A method for measuring high temperature damage distribution and evolution of ceramic matrix composite structures

Abstract

The present invention discloses a method for measuring the high-temperature damage distribution and evolution of a ceramic matrix composite material structure, comprising the following steps: selecting electrode monitoring points on the CMC structure according to the geometric features of the CMC structure; considering the electrode monitoring points as resistance nodes according to the size of the electrode monitoring points prepared on the CMC structure and the corresponding monitoring area, and obtaining a resistance network model of the CMC structure; determining the non-material resistance and the initial value of the introduced contact resistance through the initial value of the resistance response of the CMC structure obtained in the high-temperature test; constructing a data set of resistance changes and their corresponding responses in the resistance network; training a deep neural network model, substituting the resistance response of the CMC structure into the trained deep neural network model, and obtaining the high-temperature damage distribution and evolution process on the geometric space of the CMC structure. The present invention is suitable for measuring the damage distribution and evolution of CMC structures in high-temperature environments, and can solve the problem of health monitoring of CMC structures in engineering applications.

Description

Technical Field

[0001] The present invention belongs to the technical field of health monitoring of ceramic matrix composite materials, and in particular relates to a method for measuring the high-temperature damage distribution and evolution of a ceramic matrix composite material structure. Background Art

[0002] Fiber-reinforced ceramic matrix composites (Ceramic Matrix Composite, CMC) have the advantages of light weight and high strength, and they still have excellent mechanical properties in high temperature environments. Applying CMC to high-temperature connecting pin connection components of advanced aircraft engines can effectively reduce the overall weight, improve the thrust-to-weight ratio of aircraft engines, and further enhance the performance of aircraft engines.

[0003] Under long-term load-bearing conditions, the CMC pin structure is damaged, which affects the reliability and service life of the connection structure. However, the pin is buried inside the connection structure, and conventional methods are difficult to identify the CMC damage in real time, and thus cannot determine the health status of the pin. At present, there are a variety of damage evaluation technologies that can be used for CMC in the prior art, including acoustic emission (CN106770677A) and X-ray (CN108717727A), and have achieved certain application effects. However, the above methods require large and complex equipment, and cannot achieve online monitoring of the CMC structure of an aircraft engine during service. At present, when using the above technology, the structure needs to be disassembled and assembled, which is easy to cause accidental damage and has a high application cost. At the same time, the precise probes of the equipment cannot withstand the high-temperature service environment, and it is difficult to break through the difficulty of measuring the damage of the CMC structure under high temperature environment. In recent years, studies have found that the resistance change of CMC is closely related to the damage distribution, and it has the advantages of reliable electrode probes, sophisticated monitoring equipment, and in-situ real-time measurement without disassembly. Therefore, the use of damage measurement technology based on resistance monitoring is expected to solve the problem of online monitoring of the health of CMC structures under high temperature environments.

[0004] However, the existing resistance-based monitoring technology is mainly used to characterize the damage-resistance response characteristics of ceramic matrix composites, and the damage monitoring results extracted based on this single-point resistance measurement cannot be used to predict CMC structures with complex spatial geometric characteristics. Therefore, it is necessary to propose a method that can measure the damage distribution and evolution of CMC structures in high-temperature environments, which can realize the measurement of CMC structure-specific and non-uniform spatial damage, so as to solve the problem of measuring the damage distribution and evolution of high-temperature CMC components of aircraft engines. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a method for measuring the high temperature damage distribution and evolution of a ceramic matrix composite material structure in view of the above-mentioned existing deficiencies.

[0006] In order to achieve the above technical objectives, the technical solution adopted by the present invention is:

[0007] A method for measuring high temperature damage distribution and evolution of a ceramic matrix composite material structure comprises the following steps:

[0008] Step 1: According to the geometric characteristics of the CMC structure, multiple electrode monitoring points are selected on the CMC structure, and an ohmmeter is connected between two electrode monitoring points forming an electrode pair;

[0009] Step 2: According to the size of the electrode monitoring points prepared on the CMC structure and the corresponding monitoring area, the electrode monitoring points are regarded as resistance nodes, and the propagation resistance is divided to obtain the undetermined non-material resistance R0 and the introduced contact resistance R c Resistor network model of CMC structure with initial values;

