure;

[0033] Step 5: When the cluster to which the monitoring cluster model of the strain distribution of the damaged structure belongs is not in the cluster to which the benchmark cluster model established in the intact state belongs, it is considered that the structural strain distribution has changed significantly and the structure is in a damaged state.

[0034] Further, in step 1, applying temperature specifically includes:

[0035] The temperature was maintained at room temperature of 26°C, and data from each strain sensor measuring point at room temperature was collected;

[0036] The temperature of the structure was lowered to -40°C, and 6 insulation points were designed between -40°C and 60°C. Data of each strain sensor measuring point at room temperature were collected at these insulation points. Then the temperature was lowered back to room temperature 26°C to complete a temperature cycle. During the test, the box segment loading load was set to 0kN.

[0037] Further, in step 1, applying a load specifically includes:

[0038] The deformation field is constructed by applying loads to two loading points at the free end of the test loading member (5) through two actuators;

[0039] During the test, the ambient temperature was kept at 26°C, and the load cycle was loaded from 0KN to 5KN and finally returned to 0KN to complete a loading cycle.

[0040] During the loading process, four holding points of 25%, 50%, 75%, and 100% are set, and data are collected from all strain sensor measuring points at each holding point.

[0041] Further, in step 1, applying temperature and load specifically includes:

[0042] The temperature and load are applied according to the following table:

[0043]

[0044]

[0045] Beneficial effects:

[0046] The method of the present invention first classifies and simulates the structural damage of the detection object, and then respectively uses the piezoelectric active Lamb wave structure monitoring method and the strain distribution damage monitoring method to build a damage monitoring dual system, and builds a damage monitoring device to simulate aircraft structural damage monitoring under real flight conditions. It is intended to lay the foundation for the subsequent installation and application of the damage monitoring method proposed by the present invention, and has certain guiding significance for the advancement of load test flight technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope.

[0048] Figure 1 Schematic diagram of the piezoelectric sensor array layout;

[0049] Figure 2 Piezoelectric sensing damage monitoring integrated system;

[0050] Figure 3 Schematic diagram of strain path planning;

[0051] Figure 4 Similarity metric damage monitoring method for strain sensors;

[0052] Figure 5 Schematic diagram of the damage point preparation of the test loading part;

[0053] Figure 6 A schematic diagram of a typical damage monitoring device for load test flight;

[0054] No. 1 represents the loading actuator, simulating the deformation field, No. 2 represents the environmental chamber, No. 3 represents the temperature field step loading diagram, No. 4 represents the fixed constraint, No. 5 represents the test loading part, No. 6 represents the damage feature, and No. 7 represents the damage monitoring integrated system;

[0055] Figure 7 Cyclic loading diagram of test temperature field;

[0056] Figure 8 Loading curve diagram of single loading point test. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. It should be understood that the drawings in the present invention are only for the purpose of illustration and description and are used to limit the protection scope of the present invention. In addition, it should be understood that the schematic drawings are not drawn according to the actual scale or numerical value.

[0058] The embodiments described in the present invention 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 skilled in the art without creative work belong to the protection scope of the present invention.

[0059] like Figure 6 The present invention provides a typical damage monitoring device for load test flight, including: a loading actuator 1, an environmental chamber 2, a fixed constraint 4, a test loading member 5, and a damage feature 6, wherein the test loading member 5 is an aluminum alloy box section, and the upper skin of the box section is divided into three parts, wherein the upper skins at both ends are non-detachable skins, connected to the front and rear beam flanges and rib flanges by rivets; the upper skin of the middle section can be detached separately, and a bolt + support plate nut connection method is adopted, which is used to arrange a piezoelectric sensor network array during the test process to monitor the damage of the upper skin structure, and the damage feature 6 is obtained in the middle section. The upper skin of the test loader 5 is arranged in the environmental chamber 2; the middle section of the test loader 5 is arranged in the environmental chamber 2; the loading actuator 1 acts on the free end of the test loader 5; the other end of the test loader 5 is fixed by the fixed constraint 4; the environmental chamber 2 is used to realize the temperature field simulation, which is used to keep the structural temperature field to a specific value and to change the structural temperature field in a step-by-step manner; the deformation field simulation is realized by the distributed loading of the loading actuator 1, which is used to keep the structural loading load at a specific value and for step-by-step loading; the complex time-varying environment simulation is realized by the distributed loading of the loading actuator 1 and the joint loading of the environmental chamber 2.

