JP2009116404A - Alloy design support system, alloy design support program, and alloy design support method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To support a user to design alloy in order to enable the user to easily design new allow developing his/her desired metallic characteristics as much as possible. <P>SOLUTION: This alloy design support system is provided with a storage part 21 for storing prescribed manufacturing information related with the manufacturing conditions of each of a plurality of existing alloy and prescribed characteristic information related with the characteristics of the existing alloy; a difference information acquisition part 32 for acquiring difference information related with a difference between the prescribed manufacturing information of the two existing alloy for the object existing alloy as the whole part or one part of the plurality of existing alloy included in the storage part; and a map creation part 33 for creating an alloy design map by setting the object existing alloy on a map on which evaluation characteristic determined according to the prescribed characteristic information are defined as an evaluation axis, and applying the fluctuation information of the prescribed manufacturing information corresponding to the difference information obtained by the difference information acquisition part to the map. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an alloy design support system, an alloy design support program, and an alloy design support method for supporting design when a user designs a very complex alloy. It is particularly suitable for design support of aluminum alloys.

  Currently, global environmental conservation is being discussed not only in Japan but around the world. Here, the formation of a so-called recycling society has been advocated for environmental conservation, and new alloy designs are attracting attention as part of this. Aluminum alloy has various characteristics such as “high thermal conductivity, easy to process, good corrosion resistance, hard to be magnetized”, etc. and is an indispensable material for our daily life, and “light, easy to recycle” This is because it has an extremely large effect on environmental conservation as a “reproducible” material. Thus, in order to realize a recycling society, aluminum alloys are expected to increase the utilization rate and the application range as environmentally friendly materials.

  Here, a plurality of techniques for supporting complex alloy design have been proposed. The first technique is a technique disclosed in Non-Patent Document 1, and is a technique relating to so-called clustering for classifying previously developed alloys as a basis for designing a new alloy. In this technique, optimal clustering is performed based on the degree of coincidence of fluctuations of a predetermined variable between clusters formed by clustering.

The second technique is a technique disclosed in Non-Patent Document 2, and is a clustering technique as described above. Although this technique uses a genetic algorithm, since the flexibility of the clustering is relatively low, design support for the user is not sufficient.
Kazuhiro Aoyama, Akira Nishiyama, Akira Koga, Tomokazu Motegi: Experimental Data Manning and Knowledge Management System to Support Alloy Design, Proc. Kazuhiro Yoshida: Consideration of how to create a design map from a data set, The University of Tokyo Faculty of Engineering, Graduate School of System Creation, 2004

  High-performance, high-quality, low-cost new alloy design / development is indispensable in order to increase the utilization rate of alloys and expand their application range. However, the alloy design is a very complex design. That is, the alloy design is a series of steps for determining the manufacturing conditions consisting of the ratio of additive elements (composition) and processing conditions to satisfy the metal characteristics (strength, tensile strength, etc.) required according to the purpose. In general, there is no linear relationship between the manufacturing conditions and the metal characteristics, and therefore the manufacturing conditions cannot be easily determined from the alloy characteristics.

  Conventionally, the manufacturing conditions have been determined according to the individual skill of the user, for example, according to the experience and intuition of each person who is said to be an expert. However, it takes a lot of time to pass on these skills, and even in the case of an alloy design that is a series of trial and error, even a skilled person requires a lot of time and cost to design and develop a new alloy. Is done.

In the present invention, in view of the above-described problems, an alloy design support system for supporting the user's alloy design and an alloy design support system in order to make a new alloy exhibiting desired metal characteristics as easy as possible. An object is to provide a program and an alloy design support method.

  In order to solve the above problems, the present invention supports the user's alloy design from the viewpoint of so-called data mining. Data mining performs data analysis with more objectivity by applying data analysis techniques such as statistics, pattern recognition, and artificial intelligence to a large amount of information. By using this data mining, it becomes possible to derive the regularity that the user hidden behind a large amount of information has not noticed until now, so that it can efficiently support the design of the user's new alloy It becomes possible.

  Specifically, the present invention is an alloy design support system that supports a user's design of a new alloy, and for each of a plurality of existing alloys, predetermined manufacturing information regarding the manufacturing conditions of the existing alloy, and characteristics of the existing alloy A storage unit that stores predetermined characteristic information relating to, and a target existing alloy that is all or part of a plurality of existing alloys included in the storage unit, with respect to a difference between the predetermined manufacturing information of each of two existing alloys A difference information acquisition unit for acquiring difference information, and the target existing alloy are set on a map having an evaluation characteristic determined in accordance with the predetermined characteristic information as an evaluation axis, and the difference information acquisition unit is configured for the map. A map creation unit that creates an alloy design map by adding variation information of the predetermined manufacturing information corresponding to the acquired difference information.

  The alloy design support system according to the present invention provides support for alloy design based on the predetermined information stored in the storage unit, that is, the predetermined manufacturing information and the predetermined characteristic information regarding the existing alloy. That is, the alloy design support system provides design support based on the metal characteristics of an existing alloy that has been known so far and the manufacturing conditions when the existing alloy is manufactured. Here, in general, it is difficult to find a relationship that can be immediately recognized by the user, such as linearity, between the manufacturing conditions of an existing alloy and its metal characteristics. Therefore, in the alloy design support system according to the present invention, data mining useful regularity when a user makes a new alloy design from information on these existing alloys (in some cases, the amount of information is extremely large). It is derived from the point of view and presented to the user.

  The predetermined manufacturing information is information relating to manufacturing parameters set when manufacturing an existing alloy. For example, the composition amount of an existing alloy, information on the temperature and time of a processing step for manufacturing, etc. It is. The type of each parameter constituting the predetermined manufacturing information may be appropriately selected by the user as necessary and stored in the storage unit. The predetermined characteristic information is information indicating the metal characteristics of the existing alloy, and examples thereof include proof stress and tensile strength. The predetermined characteristic information is also a parameter for specifying a new alloy desired by the user when supporting the design of the alloy design support system according to the present invention. That is, the user can receive support from the alloy design support system according to the present invention when designing a new alloy having desired metal characteristics.

  Therefore, in the alloy design support system according to the present invention, the difference information acquisition unit acquires the difference information for the predetermined manufacturing information related to the existing alloy stored in the storage unit. The acquisition of the difference information is not necessarily performed for the predetermined manufacturing information of all the existing alloys stored in the storage unit, and the predetermined manufacturing information may be selected as necessary. In the present invention, the existing alloy for which the difference information is acquired is referred to as a target existing alloy. This difference information is information on the amount of production conditions that differ between the target existing alloys, that is, the amount of variation in the production conditions that has some connection with the metal characteristics.

Based on this difference information, an alloy design map is created by the map creation unit. This alloy design map assists the user when designing a new alloy. The alloy design map is a map formed with evaluation characteristics determined according to predetermined characteristic information of the target existing alloy as an evaluation axis, and is derived from the target existing alloy and the difference information related to the target existing alloy. Variation information (in some cases, the variation information may be the difference information itself) is shown at the same time. Thereby, the user can know the correlation between the target existing alloys on the evaluation axis through the variation information. As a result, it is possible to easily correlate the flow of manufacturing condition variation between the target existing alloys set in the alloy design map and the evaluation characteristics based on the evaluation axis related thereto, and thus the metal characteristics desired by the user can be obtained. It will be easier to guide the manufacturing conditions that will lead to the new alloys.

