Method and device for characterizing creep damage of ferromagnetic superalloys by magnetoacoustic testing

Abstract

The application discloses a ferromagnetic high-temperature alloy creep damage magneto-acoustic detection characterization method and device, and relates to the technical field of material detection, and can be used for effectively detecting and evaluating the creep damage state of a material without damaging the structure of the detected ferromagnetic material.

Description

Technical Field

[0001] This invention relates to the field of creep nondestructive testing technology, and in particular to a method and apparatus for magnetoacoustic detection and characterization of creep damage in ferromagnetic high-temperature alloys. Background Technology

[0002] Ferromagnetic superalloys are the main materials for critical equipment such as main steam pipes in ultra-supercritical power plant boilers. However, their long-term service in complex environments with high pressure and high temperature makes them highly susceptible to creep damage. Typically, the total creep strain of these ferromagnetic superalloys before creep fracture is only 1%-2%, which is difficult to detect and monitor. However, once creep failure or creep fracture occurs, it will cause serious safety accidents and economic losses for power plants and even the entire power system. The early creep damage evolution process usually accounts for 70%-80% of the material's total service life. Microscopically, it manifests as changes in dislocation density and distribution, subgrain and grain coarsening, and the generation of micropores and microcracks. Macroscopically, it manifests as a gradual degradation of mechanical properties. The creep evolution process exhibits time-dependent and multi-scale characteristics. Current methods can only use destructive methods such as metallographic analysis and micrometer indentation testing for evaluation, while traditional non-destructive testing methods are difficult to effectively detect and characterize creep damage at the micrometer scale. Therefore, there is an urgent need to study new non-destructive testing methods applicable to characterizing the creep evolution state of ferromagnetic superalloys. Summary of the Invention

[0003] The present invention aims to at least partially solve one of the technical problems in the related art.

[0004] Therefore, this invention proposes a magnetoacoustic detection and characterization method for creep damage in ferromagnetic high-temperature alloys. This method can effectively detect and evaluate the creep damage state of the material without damaging its structure.

[0005] Another objective of this invention is to provide a magnetoacoustic detection and characterization device for creep damage in ferromagnetic high-temperature alloys.

[0006] To achieve the above objectives, this invention provides a magnetoacoustic detection and characterization method for creep damage in ferromagnetic high-temperature alloys, comprising:

[0007] Construct a closed magnetic circuit based on the metallic material to be tested;

[0008] The magnetoacoustic response signals of metallic materials and closed magnetic circuits are detected using a magnetoacoustic detection device.

[0009] The magnetoacoustic response signal is decoupled by multiple parameters to obtain decoupled magnetoacoustic parameters, and the decoupled magnetoacoustic parameters are feature extracted to obtain magnetoacoustic parameter features.

[0010] A correspondence model between the creep damage state of metallic materials and the magnetoacoustic parametric characteristics is constructed, and the creep damage state is characterized based on the output results of the correspondence model and the creep damage state characterization model.

[0011] The magnetoacoustic detection and characterization method for creep damage in ferromagnetic high-temperature alloys according to embodiments of the present invention may also have the following additional technical features:

[0012] In one embodiment of the present invention, before constructing a closed magnetic circuit based on the metal material to be tested, the method further includes:

[0013] The surface of the metal material to be tested is pretreated.

[0014] In one embodiment of the present invention, a closed magnetic circuit is constructed using a permanent magnet and the metal material to obtain a DC magnetization field; an AC excitation coil is wound around the permanent magnet to obtain an AC magnetization field; wherein the AC magnetization field is formed simultaneously by multiple burst pulses of different frequencies.

[0015] In one embodiment of the present invention, the magnetoacoustic response signal of the metallic material and the closed magnetic circuit is detected using a magnetoacoustic detection device, including:

[0016] Micro-magnetic disturbance signals around metallic materials are detected using a highly sensitive magnetoresistive sensing module in a magnetoacoustic detection device.

[0017] The acoustic wave signal on the surface of the metal material is detected by the weak acoustic wave detection module in the magnetoacoustic detection device;

[0018] The induced electromagnetic signal in the closed magnetic circuit is detected by the induction coil pickup module in the magnetoacoustic detection device.

