A metallographic structure intelligent identification method, device and storage medium for metal materials

Abstract

The present invention discloses a metallographic structure intelligent identification method for metal materials, the steps of which include: collecting metal microstructure images; extracting component features from the metal microstructure images; and judging the constituent phases of the metal materials according to the component features. The present invention provides a metallographic structure intelligent identification method, device and storage medium for metal materials, which can identify the metallographic structure of the material to be tested.

Description

Technical Field

[0001] The present invention relates to the technical field of machine learning and metal material performance testing, and in particular to a method, device and storage medium for intelligently identifying the metallographic structure of a metal material. Background Art

[0002] Metallography refers to the chemical composition of a metal or alloy and the physical and chemical states of various components inside the alloy. The internal structure of a metal material can only be observed under a microscope. The internal structure seen under a microscope is called microstructure or metallographic structure. The metallographic structure reflects the specific morphology of the metal phase, such as martensite, austenite, ferrite, pearlite, etc. The metallographic structure may also have defects. Common defects include metallographic spheroidization and carbide aggregation. Different metallographic structures determine the comprehensive properties of metal materials, and the defects in the metallographic structure will affect the performance. Therefore, it is necessary to use metal materials reasonably on the premise of identifying the metallographic structure of metal materials. At present, people use microscopes to observe the metallographic structure of metal materials and rely on personal knowledge and experience to judge the metallographic morphology and whether there are defects. This method is subjective and highly dependent on the personal ability of the worker, so the current technology still has a lot of room for improvement. Summary of the invention

[0003] In view of the above technical problems, the present invention provides a method, device and storage medium for intelligently identifying the metallographic structure of a metal material, which can identify the metallographic structure of the material to be tested. The technical solution is as follows:

[0004] In a first aspect, an embodiment of the present invention provides a method for intelligently identifying the metallographic structure of a metal material, the steps comprising:

[0005] Collect metal microstructure images;

[0006] Extracting component features from the metal microstructure image;

[0007] The constituent phases of the metal material are determined according to the component characteristics.

[0008] In a first possible implementation manner of the first aspect of the present invention, extracting component features from the metal microstructure image includes:

[0009] Extracting all image features in the metal microstructure image;

[0010] Filtering component features from the image features according to set filtering conditions; the filtering conditions include image resolution and image pixel value;

[0011] Further identifying feature information of the component feature, the feature information comprising:

[0012] Length, angle, width, number, area and color number of grain boundaries;

[0013] The length, angle, width, quantity, area, and color number of the compounds in the crystal;

[0014] The length, angle, width, quantity, area and color number of intercrystalline compounds.

[0015] In a second possible implementation manner of the first aspect of the present invention, judging the constituent phase of the metal material according to the component characteristics further includes:

[0016] Comparing the characteristic information with the characteristic information of a pre-stored sample library, and if the characteristic information meets a preset standard of a certain metallographic structure, determining that the constituent phase of the metal material only includes the metallographic structure;

[0017] Alternatively, if the characteristic information meets the preset standards of multiple metallographic structures, it is determined that the constituent phase of the metal material includes the corresponding multiple metallographic structures.

[0018] Among them, when it is determined that the constituent phases of the metal material include corresponding multiple metallographic structures, for example, N metallographic structures including A, B, ..., N, whichever metallographic structure A, B, ..., N occupies a larger proportion in the image will be written in front, such as A accounts for 40% and B accounts for 60%, then the metal microstructure image is B+A structure.

[0019] In a third possible implementation manner of the first aspect of the present invention, the metallographic structure intelligent identification method of the metal material further includes aging rating; the aging rating is specifically as follows:

[0020] When extracting component features from the metal microstructure image, the curvature of the grain boundary line, the number of grain boundaries, the clarity of the grain boundary contour and the distribution position of the compound in the crystal are detected;

[0021] Determining the aging condition of the metal material according to the curvature of the grain boundary straight line, the number of the detected grain boundaries, the clarity of the detected grain boundary contours, and the distribution position of the intracrystalline compound;

[0022] Wherein, when the constituent phases of the metal material include multiple metallographic structures, the aging condition is rated according to the following scoring formula:

[0023]

[0024] L is the aging level; T is the characteristic score for judging the aging condition of a certain metallographic structure; n is the number of types of metallographic structures.