[0010] Step 3: Based on the resistance network model, a numerical simulation model of the resistance network model is established. Several electrode pairs are selected as monitoring channel nodes. The numerical response of the resistance network model is monitored by using an ohmmeter set at the corresponding monitoring channel node. The non-material resistance R0 and the introduced contact resistance R are determined by combining the initial value of the CMC structure resistance response obtained in the high temperature test with the established numerical simulation model. c Initial value;

[0011] Step 4: Combine the damage mode and failure law of the CMC structure with the geometric characteristics of the structure and resistor network model to construct a data set of resistance changes and their corresponding responses in the resistor network;

[0012] Step 5: Select a deep neural network architecture that can meet the needs of response resistor and resistor network data mapping, and train the deep neural network model in combination with the data set;

[0013] Step 6: During the training of the deep neural network model, the model effect in each training cycle is verified by combining the test results, and the training process is iterated repeatedly to obtain a trained deep neural network model;

[0014] Step 7. Substitute the resistance response of the CMC structure into the trained deep neural network model to obtain the evolution process of the resistance in the CMC structure with the test time, map the resistance change to the geometric space of the CMC structure, and then intuitively obtain the high-temperature damage distribution and evolution process in the geometric space of the CMC structure.

[0015] To optimize the above technical solutions, the specific measures taken also include:

[0016] In step 1, a high temperature resistant electrode is arranged at the electrode monitoring point, and the high temperature resistant electrode is connected to an ohmmeter through a wire. The high temperature resistant electrode is made of silver powder conductive glue, and the wire is made of nickel-titanium alloy wire.

[0017] In step 1, the electrode pairs are respectively arranged on both sides of the cracking, delamination or fracture position of the CMC structure.

[0018] In step 2, the electrode monitoring points are regarded as resistance nodes, the propagation resistance is divided, and the resistance value of the structural propagation resistance is determined by the following calculation equation, thereby obtaining the resistance network model of the CMC structure;

[0019]

[0020] Where R xi is the resistance value of the i-th propagation resistor in the x direction, R yj is the resistance value of the jth propagation resistor in the y direction, R zij is the propagation resistance in the z direction between the i-th electrode pair in the x direction and the j-th electrode pair in the y direction; ρ is the material resistivity of the CMC structure; L xi is the length from the i-th node to the i+1-th node in the x direction, A xi is the propagation resistance area between the i-th node and the i+1-th node in the x direction; L yj is the length from the jth node to the j+1th node in the y direction, A yj is the propagation resistance area between the jth node and the j+1th node in the y direction; l ij is the length of the CMC structure between the electrode pairs, A ij is the area of ​​the high temperature resistant electrode.

[0021] In step 3, determine the non-material resistance R0 and the introduced contact resistance R c The specific method of initial value is:

[0022] Step 1: First assume that the non-material resistance R0 and the contact resistance R c All are 0, and are brought into the numerical simulation model to calculate and obtain the theoretical response of the resistor network;

[0023] Step 2: If the response value does not reach the initial value of the CMC structure resistance response, the non-material resistance R0 and the contact resistance R c After increasing by 0.01, they were brought into the numerical simulation model again to calculate the response value;

[0024] Step 3: Compare with the initial value of the CMC structure resistance response. If it is not reached, repeat step 2 until the non-material resistance R0 and the contact resistance R c The numerical value meets the initial value of the resistance response of the CMC structure, and all the initial parameters in the resistance network model are obtained by calculation.

[0025] In step 4, the geometric characteristics of the resistor network model are as follows: the resistor network model is divided into two layers, α and β, and the two layers are connected by a CMC connection structure. The CMC connection structure is regarded as a resistor node. After the propagation resistance is divided, the propagation resistance node in the x direction is numbered as m, and the propagation resistance node in the y direction is numbered as n. The contact resistance increment matrix of the CMC connection structure is set to ΔR c , the load-bearing characteristics of the CMC connection structure are the separation of the connection plate and the breakage of the pins. Therefore, the evolution characteristics of the CMC resistance network model are: the contact resistance R in the CMC resistance network c The edge response matrix is ​​set to S e , where S e is the edge node resistance response of the resistor network model. When the contact resistance changes, the edge response also changes. The function that converts the contact resistance increment matrix into the edge response matrix is:

[0026] S e =F(R mn )=F(R0+ΔR c )

[0027] Function F(x) is the nonlinear mapping of the resistor network, R mn is the resistance value of the CMC connection structure;

[0028] If there is an inverse mapping of the resistance network, the contact resistance increment matrix of the CMC connection structure can be obtained through the edge response matrix, that is,

[0029] ΔR c =F -1 (S e )-R0

[0030] First, generate Q groups of uniformly distributed random number resistor increment matrices And bring it into the resistor network model to calculate the corresponding edge response matrix Right now:

[0031]

[0032] The above Q group resistance increment matrix and its corresponding edge response matrix are used as data sets to train the deep neural network.

