JP2021043828A - Analysis method of metal solidified crystal structure, and analyzing device of metal solidified crystal structure - Google Patents

Abstract

To provide a method of quantitatively analyzing by identifying columnar crystal area and equiaxial crystal area in a metal solidified crystal structure, and an analyzing device of metal solidified crystal structure for carrying out the method.SOLUTION: An analysis method of metal solidified crystal structure according to the present invention includes: a crystal structure image acquisition step; a crystal grain identification step; a crystal grain shape anisotropic amount calculation step; an analysis unit area setting step; a crystal grain partial area calculation step; an average shape anisotropy calculation step; an average shape anisotropic amount distribution creation step; and a columnar crystal area/equiaxial crystal area determination step.SELECTED DRAWING: Figure 1

Description

The present invention relates to a technique for analyzing the crystal structure of a solidified metal body, and in particular, a method for quantitatively analyzing a columnar crystal region and an equiaxed crystal region due to a columnar crystal-equal axis crystal transition of a solid metal solid body, and the method It relates to an analyzer for analyzing a solidified metal crystal structure for carrying out the above.

The metal material used in various mechanical devices needs to satisfy various properties (for example, mechanical properties, thermal properties, corrosion resistance) required in the operating environment of the mechanical device. In general, corrosion resistance and thermal properties are strongly related to the composition of the metal material, and mechanical properties are strongly related to the microstructure of the metal material. Moreover, the fine structure of the metal material is extremely closely related to the manufacturing method.

That is, control of the microstructure is indispensable for obtaining a metal material having desirable mechanical properties, and control of the manufacturing method is performed after clarifying the relationship with the manufacturing method in order to obtain the desired microstructure. There is a need to.

Many methods for evaluating the microstructure of metallic materials have been proposed so far, but many of them are simple numerical values of one evaluation item (for example, average crystal grain size, occupancy of a predetermined phase, etc.). The conventional simple numerical one-item index is effective when the entire observation area of the microstructure is almost even. However, when there is a large distribution or bias in the observation area of the microstructure, it is difficult to analyze the state of the distribution or bias using a simple numerical one-item index.

On the other hand, in recent years, the technique of image analysis has been dramatically improved, and it has become possible to analyze and evaluate the entire area of microstructure observation in two dimensions. For example, in Patent Document 1 (Japanese Patent Laid-Open No. 2004-192055), a composite image of a virtual environment covering a wide field of view is converted into pixels based on a conversion table set by internal parameters of the projection system using a perspective projection image as an original image. This is a wide-field division synthesis method consisting of a procedure for synthesizing by performing the above, and when an online change instruction is given to at least one internal parameter for setting the conversion table, a simple conversion table is created by the second pixel conversion procedure. A wide-field split-composite method, which comprises a procedure for synthesizing a wide-field image, is disclosed.

Japanese Unexamined Patent Publication No. 2004-192055

Metallic materials are often used as a base for solidified bodies formed by melt solidification. As described above, in order to achieve the desired mechanical properties of the solidified metal, it is necessary to control the crystal structure of the solidified metal, and in order to control the crystal structure of the solidified metal, the crystal structure and production thereof. It is necessary to control the manufacturing method after clarifying the relationship with the method. That is, it is necessary to clarify the crystal structure of the solidified metal before controlling the production method.

Here, in a solidified metal, it is known that a columnar-equal axis transition often occurs in the crystal structure during the solidification process. Since columnar crystals have anisotropy in various properties compared to equiaxed crystals, the columnar crystal region and equiaxed crystal region are quantitatively determined in order to grasp the properties of the entire metal solidified body. Need to be analyzed.

Considering the quantitative analysis of the columnar crystal region / equiaxed crystal region in the crystal structure of the solidified metal, it is difficult to grasp the whole by the conventional simple numerical one-item index. Further, according to the study by the present inventors, it is considered that even with the image analysis technique of Patent Document 1, it is difficult to quantitatively analyze the columnar crystal region / equiaxed crystal region in the crystal structure of the solidified metal.

Therefore, an object of the present invention is to provide a method for quantitatively analyzing a columnar crystal region and an equiaxed crystal region in the crystal structure of a metal solidified body, and an apparatus for analyzing the metal solidified body crystal structure for carrying out the method. To do.

(I) One aspect of the present invention is a method for analyzing the crystal structure of a solidified metal body.
A crystal structure image acquisition step for acquiring an observation image of the crystal structure, and
A crystal grain identification step for identifying crystal grains in the observation image, and
A step of calculating the amount of crystal grain shape anisotropy for calculating the amount of crystal grain shape anisotropy, and
An analysis unit area setting step for setting an analysis unit area in the observation image, and
A crystal grain partial area calculation step for calculating the partial area of the crystal grain in the analysis unit region, and
An average shape anisotropic amount calculation step for calculating the average shape anisotropic amount of the analysis unit region from the shape anisotropic amount and the partial area, and
The step of creating the average shape anisotropic amount distribution and the step of creating the average shape anisotropic amount distribution,
A columnar crystal region / equiaxed crystal region determination step for determining a columnar crystal region / equiaxed crystal region by setting a threshold value for discriminating between a columnar crystal region and an equiaxed crystal region from the distribution of the average shape anisotropic amount.
The present invention provides a method for analyzing a solidified metal crystal structure, which is characterized by having.