[0019] In one embodiment of the present invention, multi-parameter decoupling of the magnetoacoustic response signal to obtain decoupled magnetoacoustic parameters includes:

[0020] The induced electromagnetic signal is subjected to spectrum analysis to obtain the electromagnetic signal component set based on the frequency components with amplitudes greater than a preset threshold;

[0021] Time-domain analysis is performed on the micro-magnetic disturbance signal to filter out the static leakage magnetic component under DC magnetization and the AC electromagnetic component under AC magnetization, so as to obtain the first magnetoacoustic parameter.

[0022] The acoustic signal is subjected to frequency domain analysis to filter out frequency components that are the same as those in the electromagnetic signal component set, and then transformed back to the time domain to obtain the second magnetoacoustic parameter.

[0023] In one embodiment of the present invention, feature extraction is performed on the decoupled magnetoacoustic parameters to obtain magnetoacoustic parameter features, including:

[0024] The first magnetoacoustic parameter features are obtained by feature extraction on the first magnetoacoustic parameter; wherein, the first magnetoacoustic parameter features include multiple of the following: maximum amplitude, average envelope value, envelope width, and peak time interval;

[0025] The second magnetoacoustic parameter is subjected to feature extraction to obtain the second magnetoacoustic parameter features, which include multiple features such as energy intensity, root mean square voltage, pulse count, and biphasic spacing.

[0026] In one embodiment of the present invention, constructing a correspondence model between the creep damage state of a metallic material and the magnetoacoustic parametric characteristics includes:

[0027] Experimental values ​​for different degrees of creep damage in metallic materials were obtained.

[0028] Magnetoacoustic parametric characteristics under experimental values ​​of different creep damage degrees were obtained, and correlation analysis was performed between different parametric characteristics and material creep characteristics to obtain correlation analysis results;

[0029] Based on the correlation analysis results, a model was constructed to establish the correspondence between magnetoacoustic parametric characteristics and creep damage state.

[0030] In one embodiment of the present invention, the creep damage state is characterized based on the output results of the correspondence model and the creep damage state characterization model, including:

[0031] Construct a creep damage state characterization model based on a physical heuristic neural network;

[0032] The magnetoacoustic parametric features are input into the creep damage state characterization model to output the true value of the creep damage degree of the metallic material;

[0033] The model statistical output results are calculated using the correspondence model, and the creep damage state is characterized based on the true value of the creep damage degree and the model statistical output results.

[0034] In one embodiment of the present invention, the metallic material includes a ferromagnetic high-temperature alloy; the pretreatment of the surface of the metallic material to be tested includes:

[0035] The surface of the ferromagnetic high-temperature alloy to be tested is subjected to surface pretreatment, wherein the surface pretreatment includes surface cleaning and polishing.

[0036] To achieve the above objectives, another aspect of the present invention provides a magnetoacoustic detection and characterization device for creep damage in ferromagnetic high-temperature alloys, comprising:

[0037] A closed magnetic circuit construction module is used to construct a closed magnetic circuit based on the metal material to be tested.

[0038] The magnetoacoustic response signal detection module is used to detect the magnetoacoustic response signals of metallic materials and closed magnetic circuits using a magnetoacoustic detection device.

[0039] The magnetoacoustic parameter feature acquisition module is used to perform multi-parameter decoupling on the magnetoacoustic response signal to obtain decoupled magnetoacoustic parameters, and to extract features from the decoupled magnetoacoustic parameters to obtain magnetoacoustic parameter features.

[0040] The creep damage state characterization module is used to construct a correspondence model between the creep damage state of metallic materials and the magnetoacoustic parameter characteristics, and to characterize the creep damage state based on the output results of the correspondence model and the creep damage state characterization model.

[0041] The magnetoacoustic detection and characterization method and apparatus for creep damage in ferromagnetic superalloys of this invention can effectively detect and evaluate the creep damage state of materials without damaging the structure of the tested ferromagnetic material. By applying multi-frequency burst pulse excitation to the tested material, characteristic responses in different frequency bands can be excited, thereby enhancing the magnetoacoustic response intensity and improving the effectiveness of detection. Combined with high-sensitivity joint pickup of multiple types of magnetoacoustic response signals, richer response information can be obtained. After multi-parameter decoupling and feature extraction of the acquired magnetoacoustic response, further combining it with a physical heuristic neural network can achieve effective characterization of the creep damage state of ferromagnetic superalloys. It features fast detection speed, convenient operation, high reliability of characterization results, and can be applied to the creep damage detection and characterization of various ferromagnetic material components, showing broad application prospects.