[0025] In a second aspect, an embodiment of the present invention provides a device for intelligently identifying the metallographic structure of a metal material, comprising:

[0026] An image acquisition module, used for acquiring metal microstructure images;

[0027] An image recognition module, used for extracting component features from the metal microstructure image;

[0028] An image analysis module is used to determine the constituent phases of the metal material according to the component characteristics.

[0029] In a first possible implementation manner of the second aspect of the present invention, the image recognition module is further configured to:

[0030] Extracting all image features in the metal microstructure image;

[0031] Filtering component features from the image features according to set filtering conditions; the filtering conditions include image resolution and image pixel value;

[0032] Further identifying feature information of the component feature, the feature information comprising:

[0033] Length, angle, width, number, area and color number of grain boundaries;

[0034] The length, angle, width, quantity, area, and color number of the compounds in the crystal;

[0035] The length, angle, width, quantity, area and color number of intercrystalline compounds.

[0036] In a second possible implementation manner of the second aspect of the present invention, the image analysis module is further used to:

[0037] Comparing the characteristic information with the characteristic information of a pre-stored sample library, and if the characteristic information meets a preset standard of a certain metallographic structure, determining that the constituent phase of the metal material only includes the metallographic structure;

[0038] Alternatively, if the characteristic information meets the preset standards of multiple metallographic structures, it is determined that the constituent phase of the metal material includes the corresponding multiple metallographic structures.

[0039] In a third possible implementation of the second aspect of the present invention, the metallographic structure intelligent identification device of the metal material further includes a rating module; the rating module is used to:

[0040] When extracting component features from the metal microstructure image, the curvature of the grain boundary line, the number of grain boundaries, the clarity of the grain boundary contour and the distribution position of the compound in the crystal are detected;

[0041] Determining the aging condition of the metal material according to the curvature of the grain boundary straight line, the number of the detected grain boundaries, the clarity of the detected grain boundary contours, and the distribution position of the intracrystalline compound;

[0042] Wherein, when the constituent phases of the metal material include multiple metallographic structures, the aging condition is rated according to the following scoring formula:

[0043]

[0044] L is the aging level; T is the characteristic score for judging the aging condition of a certain metallographic structure; n is the number of types of metallographic structures.

[0045] In a third aspect, an embodiment of the present invention provides a device for intelligently identifying the metallographic structure of a metal material, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the method for intelligently identifying the metallographic structure of the metal material as described above when executing the computer program.

[0046] In a fourth aspect, an embodiment of the present invention provides a storage medium for a method for intelligent identification of the metallographic structure of a metal material, wherein the storage medium for the method for intelligent identification of the metallographic structure of a metal material is used to store one or more computer programs, wherein the one or more computer programs include program codes, and when the computer programs are run on a computer, the program codes are used to execute the above-mentioned method for intelligent identification of the metallographic structure of a metal material.

[0047] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0048] The present invention provides a method, device and storage medium for intelligent identification of metallographic structure of metal materials, which automatically collect aging pictures of metal materials in power plants and extract component features from the metal microstructure images to judge the constituent phases of the metal materials according to the component features, thereby realizing automatic image recognition, avoiding manual identification based on personal experience, reducing manual misjudgment, improving accuracy, and greatly reducing the workload of staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a flowchart of a method for intelligently identifying the metallographic structure of a metal material in an embodiment of the present invention;

[0050] Figure 2 is a first acquisition diagram of a metallographic structure intelligent identification method of a metal material in an embodiment of the present invention;

[0051] Figure 3 is a second acquisition diagram of a metallographic structure intelligent identification method of a metal material in an embodiment of the present invention;

[0052] Figure 4 is a third acquisition diagram of a metallographic structure intelligent identification method of a metal material in an embodiment of the present invention;

[0053] Figure 5 A schematic diagram of austenite identified by a metallographic structure intelligent identification method of a metal material in an embodiment of the present invention;

[0054] Figure 6 A schematic diagram of martensite identified by a metallographic structure intelligent identification method of a metal material in an embodiment of the present invention;