[0060] The present invention proposes a typical damage and monitoring environment construction method for load test flight. The method uses a piezoelectric active Lamb wave structure monitoring method and a strain distribution damage monitoring method to build a dual damage monitoring system, and then classifies and simulates the structural damage of the detection object, and constructs a damage monitoring environment to simulate aircraft structural damage monitoring under real flight conditions.

[0061] A typical damage and monitoring environment construction method for load test flight of the present invention comprises:

[0062] Step 1: Build a piezoelectric active lamb wave damage monitoring system

[0063] a) Piezoelectric sensor network layout. For the layout of piezoelectric sensors, the minimum distance between the excitation point and each piezoelectric sensor should be determined based on the material, structural thickness and excitation frequency range of the test load, combined with the dispersion characteristic curve of the test load material. At least the time difference of the lamb wave propagating from the excitation point to each piezoelectric sensor should be greater than one wave packet length, so that the piezoelectric sensor only receives one mode of wave signal at the same time. The piezoelectric sensor network layout of this test load is as follows: Figure 1 , where the red line of the sensor is the excitation-sensing path to be calculated. In the present invention, the Shannon continuous complex wavelet analysis method is used to analyze the Lamb wave signal and perform noise reduction processing.

[0064] b) Integration of structural damage identification and location methods.

[0065] The Gaussian mixture model damage identification algorithm is used to establish the baseline probability model of each excitation-sensing path in the healthy state and the damaged state of the structure; the probability distribution migration distance calculation method is used to quantify the migration degree between the monitoring probability model and the baseline probability model to obtain the damage characteristic parameters. Using the damage characteristic parameters of each excitation-sensing path, the probability ellipse of each excitation-sensing path is constructed according to the path probability statistical imaging method; the probability of damage occurrence at any point in the coverage area of ​​the piezoelectric sensor network is calculated, and the pixel value of each point is used for imaging to obtain the damage imaging results.

[0066] c) Integration of piezoelectric sensor damage monitoring integrated system

[0067] The system adopts modular design and is divided into core management module, waveguide excitation module, waveguide power amplification module, waveguide excitation-sensing path switching module, waveguide signal conditioning module and waveguide signal acquisition module. Figure 2 shown.

[0068] Step 2: Build a strain sensor damage monitoring system

[0069] a) Strain sensor network layout

[0070] The strain measuring points are arranged in different directions of the monitoring area. The strain curves composed of strains in different directions will be discontinuous. Therefore, in order to facilitate the measurement of curve similarity, the measuring points are divided into different measurement paths according to the actual distribution. The continuity of the paths should be ensured during the division to avoid turning points, such as Figure 3 shown.

[0071] It should be noted that although Figure 3The medium strain measuring points are arranged around the damage, but in actual applications, because the damage location is unknown, the strain measuring points should be arranged around the key damage monitoring location, and the actual measuring point arrangement method can be adjusted according to the structural form and the strain distribution after loading.

[0072] Damage Monitoring Method Based on Strain Sensor

[0073] The basic principle is to establish a damage monitoring method based on the strain distribution similarity measurement. First, the strain on the path is normalized, and then the Fréchet distance of each measuring point is calculated to determine the similarity of strain distribution.

[0074] The maximum-minimum value normalization method is adopted in the present invention, and the calculation method is as follows:

[0075]

[0076] where x min and x max are the minimum and maximum strain values ​​on the interval, respectively, 0,1 is the normalized value.