[0042] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0043] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0044] Figure 1 This is a flowchart of a magnetoacoustic detection and characterization method for creep damage in ferromagnetic high-temperature alloys according to an embodiment of the present invention;

[0045] Figure 2 This is a schematic diagram of a magnetoacoustic detection structure for creep damage in ferromagnetic high-temperature alloys according to an embodiment of the present invention.

[0046] Figure 3 This is a schematic diagram of the structure of a magnetoacoustic detection and characterization device for creep damage of ferromagnetic high-temperature alloys according to an embodiment of the present invention. Detailed Implementation

[0047] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0048] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0049] The following describes, with reference to the accompanying drawings, a method and apparatus for magnetoacoustic detection and characterization of creep damage in ferromagnetic high-temperature alloys according to embodiments of the present invention.

[0050] Figure 1 This is a flowchart of the magnetoacoustic detection and characterization method for creep damage in ferromagnetic high-temperature alloys according to an embodiment of the present invention.

[0051] like Figure 1 As shown, the method includes, but is not limited to, the following steps:

[0052] S1, Construct a closed magnetic circuit based on the metal material to be tested.

[0053] It is understood that the present invention first requires surface pretreatment of the test material, such as cleaning and polishing the surface, and applying coupling agent to the contact area with the weak acoustic wave detection module in the magnetoacoustic detection device, so that the magnetoacoustic detection device can fit tightly against the surface of the test material.

[0054] For example, the material to be tested can be a ferromagnetic high-temperature alloy.

[0055] Furthermore, a closed magnetic circuit based on the metallic material to be tested is constructed, and multi-frequency burst pulse excitation is applied to the material under test, such as... Figure 2 As shown, in this embodiment of the invention, a multi-frequency burst pulse excitation is applied to the test material 1. A permanent magnet 2 is used to form a closed magnetic circuit with the test material 1 and to provide a DC magnetization field to the test material. At the same time, an AC excitation coil 3 is wound on the permanent magnet to provide an AC magnetization field to the test material.

[0056] It is understandable that the alternating magnetization field is formed simultaneously by multiple burst pulses of different frequencies, with a frequency range of 20-100kHz.

[0057] S2, the magnetoacoustic response signal of the metallic material and the closed magnetic circuit is detected by a magnetoacoustic detection device.

[0058] First, the magnetoacoustic response signal at the test area is picked up with high sensitivity, including: using the high-sensitivity magnetoresistive sensing module in the magnetoacoustic detection device to collect the micro magnetic disturbance signal around the material under test. In this embodiment, a TMR magnetic array is used.

[0059] The weak acoustic wave detection module in the magnetoacoustic detection device is used to collect the acoustic wave signal on the surface of the material being tested. In this embodiment, a high-sensitivity resonant acoustic emission sensor is used.

[0060] The induction coil pickup module in the magnetoacoustic detection device is used to collect the induced electromagnetic signal in the closed magnetic circuit.

[0061] S3, perform multi-parameter decoupling on the magnetoacoustic response signal to obtain the decoupled magnetoacoustic parameters, and extract features from the decoupled magnetoacoustic parameters to obtain magnetoacoustic parameter features.

[0062] It is understandable that multi-parameter decoupling of the acquired magnetoacoustic response signal is based on decoupling the micro-magnetic disturbance signal around the material under test and the acoustic wave signal on the surface of the material under test from the characteristics of the induced electromagnetic signal in the closed magnetic circuit. Specifically:

[0063] Spectral analysis is performed on the induced electromagnetic signals collected in the closed magnetic circuit. Frequency components with amplitudes greater than a set threshold are selected to obtain the electromagnetic signal component set {B1(f1,A1),B2(f2,A2),…,B n (f n A n )|A i >A thre Let i = 1, 2, ..., n, where B1, f1, and A1 are the electromagnetic signal components and their frequencies and amplitudes, respectively; A thre is the set threshold, which is set to 20% of the maximum amplitude in this embodiment; n is the number of electromagnetic signal components obtained, which is 4 in this embodiment.