  From the above, according to the alloy design support system according to the present invention, it is possible to provide useful information to the user from information on the manufacturing conditions and metal properties of the existing alloy, and as a result, the user can design a new alloy. The possibility of performing easily increases.

  Here, in the alloy design support system, the map creation unit may display the variation information with a predetermined symbol visually recognizable to the user. An example of the predetermined symbol is an arrow symbol. By doing in this way, it becomes possible for a user to recognize easily the manufacturing conditions for the design of a new alloy visually.

  Further, in the alloy design support system described above, the target existing alloy is divided into a plurality of predetermined clusters on the alloy design map based on a difference tendency between the predetermined manufacturing information of the two target existing alloys in the difference information. In addition, a clustering processing unit that acquires, as cluster difference information, information on the difference between the predetermined manufacturing information between two predetermined clusters in the plurality of predetermined clusters, and the map creation unit includes: You may make it give the fluctuation | variation information of the said predetermined manufacture information corresponding to the said cluster difference information with respect to the said alloy design map which has the said predetermined cluster produced | generated by the said clustering process part.

  The alloy design map created by the map creation unit as described above is very useful for the user's new alloy design. However, as the number of target existing alloys set in the alloy design map increases, there is a possibility that more accurate design support can be provided to give more information to the user. Since there is a lot of information, there is a possibility that information on manufacturing conditions for a new alloy obtained by the user from the alloy design map will be buried. Usually, even if the target alloy is different on the map, the metal characteristics may be adjacent to each other. In such a case, information provided to the user may be aggregated.

  Therefore, clustering by the clustering processing unit is performed as an aggregation of the information. The predetermined cluster generated by the clustering includes one or a plurality of target existing alloys among all target existing alloys set in the alloy design map, thereby collecting information. Note that the generation of the predetermined cluster is performed based on the tendency of the difference between the predetermined manufacturing information of the two target existing alloys, thereby, one or a plurality of target existing alloys belonging to one predetermined cluster, The difference between the predetermined manufacturing information existing between one or a plurality of target existing alloys belonging to another predetermined cluster is approximately the same. Therefore, the map creation unit can easily recognize the manufacturing conditions of the new alloy having the metal characteristics desired by the user by giving the variation information between the predetermined clusters instead of the variation information between the individual target existing alloys. . Clustering by this clustering processing unit will be more useful as the number of target existing alloys increases.

Here, in the alloy design support system, the clustering processing unit forms a predetermined Voronoi region in which each target existing alloy belongs as a generating point in the map in which the target existing alloy is set by the map creating unit. An alloy network forming unit for forming an alloy network by a predetermined Delaunay side by connecting a target existing alloy belonging to an adjacent predetermined Voronoi region among predetermined Voronoi regions formed by the Voronoi processing unit, and the alloy network forming A map processing unit for classifying the predetermined Voronoi regions associated with each other on the basis of the alloy network formed by a unit, and dividing the predetermined Voronoi regions into the predetermined clusters. The section is divided by the sorting processing section. With respect to the alloy design map having said predetermined clusters morning, may be given the variation information.

  When clustering by the clustering processing unit on the alloy design map, a big problem may occur. In order to generate a predetermined cluster, since it is necessary to be based on the correlation between the amount of variation in manufacturing conditions between the target existing alloys and the metal characteristics, the processing amount required for generating the predetermined cluster may be extremely large. Further, if one predetermined cluster overlaps another predetermined cluster on the alloy design map, it may be difficult for the user to recognize manufacturing conditions for designing a new alloy.

  Therefore, processing by the Voronoi processing unit and the alloy network unit is performed. The Voronoi processing unit forms a predetermined Voronoi region in which each target existing alloy belongs as a generating point on the alloy design map. Since this predetermined Voronoi region is formed on the alloy design map in accordance with a Voronoi diagram forming method which is a mathematical method, an arbitrary point of one predetermined Voronoi region and an arbitrary point of another predetermined Voronoi region are the respective mother points. Further, the distance between the points is longer with the other mother point, and the areas do not overlap each other. The alloy network forming unit forms an alloy network with predetermined Delaunay sides between adjacent predetermined Voronoi regions. This Delaunay side is also based on a mathematical method.

  The predetermined Voronoi region formed in this way can be used as a unit of a predetermined cluster because it does not overlap. In addition, the alloy network formed in this way indicates a route (network) for aggregating and expanding predetermined Voronoi regions that do not overlap, so that the classification processing unit aggregates the predetermined Voronoi regions according to this network. By unfolding, it is possible to suppress an unnecessarily large amount of processing for generating a Voronoi region as a unit of a predetermined cluster. Therefore, the map creation unit can efficiently support the design of a new alloy by giving the variation information to the predetermined cluster classified by the classification processing unit.

  Here, in the alloy design support system, with respect to the number of the predetermined clusters generated by the clustering processing unit, it further includes a cluster number request receiving unit that receives a request from a user, the clustering processing unit, The target existing alloy on the alloy design map may be classified into the predetermined number of clusters received by the cluster number request receiving unit. That is, the degree of clustering in the alloy design support map can be adjusted according to the user's request.

  When the number of predetermined clusters due to clustering is small, the correlation between the manufacturing information on the alloy design map and the metal characteristics is captured macroscopically, so the user may recognize rough design guidelines. it can. On the other hand, if the number of the predetermined clusters is increased, so-called fine predetermined clusters are generated, so it becomes difficult to recognize the design guideline, but it is possible to recognize the correlation between the fluctuation information on the manufacturing conditions and the metal characteristics in detail. It becomes. Therefore, it becomes possible to change the predetermined number of clusters according to the user's request, so that the user can accurately grasp the manufacturing conditions for the desired new alloy design.

Here, in the alloy design support system up to the above, the storage device stores the plurality of existing alloys containing information similar to the premise production information that is a premise of the user's alloy design as part of the predetermined production information. An existing alloy extraction unit that extracts the target existing alloy from the existing alloy is further provided, and the difference information acquisition unit outputs the difference information to the target existing alloy extracted by the existing alloy extraction unit. You may make it acquire.

  The information on the existing alloy stored in the storage unit is a wide variety of information, and in some cases, all of the information is not necessarily required for designing a new alloy. In particular, in general alloy design, a method is adopted in which some conditions among manufacturing conditions are fixed as preconditions and other conditions are adjusted. Then, from the viewpoint of the user's design support, existing alloy information having conditions that are the same as or close to this precondition may be useful for designing the user's new alloy. Therefore, in the alloy design support system according to the present invention, information on the existing alloy stored in the storage unit by the existing alloy extraction unit from the viewpoint of similarity to the premise manufacturing information of the new metal alloy that the user intends to design. Shall be sifted. Thereby, in support of alloy design, more appropriate information can be provided to the user. Here, the premise manufacturing information is information regarding a part of manufacturing conditions determined in advance when a new alloy is to be designed.