(II) Another aspect of the present invention is an apparatus for carrying out the above-mentioned method for analyzing a solidified metal crystal structure.
A crystal structure observation image acquisition mechanism for acquiring the observation image of the metal solidified crystal structure,
A crystal grain shape anisotropy calculation mechanism for calculating the shape anisotropy of the crystal grains in the observation image,
A crystal grain partial area calculation mechanism that calculates the partial area of the crystal grain based on the analysis unit region,
An average shape anisotropy calculation mechanism that calculates the average shape anisotropy from the shape anisotropy and the partial area,
An average shape anisotropic amount distribution creation mechanism that creates a distribution of the average shape anisotropic amount, and
An average shape anisotropy binarization mechanism that classifies the average shape anisotropy into two, and
The present invention provides an analyzer for analyzing a solidified metal crystal structure, which is characterized by having.

According to the present invention, a method for quantitatively analyzing a columnar crystal region and an equiaxed crystal region in the crystal structure of a metal solidified body, and an apparatus for analyzing the metal solidified body crystal structure for carrying out the method are provided. Can be done.

It is a flow chart which shows an example of the process in the analysis method of the metal solidified body crystal structure which concerns on this invention. It is a schematic diagram which shows the example which identified each crystal grain by the crystal orientation determination for each pixel in the observation image. It is a schematic diagram which shows the example of the process of calculating the shape anisotropic amount by elliptical approximation of the outer shape of the crystal grain in the identified crystal grain. It is a schematic diagram which shows the process example at the time of performing the analysis unit area setting step and the crystal grain partial area calculation step with respect to the acquired observation image. 6 is a distribution graph of the average shape anisotropic amount showing an example of the result of performing the average shape anisotropic amount distribution creation step of the first embodiment on the observation image of FIG. 6 is a distribution map of the average shape anisotropic amount showing an example of the result of performing the average shape anisotropic amount distribution creation step of the second embodiment on the observation image of FIG. It is a columnar crystal region / equiaxed crystal region map showing the result of binarizing the calculated average shape anisotropic amount by the k-means method. 6 is a distribution map of the average shape anisotropy vector showing an example of the result of performing the average shape anisotropy distribution creation step of the third embodiment on the observation image of FIG. It is a schematic diagram which shows the schematic structure of the analysis apparatus of the metal solidified body crystal structure which concerns on this invention.

Hereinafter, embodiments of the present invention will be specifically described with reference to the drawings. The present invention is not limited to the embodiments taken up here, and can be appropriately combined with a known technique or improved based on the known technique without departing from the technical idea of the invention. ..

[First Embodiment]
FIG. 1 is a flow chart showing an example of a process in the method for analyzing a solidified metal crystal structure according to the present invention. As shown in FIG. 1, the analysis method of the metal solidified crystal structure of the present invention is roughly classified into a crystal grain feature amount analysis process, a feature amount distribution analysis process, and a columnar crystal-equal axis crystal transition analysis process. Become.

More specifically, the crystal grain feature amount analysis process includes a crystal structure image acquisition step S101, a crystal grain identification step S102, and a crystal grain shape anisotropic amount calculation step S103. The feature amount distribution analysis process includes an analysis unit region setting step S201, a crystal grain partial area calculation step S202, an average shape anisotropic amount calculation step S203, and an average shape anisotropic amount distribution creation step S204. The columnar crystal-equal axis crystal transition analysis process includes a columnar crystal region / equiaxed crystal region determination step S301 and a columnar crystal region / equiaxed crystal region identification output step S302. Hereinafter, each step will be described in more detail.

(Crystal structure image acquisition step)
This step S101 is a step of acquiring an observation image of the crystal structure of the solidified metal. As described above, one of the objects of the present invention is to quantitatively analyze the columnar crystal region and the equiaxed crystal region in the crystal structure of the solidified metal, so that the crystal structure to be observed is metal solidified. It is preferable to select a cross section including the depth direction of the solidified metal body so that the transition between columnar crystals and equiaxed crystals in the body can be easily observed.

The image acquisition method of the crystal structure is not particularly limited as long as the shape of each crystal grain can be clearly observed and photographed, and the conventional observation method / image acquisition method can be used. For example, an image observed using a scanning electron microscope or a polarizing microscope can be preferably used.

(Crystal grain identification step)
This step S102 is a step of identifying each crystal grain of the acquired observation image. The method for identifying each crystal grain is not particularly limited, and the conventional identification method / image processing method can be used. For example, backscattered electron diffraction (SEM / EBSD) method using a scanning electron microscope, a method of identifying each grain based on the RGB value of each pixel of a polarizing microscope observation image, material structure simulation (for example, phase field). A method for identifying each crystal grain by using a method (method, cell automaton method, Monte Carlo method) can be preferably used.