[0064] Time-domain analysis is performed on the micro-magnetic disturbance signal around the material under test. The static leakage magnetic component under DC magnetization and the AC electromagnetic component under AC magnetization are filtered out from the micro-magnetic disturbance signal to obtain the first magnetoacoustic parameter.

[0065] The acoustic wave signal collected from the surface of the material under test is analyzed in the frequency domain. After filtering out the frequency component signal that is the same as the electromagnetic signal component in the frequency domain, it is transformed back to the time domain to obtain the second magnetoacoustic parameter.

[0066] Furthermore, feature extraction is performed on each decoupled magnetoacoustic parameter, including: feature extraction of the first magnetoacoustic parameter, the extracted features being: maximum amplitude, average envelope, envelope width, and peak time interval;

[0067] Feature extraction was performed on the second magnetoacoustic parameter, and the extracted features were: energy intensity, root mean square voltage, pulse count, and peak spacing.

[0068] S4. Construct a correspondence model between the creep damage state of the metallic material and the magnetoacoustic parameter characteristics, and characterize the creep damage state based on the output results of the correspondence model and the creep damage state characterization model.

[0069] It is understandable that establishing a model relating the creep damage state of ferromagnetic superalloys to their magnetoacoustic parametric characteristics can be achieved through methods including, but not limited to, the following:

[0070] High-temperature creep interruption tests were conducted on ferromagnetic materials to prepare tests with different degrees of creep damage. In this embodiment, interruption sampling was carried out at 20%, 40%, 60%, 80%, and 100% of the total service life.

[0071] Based on the above steps, the first magnetoacoustic parameter characteristics and the second magnetoacoustic parameter characteristics under different creep damage degrees are obtained;

[0072] Correlation analysis was conducted between different characteristic parameters and material creep characteristics.

[0073] A model was constructed based on multiple regression analysis to establish the correspondence between key features of magnetoacoustic parameters and creep damage state.

[0074] Furthermore, a physical heuristic neural network was used to reliably characterize the creep damage state of ferromagnetic superalloys.

[0075] In one embodiment of the present invention, a creep damage state characterization model for ferromagnetic superalloys based on a physical heuristic neural network is established. The model input consists of a first magnetoacoustic parameter feature and a second magnetoacoustic parameter feature, and the output is the degree of creep damage in the ferromagnetic superalloy. The network loss function is designed as follows:

[0076]

[0077] Among them, O u O u_pred and O u_stat These represent the true value, network output, and statistical output calculated based on the creep damage state correspondence model, respectively, where λ is a complexity-related network hyperparameter. stat The weighting parameters are the statistical features of creep.

[0078] The magnetoacoustic detection and characterization method for creep damage in ferromagnetic superalloys according to embodiments of the present invention can effectively detect and evaluate the creep damage state of materials without damaging the structure of the tested ferromagnetic material. By applying multi-frequency burst pulse excitation to the tested material, characteristic responses in different frequency bands can be excited, thereby enhancing the magnetoacoustic response intensity and improving the effectiveness of detection. Combined with high-sensitivity joint pickup of multiple types of magnetoacoustic response signals, richer response information can be obtained. After multi-parameter decoupling and feature extraction of the acquired magnetoacoustic response, further combining it with a physical heuristic neural network can achieve effective characterization of the creep damage state of ferromagnetic superalloys. The method features fast detection speed, convenient operation, high reliability of characterization results, and can be applied to the creep damage detection and characterization of various ferromagnetic material components, showing broad application prospects.

[0079] To achieve the above embodiments, such as Figure 3 As shown, this embodiment also provides a magnetoacoustic detection and characterization device 10 for creep damage of ferromagnetic high-temperature alloys. The device 10 includes a closed magnetic circuit construction module 100, a magnetoacoustic response signal detection module 200, a magnetoacoustic parameter feature acquisition module 300, and a creep damage state characterization module 400.

[0080] Closed magnetic circuit construction module 100 is used to construct a closed magnetic circuit based on the metal material to be tested;

[0081] The magnetoacoustic response signal detection module 200 is used to detect the magnetoacoustic response signals of metallic materials and closed magnetic circuits using a magnetoacoustic detection device.