  And in the said alloy design support system, the said existing alloy extraction part is a correlation with the said predetermined manufacturing information and the said predetermined characteristic information of the target existing alloy extracted temporarily about the aptitude in the target existing alloy extracted temporarily. You may make it judge based on. That is, even if the existing alloy extracting unit extracts the target existing alloy, it can finally extract the target existing alloy suitable for the user's design support by judging the suitability of the target alloy each time. Here, the criterion for determining the suitability is the correlation between the predetermined manufacturing information of the target existing alloy and the predetermined characteristic information that are temporarily extracted. If any contradiction can be found in this correlation, the suitability of the target existing alloy is denied.

  Here, in the alloy design support system described above, the predetermined manufacturing information includes information corresponding to a plurality of predetermined manufacturing items related to alloy manufacturing conditions, and in that case, the difference information acquisition unit includes The difference information is acquired corresponding to each of the predetermined manufacturing items included in the predetermined manufacturing information, and the map creating unit is configured to correspond to the predetermined manufacturing items included in the predetermined manufacturing information, You may make it create. By creating an alloy design map for each predetermined manufacturing item, the user can grasp the relationship between the fluctuation of the predetermined manufacturing item and the metal characteristics of the alloy, thereby providing more efficient alloy design support.

  The predetermined manufacturing item of the predetermined manufacturing information may include at least a composition amount of a composition metal composing the existing alloy and a manufacturing process for manufacturing the existing alloy. That is, in the actual alloy design, since it is indispensable to determine the composition amount of the composition and the manufacturing process, it is preferable to create an alloy design map for each item.

  The alloy design support system according to the present invention is particularly useful for alloy design support related to wrought material. That is, it means that the existing alloy is an alloy related to the wrought material. This is because the wrought material is an alloy that must consider both the composition of the alloy and its manufacturing process. Of course, it goes without saying that the alloy design support system according to the present invention is also useful for alloys other than wrought materials, for example, alloys related to cast materials whose manufacturing process is relatively simple.

Next, in order to solve the above problems, the present invention may be grasped from the aspect of a program that operates a computer. Specifically, the present invention relates to a computer program for supporting a user's alloy design, and relates to predetermined manufacturing information regarding manufacturing conditions of the existing alloy and characteristics of the existing alloy for each of a plurality of existing alloys. In a computer having a storage device for storing predetermined characteristic information, the predetermined manufacturing information of each of the two existing alloys with respect to a target existing alloy that is all or part of the plurality of existing alloys included in the storage device A difference information acquisition step for acquiring difference information relating to a difference, and the target existing alloy is set on a map having an evaluation characteristic determined according to the predetermined characteristic information as an evaluation axis, and the difference information is acquired with respect to the map A map creation process for creating an alloy design map by adding variation information of the predetermined manufacturing information corresponding to the difference information acquired by the unit Tsu and up, is an alloy design support for the program to be executed by the. Thereby, the user can receive the support for the above-mentioned efficient alloy design via the computer by which the said program is executed. In addition, the technical idea included in the invention specifying items related to the alloy design support system disclosed above can also be applied to the alloy design support program according to the present invention.

  Furthermore, in order to solve the above-mentioned problems, the present invention may be grasped from the aspect of a method for assisting in designing a new alloy. Specifically, the present invention relates to an alloy design support method for supporting a user's alloy design, and for each of a plurality of existing alloys, predetermined manufacturing information relating to the manufacturing conditions of the existing alloy, and characteristics of the existing alloy The difference information about the difference between the predetermined manufacturing information of each of the two existing alloys is acquired with respect to the target existing alloy that is all or part of the plurality of existing alloys included in the storage device that stores the predetermined characteristic information regarding Difference information acquisition step, and the target existing alloy is set on a map having an evaluation characteristic determined according to the predetermined characteristic information as an evaluation axis, and the difference acquired by the difference information acquisition unit with respect to the map A map creation step of creating an alloy design map by giving variation information of the predetermined manufacturing information corresponding to the information. Thereby, the user can receive support for the above-mentioned efficient alloy design through the alloy design support method. In addition, the technical idea included in the invention specifying items related to the alloy design support system disclosed above can be applied to the alloy design support method according to the present invention.

  It is possible to support the user's alloy design in order to make a new alloy exhibiting desired metal characteristics as easy as possible.

  Here, an alloy design support system according to the present invention will be described with reference to the accompanying drawings. In addition, the said Example shows an example of the alloy design assistance system which concerns on this invention, and does not limit the scope of the right of this invention. In addition, the alloy design support system allows a computer to execute a predetermined program so that its function is exhibited. As a result, the user can efficiently support the design of a new alloy. Here, FIG. 1 is a diagram showing a schematic configuration of the alloy design support system 1 according to the present invention and functional blocks representing the function performed by each component. This functional block is realized by a program executed in the computer main body 3, but may be realized by a dedicated processor or the like instead.

  A schematic configuration of the alloy design support system 1 according to the present invention will be described below. The alloy design support system 1 includes a storage device 2 composed of a hard disk drive or the like, a computer main body 3 having an arithmetic processing device such as a CPU and capable of various arithmetic processing, and user instructions and requests to the computer main body 3. Are input devices 4 such as a keyboard and a display 5 for displaying calculation results by the computer main body 3. In the storage device 2, information related to the existing alloy that has been accumulated so as to support the design of a new alloy by the user is stored in the alloy information storage unit 21. Information on the existing alloy will be described later. In addition, the input device 4 is provided with a clustering input unit 41 for inputting a request regarding clustering from a user described later. The display 5 is provided with a display unit 51 for displaying a calculation result by the computer main body 3.

  Here, in the computer main body 3, the target existing alloy extraction unit that extracts, as the target existing alloy, the existing alloy necessary for the user's new alloy design support from the information on the existing alloy stored in the alloy information storage unit 31. 31 is provided. And the difference information acquisition part 32 acquires the difference information regarding the manufacturing information and characteristic information which they have about the object existing alloy extracted by this object existing alloy extraction part 31. Thereafter, according to the acquired difference information, an alloy design map is created by the map creation unit 33, and the alloy design map is displayed on the screen of the display 5 by the display unit 51. The map creation unit 33 is provided with a clustering processing unit 35, and the clustering processing unit 35 includes a Voronoi processing unit 36, an alloy network forming unit 37, and a classification processing unit 38. Detailed description thereof will be described later.

  Here, processing for design support of a new alloy by the computer main body 2 will be described with reference to FIG. This process is performed when a design support request is issued from the user via the input device 4. Here, before describing the alloy design support process shown in FIG. 2, information on the existing alloy stored in the storage device 2 (hereinafter referred to as “existing alloy information”) will be described with reference to FIG. . As shown in FIG. 3, the existing alloy information includes information on the manufacturing conditions of the existing alloy and information on the metal characteristics of the existing alloy for each existing alloy, and further information on the manufacturing conditions includes the existing alloy information. It consists of information on the composition amount of the composition constituting the alloy and information on the manufacturing process set in manufacturing the existing alloy, and each information stores a corresponding numerical value. Thus, the existing alloy information is information in which the manufacturing conditions of the existing alloy and the metal characteristics are stored in association with each other.