To explain an example, in the SEM / EBSD method, when the angular difference in crystal orientation between adjacent pixels of the acquired observation image is within a predetermined range (for example, within 15 °), it is determined that each crystal is the same crystal grain. Grains can be identified. FIG. 2 is a schematic diagram showing an example in which each crystal grain is identified by determining the crystal orientation for each pixel in the observation image. In FIG. 2, it is determined that five crystal grains are present in the observation image. At this time, an index for identifying each crystal grain (for example, crystal grain G ₁ to crystal grain G ₅ ) is given.

(Crystal grain shape anisotropic amount calculation step)
This step S103 is a step of calculating the amount of shape anisotropy of each of the identified crystal grains. FIG. 3 is a schematic diagram showing an example of a process of calculating the amount of shape anisotropy by elliptical approximation of the outer shape of the identified crystal grain.

As shown in FIG. 3, when crystal grains G _i (i is a natural number) are identified in the observation image, when the xy Cartesian coordinate system is applied to the observation image, the outer shape of the crystal grains _{G i is the crystal grains.} It is represented by a set _{of each vertex (coordinates (x j} , y _j )) on the outer circumference of the set of pixels that define G _i.

Next, the outer shape of the crystal grain G _i is approximated to an ellipse defined by the following equation (1). The coefficients A _i , B _i , C _i , D _i , and E _i in Eq. (1) are selected so that the error Error (i) specified in Eq. (2) is minimized.

_{Next, the semimajor axis a i} and the _{minor axis diameter b i} are obtained from the approximate ellipse defined by the equation (1), and their ratio is defined by the equation (3) as the shape anisotropic amount p _{i of the crystal grain G i.} To do. From the definition of equation (3), the _{smaller the value of the shape anisotropy amount p i} , the larger the shape anisotropy.

The above process is performed for _{each grain G i} identified in the observation image.

In the above, it was determined from the ratio of the crystal grains G major axis diameter of the shape anisotropy amount p _i the approximation ellipses _i a _i and minor axis b _i, the present invention is not limited thereto. Shape anisotropy amount p _i of grain G _i may be any other means as long as can properly define. For example, an circumscribed ellipse or an inscribed ellipse with respect to the outer shape of the crystal grain G _{i may be determined, and the shape anisotropic amount p i} may be obtained from the semi-major axis-minor axis diameter ratio thereof. Further, the circumscribed rectangle or the inscribed rectangle with respect to the outer shape of the crystal grain G _{i may be determined, and the shape anisotropic amount p i} may be obtained from the long side short side ratio thereof.

(Analysis unit area setting step)
This step S201 is a step of setting the analysis unit area in the observed image. The analysis unit region R _n (n is a natural number) is a basic unit for analyzing the crystal structure of the entire observation image, and is divided into two or more regions so as to cover the entire observation image. Further, it is preferable that the minimum area of the analysis unit region R _n is set to be an area equal to or larger than the minimum crystal grain in the observed image.

As the division direction, it is preferable to set the _{analysis unit region R n} so as to divide at least the depth direction of the solidified metal body in order to facilitate the analysis of the columnar crystal-isoaxial crystal transition in the crystal structure. .. The shape of the analysis unit region R _n is not particularly limited, but from the viewpoint of ease of calculation in the next step and thereafter, it is preferable to divide the shape into the same shape, and a rectangular shape in which the shape can be easily defined by a mathematical formula is more preferable.

This step S201 may be before the next step S202, and does not necessarily have to be after step S103. For example, it may be performed immediately after step S101 or step S102.

(Crystal grain partial area calculation step)
This step S202 is a step of calculating the partial area of each crystal grain G _i within each analysis unit region R _n. FIG. 4 is a schematic diagram showing a process example when performing the analysis unit region setting step and the crystal grain partial area calculation step on the acquired observation image. In FIG. 4, an observation image of 9 mm parallel to the surface of the solid metal body and 3 mm in the depth direction of the solid metal body was used, and an xy Cartesian coordinate system with the lower left corner of the observation image as the origin was applied.

Further, in FIG. 4, as the analysis unit area setting step S201, a rectangular analysis in which the width (x-axis direction) is the same as the width of the observed image and the height (y-axis direction) is the height of the observed image divided into 10. An example in which the unit area R _n (n = 1 to 10) is set is shown. _{In the crystal grain partial area calculation step S202, the approximate ellipse of the crystal grain G i} derived in the crystal grain shape anisotropic amount calculation step S103 (see equation (1)) is the analysis unit region R set in the analysis unit region setting step S201. Calculate the partial area A _{in that occupies in n.}

The above process is performed for each analysis unit region R _n and each grain G _i .

(Step for calculating average shape anisotropic amount)
This step S203, the average shape anisotropy amount p _n of the analysis unit regions R _n within an area weighted average of the shape anisotropy amount from the shape anisotropy amount p _i and the partial area A _in each grain G _i (analysis unit region R _{n (ave))} ) is a step to calculate. The calculation of the average shape anisotropic amount p _{n (ave)} can be expressed by a mathematical formula as shown in the following equation (4).