[0033] In step 5, the selected deep neural network architecture is the GANs architecture.

[0034] The beneficial effects of the present invention are:

[0035] 1. The present invention proposes a method for establishing a CMC structure resistor network model, which realizes the prediction of the resistor network evolution process, thereby being able to more intuitively obtain the spatial geometric characteristics of the CMC structure damage distribution and improve the damage location accuracy of the CMC structure in engineering applications;

[0036] 2. The CMC high temperature damage distribution and evolution measurement method proposed in the present invention is universal. By selecting a suitable arrangement of resistance nodes, it can meet the monitoring requirements of various components with complex geometric features such as aircraft engine flame tubes, turbine guide vanes, tail nozzle adjustment plates, etc. It has a wide range of applications and high prediction accuracy.

[0037] 3. The high-temperature damage distribution and evolution measurement method of CMC structure proposed in the present invention expands the resistance-based monitoring technology from one dimension to two dimensions, thereby reversely mapping the damage distribution characteristics of the hot end components of aircraft engines, providing a reference for engineering personnel to inspect and repair the hot end components of aircraft engines. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic diagram of the monitoring electrode setup of the CMC connection structure;

[0039] Figure 2 It is the resistance monitoring value of the CMC connection structure during load bearing;

[0040] Figure 3 It is the resistance network model of CMC connection structure;

[0041] Figure 4 is the resistor network simulation model;

[0042] Figure 5 It is the algorithm for determining the initial contact resistance of the CMC connection structure;

[0043] Figure 6 is the relationship between the edge response of the CMC connection structure and time;

[0044] Figure 7 It is the deep neural network architecture established;

[0045] Figure 8 It is the predicted result of the resistance increment of the CMC connection structure. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated 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 intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0047] Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. In addition, it can also be understood that although the efforts made in this development process may be complicated and lengthy, for ordinary technicians in this field related to the content disclosed in this application, some changes in design, manufacturing or production based on the technical content disclosed in this application are just conventional technical means, and should not be understood as insufficient content disclosed in this application.

[0048] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0049] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantitative limitation, and may represent the singular or plural. The terms "include", "comprise", "have" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or units (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" / "several" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships, for example, "A and / or B" can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0050] This embodiment takes the CMC pin connection structure in the hot end component of an aircraft engine as an example, and further describes the present invention in detail in conjunction with the accompanying drawings.

[0051] like Figure 1 As shown, for the CMC pin connection structure, a total of 3 pairs of electrode monitoring points are designed in this embodiment, which are located at the two vertices of the connection area and the center pin respectively. The structural member is symmetrical along the central axis of the x-direction, and the two electrodes along the z-direction form a group of electrode pairs, which are used to monitor the separation state between the plates of the connection structure. After the connection structure fails, the connection plate separates to break the circuit, and the final resistance response when the structure fails is obtained. Among them, channel 1 is located at the end of plate A in the structural connection area, channel 2 is located at the end of plate B in the structural connection area, and channel 3 is located at the center pin of the structural connection area. The resistance monitoring values ​​during the test are shown as follows. Figure 2 shown.

[0052] like Figure 3 As shown in the figure, according to the geometric characteristics and failure modes of the structure, the present invention establishes a resistance network model of the CMC connection structure. The resistance network model is divided into two layers, α and β, corresponding to the 3mm thick connection plate and 2.5mm thick connection plate of the connection structure respectively. The connection area is 30mm long and 20mm wide, and the connection plate node spacing is 5mm. A total of 83 resistors are divided, of which 24 are α plate propagation resistors, 24 are β plate propagation resistors, and 35 are inter-plate contact resistors. R αxi and R βxi is the propagation resistance of the connecting plate in the x direction, R αyj and R βyj is the propagation resistance of the connecting plate in the y direction, R pin is the resistance of the pin in z direction, calculated using formula (1). mn (m=1,2,...,5,n=1,2,...,7) is the contact resistance between plates, and its value is affected by the damage state of the connection structure.