  The alloy design system 1 according to the present invention uses this existing alloy information to support the design of a new alloy having the characteristics required by the user. Therefore, the metal characteristics that can be requested by the user are the same as or derived from the metal characteristics of the information stored in the existing alloy information. The existing alloy information shown in FIG. 3 is an example, and the information structure is not limited to this example. In the following description, the structure of the existing alloy information is appropriately adjusted so that the description is simple and easy to understand.

  Returning to FIG. 2, in the alloy design support process, first, in step S <b> 101, the target existing alloy information extraction process is performed by the target existing alloy extraction unit 31. In the present invention, this processing is referred to as “grouping”. This grouping process is a process for extracting information useful for the user's alloy design support from the existing alloy information stored in the alloy information storage unit 21. Since the alloy information storage unit 21 stores a large amount of information on existing alloys, not all of them are necessary for design support of new alloys desired by the user. In particular, in order to acquire “difference information”, which will be described later, it is necessary to acquire difference information useful for the user, the grouping process in S101 is a useful process. The details will be described later. When the process of S101 ends, the process proceeds to S102.

  In S102, the difference information of the target existing alloy extracted in S101 is acquired. The difference information acquisition process is performed by the difference information acquisition unit 32. The difference information according to the present invention is information on manufacturing conditions between two existing alloys, which is considered to be a factor for deriving a difference in properties between one existing alloy having different properties and another existing alloy. Represents the “difference”. In other words, the existence of this “difference” is a major factor that determines the metal characteristics of the alloy, so by presenting the difference information regarding the “difference” to the user, the user can support the design of a new alloy desired by the user. It can be done.

An example of obtaining the difference information will be described based on FIGS. 4 (a) and 4 (b). FIG. 4A is a diagram showing difference information between the existing alloy number 6 and the existing number 8 alloy information in the existing alloy information shown in FIG. Between the existing alloy No. 6 and the existing alloy No. 8, among the information on the manufacturing conditions, the information on the manufacturing process is the same, but the information on the composition is different, and the metal characteristics of both There are also differences. Therefore, when looking at the difference between the No. 6 existing alloy and the No. 8 existing alloy, it is said that “the composition 1 increases and the composition 4 decreases and the characteristics 1 and 2 increase”. Difference information can be found.

  Similarly, FIG. 4B is a diagram showing difference information between the existing alloy number 9 and the existing number 11 alloy information in the existing alloy information shown in FIG. Between the existing alloy No. 9 and the existing alloy No. 11, among the information on the manufacturing conditions, the information on the manufacturing process is the same, but the information on the composition is different, and the metal characteristics of both There are also differences. Therefore, looking at the difference between the No. 9 existing alloy and the No. 11 existing alloy, it is said that “the composition 1 increases and the composition 4 decreases and the characteristics 1 and 2 increase”. Difference information similar to that shown in FIG. 4A can be found.

  Moreover, although the difference information shown in FIG. 4 shows the correlation with the difference of the information regarding a composition among the information of manufacturing conditions, and a metal characteristic, the difference of the information regarding a manufacturing process among the information of manufacturing conditions is also similar. Difference information indicating a correlation with metal characteristics is also acquired by the difference information acquisition unit 32. By presenting this difference information together to the user, the user can easily and reliably design a new alloy. Then, the process of S103 is demonstrated regarding presentation of the difference information to a user next.

  In S103, an alloy design map is created based on the difference information acquired in S102. The alloy design map creation process is executed by the map creation unit 33. Here, the correlation between the difference information and the alloy design map will be described with reference to FIG. The difference information shown in FIG. 5 is not the existing alloy information shown in FIG. 3 but the difference information related to the newly set target existing alloys A, B, and C in order to simplify the explanation. And this difference information shows the difference of the information regarding the composition 1, the composition 2, and the process 1 as information regarding manufacturing conditions. For example, the difference information when the alloy B is viewed from the alloy A is information that the numerical value of the composition 1 is increased, but the numerical values of the composition 2 and the process 1 are decreased. Thus, all the difference information (six kinds) among the three types of target existing alloys is shown in a matrix in FIG.

  Here, as shown in FIG. 5, the alloy design map according to the present invention is a map formed by evaluation axes based on two metal characteristics of tensile strength and proof stress, and items of manufacturing conditions related to difference information (hereinafter, “ In the present example, it is prepared for each of the composition 1, the composition 2, and the step 1), and three existing alloys A to C are plotted on each map. And in each map, the difference information regarding each manufacture item between each existing alloy is shown as an "arrow" which connects between the existing alloys. The direction of the arrow is set so that the value of each manufacturing item decreases. For example, in the example shown in FIG. 5, for the manufacturing item “composition 1”, when the composition 1 is reduced from the existing alloy B, the existing alloy A and C approaches, and when the composition is further reduced from the existing alloy C, the existing alloy A The three-way correlation of approaching is shown. The correlation between the three items is similarly shown for the manufacturing item “Composition 2” and the manufacturing item “Step 1”. These three results are displayed on the screen of the display 5 via the display unit 51 by the process of S104.

  As described above, according to the present alloy design support process, the user can estimate the manufacturing conditions of the alloy he / she desires from the relationship between the manufacturing conditions of the existing alloy and the metal characteristics, and thus the alloy design support system 1. The effect of design support can be enjoyed.

Here, the extraction process (grouping process) by the target existing alloy extraction unit 31 will be described in detail with reference to FIGS. FIG. 6 is a flowchart of the grouping process, FIG. 7 is a diagram showing a change in the group of the target existing alloy extracted by the grouping process, and FIG. 8 shows that the group grouped in the grouping process is appropriate. It is a figure which shows the procedure for discriminating whether or not.

  The grouping process by the target existing alloy extraction unit 31 is performed in consideration of the convenience in the actual alloy design. When actually designing an alloy, the user often considers the composition and the manufacturing process separately among the manufacturing conditions. One is fixed because there is a temporal order between the adjustment of the composition and the change of the manufacturing process condition, that is, the order of the process is determined after determining the amount of the composition. This is because it is efficient to determine the other under the above assumption. As an example, when determining a composition, a user sets process conditions in advance, and designs a new alloy by repeatedly adjusting the composition under the same process conditions. Similarly, a new alloy is designed by setting a composition in advance and repeatedly adjusting the process conditions under the same composition. In the present invention, these previously determined conditions are referred to as “preconditions”.

  In consideration of the above, in designing an actual new alloy, a process is performed in which one of the manufacturing conditions is fixed and the other is determined after fixing one of the composition conditions. Then, it is thought that it is preferable to follow the flow of this process also in the alloy design support system 1 according to the present invention. That is, in order for the alloy design support system 1 to assist the user with the alloy design map, one of the composition and the process condition is fixed as a precondition, and then the variation of the other and the metal characteristics of the alloy are determined. It is necessary to show this correlation on the alloy design map. In other words, the difference information to be displayed in the alloy design map is the difference information of the composition between the existing alloys when the process conditions are the same, or the difference information of the process conditions between the existing alloys when the composition is the same. It is preferable.