[0077] The Fleche distance method is selected to measure the similarity of the strain distribution curves. This method is a classic path space similarity description method. The calculation process is as follows:

[0078] Let the binary (S, d) be a metric space, where d is a metric function on S. Let A and B be two strain sensor distribution paths on S, i.e. A: [0, 1] → S, B: [0, 1] → S. Let α and β be two parameterized functions of the unit interval, i.e. α: [0, 1] → S, β: [0, 1] → S. Then the Fréchet distance F(A, B) between path A and path B is defined as:

[0079] F(A,B)=inf α,β max t∈[0,1] {d(A(α(t)),B(β(t)))} (2)

[0080] Assume that t is a time point, the sampling point on path A at this moment is A(α(t)), the sampling point on path B is B(β(t)), and the Euclidean distance is defined as d(A(α(t)), B(β(t))). In each sampling, t discretely traverses the interval [0,1], and the maximum distance max under this sampling is obtained. t∈[0,1] {d(A(α(t)),B(β(t)))}. The Fréchet distance is the value under the sampling method that minimizes the maximum distance.

[0081] c) Use clustering ideas to determine whether the structure is damaged. For aircraft structures, they are generally regarded as linear elastic structures in an undamaged state, that is, the external load is proportional to the structural strain. Therefore, when the external load action points are the same but the values ​​are different, the normalized curve shape of the strain distribution on the measurement path should be consistent, that is, the Fréchet distance is the same. Perform cluster analysis on the data points and establish a clustering model that characterizes the strain distribution of undamaged structures under different loading conditions. When the cluster to which the monitoring clustering model of the strain distribution of the damaged structure belongs is not in the cluster to which the benchmark clustering model established in the undamaged state belongs, it is considered that the structural strain distribution has changed significantly and the structure is in a damaged state. See the method flow chart for details. Figure 4 .

[0082] Step 3: Three typical forms of damage, namely nail removal, cracks and corrosion pits, are produced on the real aircraft structure test loading parts for damage simulation.

[0083] The test loading part is an aluminum alloy box section. The upper skin of the box section adopts a block design, which is divided into 3 parts in total. The upper skin at both ends is non-detachable and connected to the front and rear beam flanges and rib flanges by rivets; the upper skin in the middle section can be removed separately and connected by bolts + support plate nuts. It is used to arrange the piezoelectric sensor network array during the test to monitor the damage of the upper skin structure. The damage is made in the upper skin in the middle section, such as Figure 5 .

[0084] a) Cracks are produced on the structural skin by on-site cutting. In the crack damage monitoring test, cracks of specified length (unit: mm) are cut at the edge of the bolt hole. The crack direction is parallel to the rib direction.

[0085] b) In the nail pull failure damage monitoring test, the nail pull failure damage was simulated by loosening the bolts. A bolt connecting the upper skin and the wing rib of the metal wing box structure was pulled out to induce nail pull failure damage.

[0086] c) The pits are made by grinding on the structure. In the pit damage monitoring test, a certain area on the outer surface of the skin of the metal wing box structure is selected and the pit damage is ground by a grinder.

[0087] Constructing a damage environment monitoring device:

[0088] The damage monitoring environment is constructed. The damage monitoring environment includes three parts: temperature field simulation, deformation field simulation, and complex time-varying environment simulation. Figure 6 shown.

[0089] a) Temperature field simulation is achieved using a laboratory environmental chamber, which is used to keep the structural temperature field at a specific value and to change the structural temperature field in a step-by-step manner.

[0090] b) The deformation field simulation is achieved by distributing the loading of the loading actuator, so that the structural loading load is kept at a specific value and step-by-step loading is achieved.

[0091] c) Complex time-varying environment simulation is a complex environment when temperature-load coupling loading occurs.

[0092] The environmental simulation system of the present invention can simulate the time-varying temperature environment and structural load environment under real flight conditions.

[0093] a) Dynamic temperature field construction

[0094] Dynamic temperature field construction adopts Figure 6 The test temperature field cycle is as follows: Figure 7 As shown. First, the temperature is maintained at room temperature (26°C), and the data of each excitation-sensing path and strain sensor measuring point at room temperature are scanned and collected. Then, the temperature of the structure is reduced to -40°C, and 6 insulation points are designed between -40°C and 60°C. The data of each excitation-sensing path and strain sensor measuring point are scanned and collected at these insulation points. Then, the temperature is reduced back to room temperature (26°C) to complete a temperature cycle. During the test, the box section loading load is set to 0kN, and the guided wave signal affected by dynamic temperature is continuously collected. The temperature change rate is 0.5°C / min. The average time to complete a temperature cycle is 6.5 hours. The structural temperature in the test is measured by platinum resistance.