[0053]

[0054] Where ρ is the material resistivity, L α With L β A is the resistance length between the connection plate nodes; α With A β is the resistance area between the connection plate nodes; L pin is the pin length, A pin is the pin area, i is the x-direction propagation resistor number, j is the y-direction propagation resistor number, m is the x-direction propagation resistor node number, and n is the y-direction propagation resistor node number. Since the connecting plate is almost undamaged during the CMC connection structure load-bearing process, the x-direction and y-direction propagation resistances of the connecting plate in the embodiment resistance network model remain unchanged. Table 1 lists the calculation parameters of the resistance network model in this embodiment.

[0055] Table 1 Resistor network model parameters

[0056]

[0057]

[0058] Furthermore, for different connection structures, the initial inter-plate contact pressure and deposition state are different, and the initial contact resistance is also different. Therefore, it is necessary to determine the initial value of the inter-plate contact resistance in this embodiment. Figure 4 The figure shows the established resistance network simulation model of the CMC connection structure. An ohmmeter is set in the resistance network model to obtain the resistance response of the corresponding monitoring channel in the high temperature test. This embodiment determines the initial contact resistance of the connection structure by monitoring the initial value of the response and combining it with the established resistance network simulation model. Considering the different bonding states of the wire, conductive glue and test piece when making high-temperature electrodes, 5% of the initial external resistance is compensated based on experience, and the initial resistance network response of channels 1 and 2 is determined to be 4.91Ω. Figure 5 The process shown is as follows: ① first assume that the contact resistance value is 0, and bring it into the resistance network simulation model for calculation to obtain the theoretical response of the resistance network; ② if the response value does not reach the initial response value in the test, increase the contact resistance by 0.01 and then substitute it into the resistance network model again to calculate the response value; ③ compare it with the initial value of the test response. If it is not reached, repeat step ② until the non-contact resistance value meets the initial value of the test response obtained in the test. The contact resistance of the CMC connection structure in the initial state in this embodiment is calculated to be 176.50Ω.

[0059] Next, a deep neural network is trained to predict the evolution of the resistance network of the CMC connection structure based on the experimental monitoring response. The incremental matrix of the contact resistance (including the pin resistance) of the CMC connection structure is set to ΔR c , the edge response matrix is ​​set to S e The specific meaning of edge response is Figure 4 The five ohmmeter values ​​on the left are S1 to S5. The edge response changes when the contact resistance changes. Figure 5 The connection structure resistance network model shown can transform the contact resistance increment matrix into a function of the edge response matrix, that is, formula (2):

[0060] S e =F(R mn )=F(R0+ΔR c ) (2)

[0061] Among them, R0 is the initial contact resistance in the undamaged state, and the initial value has been determined. ΔR c The expression is:

[0062]

[0063] Since the CMC pin connection structure is symmetrical along the central axis, it is assumed that the damage state of the structure is also symmetrical along the central axis. c Simplified to:

[0064]

[0065] Edge response matrix S e The expression is:

[0066]

[0067] The function F(x) is essentially a nonlinear mapping of the resistor network. It is given by the contact resistance increment matrix ΔR of the CMC connection structure. c , the edge response matrix S corresponding to the damage state can be obtained e If there is an inverse mapping of the resistance network, the contact resistance increment matrix of the connection structure can be obtained through the edge response matrix, that is, formula (3):

[0068] ΔR c =F -1 (S e )-R0 (3)

[0069] Deep neural networks have powerful nonlinear function approximation capabilities. According to the universal approximation theorem, neural networks can approximate any continuous function. In this embodiment, generative adversarial networks (GANs) are used to learn the patterns in the contact resistance increment matrix and edge response matrix data to find the potential mapping relationship. First, 500,000 sets of uniformly distributed random number resistance increment matrices ΔR are generated. i c , and bring in Figure 4 The corresponding edge response S is calculated in the resistor network model shown in i e , that is, formula (4):

[0070]

[0071] The above 500,000 sets of resistance increment matrices and their corresponding edge responses are used as data sets to train a deep neural network.