  Here, the existing alloy information stored by the alloy information storage unit 21 is various and a large amount of information. However, the number of information in which the information on the conditions of the composition or the process conditions as the preconditions completely matches is compared. It seems that there is little. Therefore, there is a possibility that it is difficult to provide reliable design support to the user by using only the existing alloy information that completely matches. Therefore, it is conceivable to increase the number of existing alloy information stored by conducting many alloy experiments for information expansion, but this is not preferable because it requires a lot of time and cost.

  Based on the above, the grouping process by the target existing alloy extraction unit 31 is performed, and specifically, in the existing alloy information, even if the composition condition or process condition as a precondition is not completely the same, If it is “similar”, it is considered that the process of acquiring the difference information may be performed in the same manner as in the case of being the same. In the grouping process, existing alloys with similar compositions or process conditions are collected into one group, and existing alloys in the same group are regarded as having the same composition or process conditions, thereby obtaining difference information. Presentation to the user is performed.

Therefore, in this embodiment, in order to determine the similarity and perform the grouping process, a SOM (self-organization map) method, which is one of the clustering methods using the neural network method, is widely used as a data mining method. Use. SOM is a neural network that models the neural function of the human cerebral cortex. A neural network is an information processing mechanism that mimics the mechanism of the human brain. The basic model of the neuron is expressed by the following equation.
mi (t + 1) = mi (t) + hci (t) [x (t) -mi (t)]
This outputs 1 when the linear weighted sum Σx _i w _i of a large number of inputs xi is greater than a certain threshold value, and 0 otherwise.

This SOM is a kind of hierarchical neural network and has two layers. The first layer is an n-dimensional input layer x (t), and the second layer is called a competitive layer and is generally a two-dimensional array for visually viewing the output. The vector of the competitive layer is expressed by a reference vector mi (t) and has n-dimensional elements as in the input layer. This SOM learning is “unsupervised competitive learning”. In SOM, Euclidean distance is used for learning. The procedure for learning the SOM is shown below.
1) Randomly determine the elements of all reference vectors mi.
2) Give the input vector x (t). At this time, if a neuron i that minimizes the Euclidean distance | x-mi | with x (t) is searched for and that neuron is c, the following equation is established.
| x-mc | = min | x-mi |
3) A neuron having a reference vector mc is a winner unit.
4) The winner unit and the units in the vicinity Nc around it learn the input vector according to the following equation.
mi (t + 1) = mi (t) + hci (t) [x (t) -mi (t)]
However, hci is a neighborhood function. If i ∈ Nc, hci = a (t), otherwise hci =
Defined as 0. Note that a (t) is a learning rate coefficient.
5) Learning is performed by repeating 2) to 4) T times. The neighborhood size is expressed as a function of time of Nc = Nc (t), and the size is reduced with learning.
6) By repeating steps 2) to 5) for all input vectors, units similar to each input vector are gathered.

  In this embodiment, all existing alloy information stored in the alloy information storage unit 21 is grouped by this SOM to group a large amount of existing alloy information into a plurality of groups from the viewpoint of similarity. can do. In other words, this grouping process is a process of extracting existing existing alloy information appropriate for providing user design support as a pre-stage of processing for obtaining difference information and creating an alloy design map as target existing alloy information. It is. FIG. 7 schematically shows how the number of groups is changed by the grouping process based on the “similarity” of the existing alloy information. The preconditions for judging similarity determine how all existing alloys A to E are grouped. FIG. 7 shows an example of the change in the number of groups of existing alloys depending on the degree of grouping.

  In this way, the grouping process using “similarity” as a criterion is performed by SOM, but the preconditions are not always completely matched even for information on existing alloys belonging to the same group due to similarity. In spite of the fact that they are judged to be similar, the slight difference that exists there can affect the metal properties. In other words, the similarity spread must be appropriate in order to accurately derive the difference information between the other manufacturing conditions and the metal characteristics, assuming that the preconditions are the same. Here, in the grouping process by SOM, when the number of generated groups increases, the number of existing alloys belonging to the group decreases, so that the spread of similarities narrows and very similar existing alloys gather. However, the number of target existing alloys at the time of acquiring the difference information is reduced, and it becomes difficult to perform appropriate design support for the user.

  Therefore, a grouping process according to the present embodiment in consideration of the above problem will be described with reference to FIG. In this embodiment, it is assumed that the precondition is the composition amount of the alloy composition, and the user receives design support from the alloy design support system regarding the process conditions. Therefore, in this case, a grouping process is performed for the similarity of the composition conditions. First, in S201, the number of provisional groups is determined. In order to increase the accuracy of the design support based on the difference information, the number of groups should be small. Therefore, the initial value of the number of groups is set to 2, and the following processing is performed. When the process of S201 ends, the process proceeds to S202.

  In S202, grouping processing is performed by the SOM described above. At this time, the number of groups generated is the number of groups determined in the previous step. Thereafter, in S203, the group most similar to the precondition set in advance among the groups grouped in S202 is selected as a candidate group. When the process of S203 ends, the process proceeds to S204.

  Here, in S204, the candidate group selected in S203 is divided (grouped) into two groups by SOM. Note that the grouping in S204 is not a grouping that is directly required to acquire the difference information for supporting the design of the new alloy, but the candidate group selected in S203 acquires the difference information. This is a grouping for determining whether or not the group is appropriate. The processes of S204 to S206 are processes for determining the suitability of this candidate group, and the situation will be described with reference to FIG. Here, the grouping by the process of S204 is a process of dividing the upper left candidate group shown in FIG. 8 into two groups of P and Q as shown in the upper right side of the figure.

  Next, in S205, a temporary alloy design map is created for each of the groups divided in S204. This alloy design map is a map created in order to determine the suitability of the candidate group as described above, and the creation is the same as that by the map creation unit 33 described above. A provisional alloy design map created for each group divided into P and Q is shown in the lower part of FIG. When the processing of S205 ends, the process proceeds to S206.

  In S206, it is determined whether or not contradictory regularity exists in the two provisional alloy design maps created in S205. In this example, the provisional alloy design map shows the correlation between the variation of the composition X and the characteristics A and B. Here, in group P, the regularity that the characteristic B decreases as the composition X decreases, while in the group Q, the regularity that the characteristic B increases as the composition X decreases can be found. Both rules are contradictory. Therefore, the candidate group selected in S203 is considered to belong to the existing alloy collected by the similarity having a relatively wide spread, and in order to acquire the difference information by the difference information acquisition unit 32, It is considered an inappropriate group. Therefore, in such a case, the determination in S206 is negative and the process proceeds to S208. In S208, the number of provisional groups is incremented by one from the current number, and the processing from S202 onward is performed.