[0055] Figure 7 This is a schematic diagram of granular bainite identified by a metallographic structure intelligent identification method of a metal material in an embodiment of the present invention;

[0056] Figure 8 It is a schematic diagram of grain boundaries, grains, and intracrystalline compounds identified by a metallographic structure intelligent identification method of a metal material in an embodiment of the present invention;

[0057] Fig. 9 is a schematic diagram of the aging process of a metal material described in a method for intelligent identification of metallographic structure of a metal material in an embodiment of the present invention;

[0058] Fig.10 This is a schematic diagram of a metallographic structure intelligent identification device for a metal material according to an embodiment of the present invention;

[0059] Among them, 1. grain boundary; 2. grain; 3. intracrystalline compound. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] See also Figure 1 The present invention provides an exemplary embodiment, a method for intelligently identifying the metallographic structure of a metal material, the steps comprising:

[0062] S101, collecting metal microstructure images; in this embodiment, pictures of power station metal materials aged at level one, level two, level three to level five are used;

[0063] It can be understood that the metal microstructure image is an image of the metal structure obtained by grinding, polishing and corroding the surface of the metal material and then observing it under a metallographic microscope.

[0064] See also Figure 2 , 3 4. Different alloy contents and different heat treatment processes will present different metallographic organization diagrams. According to the morphological characteristics of each component (component definition is the element or stable compound that makes up the alloy) in the diagram, it is possible to identify which phases the metal material is composed of (such as ferrite phase + pearlite phase).

[0065] S102, extracting component features from the metal microstructure image;

[0066] It is to be understood that the component characteristics, component, refer to the elements or stable compounds that make up the metal.

[0067] S103, determining the constituent phases of the metal material according to the component characteristics.

[0068] It is understandable that the microstructural composition of the metal material can be intelligently identified based on the morphology, quantity, orientation, size, distribution, etc. of the component characteristics.

[0069] The component features include:

[0070] Ferrite outline, pearlite outline, intergranular carbide outline (round, strip, lamellar, etc.);

[0071] Intergranular carbide content, distribution position of carbides (at grain boundaries, between grains);

[0072] The number and outline of island bainite, the number and outline of troostite, the number and outline of fine plate martensite, the number and outline of coarse plate martensite, and the number and outline of austenite.

[0073] The step of extracting component features from the metal microstructure image comprises:

[0074] Extracting all image features in the metal microstructure image;

[0075] Filtering component features from the image features according to set filtering conditions; the filtering conditions include image resolution and image pixel value;

[0076] Further identifying feature information of the component feature, the feature information comprising:

[0077] Length, angle, width, number, area and color number of grain boundaries;

[0078] The length, angle, width, quantity, area, and color number of the compounds in the crystal;

[0079] The length, angle, width, quantity, area and color number of intercrystalline compounds.

[0080] It is understandable that by filtering according to the filtering conditions, we can rule out interferences such as dirt and scratches in the pictures that may be caused by improper collection, and now I only need information related to the metallographic structure.

[0081] It can be understood that the carbide contour and image recognition are one of the shapes including round, strip, and lamellar; the distribution position of the carbide is at the grain boundary or between grains.

[0082] Comparing the characteristic information with the characteristic information of a pre-stored sample library, and if the characteristic information meets a preset standard of a certain metallographic structure, determining that the constituent phase of the metal material only includes the metallographic structure;

[0083] Alternatively, if the characteristic information meets the preset standards of multiple metallographic structures, it is determined that the constituent phase of the metal material includes the corresponding multiple metallographic structures.

[0084] It can be understood that, through machine learning, a large number of images of calibrated metallographic structures are learned using the above-mentioned method of extracting component features, and a database is obtained as a comparison standard.

[0085] The determining the constituent phase of the metal material according to the component characteristics comprises:

[0086] If the characteristic information meets the preset standard of ferrite, the metallographic phase is further determined to be ferrite; it can be understood that the interstitial solid solution formed by carbon dissolving in the interstitial space of the α-Fe lattice is called ferrite, which has a bcc structure and is distributed in equiaxed polygonal grains.

[0087] If the characteristic information meets the preset standard of austenite, the metallographic phase is further determined to be austenite. It can be understood that the interstitial solid solution formed by carbon dissolving in the interstitial space of the γ-Fe lattice is called austenite, which has a face-centered cubic structure.