[0095] b) Dynamic deformation field construction

[0096] The construction of the deformation field is achieved by applying loads to the wing tip loading joints by the actuator, which are named loading points 1 and 2 respectively. During the test, the ambient temperature is kept at 26°C, and the load cycle is loaded from 0KN to 5KN and finally back to 0KN to complete a loading cycle. Figure 8 During the loading process, four holding points of 25%, 50%, 75%, and 100% are set. All excitation-sensing paths are scanned at each holding point. During the loading process, the guided wave signals are continuously scanned and collected, and data is also collected at all strain sensor measuring points.

[0097] c) Dynamic temperature-load coupling field construction

[0098] The temperature-load coupling test cycle is a comprehensive test based on the dynamic temperature impact test and the loading test. The load is carried out at the temperature cycle insulation point of the dynamic temperature impact test. Figure 7 and Figure 8 The piezoelectric sensor network is scanned and waveguide data is collected at each load holding point and loading process, and data is also collected at all strain sensor measurement points.

[0099] Temperature-load coupling loading cycle table

[0100] Load Step Single point loading load cycle / KN Temperature field loading cycle / ℃ 1 0 26 2 0 -40 3 1.25 -20 4 2.5 0 5 3.75 20 6 5 40 7 0 60 8 0 26

[0101] After the above steps, a typical damage and monitoring environment simulation system for aircraft structures for load test flight can be constructed.

Claims

1. A typical damage monitoring device for load test flight, characterized in that: include: A loading actuator (1), an environmental chamber (2), a fixed constraint (4), a test load (5), and a damage feature (6), wherein: The test loading member (5) is an aluminum alloy box section, and the upper skin of the box section is divided into three parts, wherein the upper skins at both ends are non-detachable skins, connected to the front and rear beam flanges and rib flanges by rivets; the upper skin of the middle section can be removed separately, and is connected by bolts + support plate nuts, and is used to arrange a piezoelectric sensor network array during the test process to monitor the damage of the upper skin structure, and the damage characteristics (6) are obtained in the upper skin of the middle section; the middle section of the test loading member (5) is arranged in an environmental chamber (2); the loading actuator (1) acts on the free end of the test loading member (5); the other end of the test loading member (5) is fixed by a fixed constraint (4); the environmental chamber (2) is used to realize temperature field simulation, which is used to keep the structural temperature field at a specific value and to change the structural temperature field in a step-by-step manner; the deformation field simulation is realized by distributed loading of the loading actuator (1), which is used to keep the structural loading load at a specific value and to load in a step-by-step manner; the complex time-varying environment simulation is realized by distributed loading of the loading actuator (1) and joint loading of the environmental chamber (2).

2. A typical damage monitoring method for load test flight, characterized in that: The method is performed based on the typical damage monitoring device for load test flight according to claim 1, and the method comprises: Step 1: Arrange a piezoelectric sensor network in the middle section of the test loading member (5) and apply temperature and / or load, wherein the piezoelectric sensor network is in an array type; Step 2: Based on the piezoelectric sensor network data of the healthy structural state, a baseline probability model of each excitation-sensing path in the healthy state of the structure is established through the Gaussian mixture model damage identification algorithm; Step 3: constructing a damage feature (6) on the test loading component (5): the damage feature (6) includes three types of damage, namely bolt pullout, prefabricated cracks, and prefabricated pits, and applying temperature and / or load; Step 4: Based on the piezoelectric sensor network data under the structural damage state, a monitoring probability model of each excitation-sensing path under the structural damage state is established through the Gaussian mixture model damage identification algorithm; Step 5: Use the probability distribution migration distance calculation method to quantify the migration degree between the monitoring probability model and the benchmark probability model; The migration degree is converted into a damage characteristic parameter, and the damage characteristic parameters of all excitation-sensing paths are compared. If there is a damage characteristic parameter greater than a damage alarm threshold, a damage alarm is generated, otherwise no damage alarm is generated; Step 6: Using the damage characteristic parameters of each excitation-sensing path, construct the probability ellipse of each excitation-sensing path according to the path probability statistical imaging method; calculate the probability of damage occurrence at any point in the coverage area of ​​the piezoelectric sensor network, and use it as the pixel value of each point for imaging to obtain the damage imaging result.

