JPH0452223A - Device for deciding whether production of iron and steel products having required toughness is possible or not - Google Patents

Abstract

PURPOSE:To decide whether the production of steels having required toughness is possible or not by previously preparing the plural membership functions in correspondence to the values of the required specifications concerning the toughness characteristics and the fluctuation values thereof in the production thereof, selecting the membership function suitable for the conditions of this time and using the functions for the decision. CONSTITUTION:An input means which inputs the information on the fraction transition temp. and absorption energy relating to the required specifications of the object to be decided; a condition section including >=1 deciding elements for contrasting the required specifications and the information on the actual results; a deciding rule holding means: a membership function holding means which includes the deciding element membership functions indicating the degrees relating to the results of the comparison of the deciding elements and the consequence part membership functions indicating the degrees relating to the consequence parts and holds at least plural sets of the above-mentioned deciding element membership functions are provided. The values indicating the degrees of the condition parts are determined by contrasting the inputted information on the required specifications and the information on the actual results by utilizing the selected deciding element membership functions and the consequence part membership functions are corrected according to the resulted values. Whether the production is possible or not is decided in accordance with the corrected consequence part membership function.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to determining whether or not steel products can be manufactured, and particularly relates to determining whether or not toughness properties can be manufactured.

[Prior Art] Steel products have been the most widely used materials in the past, and various necessary functional conditions are presented by users to manufacturers in the form of required specifications depending on the form of use. Manufacturers must determine whether or not the product can be manufactured for such required specifications, and provide the user with the results of the determination that it is possible to manufacture, cannot be answered without conducting an accuracy test, or cannot be manufactured. .

In addition, among this type of manufacturing feasibility determination, the manufacturing feasibility determination regarding toughness characteristics is required due to the increasing sophistication of required toughness characteristics for steel products due to the increase in pipeline construction in cold regions and large ship construction. The reality is that its importance is increasing.

As devices that support this type of manufacturing feasibility determination, there have traditionally been databases and search systems that accumulate data on past manufacturing results or manufacturing feasibility review results.
It is used. However, this system only supports the acquisition of reference data, and is not of much help unless there is a past record of specifications that are exactly the same as the required specifications to be determined. It is rare that the required specifications completely match past results, and usually they do not, so in such cases, experts use the above system to extract past performance data similar to the required specifications. The reality is that required specifications are compared with actual results, and the technical and experiential knowledge of experts is used to determine whether or not a product with toughness characteristics can be manufactured.

[Problems to be Solved by the Invention] Such conventional determination methods always require highly experienced experts, take a long time to perform the determination, and may also cause variations in determination results due to individual differences.

Therefore, it is an object of the present invention to automate the manufacturing feasibility determination of toughness characteristics as described above and to improve the reliability of one determination.

[Means for Solving Problem 11II] In order to solve the above problem, the present invention provides information on at least temperature and absorbed energy regarding the required specifications to be determined, as well as past performance of specifications that match or are similar to the required specifications. an input means for inputting information on at least fracture surface transition temperature and absorbed energy; a condition section including one or more determination elements for comparing information on required specifications inputted by the input means and information on actual performance; and the condition section Judgment rule holding means holding a judgment rule comprising a consequent part indicating a relationship between the result of the judgment and the manufacturability judgment result; a judgment element membership function indicating a degree related to the comparison result of the judgment element, and a degree related to the consequent part. and a member that holds at least a plurality of sets of the determination element membership functions corresponding to differences in at least one of the numerical value of the requirement specification and the manufacturing variation value thereof. and selecting a set of determination element membership functions on the membership function holding means according to the content of at least one of the numerical value of the required specification and the manufacturing variation value thereof; Using the judgment element membership function obtained, compare the input requirement specification information with the actual information to find a value indicating the degree of the condition part, and then set the consequence part membership function according to the obtained value. Manufacturability determining means is provided for determining whether manufacturing is possible based on the modified consequent membership function.

[For example, the determination of whether manufacturing is possible for the action toughness property is based on the following criteria: ``If the required temperature is slightly higher than the actual fracture surface transition temperature, and the required average absorbed energy is smaller than the actual absorbed energy, it is probably possible to manufacture it.'' This is done based on judgment rules using linguistic information with ambiguous boundaries. Moreover, the ambiguous boundary itself changes according to the contents of the required specifications (high or low required temperature, etc.) and manufacturing variations in actual toughness characteristics. Therefore, in the present invention, the idea of fuzzy inference is applied to configure a toughness characteristic manufacturing feasibility determination device. In addition, in order to deal with changes in the boundaries, we have set up multiple sets of membership functions that indicate the degree of ambiguity, and select the membership function that is most suitable for the specifications being considered, depending on the contents of the required specifications and manufacturing variations in actual toughness characteristics. A function is selected and a determination is made based on the selected membership function.