The process is performed for each analysis unit region R _n .

(Step for creating average shape anisotropic amount distribution)
This step S204 is a step of creating a distribution of the average shape anisotropy amount p _{n (ave) of the analysis unit region R n} calculated in the previous step S203. FIG. 5 is a distribution graph of the average shape anisotropic amount showing an example of the result of performing the average shape anisotropic amount distribution creation step of the first embodiment on the observation image of FIG. Here, the average shape anisotropy weight p _{n (ave)} can be considered because it is a feature of the entire analysis unit region R _n, and there is a representative point in the center of gravity of the analysis unit region R _n. Therefore, in FIG. 5, the plot position (y coordinate) on the vertical axis is set to the center position of the height _{of each analysis unit region R n.}

As shown in FIG. 5, "p _{n (ave)} <0.25" in "y = 0.15 to 2.25", but "p _{n (ave)} >0.4" in "y = 2.55 to 2.85". It can be seen that there is a rapid change between "y = 2.25 and 2.55".

(Columnar crystal region / equiaxed crystal region determination step)
In this step S301, a threshold value for discriminating between the columnar crystal region and the equiaxed crystal region is set from the distribution of the average shape anisotropic amount, and the columnar crystal region / equiaxed crystal region is quantitatively determined based on the threshold value. It is a step to judge. In the example of FIG. 5, as described above, the average shape anisotropy amount p _{n (ave)} suddenly changes between “y = 2.25 to 2.55” (adjacent average shape anisotropy amount p _{n (ave)).} The difference between them is the largest). Therefore, for example, a threshold value can be set at "y = 2.4", which is the midpoint between "y = 2.25" and "y = 2.55".

Comparing the distribution graph of the average shape anisotropic amount in FIG. 5 with the observation image in FIG. 4, it can be said that the threshold value of “y = 2.4” well represents the boundary between the columnar crystal region and the equiaxed crystal region. .. In other words, by creating the distribution of the average shape anisotropic amount p _{n (ave)} in step S204 and setting the discrimination threshold value of the columnar crystal region / equiaxed crystal region in step S301, the columnar crystal region in the entire observation image. / It can be seen that the equiaxed crystal region can be quantitatively determined.

(Columnar crystal region / equiaxed crystal region output step)
This step S302 is a step of outputting the determination result of the columnar crystal region / equiaxed crystal region obtained in the previous step S301. In the present invention, the output means performed in step S302 is not particularly limited, and the output format / output method desired by the user (for example, display screen display, paper printing, data output) can be appropriately used.

In the analysis method of the metal solidified body crystal structure of the first embodiment, since the analysis unit region R _n was set so as to divide only the depth direction of the metal solidified body in the analysis unit region setting step S201 (in FIG. 4). _{, The analysis unit region R n} having the height obtained by dividing the height direction of the observation image by the same width as the width of the observation image was set for the observation image in which the depth direction of the metal solidified body is the height direction. It is suitable for grasping the overall tendency of the solidified metal in the depth direction (one-dimensional quantitative determination). In addition, since the total number of divisions of the analysis unit region R _n is relatively small, there is an advantage that the number of calculations can be reduced (calculation time can be shortened / calculation results can be obtained in a short time).

[Second Embodiment]
In the method for analyzing the metal solidified crystal structure of the second embodiment, in the analysis unit region setting step S201, the analysis unit region is set finer / smaller than that of the first embodiment, and the entire observed image is two-dimensionally quantitatively determined. It differs in that it does. Hereinafter, the points different from the first embodiment will be mainly described.

First, the crystal structure image acquisition step S101 to the crystal grain shape anisotropic amount calculation step S103 are the same as those in the first embodiment.

On the other hand, in the analysis unit area setting step S201, as described above, the analysis unit area is set more finely than in the first embodiment. For example, for the observation image of FIG. 4, a rectangular analysis unit region R _{n, m} (n = 1 to 10, m = 1 to 30) divided into 10 in the y-axis direction and 30 in the x-axis direction is set. To do.

Next, in the same manner as in the first embodiment, in the crystal grain partial area calculation step S202, _{the approximate ellipse of the crystal grain G i} derived in the crystal grain shape anisotropic amount calculation step S103 (see equation (1)) is the analysis unit. The partial area A _{in, m occupied in the analysis unit area R n, m} set in the area setting step S201 is calculated.

Next, step S204 for creating the average shape anisotropic amount distribution is performed. To calculate the average shape anisotropic amount p _{n, m (ave) in} the analysis unit region R _{n, m , “A in} ” in Eq. (4) may be read as “A _{in, m”.} Further, in the second embodiment, the _{average shape anisotropic amount p n, m (ave)} calculated from the two-dimensional evaluation of the entire observed image by setting the _{analysis unit regions R n, m relatively finely.} It is easy to see if the color and / or shade is applied to the notation.