[0088] See also Figure 5 , austenite, the grain boundaries are relatively straight and present irregular polygons.

[0089] Further, on the basis of identifying the ferrite, the metal material image is magnified; if a mixture inside the ferrite or between adjacent ferrites is detected, the constituent phase of the metal material is determined to be bainite;

[0090] See also Figure 6 , bainite is also divided into different types: large blocks or strips of ferrite with many small islands distributed in the complex phase structure and are determined to be granular bainite;

[0091] Upper bainite is distributed in bundles. It is a mixture of parallel ferrite and intermittent strip-shaped cementite sandwiched in between. The typical upper bainite is feather-shaped, with the grain boundary as the symmetry axis. Due to different orientations, the feathers can be symmetrical or asymmetrical. The ferrite feathers can be needle-shaped, dot-shaped, or block-shaped. Depending on the carbon content of the alloy, the feathers change from fuzzy to clear, and from thin needles to thick needles.

[0092] Lower bainite, the spatial morphology is biconvex lens-shaped, black needle-shaped or bamboo-leaf-shaped, with a certain angle between the needles and the same color;

[0093] Furthermore, when the constituent phase of the metal material is identified as ferrite, the method further comprises the steps of:

[0094] The carbide content is analyzed; when the carbide content is less than 6.69%, the metallographic phase of the metal material is determined to be pearlite.

[0095] Furthermore, when the constituent phase of the metal material is identified as pearlite, the method further comprises the steps of:

[0096] The metal material image is magnified; if a lamellar structure is identified, it is determined to be troostite and the number of troostites is counted.

[0097] In particular, the constituent phase of the metal material is determined according to the component characteristics, and when the metallographic phase cannot be determined to be ferrite or austenite, the metallographic phase is determined to be martensite.

[0098] It is understood that the supersaturated solid solution of carbon in α-Fe is called martensite, that is, the solid solution of carbon in γ-Fe (austenite) is transformed into the solid solution of carbon in α-Fe (ferrite).

[0099] See also Figure 7 , martensite, among which lath martensite is composed of many groups of laths arranged parallel to each other, each lath is a single crystal, they are generally alternated by small-angle grain boundaries, showing a high density of dislocations, these dislocations are unevenly distributed and entangled with each other to form a cellular substructure, lamellar martensite (or needle-shaped martensite) is needle-shaped or bamboo-leaf-shaped, the martensite sheets are not parallel, but have a certain intersection angle, and some needles are arranged at an angle of 120°.

[0100] After the martensite is identified, the angle between adjacent martensite is measured by an angle measurement module.

[0101] Furthermore, when the constituent phase of the metal material is identified as martensite, the martensite metallographic structure can be divided into fine plate-like martensite and coarse plate-like martensite by identifying the structure shape.

[0102] In particular, the constituent phases of the metal material are determined according to the component characteristics, and the structure of ferrite in the form of lamellae, feathers or triangles appearing near the austenite grain boundary is identified as the Widmanstatten structure.

[0103] The present invention also provides a preferred embodiment, after extracting the component features from the metal microstructure image, further comprising:

[0104] Whether the metallographic structure of the metal material has defects is determined based on the component characteristics.

[0105] The step of judging whether the metallographic structure of the metal material has defects according to the component characteristics is specifically as follows:

[0106] Detect intergranular carbides, and when the intergranular carbide aggregation exceeds a preset mark, it is determined that the metallographic structure has defects; or, when object structures other than metals and carbides are detected, it is determined that the metallographic structure has defects.

[0107] The present invention also provides a preferred embodiment, wherein the metallographic structure intelligent identification method of the metal material further comprises:

[0108] When the metallographic phase of the metal material is identified as pearlite, spheroidization rating is performed.