Thereby, it is possible to automate the determination of whether or not the product can be manufactured based on toughness characteristics, and the reliability of the determination can also be improved.

Other objects and features of the present invention will become apparent from the following description of embodiments with reference to the drawings.

[Example] FIG. 1 shows an example of an apparatus for implementing the present invention.

Referring to FIG. 1, this device includes a computer main body 100, an input device (keyboard, etc.) 101 connected to it, and an output device! (printer, etc.) 102. On the storage device of the computer main body 100, a user interface 103. Inference engine 104゜Knowledge base 105. A membership function database 106 and a manufacturing performance database 107 are arranged as software.

Input device 101, output device 102, and inference engine 10
4, data is exchanged via the user interface 103.

Information on required specifications to be considered is input from the input device 101 and passed to the inference engine 104 via the user interface 103. The inference engine 104 executes inference based on various rules registered in advance on the knowledge base 105 and outputs the final inference result, that is, the manufacturing feasibility determination result, to the output device 102 via the user interface 103. do.

FIG. 2 shows an example of data on the manufacturing record database 107, and FIG. 3 shows an example of required specifications input from the input device 101.

The manufacturing record database 107 organizes and accumulates data on past manufacturing records, and as shown in Figure 2, it includes product names, mold sizes, standard names, etc. that define product specifications, and impact test data for the products. Contains. The impact test data also includes fracture surface transition temperature, average absorbed energy, and minimum absorbed energy, which define the toughness characteristics of the product. Furthermore, in this embodiment, the fracture surface transition temperature, average absorbed energy, and minimum absorbed energy each include their values and manufacturing variations. For example, in the example shown in Figure 2, the fracture surface transition temperature is -54
℃, the variation value is 10℃, the average absorbed energy is 15Kgf-m, and the variation is 5Kgf1.

All of these are performance data obtained by testing manufactured products, and are classified and registered by standard name.

On the other hand, the required specifications include, for example, the product name, mold size, standard name, impact direction (L direction, C direction, etc.) as shown in Figure 3.
, shock temperature, average absorbed energy.

and the minimum absorbed energy. However, only one of the average absorbed energy and the minimum absorbed energy may be specified. In addition, impact direction, impact temperature,
Regarding the average absorbed energy and minimum absorbed energy,
In some cases, the characteristics are specified by standards, but for characteristics that are not specified in the required specifications, the information specified by the standard name is obtained from a specified standards database (not shown) and used as the required specifications. adopt.

When determining whether manufacturing is possible, after inputting required specification data from the input device 101, the inference engine 104 first obtains performance data of specifications similar to the required specifications from the manufacturing performance database 107. Specifically, the manufacturing performance database 107 is searched to find performance data whose mold size and standard name are the same as or most similar to the required specifications.

There are three types of performance information used in actual manufacturing feasibility determination: fracture surface transition temperature, average absorbed energy, and minimum absorbed energy. However, if the average absorbed energy is not specified in the required specifications, the average absorbed energy of the actual data is not used for the judgment, and if the minimum absorbed energy is not specified in the required specifications, the minimum absorbed energy of the actual data is not used for the judgment. do not use. That is, at least one of the average absorbed energy and the maximum tJs absorbed energy and the fracture surface transition temperature are used for the determination.

Actually, the following judgment rules are written on the knowledge base 105.

(1) If the required temperature is higher than the actual fracture surface transition temperature, and the required average absorbed energy is smaller than the actual average absorbed energy, it is possible to manufacture.

(2) If, The required temperature is slightly higher than the phantom eruption surface transition temperature.

and.

If the required average absorbed energy is lower than the actual average absorbed energy (Obama), then it is probably possible to manufacture.

(3) If, The required temperature is actually [! smaller than the transfer temperature.

And, if the required average absorbed energy is slightly larger than the actual average absorbed energy (1), it cannot be manufactured.

As can be seen from the above example, the determination rule in this embodiment includes ambiguous elements. Therefore, a large number of membership functions indicating the degree of these elements are prepared on the membership function database 106. In this example, the membership functions are roughly divided into two types: one-determination element membership functions and consequent part membership functions. These details will be explained later.

The procedure of determination processing based on various rules on the knowledge base 105 is shown in FIG. 4. The contents of each processing step will be explained in sequence with reference to FIG.

401: First, the amount of deviation of the content of the characteristic item of the required specification with respect to the content of the characteristic item of the performance data is determined.