FIG. 6 is a distribution map of the average shape anisotropic amount showing an example of the result of performing the average shape anisotropic amount distribution creation step of the second embodiment on the observation image of FIG. As shown in FIG. 6, the average shape anisotropic amount distribution map has an advantage that it is easier to understand the state of the entire observed image as compared with the average shape anisotropic amount distribution graph of FIG. For example, in FIG. 6, the color from the top to the second stage is darker than that of the third and subsequent stages, and the average shape anisotropy amount p _{n, m (ave)} is large (that is, the shape anisotropy is small). Can be read.

The “x” in the upper left corner of the average shape anisotropic amount distribution map indicates a defective region. As can be seen from FIG. 4, since this region is a region in which the crystal grains G _i do not exist in the observation image, the partial areas A _{in and m} of the crystal grains G _i become zero in the region, and the equation (4) ) Cannot calculate the average shape anisotropy amount p _{n, m (ave).} In other words, in the present invention, a region in which the average shape anisotropic amount cannot be calculated by the formula (4) is treated as a defective region.

The average shape anisotropic amount distribution map of FIG. 6 has an advantage that the state of the entire observed image is easier to understand than the average shape anisotropic amount distribution graph of FIG. 5, but the columnar crystal region / equiaxed crystal region can be discriminated. There is room for improvement in that it tends to be ambiguous (depending on the subjectivity of the observer). Therefore, in the columnar crystal region / equiaxed crystal region determination step S301 of the second embodiment, the average shape anisotropic amount p _{n, m (ave)} calculated in the previous step S204 by the k-means method is classified into two. Set the threshold value for the columnar / equiaxed region.

The k-means method is a method of classifying a large number of data into k clusters, and is generally performed in the following flow. (A) Randomly allocate clusters for each data. (B) Calculate the center of gravity of each cluster based on the allocated data. The calculation of the center of gravity of the cluster often uses the arithmetic mean of each element of the assigned data. (C) Obtain the distance between each data and the center of gravity of each cluster, and reassign each data to the cluster with the closest center of gravity. (D) When all the data cluster allocations do not change in the above process, or when the amount of change falls below a predetermined threshold value, it is determined that the data clusters have converged and the process ends. If not, the center of gravity is recalculated from the newly allocated cluster and the above process is repeated.

In the present embodiment, each calculated average shape anisotropic amount p _{n, m (ave)} is classified into two clusters (here, a columnar crystal region and an equiaxed crystal region). Regarding the above (a) of the k-means method, for example, random number generation of a computer can be used. Regarding the threshold value of (d) above, for example, 1 × 10 ^-3 to 1 × 10 ^-6 may be appropriately used. Further, the series of the above steps (a) to (d) may be performed a plurality of times, and the result showing the value in which the amount of change in (d) is the smallest may be adopted. A larger value of the cluster center of gravity means a smaller shape anisotropy, so it is determined to be an equiaxed crystal region, and a smaller value of the cluster center of gravity means a larger shape anisotropy. It is determined to be an area.

FIG. 7 is a columnar crystal region / equiaxed crystal region map showing the result of binarizing the calculated average shape anisotropic amount p _{n, m (ave) by the k-means method.} As the conditions of the k-means method, the threshold value of (d) was set to 1 × 10 ^-5, and the series of (a) to (d) was repeated 5 times to show the value in which the amount of change in (d) was the smallest. The result was adopted.

As shown in FIG. 7, the columnar crystal region / equiaxed crystal region map has the advantage that the state of the entire observed image can be understood, and in addition, the columnar crystal region / equiaxed crystal region map is compared with the average shape anisotropic amount distribution map of FIG. There is an advantage that the equiaxed crystal region can be discriminated more clearly.

In the above explanation, each average shape anisotropic amount p _{n, m (ave)} calculated by the k-means method is classified into two, but as long as each average shape anisotropic amount p _{n, m (ave)} can be appropriately classified into two categories. , The k-means method is not limited. For example, a spectral clustering method, a ward hierarchical clustering method, or the like can be appropriately used.

After the columnar crystal region / equiaxed crystal region determination step S301, as in the first embodiment, as the columnar crystal region / equiaxed crystal region output step S302, the obtained columnar crystal region / equiaxed crystal region determination result is obtained. The output format / output method desired by the user may be used for output.

[Third Embodiment]
The method for analyzing the metal solidified crystal structure of the third embodiment is the average of the analysis unit regions as in the first embodiment and the second embodiment in the average shape anisotropic amount calculation step S203 and the average shape anisotropic amount distribution creation step S204. The difference is that the amount of shape anisotropy is not calculated and displayed as a scalar, but is calculated and displayed as a vector. Hereinafter, the points different from the second embodiment will be mainly described.

First, the crystal structure image acquisition step S101 to the crystal grain partial area calculation step S202 may be performed in the same manner as in the second embodiment.

On the other hand, in the average shape anisotropy calculation step S203, as described above, the average shape anisotropy of the analysis unit regions R _{n and m} is not calculated as a scalar, but a vector (average shape anisotropy vector p ^vec _n, Calculate as _{m (ave)).} In order to define a vector, it is necessary to specify the magnitude and direction.