[0109] Specifically, identifying spherical carbides in the metal material image;

[0110] If a small portion of the carbides are in the form of clusters, the spheroidization rate is determined to be not less than 95%, and the spheroidization rating is graded as level one;

[0111] If the proportion of lumpy carbides exceeds 30%, the spheroidization rate is judged to be greater than 90% and less than 95%, and the spheroidization rating is level 2;

[0112] If most of the carbides are in the form of clusters and spheres, the spheroidization rate is determined to be greater than 80% and less than 90%, and the spheroidization rating is grade three;

[0113] If most of the carbides are in the form of lumps and flocs, the spheroidization rate is determined to be greater than 70% and less than 80%, and the spheroidization rating is level four;

[0114] If the carbides are dispersed in the form of worms, balls, lumps and floccules, the spheroidization rate is determined to be greater than 60% and less than 70%, and the spheroidization rating is level five;

[0115] If the carbides are distributed in aggregated worm-like, flake-like, spherical, lumpy, or flocculent shapes, the spheroidization rate is determined to be less than 60%, and the spheroidization rating is grade six.

[0116] The metallographic structure intelligent identification method of the metal material also includes aging rating;

[0117] Specifically, when extracting component features from the metal microstructure image, the curvature of the grain boundary straight line, the number of grain boundaries, the clarity of the grain boundary contour and the distribution position of the compound in the crystal are detected;

[0118] The aging condition of the metal material is comprehensively judged based on the test results of the above tests.

[0119] Preferably, the aging condition is expressed by rating. When the constituent phases of the metal material include multiple metallographic structures, the rating can be calculated by the following score formula:

[0120]

[0121] L---aging level; T is the characteristic score (1.0~5.0) for judging the aging condition of a certain metallographic structure; n is the number of types of metallographic structures.

[0122] See also Figure 8 It can be understood that the grain boundary refers to the interface between grains;

[0123] The intracrystalline compound refers to an intracrystalline solid solution, a metal compound, and a mechanical mixture.

[0124] See also Fig. 9 The present invention also provides a specific embodiment to illustrate that, from the perspective of morphology, the grain boundary feature develops from an ideal straight line state to a curved deformation, from a line to a point, from a single crystal boundary to a twin crystal boundary, and from a large number to a small number, from a clear visible to a blurred one; from the perspective of morphology, the grain feature develops from a clear outline, uniform distribution, and uniform size to a blurred boundary, a segregated distribution, and a different size; for intracrystalline compounds, this characteristic morphology develops from aggregation to boundary dispersion, from strips to points, from black entities to white virtual bodies, the number of strip carbides changes from large to small, and the number of black round points changes from a small number to a large number, and from a small number to a large number. The above extracted features are scored from 1.0 to 5.0 from no aging (level 1) to severe aging (level 5), which constitute the subsequent rating score.

[0125]

[0126] The present invention intends to judge the aging level of the material by dividing the scores of the metal microstructure diagram, thus avoiding the errors caused by subjective judgment and reducing human errors.

[0127] The present invention provides an exemplary embodiment, a device for intelligently identifying the metallographic structure of a metal material, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the method for intelligently identifying the metallographic structure of the metal material as described above when executing the computer program.

[0128] See also Fig.10 The present invention provides an exemplary embodiment, a metallographic structure intelligent identification device for metal materials, comprising:

[0129] An image acquisition module 201 is used to acquire metal microstructure images;

[0130] It can be understood that the metal microstructure image is an image of the metal structure obtained by grinding, polishing and corroding the surface of the metal material and then observing it under a metallographic microscope.

[0131] See also Figure 2 , 3 4. Different alloy contents and different heat treatment processes will present different metallographic organization diagrams. According to the morphological characteristics of each component (component definition is the element or stable compound that makes up the alloy) in the diagram, it is possible to identify which phases the metal material is composed of (such as ferrite phase + pearlite phase).

[0132] An image recognition module 202, used to extract component features from the metal microstructure image;

[0133] The image analysis module 203 is used to determine the constituent phases of the metal material according to the component characteristics.

[0134] It is understandable that the microstructural composition of the metal material can be intelligently identified based on the morphology, quantity, orientation, size, distribution, etc. of the component characteristics.

[0135] The component features include:

[0136] Ferrite outline, pearlite outline, intergranular carbide outline (round, strip, lamellar, etc.);

[0137] Intergranular carbide content, distribution position of carbides (at grain boundaries, between grains);

[0138] The number and outline of island bainite, the number and outline of troostite, the number and outline of fine plate martensite, the number and outline of coarse plate martensite, and the number and outline of austenite.