[0072] In this embodiment, the number of monitoring channels of the CMC structure in a high temperature environment is limited by the physical volume of the electrode and environmental conditions. In general, the actual number of monitoring channels is smaller than the number of edge responses. Therefore, it is necessary to first establish the experimental monitoring response S of channel 1 and channel 2. c With edge response S e The mapping relationship between them is formula (5):

[0073] S e =H(S c) (5)

[0074] in

[0075]

[0076] This embodiment uses a multi-layer perceptron (MLP) to learn the complex mapping relationship between the test monitoring response and the edge response, that is, to predict the edge responses S2, S3 and S4 through the test monitoring responses S1 and S5. The data sets are split into input monitoring responses and output edge responses. and Right now

[0077]

[0078] The MLP network structure proposed in this embodiment includes 1 input layer, 2 hidden layers and 1 output layer. The input is a [2,1] matrix, and the input layer contains 2 neurons. Each hidden layer of the model contains 5 neurons, and the nonlinear activation function ReLU is used. The output layer contains 3 neurons, and the output is a [3,1] matrix. Figure 6 As shown, based on this model, the time-dependent relationship of the edge response can be predicted by experimentally monitoring the responses of channels 1 and 2.

[0079] The time-varying CMC connection structure resistance network edge response matrix S is obtained based on the experimental monitoring response e After that, the deep neural network model G(x) is used to predict the resistance increment matrix ΔR of the structure that changes with time during the load-bearing process. c . Figure 7 The following is the GANs architecture. The generator learns the edge response matrix S from the dataset. e To the resistance increment matrix ΔR c The first layer of the generator network is the input layer, and the input data is a [5,1] matrix, which corresponds to the edge response matrix S e There are 5 transposed convolution layers after the input layer, which are used to restore the low-dimensional edge response feature map to the high-dimensional resistance increment feature map. A batch normalization layer (BN) and a Dropout layer are added after each transposed convolution layer to speed up training and avoid overfitting. The output data of the generator network is a [5,4] matrix, which corresponds to the resistance increment matrix ΔR c The data format.

[0080] like Figure 8 As shown, this embodiment obtains the CMC connection structure resistance network resistance increment ΔR that changes with the test time. c . Where ΔR 34The increase in resistance is caused by the damage to the pins, and the rest are caused by the pressure reduction and gap generation between the connecting plates. At the beginning of the test, the CMC connection structure is in the elastic stage, and the structural resistance increment is almost unchanged. As the test load gradually increases, the structural damage gradually expands, and the structural resistance increment gradually increases. Due to the different damage states at different positions in the connection area, the growth trend of the resistance increment is also different. The damage distribution characteristics in the structural geometric space can be intuitively obtained through the changing trend of the resistance increment. Therefore, this embodiment finally realizes the high-temperature damage distribution and evolution measurement of the CMC connection structure based on resistance monitoring.

[0081] The above are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims

1. A method for measuring high temperature damage distribution and evolution of a ceramic matrix composite structure, characterized by: The following steps are involved: Step 1: According to the geometric characteristics of the CMC structure, multiple electrode monitoring points are selected on the CMC structure, and an ohmmeter is connected between two electrode monitoring points forming an electrode pair; Step 2: According to the size of the electrode monitoring points prepared on the CMC structure and the corresponding monitoring area, the electrode monitoring points are regarded as resistance nodes, and the propagation resistance is divided to obtain the undetermined non-material resistance R0 and the introduced contact resistance R c Resistor network model of CMC structure with initial values; Step 3: Based on the resistance network model, a numerical simulation model of the resistance network model is established. Several electrode pairs are selected as monitoring channel nodes. The numerical response of the resistance network model is monitored by using an ohmmeter set at the corresponding monitoring channel node. The non-material resistance R0 and the introduced contact resistance R are determined by combining the initial value of the CMC structure resistance response obtained in the high temperature test with the established numerical simulation model. c Initial value; Step 4: Combine the damage mode and failure law of the CMC structure with the geometric characteristics of the structure and resistor network model to construct a data set of resistance changes and their corresponding responses in the resistor network; Step 5: Select a deep neural network architecture that can meet the needs of response resistor and resistor network data mapping, and train the deep neural network model in combination with the data set; Step 6: During the training of the deep neural network model, the model effect in each training cycle is verified by combining the test results, and the training process is iterated repeatedly to obtain a trained deep neural network model; Step 7. Substitute the resistance response of the CMC structure into the trained deep neural network model to obtain the evolution process of the resistance in the CMC structure with the test time, map the resistance change to the geometric space of the CMC structure, and then intuitively obtain the high-temperature damage distribution and evolution process in the geometric space of the CMC structure.