As the calculation method of the average shape anisotropy vector p ^vec _{n, m (ave)} , the analysis unit region R _{n, m} is constructed in the same way as the calculation method of the average shape anisotropy amount p _{n, m (ave).} deriving a direction of the vector in each crystal grain G _i approximation surface load synthesize major axis of the ellipse of (partial area a _in each grain G _i, to synthesize a vector having a length ratio of _m) mETHOD be able to. As the magnitude (norm) of the average shape anisotropy vector p ^vec _{n, m (ave)} , it is desirable that the larger the shape anisotropy, the larger the value. _{It is possible to adopt 1 / p n, m (ave)} (corresponding to the average shape aspect ratio of the analysis unit region R _{n, m} ), which is the reciprocal of the square quantities p _{n, m (ave).}

Further, in vector synthesis, it is necessary to determine the positive and negative directions of the vector. For example, the heat removal direction (negative direction of the y-axis in FIG. 4) when the solidified metal solidifies is set to the positive direction of the vector. The solidification direction of the solidified metal (the positive direction of the y-axis in FIG. 4) may be aligned with the positive direction of the vector. The heat removal direction and the solidification direction when the metal solidified body is solidified are usually opposite to each other.

Next, step S204 for creating the average shape anisotropic amount distribution is performed. FIG. 8 is a distribution map of the average shape anisotropy vector showing an example of the result of performing the average shape anisotropy distribution creation step of the third embodiment on the observation image of FIG. In this example, at the stage of the previous step S203, the analysis unit areas R _{n and m} are divided into 10 in the y-axis direction and 30 in the x-axis direction as in FIG. In vector synthesis for calculating the direction of the average shape anisotropy vector p ^vec _{n, m (ave)} , the heat removal direction (negative direction of the y-axis) when the metal solidified body solidifies is set to the positive direction of the vector. I prepared it.

Further, based on the same concept as in the first embodiment, since the average shape anisotropy vector p ^vec _{n, m (ave)} is a feature quantity of the entire analysis unit region R _{n, m , the analysis unit region R n,} It can be considered that there is a representative point at the position of the center of gravity of _m. Therefore, in FIG. 7, the starting point of each vector is the position of the center of gravity of each analysis unit region.

As shown in FIG. 8, in the average shape anisotropy vector distribution map, in addition to the information of the average shape anisotropy distribution map of FIG. 6, the state of anisotropy of the crystal grains (when the metal solidified body solidifies). There is an advantage that it is easier to understand the heat removal direction). For example, in FIG. 8, since the vector length from the top to the second stage is shorter than that from the third stage onward, it can be read that the shape anisotropy is small. In addition, the heat flow when the metal solidified body solidifies can be read from the direction and concentration of the vector.

However, the average shape anisotropy vector distribution map of FIG. 8 has room for improvement in the same points as in FIG. 6 (the point that the discrimination between the columnar crystal region / equiaxed crystal region depends on the subjectivity of the observer). Therefore, also in this embodiment, it is preferable to perform the same columnar crystal region / equiaxed crystal region determination step S301 as in the second embodiment after the average shape anisotropic amount distribution creation step S204. As a result, a columnar crystal region / equiaxed crystal region map as if FIG. 7 and FIG. 8 are combined can be obtained (not shown).

After the columnar crystal region / equiaxed crystal region determination step S301, the user uses the obtained columnar crystal region / equiaxed crystal region map as the columnar crystal region / equiaxed crystal region output step S302 as in the second embodiment. You can output in the output format / output method you desire.

In the above description, the average shape anisotropic amount calculation step S203 to the columnar crystal region / equiaxed crystal region determination step S301 have been described based on the second embodiment, but the present embodiment is not limited thereto. Steps S203 to S301 may be performed based on the first embodiment.

[Fourth Embodiment]
In this embodiment, an analyzer for analyzing a solidified metal crystal structure for carrying out the methods of the first to third embodiments will be described. FIG. 9 is a schematic view showing a schematic configuration of an analyzer for analyzing a solidified metal crystal structure according to the present invention.

The metal solidified body crystal structure analysis device 100 is a device that quantitatively analyzes the columnar crystal region and the equiaxed crystal region due to the columnar crystal-equal axis crystal transition of the metal solidified body. The observation image data acquisition unit 10 that acquires the observation image of the crystal structure of the crystal structure, and the feature amount analysis of the crystal grains, the feature amount distribution analysis of the crystal structure, and the quantitative analysis of the columnar crystal-equal axis crystal transition are performed from the observation image data. It is composed of an observation image data analysis unit 20, a data storage unit 30 that stores observation image data and various analysis data, and an input / output unit 40 that inputs analysis conditions and outputs analysis results. Hereinafter, the internal configuration of each part will be described in more detail.