[0082] The magnetoacoustic parameter feature acquisition module 300 is used to perform multi-parameter decoupling on the magnetoacoustic response signal to obtain decoupled magnetoacoustic parameters, and to extract features from the decoupled magnetoacoustic parameters to obtain magnetoacoustic parameter features.

[0083] The creep damage state characterization module 400 is used to construct a correspondence model between the creep damage state of the metallic material and the magnetoacoustic parameter characteristics, and to characterize the creep damage state based on the output results of the correspondence model and the creep damage state characterization model.

[0084] The magnetoacoustic detection and characterization device for creep damage of ferromagnetic superalloys according to embodiments of the present invention can effectively detect and evaluate the creep damage state of materials without damaging the structure of the tested ferromagnetic material. By applying multi-frequency burst pulse excitation to the tested material, characteristic responses in different frequency bands can be excited, thereby enhancing the magnetoacoustic response intensity and improving the effectiveness of detection. Combined with high-sensitivity joint pickup of multiple types of magnetoacoustic response signals, richer response information can be obtained. After multi-parameter decoupling and feature extraction of the acquired magnetoacoustic response, further combining it with a physical heuristic neural network can achieve effective characterization of the creep damage state of ferromagnetic superalloys. It features fast detection speed, convenient operation, high reliability of characterization results, and can be applied to the creep damage detection and characterization of various ferromagnetic material components, showing broad application prospects.

[0085] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0086] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A magnetoacoustic detection and characterization method for creep damage in ferromagnetic superalloys, characterized in that, The method includes the following steps: A closed magnetic circuit based on the ferromagnetic high-temperature alloy material to be tested is constructed. The closed magnetic circuit is composed of a permanent magnet and the ferromagnetic high-temperature alloy material to provide a DC bias magnetization field. An AC excitation coil is wound on the permanent magnet, and a burst pulse excitation signal containing multiple different frequency components is applied to the AC excitation coil to form a multi-frequency superimposed AC magnetization field in the closed magnetic circuit. The magnetoacoustic response signals of the ferromagnetic high-temperature alloy material and the closed magnetic circuit are detected by a magnetoacoustic detection device. The magnetoacoustic response signals include: micro-magnetic disturbance signals around the ferromagnetic high-temperature alloy material picked up by a high-sensitivity magnetoresistive sensing module, acoustic wave signals on the surface of the ferromagnetic high-temperature alloy material picked up by a weak acoustic wave detection module, and induced electromagnetic signals in the closed magnetic circuit picked up by an induction coil pickup module. The magnetoacoustic response signal is decoupled using multiple parameters to obtain decoupled magnetoacoustic parameters. The multi-parameter decoupling includes: performing spectral analysis on the induced electromagnetic signal to extract frequency components with amplitudes greater than a preset threshold to form an electromagnetic signal component set; filtering out the static leakage magnetic component under DC bias magnetization and the AC electromagnetic component under AC magnetization from the micro-magnetic disturbance signal based on the electromagnetic signal component set to obtain a first magnetoacoustic parameter; and performing frequency domain analysis on the acoustic signal to filter out component signals with the same frequency as the electromagnetic signal component set in the frequency domain and transforming it back to the time domain to obtain a second magnetoacoustic parameter. The decoupled magnetoacoustic parameters are subjected to feature extraction to obtain magnetoacoustic parameter features. Feature extraction includes: extracting multiple features from the first magnetoacoustic parameter, such as the maximum amplitude, the average envelope value, the envelope width, and the peak time interval, as the first magnetoacoustic parameter features; and extracting multiple features from the second magnetoacoustic parameter, such as the energy intensity, the root mean square voltage, the pulse count, and the bi-peak spacing, as the second magnetoacoustic parameter features. A model is constructed to establish a correspondence between the creep damage state of ferromagnetic superalloy materials and the magnetoacoustic parameter characteristics. The correspondence model is established by obtaining the magnetoacoustic parameter characteristics of ferromagnetic superalloy materials under experimental values ​​of different creep damage degrees, performing correlation analysis on different magnetoacoustic parameter characteristics and material creep characteristics, and constructing the model based on the correlation analysis results and using a multivariate regression analysis method. The creep damage state is characterized based on the output results of the correspondence model and the creep damage state characterization model. The creep damage state characterization model is a model built based on a physical heuristic neural network. The corresponding input is magnetoacoustic parameter features, and the corresponding loss function includes the error term between the network output and the true value of the creep damage degree, as well as the constraint term between the network output and the statistical output of the correspondence model.