[0139] The image recognition module is further used for:

[0140] Extracting all image features in the metal microstructure image;

[0141] Filtering component features from the image features according to set filtering conditions; the filtering conditions include image resolution and image pixel value;

[0142] Further identifying feature information of the component feature, the feature information comprising:

[0143] Length, angle, width, number, area and color number of grain boundaries;

[0144] The length, angle, width, quantity, area, and color number of the compounds in the crystal;

[0145] The length, angle, width, quantity, area and color number of intercrystalline compounds.

[0146] The image analysis module is also used for:

[0147] If the characteristic information meets the preset standard of ferrite, the metallographic phase is further determined to be ferrite; it can be understood that the interstitial solid solution formed by carbon dissolving in the interstitial space of the α-Fe lattice is called ferrite, which has a bcc structure and is distributed in equiaxed polygonal grains.

[0148] If the characteristic information meets the preset standard of austenite, the metallographic phase is further determined to be austenite. It can be understood that the interstitial solid solution formed by carbon dissolving in the interstitial space of the γ-Fe lattice is called austenite, which has a face-centered cubic structure.

[0149] See also Figure 5 , austenite, the grain boundaries are relatively straight and present irregular polygons.

[0150] The image analysis module is also used for:

[0151] Further, on the basis of identifying the ferrite, the metal material image is magnified; if a mixture inside the ferrite or between adjacent ferrites is detected, the constituent phase of the metal material is determined to be bainite;

[0152] See also Figure 6 , bainite is also divided into different types: large blocks or strips of ferrite with many small islands distributed in the complex phase structure and are determined to be granular bainite;

[0153] Upper bainite is distributed in bundles. It is a mixture of parallel ferrite and intermittent strip-shaped cementite sandwiched in between. The typical upper bainite is feather-shaped, with the grain boundary as the symmetry axis. Due to different orientations, the feathers can be symmetrical or asymmetrical. The ferrite feathers can be needle-shaped, dot-shaped, or block-shaped. Depending on the carbon content of the alloy, the feathers change from fuzzy to clear, and from thin needles to thick needles.

[0154] Lower bainite, the spatial morphology is biconvex lens-shaped, black needle-shaped or bamboo-leaf-shaped, with a certain angle between the needles and the same color;

[0155] Furthermore, when the constituent phase of the metal material is identified as ferrite, the method further comprises the steps of:

[0156] The carbide content is analyzed; when the carbide content is less than 6.69%, the metallographic phase of the metal material is determined to be pearlite.

[0157] Furthermore, when the constituent phase of the metal material is identified as pearlite, the method further comprises the steps of:

[0158] The metal material image is magnified; if a lamellar structure is identified, it is determined to be troostite and the number of troostites is counted.

[0159] In particular, the constituent phase of the metal material is determined according to the component characteristics, and when the metallographic phase cannot be determined to be ferrite or austenite, the metallographic phase is determined to be martensite.

[0160] It is understood that the supersaturated solid solution of carbon in α-Fe is called martensite, that is, the solid solution of carbon in γ-Fe (austenite) is transformed into the solid solution of carbon in α-Fe (ferrite).

[0161] See also Figure 7 , martensite, among which lath martensite is composed of many groups of laths arranged parallel to each other, each lath is a single crystal, they are generally alternated by small-angle grain boundaries, showing a high density of dislocations, these dislocations are unevenly distributed and entangled with each other to form a cellular substructure, lamellar martensite (or needle-shaped martensite) is needle-shaped or bamboo-leaf-shaped, the martensite sheets are not parallel, but have a certain intersection angle, and some needles are arranged at an angle of 120°.

[0162] After the martensite is identified, the angle between adjacent martensite is measured by an angle measurement module.

[0163] Furthermore, when the constituent phase of the metal material is identified as martensite, the martensite metallographic structure can be divided into fine plate-like martensite and coarse plate-like martensite by identifying the structure shape.

[0164] In particular, the constituent phases of the metal material are determined according to the component characteristics, and the structure of ferrite in the form of lamellae, feathers or triangles appearing near the austenite grain boundary is identified as the Widmanstatten structure.