2. The method for measuring high temperature damage distribution and evolution of a ceramic matrix composite material structure according to claim 1, characterized in that: In step 1, a high temperature resistant electrode is arranged at the electrode monitoring point, and the high temperature resistant electrode is connected to an ohmmeter through a wire. The high temperature resistant electrode is made of silver powder conductive glue, and the wire is made of nickel-titanium alloy wire.

3. The method for measuring high temperature damage distribution and evolution of a ceramic matrix composite material structure according to claim 1, characterized in that: In step 1, the electrode pairs are respectively arranged on both sides of the cracking, delamination or fracture position of the CMC structure.

4. The method for measuring high temperature damage distribution and evolution of a ceramic matrix composite material structure according to claim 3 is characterized by: In step 2, the electrode monitoring points are regarded as resistance nodes, the propagation resistance is divided, and the resistance value of the structural propagation resistance is determined by the following calculation equation, thereby obtaining the resistance network model of the CMC structure; Where R xi is the resistance value of the i-th propagation resistor in the x direction, R yj is the resistance value of the jth propagation resistor in the y direction, R zij is the propagation resistance in the z direction between the i-th electrode pair in the x direction and the j-th electrode pair in the y direction; ρ is the material resistivity of the CMC structure; L xi is the length from the i-th node to the i+1-th node in the x direction, A xi is the propagation resistance area between the i-th node and the i+1-th node in the x direction; L yj is the length from the jth node to the j+1th node in the y direction, A yj is the propagation resistance area between the jth node and the j+1th node in the y direction; l ij is the length of the CMC structure between the electrode pairs, A ij is the area of ​​the high temperature resistant electrode.

5. The method for measuring high temperature damage distribution and evolution of a ceramic matrix composite material structure according to claim 4, characterized in that: In step 3, determine the non-material resistance R0 and the introduced contact resistance R c The specific method of initial value is: Step 1: First assume that the non-material resistance R0 and the contact resistance R c All are 0, and are brought into the numerical simulation model to calculate and obtain the theoretical response of the resistor network; Step 2: If the response value does not reach the initial value of the CMC structure resistance response, the non-material resistance R0 and the contact resistance R c After increasing by 0.01, they were brought into the numerical simulation model again to calculate the response value; Step 3: Compare with the initial value of the CMC structure resistance response. If it is not reached, repeat step 2 until the non-material resistance R0 and the contact resistance R c The numerical value meets the initial value of the resistance response of the CMC structure, and all the initial parameters in the resistance network model are obtained by calculation.

6. The method for measuring high temperature damage distribution and evolution of a ceramic matrix composite material structure according to claim 1, characterized in that: In step 4, the geometric characteristics of the resistor network model are as follows: the resistor network model is divided into two layers, α and β, and the two layers are connected by a CMC connection structure. The CMC connection structure is regarded as a resistor node. After the propagation resistance is divided, the propagation resistance node in the x direction is numbered m, and the propagation resistance node in the y direction is numbered n. The bearing characteristics of the CMC connection structure are the separation of the connection plate and the breakage of the pin. Therefore, the evolution characteristics of the CMC resistor network model are: the contact resistance R in the CMC resistor network c The contact resistance increment matrix of the CMC connection structure is set as ΔR c , the edge response matrix is ​​set to S e , where S e is the edge node resistance response of the resistor network model. When the contact resistance changes, the edge response also changes. The function that converts the contact resistance increment matrix into the edge response matrix is: S e =F(R mn )=F(R0+ΔR c ) Function F(x) is the nonlinear mapping of the resistor network, R mn is the resistance value of the CMC connection structure; If there is an inverse mapping of the resistance network, the contact resistance increment matrix of the CMC connection structure can be obtained through the edge response matrix, that is, ΔR c =F -1 (S e )-R0 First, generate Q groups of uniformly distributed random number resistor increment matrices And bring it into the resistor network model to calculate the corresponding edge response matrix Right now: The above Q group resistance increment matrix and its corresponding edge response matrix are used as data sets to train the deep neural network.

7. The method for measuring high temperature damage distribution and evolution of a ceramic matrix composite material structure according to claim 1, characterized in that: In step 5, the selected deep neural network architecture is the GANs architecture.

Images