  If the determination in S206 is affirmative, the candidate group selected in S203 is considered appropriate. In this case, the process proceeds to S207, where the candidate group is extracted as the target group, The target existing alloy extraction process which is a grouping process is terminated. The information on the existing alloy belonging to the target group extracted in this process is passed to the process of S102 shown in FIG. 2 as the information on the target existing alloy. Thereby, based on the existing alloy information which has appropriate similarity, an alloy design support is performed to a user.

  Here, as described above, in the alloy design map, the correlation between the variation of the manufacturing condition based on the difference information and the metal characteristic is presented to the user. For example, when there are five target existing alloys, as shown in FIG. 9 (a), in the alloy design map, the five points corresponding to the target existing alloy and the points are connected to each other. The straight line with an arrow which shows the fluctuation | variation of manufacturing conditions (a predetermined composition in Fig.9 (a)) is shown. Thereby, the user can estimate how a new alloy having desired characteristics A and B can be designed by changing the predetermined composition.

  The design support by this alloy design map is naturally useful for the user's design support because as the number of target existing alloys displayed on the map increases, more information is given to the user. However, since the straight line with the arrow is disturbed due to an increase in the number, there is a high possibility that the user is difficult to read a certain regularity from the alloy design map.

Therefore, the concept of cluster was introduced into the alloy design map created by the alloy design support system 1 according to this example. A cluster is defined as a group of existing alloys having similar difference information from existing alloys belonging to other clusters. Therefore, in the alloy design map, alloys belonging to the same cluster tend to have the same arrow direction based on difference information from alloys belonging to other clusters. And it can be said that the more consistent the directions of the arrows between the existing alloys belonging to different clusters are, the higher the reliability of the rule found therefrom.

  Therefore, in this embodiment, as shown in FIG. 10, the direction of the arrow based on the difference information between the existing alloys as much as possible is converted into the direction of the arrow between the clusters, and the change in the manufacturing conditions between the existing alloys is sublimated. This is expressed as a change in manufacturing conditions between clusters (state shown on the right side of FIG. 10). As a result, since the number of arrows shown in the alloy design map is reduced, the alloy design map is easy to see, and the difference information of the manufacturing conditions between the clusters is indicated by the arrows across the clusters which are the variation information. Compared to the case indicated by the arrow between existing alloys, the reliability of the rules found therefrom is considered to be improved, which contributes to the user's support for designing a new alloy.

  In the alloy design map in which the concept of clusters is introduced, the user can overlook the correlation between the manufacturing conditions and the metal characteristics between the clusters, and thus the user can easily design a new alloy. Therefore, based on FIG. 11, an example of a new alloy design support based on an alloy design map in which the concept of clusters is introduced will be shown.

  In the example shown in FIG. 11, the user aims to design a new alloy that has both excellent properties A and B. Here, the compositions added to the new alloy are composition X and composition Y. Therefore, an alloy design map having the characteristics A and B as evaluation axes is created for each of the composition X and the composition Y. The alloy design map shows 11 target alloys to be designed and 11 target existing alloys for design support. Therefore, when the variation information of the composition X between the clusters and the variation information of the composition Y are displayed by the alloy design support system 1 according to the present invention, the results are as shown in FIGS. 11B and 11C. . Therefore, based on these alloy design maps, the user can easily guess that it is only necessary to lower the composition X of the base alloy and raise the composition Y as the manufacturing conditions in order to obtain a target new alloy. Can do.

  A process for creating an alloy design map in which the concept of a cluster that enables design support by the user is introduced will be described with reference to FIGS. The alloy design map creation process is executed by the map creation unit 33 and the clustering processing unit 35. FIG. 12 is a flowchart for creating the cluster on the alloy design map.

  First, in S301, a plot of the target existing alloy that is the target for obtaining the difference information in S102 shown in FIG. 2 is performed on the alloy design map. In this embodiment, the alloy design map is a two-dimensional map having characteristics X and Y as evaluation axes as shown in FIGS. And the target existing alloys plotted there are six alloys A to F. After the process of S301, the process proceeds to S302.

In S302, a predetermined Voronoi region is created on the alloy design map plotted in S301. This process is executed by the Voronoi processing unit 36 shown in FIG. Here, the predetermined Voronoi region is a region divided according to which generating point is close to the point on the alloy design map when the target existing alloy plotted in S301 is used as the generating point in the alloy design map. A well-known mathematical method is used to calculate the Voronoi region. In the case of a two-dimensional Euclidean plane as in the alloy design map of the present embodiment, the boundary line of the Voronoi region becomes a part of a straight bisector between the respective mother points. Therefore, the predetermined Voronoi regions derived by the process of S302 are as shown in FIG. 13, and one target existing alloy belongs to each predetermined Voronoi region. When the process of S302 ends, the process proceeds to S303.

  In S303, in accordance with the predetermined Voronoi region created in S302, an alloy network is formed by Delaunay sides connecting the base points of adjacent predetermined Voronoi regions. This process is executed by the alloy network forming unit 37 shown in FIG. FIG. 14 shows the adjacent relationship of the target existing alloy by the alloy network via the Delaunay side.

  Here, the predetermined Voronoi region created in S302 is a unit of clusters (hereinafter referred to as “unit cluster”) created by the clustering processing unit 35 in this embodiment from the mathematical significance. Since this predetermined Voronoi region does not overlap with other predetermined Voronoi regions, it is very useful as a unit cluster. In addition, the alloy network formed in S303 connects the base points of the Voronoi region, and thus becomes a route for aggregation and development of the unit clusters. Here, the aggregation of unit clusters means that unit clusters are combined into one large cluster, and the reverse movement of unit clusters is the opposite. However, this unit cluster aggregation is performed according to the alloy network. If there are multiple alloy networks (if there are three or more predetermined Voronoi regions, there will always be multiple alloy networks) Which Delaunay side to follow depends on the processing in S304.

  In S304, for the aggregation of unit clusters according to the Delaunay sides constituting the alloy network, the priority of the aggregation is calculated using the displacement matrix method. The alloy design map created by the alloy design support system 1 according to the present invention provides a design support for the user by showing a tendency of change in manufacturing conditions between clusters. Therefore, even in the priority order of the aggregation, the priority of the aggregation is based on the strength of the tendency of the change in the manufacturing conditions between the clusters, that is, the degree of coincidence of the manufacturing conditions between the target existing alloys belonging to each cluster. The ranking is preferably calculated. The displacement matrix method is an evaluation based on an evaluation formula according to the standard.

  Here, the procedure for calculating the “matching degree of change in manufacturing conditions” between target existing alloys belonging to two clusters by the displacement matrix method will be described with reference to FIG. The two clusters are cluster A and B, and the target existing alloys belonging to the respective clusters are ai (i = 1, 2) and bj (j = 3, 4, 5). Further, it is assumed that each target existing alloy has manufacturing conditions depending on the composition or process conditions. Here, regarding the certain manufacturing condition, when the difference information is acquired between the target existing alloys belonging to each of the two clusters, it is counted whether the difference is an increase, a decrease, or the same value. Therefore, the total number of counts (the total number obtained by adding the number in the case of increase (hereinafter referred to as p) and the number in the case of decrease (hereinafter referred to as q) to the same value (hereinafter referred to as r)) is The number is obtained by multiplying the number of target existing alloys belonging to each cluster (in this embodiment, i × j). The state shown in FIG. 15 is p = 5, q = 1, and r = 0.