[0165] The present invention also provides a preferred embodiment, wherein the metallographic structure intelligent identification device of the metal material further comprises a screening module;

[0166] The screening module is used to determine whether the metallographic structure of the metal material has defects according to the component characteristics.

[0167] The step of judging whether the metallographic structure of the metal material has defects according to the component characteristics is specifically as follows:

[0168] Detect intergranular carbides, and when the intergranular carbide aggregation exceeds a preset mark, it is determined that the metallographic structure has defects; or, when object structures other than metals and carbides are detected, it is determined that the metallographic structure has defects.

[0169] The present invention also provides a preferred embodiment, wherein the metallographic structure intelligent identification device of the metal material further comprises:

[0170] The rating module is used for performing spheroidization rating when the metallographic phase of the metal material is identified as pearlite.

[0171] Specifically, identifying spherical carbides in the metal material image;

[0172] If a small portion of the carbides are in the form of clusters, the spheroidization rate is determined to be not less than 95%, and the spheroidization rating is graded as level one;

[0173] If the proportion of lumpy carbides exceeds 30%, the spheroidization rate is judged to be greater than 90% and less than 95%, and the spheroidization rating is level 2;

[0174] If most of the carbides are in the form of clusters and spheres, the spheroidization rate is determined to be greater than 80% and less than 90%, and the spheroidization rating is grade three;

[0175] If most of the carbides are in the form of lumps and flocs, the spheroidization rate is determined to be greater than 70% and less than 80%, and the spheroidization rating is level four;

[0176] If the carbides are dispersed in the form of worms, balls, lumps and floccules, the spheroidization rate is determined to be greater than 60% and less than 70%, and the spheroidization rating is level five;

[0177] If the carbides are distributed in aggregated worm-like, flake-like, spherical, lumpy, or flocculent shapes, the spheroidization rate is determined to be less than 60%, and the spheroidization rating is grade six.

[0178] The rating module is specifically used for:

[0179] When extracting component features from the metal microstructure image, the curvature of the grain boundary line, the number of grain boundaries, the clarity of the grain boundary contour and the distribution position of the compound in the crystal are detected;

[0180] The aging condition of the metal material is comprehensively judged based on the test results of the above tests.

[0181] Preferably, the aging condition is expressed by rating. When the constituent phases of the metal material include multiple metallographic structures, the rating can be calculated by the following score formula:

[0182]

[0183] L---aging level; T is the characteristic score (1.0~5.0) for judging the aging condition of a certain metallographic structure; n is the number of types of metallographic structures.

[0184] See also Figure 8 It can be understood that the grain boundary refers to the interface between grains;

[0185] The intracrystalline compound refers to an intracrystalline solid solution, a metal compound, and a mechanical mixture.

[0186] The present invention provides an exemplary embodiment, a storage medium for a method for intelligently identifying the metallographic structure of a metal material, wherein the storage medium for the method for intelligently identifying the metallographic structure of a metal material is used to store one or more computer programs, wherein the one or more computer programs include program codes, and when the computer programs are run on a computer, the program codes are used to execute the above-mentioned method for intelligently identifying the metallographic structure of a metal material.

[0187] The computer-readable medium of the embodiment of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. More specific examples of computer-readable storage media include at least (non-exhaustive list) the following: an electrical connection portion (electronic device) with one or more wirings, a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and editable read-only memory (EPROM or flash memory), an optical fiber device, and a portable read-only memory (CDROM). In addition, the computer-readable storage medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting or processing in other suitable ways as necessary, and then stored in a computer memory.

[0188] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0189] A person skilled in the art may understand that all or part of the steps in the above-mentioned embodiment method may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0190] In addition, each functional unit in each embodiment of the present application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into one module. The above integrated module can be implemented in the form of hardware or in the form of software functional modules.

[0191] The present invention provides a method, device and storage medium for intelligent identification of metallographic structure of metal materials, which automatically collect aging pictures of metal materials in power plants and extract component features from the metal microstructure images to judge the constituent phases of the metal materials according to the component features, thereby realizing automatic image recognition, avoiding manual identification based on personal experience, reducing manual misjudgment, improving accuracy, and greatly reducing the workload of staff.