That is, for each of the temperature, average absorbed energy, and maximum/J% absorbed energy, the deviation amount is determined by calculating the deviation amount = required specification value - actual value. Note that the process for the average absorbed energy or minimum absorbed energy is omitted if it does not exist in the required specifications.

402: Apply each of the deviation amounts obtained in 401 to the corresponding determination element membership function to obtain a value indicating the magnitude (large, approximately equal, small, etc.) of each determination element. For example, when using the determination element membership function -1 shown in Figure 6a for temperature, if the deviation amount of the required temperature from the actual fracture surface transition temperature is 14°C, it is assumed that the required temperature is equal to the actual fracture surface transition temperature. On the other hand, the degree to which it can be said to be larger is 0.4, the degree to which it can be said to be slightly larger is 0.6, and the other degrees (smaller, slightly smaller, approximately equal) are all 0.

The determination element membership function used here is selected from a plurality of determination element membership functions. This selection will be explained in detail later.

403: Here, the value of the condition part of the determination rule is found. When the conditional part is composed of a single judgment element, the degree value of that judgment element (the value obtained in 402) itself is adopted as the value of the conditional part, and the conditional part is composed of two or more judgment elements. In this case, the minimum value of the degree values of those determination elements is adopted as the value of the condition part.

For example, the degree to which it can be said that the required temperature is greater than the actual fracture surface transition temperature shown in (1) of the above judgment rule is 0.4,
If the degree to which it can be said that "the requested average absorbed energy is smaller than the actual average absorbed energy" is 1.0, the value of the condition part of this determination rule is 0.4, which is the smaller of the two.

404: First, a consequent membership function corresponding to the consequent of each determination rule is obtained from the membership function database 106, and is used as a prototype consequent membership function. Next, this prototype consequent membership function is reduced and modified in the vertical direction according to the value of its conditional part, and is used as the consequent membership function of the judgment rule.

For example, in (1) of Fig. 6b, the prototype consequent membership function corresponding to the decision rule "manufacturable" is shown by a dotted line, and the condition part of the decision rule "if... then" is shown as a dotted line. When the value is 0.4, the resultant membership function is modified as shown by the solid line.

405, 406: Here, all the consequent membership functions of each determination rule are superimposed, and the center of gravity position with respect to the horizontal axis of the combined result is determined.

For example, in (3) of FIG. 6b, the consequent membership function of "manufacturable" and the consequential part membership function of "probably possible" are combined, and their centroid positions are determined. It should be noted that the method for determining the center of gravity position is well known, so a description thereof will be omitted.

407: Here, the prototype consequent membership function having the maximum value at the center of gravity position determined in 406 is determined, and the determination is made based on this. For example, in (3) of Figure 6a, "manufacturable", "probably possible", "cannot say", "
Among the five functions of "probably not possible" and "not possible", the value of the "manufacturable" function is large at the center of gravity W (Δ in the figure), so "manufacturable" is adopted as the determination result.

FIG. 7 shows the shape of a set of determination element membership functions (an example of required temperature-actual fracture surface transition temperature) and the contents of a table in the memory corresponding to the shape.

As can be seen from FIG. 7, in this example, it is small.

Slightly smaller, approximately equal, slightly larger, and larger 5
constitute a set of membership functions.

The actual value of the function can be calculated based on the contents of the table. For example, deviation amount X is deviation amount A
and B, "a slightly smaller value of j is slightly smaller i = t x (x-A/ (B A)
).

In the example shown in Figure 7, the membership function table includes the manufacturing variation value (d) of the required temperature (T) and the actual fracture surface transition temperature.
T) information is included. In this embodiment, a plurality of membership functions are provided, and each membership function corresponds to, for example, a classification of the required temperature (T) and the manufacturing variation value (dT) of the actual fracture surface transition temperature. .

In reality, a total of 15 sets (5 x 3 types) of membership functions related to the required temperature-actual fracture surface transition temperature are prepared, with 5 types for the required temperature and 3 types for the variation. . In addition, as a membership function regarding the requested average absorbed energy and the actual average absorbed energy, we have a total of 9 independent functions for 3 types of request values and 3 types of dispersion.
Sets (3 x 3 types) are available. Furthermore, nine sets of membership functions regarding required minimum absorbed energy and actual minimum absorbed energy are prepared. As for the consequent membership functions, only one set shown in FIG. 8 is prepared.

Therefore, when using a one-judgment element membership function, an appropriate membership function is selected from among a large number of functions depending on the conditions at that time, that is, the classification of the required value and the variation value. When using the membership function of required temperature-actual fracture surface transition temperature, the process shown in FIG. 5 is actually executed. The contents of each step will be explained below.