3. The typical damage monitoring method for load test flight according to claim 2 is characterized in that: In step 1, applying temperature specifically includes: The temperature was maintained at room temperature of 26°C, and data of each excitation-sensing path at room temperature was scanned and collected; The temperature of the structure was lowered to -40°C, and 6 insulation points were designed between -40°C and 60°C. The data of each excitation-sensing path were scanned and collected at these insulation points. Then the temperature was lowered back to room temperature 26°C to complete a temperature cycle. During the test, the box segment loading load was set to 0kN.

4. The typical damage monitoring method for load test flight according to claim 2 is characterized in that: In step 1, applying temperature specifically includes: The rate of temperature change is 0.5℃ / min, and the average time to complete a temperature cycle is 6.5 hours. The structural temperature in the test is measured using a platinum resistance.

5. The typical damage monitoring method for load test flight according to claim 2 is characterized in that: In step 1, the load is applied, including: The deformation field is constructed by applying loads to two loading points at the free end of the test loading member (5) through two actuators; During the test, the ambient temperature was kept at 26°C, and the load cycle was loaded from 0KN to 5KN and finally returned to 0KN to complete a loading cycle. During the loading process, four holding points of 25%, 50%, 75%, and 100% are set. All excitation-sensing paths are scanned at each holding point. During the loading process, lamb wave signals are continuously scanned and collected.

6. The typical damage monitoring method for load test flight according to claim 2 is characterized in that: In step 1, temperature and load are applied, including: The temperature and load are applied according to the following table: 。 7. A typical damage monitoring method for load test flight, characterized in that: The method is performed based on the typical damage monitoring device for load test flight according to claim 1, and the method comprises: Step 1: planning a strain monitoring path according to the structural form of the test loading member (5) and the strain distribution after loading, arranging strain sensor measuring points along the strain monitoring path to form a strain sensor network, and applying temperature and / or load; Step 2: Based on the strain sensor network data of the healthy structure, the Fréchet distance between each measuring point in the healthy state of the structure is calculated by the strain distribution similarity measurement method, and a cluster analysis is performed to establish a benchmark clustering model of strain distribution of the undamaged structure; Step 3: constructing a damage feature (6) on the test loading component (5): the damage feature (6) includes three types of damage, namely bolt pullout, prefabricated cracks, and prefabricated pits, and applying temperature and / or load; Step 4: Based on the strain sensor network data of the structural damage state, the Fréchet distance between each measuring point under the structural damage state is calculated by the strain distribution similarity measurement method, and a cluster analysis is performed to establish a monitoring clustering model for the strain distribution of the damaged structure; Step 5: When the cluster to which the monitoring cluster model of the strain distribution of the damaged structure belongs is not in the cluster to which the benchmark cluster model established in the intact state belongs, it is considered that the structural strain distribution has changed significantly and the structure is in a damaged state.

8. The typical damage monitoring method for load test flight according to claim 7 is characterized in that: In step 1, applying temperature specifically includes: The temperature was maintained at room temperature of 26°C, and data from each strain sensor measuring point at room temperature was collected; The temperature of the structure was lowered to -40°C, and 6 insulation points were designed between -40°C and 60°C. Data of each strain sensor measuring point were collected at these insulation points, and then the temperature was lowered back to room temperature 26°C to complete a temperature cycle. During the test, the loading load of the box section was set to 0kN.

9. The typical damage monitoring method for load test flight according to claim 7 is characterized in that: In step 1, the load is applied, including: The deformation field is constructed by applying loads to two loading points at the free end of the test loading member (5) through two actuators; During the test, the ambient temperature was kept at 26°C, and the load cycle was loaded from 0KN to 5KN and finally returned to 0KN to complete a loading cycle. During the loading process, four holding points of 25%, 50%, 75%, and 100% are set, and data are collected from all strain sensor measuring points at each holding point.

10. The typical damage monitoring method for load test flight according to claim 7, characterized in that: In step 1, temperature and load are applied, including: The temperature and load are applied according to the following table: 。

Images