  Then, according to the number of p, q, and r, the “matching degree of change in manufacturing conditions” between both clusters may be calculated. As an example of the calculation formula, the degree of coincidence may be determined based on the absolute value of the difference between the number p in the case of increase and the number in the case of decrease. That is, it is considered that either the value of p or q is biased higher as the degree of coincidence is higher. In the example shown in FIG. 15, when two clusters out of four clusters are aggregated, there are six combinations as shown in FIG. Therefore, in these six ways, the degree of coincidence is calculated according to the number of p, q, and r, and two clusters having the highest combination among them are preferentially aggregated. As the evaluation formula, various other evaluation formulas can be used.

In addition, the calculation of the “matching degree of change in manufacturing conditions” follows the route of the alloy network. Therefore, in the example shown in FIG. 14, when it is assumed that the alloy A and the alloy B are first aggregated, the degree of change coincidence between the cluster after the aggregation and each of the other unit clusters is calculated. Assuming that the alloy C and the alloy C are aggregated, assuming that the alloy A and the alloy E are aggregated, assuming that the alloy C and the alloy D are aggregated,... In the combination, the “matching degree of change in manufacturing conditions” is calculated. A high aggregation priority is given to the two clusters having the highest degree of coincidence among them. When the two clusters are aggregated into one cluster, the one cluster inherits the adjacent relationship by the alloy network with the other clusters that the original two clusters had.

  Thus, when the priority of aggregation of each cluster is calculated, the dendrogram information shown in FIG. 16 is created by the processing of S305. The dendrogram is binary tree information in which each terminal node represents each target (unit cluster including the target existing alloy) and the cluster formed by aggregation is represented by a non-terminal node. Since this dendrogram information clearly describes the process in which the unit clusters are aggregated, according to this dendrogram information, it is possible to form an appropriate cluster that reflects the “matching degree of change in manufacturing conditions”.

  When the process of S305 is completed, various information including the dendrogram information created by the alloy design map creation process is transferred to the process of S104 shown in FIG. Here, the details of the display process of the alloy design map which is the process of S104 will be described based on FIG. First, in S401, the number of clusters requested by the user is accepted. This process is executed when a user request input via the clustering input unit 41 of the input device 4 is received by the clustering receiving unit 34. After the process of S401 is completed, the process proceeds to S402.

  In S402, it is defined which predetermined cluster is to be formed according to the requested number of clusters accepted in S401. This processing is performed by the classification processing unit 38 based on the dendrogram information created by the alloy design map creation processing, so that an appropriate cluster reflecting the “matching degree of change in manufacturing conditions” is defined. For example, in the example shown in FIG. 18, when the number of target existing alloys plotted on the alloy design map is six, if the number of clusters requested by the user is six, clusters are formed as shown on the left side of FIG. If the number of defined and requested clusters is four, the clusters are defined as shown on the right side of FIG. This cluster aggregation (eg, aggregation of predetermined Voronoi regions of alloy B and alloy C) is based on the previously calculated dendrogram information. When the process of S402 ends, the process proceeds to S403.

In S403, variation information indicating changes in manufacturing conditions between the clusters defined in S402 is given in the form of an arrow or the like. In the present embodiment, the shape of the arrow to be applied, the presence or absence thereof, the type of straight line connecting both clusters, and the like are changed according to the degree of change in manufacturing conditions. This makes it easier for the user to visually recognize changes in manufacturing conditions between clusters in the alloy design map, and the user's support for alloy design is more efficiently performed. An example of selection of the arrow shape and the like is shown in Table 1. P, q, and r shown in Table 1 are synonymous with parameters representing increase, decrease, and equivalence in the difference information shown above.

  When the process of S403 ends, the process proceeds to S404. In S404, the clusters in the alloy design map defined in S402, the variation information given in S403, etc. are passed to the display unit 51 and displayed on the screen of the display 5. In consideration of the display result, the user can perform efficient alloy design as described above.

  Here, in the alloy design system 1 according to the present invention, the number of clusters shown in the alloy design map can be arbitrarily adjusted within a possible range. This is because dendrogram information is created by the alloy design map creation process. As a result, if the user wants to grasp the correlation between the change in the manufacturing conditions of the target existing alloy and the metal properties macroscopically, it is sufficient to reduce the number of clusters, and the more detailed correlation after understanding the macroscopic correlation. If you want to know, you can increase the number of clusters. As a result, the user can set an appropriate design guideline for a new alloy.

It is a figure showing the schematic structure of the alloy design assistance system which concerns on the Example of this invention. 2 is a flowchart showing a flow of processing for supporting a user's alloy design in the alloy design support system shown in FIG. 1. It is a figure which shows the structure of the existing alloy information stored in the memory | storage device in the alloy design assistance system which concerns on the Example of this invention. In the alloy design support system which concerns on the Example of this invention, it is a figure which shows an example of the difference information of the existing alloy information. It is a figure which shows an example of the alloy design map shown with respect to a user in the alloy design assistance system which concerns on the Example of this invention. In the process shown in FIG. 2, it is a flowchart which shows the flow of the process for extracting the existing alloy used as the object from which difference information is acquired. It is a figure which shows the mode of the grouping performed by the process shown in FIG. It is a figure which shows notionally the method for judging the appropriateness of the group obtained by the grouping performed by the process shown in FIG. It is a figure which shows an example of the alloy design map shown with respect to a user in the alloy design assistance system which concerns on the Example of this invention. It is a 1st figure when the concept of a cluster is introduced into the alloy design map created with the alloy design support system which concerns on the Example of this invention. It is a 2nd figure when the concept of a cluster is introduced into the alloy design map created with the alloy design support system which concerns on the Example of this invention. 3 is a flowchart showing a flow of processing for creating an alloy design map in the processing shown in FIG. 2. FIG. 13 is a diagram of an alloy design map in which clusters created using Voronoi regions are introduced in the process shown in FIG. 12. It is a figure which shows the alloy network formed using the Voronoi area | region created in FIG. It is a figure explaining aggregation of the clusters based on the coincidence degree of the change of manufacturing conditions in the alloy design map in which the clusters of the alloy design support system according to the embodiment of the present invention are introduced. It is a figure which shows the dendrogram information which shows the process of aggregation of clusters in the alloy design map in which the cluster of the alloy design support system which concerns on the Example of this invention was introduced. 3 is a flowchart showing a flow of processing for displaying an alloy design map in the processing shown in FIG. 2. In the alloy design support system which concerns on the Example of this invention, it is a figure which shows the state at the time of changing the number of clusters on an alloy design support map by a user's request | requirement.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Alloy design support system 2 ... Storage device 3 ... Computer body 4 ... Input device 5 ... Display 21 ... Alloy information storage unit 31 ... Target existing alloy extraction unit 32 ... Difference information acquisition unit 33 ... Map creation unit 34 ... Clustering reception unit 35 ... Clustering processing unit 36 ... Voronoi processing unit 37 ...・ Alloy network forming unit 38...