The image data acquisition unit 10 has a crystal structure observation image acquisition mechanism 11 that captures an observation image of the metal solidified crystal structure, and is responsible for the crystal structure image acquisition step S101. The crystal structure observation image acquisition mechanism 11 may include microscopes (microscope, control device for the microscope, analysis device for microscope observation data, etc.) for observing the crystal structure itself, or observed with a microscope of another device. It may be a mechanism for capturing image data of the crystal structure.

The captured observation image data is stored in the data storage mechanism 31 of the data storage unit 30. The data storage mechanism 31 is not particularly limited as long as necessary data can be stored, and a conventional data storage device (for example, random access memory (RAM), hard disk (HD), solid state drive (SSD), etc.) is appropriately used. it can.

The observation image data analysis unit 20 calculates the crystal grain identification mechanism 21 for identifying each crystal grain in the acquired observation image, and the approximate shape of each identified crystal grain, and calculates the amount of shape anisotropy from the approximate shape. Each analysis is performed from the shape anisotropy calculation mechanism 22, the crystal grain partial area calculation mechanism 23 that calculates the partial area of each crystal grain based on a separately set analysis unit region, and the shape anisotropy amount and partial area of each crystal grain. An average shape anisotropy calculation mechanism 24 that calculates the average shape anisotropy of a unit region, an average shape anisotropy distribution creation mechanism 25 that creates a distribution of the average shape anisotropy, and an average shape that binarizes each average shape anisotropy. Anisotropy binarization mechanism 26 and an average shape anisotropy vector calculation mechanism 27 that calculates the average shape anisotropy vector of each analysis unit region from the major axis direction and partial area of the approximate shape of each crystal grain. Have. The result calculated in each mechanism is stored in the data storage mechanism 31.

The observation image data analysis unit 20 basically has a crystal grain identification step S102, a crystal grain shape anisotropic amount calculation step S103, a crystal grain partial area calculation step S202, an average shape anisotropic amount calculation step S203, and an average shape anisotropic amount distribution creation step S204. , And the columnar crystal region / equiaxed crystal region determination step S301.

As long as the required analysis is properly performed, the device configuration of each mechanism of the observation image data analysis unit 20 is not particularly limited, and a conventional analysis device (for example, a computer) can be used as appropriate.

When the crystal structure observation image acquisition mechanism 11 of the image data acquisition unit 10 itself has a function of identifying each crystal grain of the observation image, the crystal grain identification mechanism 21 of the observation image data analysis unit 20 is not an indispensable configuration. Further, in the case of an apparatus configuration that does not require vector analysis as shown in the third embodiment, the average shape anisotropy vector calculation mechanism 27 of the observation image data analysis unit 20 is not an indispensable configuration.

The input / output unit 40 outputs the analysis result and the input mechanism 41 that inputs the analysis conditions (for example, _{the size of the analysis unit region R n} and the condition of the identification threshold of the columnar crystal region / equiaxed crystal region). It has an output mechanism 42, and is responsible for the analysis unit region setting step S201 and the columnar crystal region / equiaxed crystal region output step S302. The input various analysis conditions and the results of the columnar crystal region / equiaxed crystal region determination are stored in the data storage mechanism 31.

As long as the required input and the desired output can be obtained, the device configuration of the input mechanism 41 and the output mechanism 42 is not particularly limited, and conventional input / output devices (for example, keyboard, display, printer) can be appropriately used.

The above-described embodiments have been described for the purpose of assisting the understanding of the present invention, and the present invention is not limited to the specific configurations described. For example, it is possible to replace a part of the configuration of the embodiment with the configuration of the common general technical knowledge of those skilled in the art, and it is also possible to add the configuration of the common general technical knowledge of the person skilled in the art to the configuration of the embodiment. In addition, the embodiments may be combined as appropriate. That is, the present invention can delete, replace, or add another configuration to a part of the configuration of the embodiment of the present specification without departing from the technical idea of the invention. ..

100 ・ ・ ・ Analysis device for metal solidified crystal structure,
10 ... Observation image data acquisition unit, 11 ... Crystal structure observation image acquisition mechanism,
20 ... Observation image data analysis unit, 21 ... Crystal grain identification mechanism,
22 ... Crystal grain shape anisotropic amount calculation mechanism, 23 ... Crystal grain partial area calculation mechanism,
24 ... Average shape anisotropic amount calculation mechanism, 25 ... Average shape anisotropic amount distribution creation mechanism,
26 ... Average shape anisotropy binarization mechanism, 27 ... Average shape anisotropy vector calculation mechanism,
30 ... Data storage unit, 31 ... Data storage mechanism,
40: Input / output unit, 41: Input mechanism, 42: Output mechanism.