501: Search the database 106 for membership functions assigned to conditions corresponding to the input information at that time, that is, the required temperature and the actual fracture surface transition temperature, as search keys. In other words, T and dT shown in the table ((2) in FIG. 7) showing each of the many membership functions registered on the database 106 are checked, and a table in which they match the input information is searched. do.

502: As a result of the search in 501, if a table matching the search key exists, the process proceeds to 503; otherwise, the process proceeds to 504.

503: Obtain the contents of a table indicating one membership function found as a result of the search, and restore the membership function based on the data. In other words, as shown in FIG. 7, each table shows the deviation amount A,
The values (1 or 0) at positions B, C, D, and 0 are shown, and the slope of the function is constant within each range of A-B, B-0, O-C, and C-D. Therefore, based on these conditions, the value of the function at any deviation amount can be determined by calculation.

504: If there is no table with conditions matching the search key, execute this process. Here, the first and second
Select the second two sets of tables with similar membership functions.

505: Based on the two tables selected in 504, a weighted average is calculated according to the distance between the conditions of each table and the value of the search key.

506: Based on the value obtained in 505, the membership function is restored in the same manner as in 503.

In other words, this embodiment has tables showing 15 sets of membership functions regarding the required temperature-actual fracture surface transition temperature, and the condition (T) of each table is provided.

Even for conditions that do not match dT), an accurate membership function suitable for the condition can be obtained by calculation.

Please refer to FIG. 9, which shows an example of the specific contents of a table showing membership functions on the database 106.

In the above example, multiple membership functions were prepared according to the differences in both the required specifications and the actual dispersion values, but it is also possible to simply switch the membership functions in response to differences in either one. , a sufficient effect can be expected in accurately determining whether manufacturing is possible.

[Effects of the Invention] As described above, according to the present invention, ambiguous determination rules are processed by applying the concept of fuzzy inference, so automation of manufacturing feasibility determination regarding toughness properties is realized. Moreover, multiple membership functions are prepared in advance in response to the values of the required specifications and their manufacturing variations, and the membership function suitable for the conditions at that time is selected and used for determination. Highly accurate judgment results can be obtained under these conditions.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an apparatus according to an embodiment. FIGS. 2 and 3 are memory maps showing the content of data in the manufacturing record database and the content of required specification data, respectively. FIG. 4 is a flowchart showing the process for determining whether or not the device shown in FIG. 1 can be manufactured, and FIG. 5 is a flowchart showing the process related to membership function selection. Figures 6a and 6b are graphs showing examples of membership functions. 7, 8, and 9 are maps showing the contents of tables corresponding to membership functions on the database. 100: Computer body 101: Input device 102: Output device 103: User interface 104: Inference engine (manufacturability determination means) 1o5: Knowledge base (determination rule holding means) 106: Membership function database (membership function holding means) 107 ：Manufacturing results database diagram (1) Menu layout (2) Membership function storage table Judgment element Membership function and its storage example in the computer (required temperature vs. actual results) In the case of fracture surface transition temperature) Figure 6a Machi Memorial Mechanism 19mi (2) 'l'7
Is it true? fr man to Y men no g ° - powder file Sekishiki voice 6b figure (1) Hanbunryiki kana ゛ j no 2 - Ibe men ni zog counterfeit) comparison 9zu (2) publishing exchange no tun ν'' 1 no Tsuki Risa rqme”/l z'
-Sisig 5! Payment J-2 Procedural Amendment (Voluntary) August 6, 1990 6 Contents of the Amendment (1) Correct the "actual absorbed energy" and "actual average absorbed energy" in line 5 of page 6 of the specification. of

Claims (1)

[Scope of Claims] Input means for inputting at least information on temperature and absorbed energy regarding the required specifications to be determined, and at least information on fracture surface transition temperature and absorbed energy regarding past performance of specifications that match or are similar to the required specifications. ; a condition section including one or more determination elements for comparing information on required specifications inputted by the input means and information on actual performance;
Judgment rule holding means that holds a judgment rule including a consequence part that shows the relationship between the result of the condition part and the manufacturability judgment result; a judgment element membership function that shows the degree regarding the comparison result of the judgment element; and a consequential part membership function indicating the extent to which the part is concerned, and retains at least a plurality of sets of the judgment element membership functions corresponding to differences in at least one of the numerical value of the required specification and the manufacturing variation value thereof. membership function holding means; and selecting a set of determination element membership functions on the membership function holding means according to the contents of at least one of the numerical value of the required specification and the manufacturing variation value thereof; , Using the selected judgment element membership function, compare the input requirement specification information and actual information to find a value indicating the degree of the condition part, and select the result part member according to the obtained value. 1. An apparatus for determining whether or not the toughness characteristics of a steel product can be manufactured.