Claims (15)

For each of a plurality of existing alloys, a storage unit that stores predetermined manufacturing information related to manufacturing conditions of the existing alloy and predetermined characteristic information related to characteristics of the existing alloy;
For a target existing alloy that is all or part of a plurality of existing alloys included in the storage unit, a difference information acquisition unit that acquires difference information regarding the difference between the predetermined manufacturing information of each of two existing alloys;
The target existing alloy is set on a map having an evaluation characteristic determined according to the predetermined characteristic information as an evaluation axis, and the predetermined manufacturing corresponding to the difference information acquired by the difference information acquisition unit with respect to the map A map creation unit that creates an alloy design map by giving information fluctuation information;
An alloy design support system comprising:   The alloy design support system according to claim 1, wherein the map creation unit displays the variation information with a predetermined symbol visually recognizable to a user. Based on the difference tendency of the predetermined manufacturing information of the two target existing alloys in the difference information, the target existing alloy is divided into a plurality of predetermined clusters on the alloy design map, and two in the plurality of predetermined clusters A clustering processing unit that acquires information on the difference in the predetermined manufacturing information between the two predetermined clusters as cluster difference information;
The map creation unit gives variation information of the predetermined manufacturing information corresponding to the cluster difference information to the alloy design map having the predetermined cluster generated by the clustering processing unit.
The alloy design support system according to claim 1 or 2, characterized in that: The clustering processing unit
In the map in which the target existing alloy is set by the map creating unit, a Voronoi processing unit that forms a predetermined Voronoi region to which each target existing alloy belongs as a generating point,
Among the predetermined Voronoi regions formed by the Voronoi treatment unit, an alloy network forming unit that forms an alloy network with predetermined Delaunay sides by connecting target existing alloys belonging to adjacent predetermined Voronoi regions,
Based on the alloy network formed by the alloy network forming unit, the predetermined Voronoi regions associated with each other are aggregated and expanded, thereby classifying the predetermined cluster,
Have
The map creation unit gives the variation information to the alloy design map having the predetermined cluster divided by the classification processing unit,
The alloy design support system according to claim 3, wherein: Regarding the number of the predetermined clusters generated by the clustering processing unit, a cluster number request receiving unit that receives a request from a user is further provided,
The clustering processing unit classifies the target existing alloy on the alloy design map into the predetermined cluster number received by the cluster number request receiving unit.
The alloy design support system according to claim 3 or 4, wherein An existing alloy containing information similar to the premise manufacturing information which is a premise of the user's alloy design is extracted as the target existing alloy from the plurality of existing alloys stored in the storage device. An existing alloy extraction part is further provided,
The difference information acquisition unit acquires the difference information for the target existing alloy extracted by the existing alloy extraction unit,
The alloy design support system according to any one of claims 1 to 5, wherein The existing alloy extraction unit determines suitability in the temporarily extracted target existing alloy based on the correlation between the predetermined manufacturing information and the predetermined characteristic information of the temporarily extracted target existing alloy.
The alloy design support system according to claim 6. The predetermined manufacturing information has information corresponding to a plurality of predetermined manufacturing items related to alloy manufacturing conditions,
The difference information acquisition unit acquires the difference information corresponding to each of the predetermined manufacturing items included in the predetermined manufacturing information,
The alloy design map according to any one of claims 1 to 7, wherein the map creation unit creates the alloy design map in correspondence with each of the predetermined manufacturing items included in the predetermined manufacturing information. Support system. The predetermined manufacturing item of the predetermined manufacturing information includes at least a composition amount of a composition metal composing the existing alloy and a manufacturing process for manufacturing the existing alloy.
The alloy design support system according to claim 8. The existing alloy is an alloy related to a wrought material,
The alloy design support system according to any one of claims 1 to 9, wherein A computer program for supporting a user's alloy design, which stores, for each of a plurality of existing alloys, predetermined manufacturing information regarding manufacturing conditions of the existing alloy and predetermined characteristic information regarding characteristics of the existing alloy In the computer with the device,
A difference information acquisition step for acquiring difference information regarding a difference between the predetermined manufacturing information of each of two existing alloys for a target existing alloy that is all or part of a plurality of existing alloys included in the storage device;
The target existing alloy is set on a map having an evaluation characteristic determined according to the predetermined characteristic information as an evaluation axis, and the predetermined manufacturing corresponding to the difference information acquired by the difference information acquisition unit with respect to the map A map creation step for creating an alloy design map by giving information fluctuation information,
An alloy design support program characterized in that In the computer,
Based on the difference tendency of the predetermined manufacturing information of the two target existing alloys in the difference information, the target existing alloy is divided into a plurality of predetermined clusters on the alloy design map, and two in the plurality of predetermined clusters An alloy design support program for further executing a clustering process step of acquiring information on a difference of the predetermined manufacturing information between the two predetermined clusters as cluster difference information,
In the map creating step, variation information of the predetermined manufacturing information corresponding to the cluster difference information is given to the alloy design map having the predetermined cluster generated in the clustering processing step.
The alloy design support program according to claim 11, wherein: In the clustering processing step, the computer
In the map in which the target existing alloy is set in the map creating step, a Voronoi processing step of forming a predetermined Voronoi region to which each target existing alloy belongs as a generating point;
Of the predetermined Voronoi regions formed in the Voronoi treatment step, connecting the target existing alloy belonging to the adjacent predetermined Voronoi region, forming an alloy network by a predetermined Delaunay side, and
Based on the alloy network formed in the alloy network formation step, an alloy design support for executing a partitioning processing step of partitioning into the predetermined cluster by aggregating and expanding the predetermined Voronoi regions associated with each other A program for
In the map creation step, the variation information is given to the alloy design map having the predetermined cluster divided in the classification processing step.
The alloy design support program according to claim 12, wherein: In the computer,
An existing alloy containing information similar to the premise manufacturing information which is a premise of the user's alloy design is extracted as the target existing alloy from the plurality of existing alloys stored in the storage device. An alloy design support program for further executing the existing alloy extraction step,
In the difference information acquisition step, the difference information is acquired for the target existing alloy extracted in the existing alloy extraction step.
The alloy design support program according to any one of claims 11 to 13. An alloy design support method for supporting a user's alloy design,
For each of a plurality of existing alloys, all or part of the plurality of existing alloys included in a storage device that stores predetermined manufacturing information related to manufacturing conditions of the existing alloy and predetermined characteristic information related to characteristics of the existing alloy For the target existing alloy, a difference information acquisition step for acquiring difference information regarding the difference between the predetermined manufacturing information of each of the two existing alloys;
The target existing alloy is set on a map having an evaluation characteristic determined according to the predetermined characteristic information as an evaluation axis, and the predetermined manufacturing corresponding to the difference information acquired by the difference information acquisition unit with respect to the map A map creation step for creating an alloy design map by giving information fluctuation information,
An alloy design support method, comprising:

Images