Claims (11)

This is a method for analyzing the crystal structure of a solid metal body.
A crystal structure image acquisition step for acquiring an observation image of the crystal structure, and
A crystal grain identification step for identifying crystal grains in the observation image, and
A step of calculating the amount of crystal grain shape anisotropy for calculating the amount of crystal grain shape anisotropy,
An analysis unit area setting step for setting an analysis unit area in the observation image, and
A crystal grain partial area calculation step for calculating the partial area of the crystal grain in the analysis unit region, and
An average shape anisotropic amount calculation step for calculating the average shape anisotropic amount of the analysis unit region from the shape anisotropic amount and the partial area, and
The step of creating the average shape anisotropic amount distribution and the step of creating the average shape anisotropic amount distribution,
A columnar crystal region / equiaxed crystal region determination step for determining a columnar crystal region / equiaxed crystal region by setting a threshold value for discriminating between a columnar crystal region and an equiaxed crystal region from the distribution of the average shape anisotropic amount.
A method for analyzing a solidified metal crystal structure. In the method for analyzing a solidified metal crystal structure according to claim 1,
The crystal grain shape anisotropic amount calculation step is characterized in that the outer shape of the crystal grain is approximated to an ellipse and the shape anisotropic amount is calculated from the ratio of the minor axis diameter to the major axis diameter of the ellipse. A method for analyzing the crystal structure of a solidified metal. In the method for analyzing a solidified metal crystal structure according to claim 2.
The observation image is an observation image of a cross section including the depth direction of the solidified metal body.
When the depth direction is applied to the y-axis direction of the xy Cartesian coordinate system, the analysis unit region setting step is the length obtained by dividing the y-axis length of the observation image into a plurality of the analysis unit regions and the above. A method for analyzing a solid metal crystal structure, which is a step of setting a rectangular region having a length in the x-axis direction of an observation image. In the method for analyzing a solidified metal crystal structure according to claim 2.
The observation image is an observation image of a cross section including the depth direction of the solidified metal body.
When the depth direction is applied to the y-axis direction of the xy Cartesian coordinate system, the analysis unit region setting step is the length obtained by dividing the length of the observation image in the y-axis direction into a plurality of the analysis unit regions and the above. A method for analyzing a metal solidified crystal structure, which is a step of setting a rectangular region having a length obtained by dividing the length in the x-axis direction of an observation image into a plurality of parts. In the method for analyzing a solidified metal crystal structure according to claim 4,
A method for analyzing a metal solidified crystal structure, which comprises setting the threshold value in the columnar crystal region / equiaxed crystal region analysis step by the k-means method. In the method for analyzing a solidified metal crystal structure according to claim 4 or 5.
The average shape anisotropic amount in the average shape anisotropic amount calculation step is a vector.
The vector is an area load average obtained from the reciprocal of the shape anisotropic amount and the area load average obtained from the partial area, and the major axis direction of the ellipse and the partial area that approximate the outer shape of the crystal grains. A method for analyzing a metal solidified crystal structure, which comprises the direction of. An apparatus for carrying out the method for analyzing a solidified metal crystal structure according to any one of claims 1 to 5.
A crystal structure observation image acquisition mechanism for acquiring the observation image of the metal solidified crystal structure,
A crystal grain shape anisotropy calculation mechanism for calculating the shape anisotropy of the crystal grains in the observation image,
A crystal grain partial area calculation mechanism that calculates the partial area of the crystal grain based on the analysis unit region,
An average shape anisotropy calculation mechanism that calculates the average shape anisotropy from the shape anisotropy and the partial area,
An average shape anisotropic amount distribution creation mechanism that creates a distribution of the average shape anisotropic amount, and
An average shape anisotropy binarization mechanism that classifies the average shape anisotropy into two, and
An analyzer for analyzing a solidified metal crystal structure. In the analysis apparatus for the crystal structure of a solidified metal according to claim 7.
An analysis device for a metal solidified crystal structure, further comprising a crystal grain identification mechanism for identifying the crystal grains in the observation image. An apparatus for carrying out the method for analyzing a solidified metal crystal structure according to claim 6.
A crystal structure observation image acquisition mechanism for acquiring the observation image of the metal solidified crystal structure,
A crystal grain shape anisotropy calculation mechanism for calculating the shape anisotropy of the crystal grains in the observation image,
A crystal grain partial area calculation mechanism that calculates the partial area of the crystal grain based on the analysis unit region,
An average shape anisotropy calculation mechanism that calculates the average shape anisotropy from the shape anisotropy and the partial area,
An average shape anisotropic amount distribution creation mechanism that creates a distribution of the average shape anisotropic amount, and
An average shape anisotropy binarization mechanism that classifies the average shape anisotropy into two, and
An average shape anisotropy vector calculation mechanism that calculates an average shape anisotropy vector from the long axis direction of the ellipse that approximates the outer shape of the crystal grain, the partial area, and the average shape anisotropy amount.
An analyzer for analyzing a solidified metal crystal structure. In the analysis apparatus for the crystal structure of a solidified metal according to claim 9,
An analysis device for a metal solidified crystal structure, further comprising a crystal grain identification mechanism for identifying the crystal grains in the observation image. The analyzer for analyzing a solidified metal crystal structure according to any one of claims 7 to 10.
An input mechanism for setting and inputting the analysis unit area, and
An analysis device for a solidified metal crystal structure, which further has an output mechanism for outputting analysis results.

Images