[0192] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for intelligently identifying the metallographic structure of a metal material, characterized in that the steps include: Collect metal microstructure images; Extracting component features from the metal microstructure image; The constituent phase of the metal material is determined according to the component characteristics, and an aging rating is performed according to the component characteristics; wherein the aging rating is specifically as follows: When extracting component features from the metal microstructure image, the curvature of the grain boundary line, the number of grain boundaries, the clarity of the grain boundary contour and the distribution position of the compound in the crystal are detected; Wherein, when extracting component features from the metal microstructure image, the feature information is compared with the feature information of the pre-stored sample library. If the feature information meets the preset standard of a certain metallographic structure, it is determined that the constituent phase of the metal material only includes the metallographic structure; or, if the feature information meets the preset standards of multiple metallographic structures, it is determined that the constituent phase of the metal material includes the corresponding multiple metallographic structures; the pre-stored sample library is obtained by learning the image of the calibrated metallographic structure; extracting component features from the metal microstructure image is specifically as follows: Extracting all image features in the metal microstructure image; Filtering component features from the image features according to set filtering conditions; the filtering conditions include image resolution and image pixel value; Further identifying feature information of the component feature, the feature information includes: Length, angle, width, number, area and color number of grain boundaries; The length, angle, width, quantity, area, and color number of the compounds in the crystal; The length, angle, width, quantity, area and color number of intergranular compounds; Determining the aging condition of the metal material according to the curvature of the grain boundary straight line, the number of the detected grain boundaries, the clarity of the detected grain boundary contours, and the distribution position of the intracrystalline compound; Wherein, when the constituent phases of the metal material include multiple metallographic structures, the aging condition is rated according to the following scoring formula: L is the aging level; T is the characteristic score for judging the aging condition of a certain metallographic structure; n is the number of types of metallographic structures.

2. An intelligent identification device for metallographic structure of metal materials, characterized in that: include: An image acquisition module, used for acquiring metal microstructure images; An image recognition module, used for extracting component features from the metal microstructure image; An image analysis module is used to determine the constituent phase of the metal material according to the component characteristics, and to perform aging rating according to the component characteristics; wherein the aging rating is specifically as follows: When extracting component features from the metal microstructure image, the curvature of the grain boundary line, the number of grain boundaries, the clarity of the grain boundary contour and the distribution position of the compound in the crystal are detected; Wherein, when extracting component features from the metal microstructure image, the feature information is compared with the feature information of the pre-stored sample library. If the feature information meets the preset standard of a certain metallographic structure, it is determined that the constituent phase of the metal material only includes the metallographic structure; or, if the feature information meets the preset standards of multiple metallographic structures, it is determined that the constituent phase of the metal material includes the corresponding multiple metallographic structures; the pre-stored sample library is obtained by learning the image of the calibrated metallographic structure; extracting component features from the metal microstructure image is specifically as follows: Extracting all image features in the metal microstructure image; Filtering component features from the image features according to set filtering conditions; the filtering conditions include image resolution and image pixel value; Further identifying feature information of the component feature, the feature information includes: Length, angle, width, number, area and color number of grain boundaries; The length, angle, width, quantity, area, and color number of the compounds in the crystal; The length, angle, width, quantity, area and color number of intergranular compounds; Determining the aging condition of the metal material according to the curvature of the grain boundary straight line, the number of the detected grain boundaries, the clarity of the detected grain boundary contours, and the distribution position of the intracrystalline compound; Wherein, when the constituent phases of the metal material include multiple metallographic structures, the aging condition is rated according to the following scoring formula: L is the aging level; T is the characteristic score for judging the aging condition of a certain metallographic structure; n is the number of types of metallographic structures.

3. An intelligent identification device for metallographic structure of metal materials, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the metallographic structure intelligent identification method of the metal material as claimed in claim 1 when executing the computer program.

4. A storage medium for a method for intelligently identifying the metallographic structure of a metal material, characterized in that: The storage medium of the metal material metallographic structure intelligent identification method is used to store one or more computer programs, and the one or more computer programs include program codes. When the computer program is run on a computer, the program code is used to execute the metallographic structure intelligent identification method of the metal material described in claim 1.

Images