2. The method according to claim 1, characterized in that, Before constructing a closed magnetic circuit based on the metal material to be tested, the method further includes: The surface of the metal material to be tested is pretreated.

3. The method according to claim 2, characterized in that, The metallic material includes ferromagnetic high-temperature alloys; The pretreatment of the surface of the metal material to be tested includes: The surface of the ferromagnetic high-temperature alloy to be tested is subjected to surface pretreatment, wherein the surface pretreatment includes surface cleaning and polishing.

4. A magnetoacoustic detection and characterization device for creep damage in ferromagnetic high-temperature alloys, characterized in that, include: A closed magnetic circuit construction module is used to construct a closed magnetic circuit based on the ferromagnetic high-temperature alloy material to be tested. The closed magnetic circuit is composed of a permanent magnet and the ferromagnetic high-temperature alloy material to provide a DC bias magnetization field. An AC excitation coil is wound on the permanent magnet, and a burst pulse excitation signal containing multiple different frequency components is applied to the AC excitation coil to form a multi-frequency superimposed AC magnetization field in the closed magnetic circuit. A magnetoacoustic response signal detection module is used to detect the magnetoacoustic response signal of the ferromagnetic high-temperature alloy material and the closed magnetic circuit using a magnetoacoustic detection device; The magnetoacoustic response signal includes: a micro-magnetic disturbance signal around the ferromagnetic high-temperature alloy material picked up by a highly sensitive magnetoresistive sensing module; an acoustic wave signal on the surface of the ferromagnetic high-temperature alloy material picked up by a weak acoustic wave detection module; and an induced electromagnetic signal in the closed magnetic circuit picked up by an induction coil pickup module; a magnetoacoustic parameter feature acquisition module is used for... The magnetoacoustic response signal is decoupled using multiple parameters to obtain decoupled magnetoacoustic parameters. The multi-parameter decoupling includes: performing spectral analysis on the induced electromagnetic signal to extract frequency components with amplitudes greater than a preset threshold to form an electromagnetic signal component set; filtering out the static leakage magnetic component under DC bias magnetization and the AC electromagnetic component under AC magnetization from the micro-magnetic disturbance signal based on the electromagnetic signal component set to obtain a first magnetoacoustic parameter; and performing frequency domain analysis on the acoustic signal to filter out component signals with the same frequency as the electromagnetic signal component set in the frequency domain and transforming it back to the time domain to obtain a second magnetoacoustic parameter. The decoupled magnetoacoustic parameters are subjected to feature extraction to obtain magnetoacoustic parameter features. Feature extraction includes: extracting multiple features from the first magnetoacoustic parameter, such as the maximum amplitude, the average envelope value, the envelope width, and the peak time interval, as the first magnetoacoustic parameter features; and extracting multiple features from the second magnetoacoustic parameter, such as the energy intensity, the root mean square voltage, the pulse count, and the bi-peak spacing, as the second magnetoacoustic parameter features. The creep damage state characterization module is used to construct a correspondence model between the creep damage state of ferromagnetic superalloy materials and the magnetoacoustic parameter characteristics. The correspondence model is established in the following way: the magnetoacoustic parameter characteristics of ferromagnetic superalloy materials under experimental values ​​of different creep damage degrees are obtained, the correlation between different magnetoacoustic parameter characteristics and material creep characteristics is analyzed, and the model is constructed based on the correlation analysis results and using a multivariate regression analysis method. The creep damage state is characterized based on the output results of the correspondence model and the creep damage state characterization model. The creep damage state characterization model is a model built based on a physical heuristic neural network. The corresponding input is magnetoacoustic parameter features, and the corresponding loss function includes the error term between the network output and the true value of the creep damage degree, as well as the constraint term between the network output and the statistical output of the correspondence model.

Images