JPS61131814A - Optimum sawing method of shape steel - Google Patents

Abstract

PURPOSE:To keep a sawing yield rate ever so high throughout a roll chance as a whole, by determining required sawing numbers at every ordering size, while leading in plural parameters such as required sawing residual numbers and the like, and determining an assortment so as to minimize a cutoff quantity. CONSTITUTION:Actual extension length of the product rolled by a finish rolling mill 1 is measured by a length gauge 2 and it is inputted into a calculation system 3. Since there are partially some varieties a rolling mill 14 comes to the final stand, as for these varieties, the actual extension length is inputted into the calculation system 3 from a length gauge 4. On the other hand, a charging order file for blooms is inputted into the calculation system 3 from a heating furnace charging bed 5 while ordering dimensional length at every roll chance and the number are inputted into the calculation system 3 from an input terminal 6. Seven kinds of optimum sawing plans are made out from these information data in advance, from which a kind of the most suitable optimum sawing plan for the actual extension length is selected, and a cutting instruction is fed to hot sawing cutters 7 and 8 and a sizing machine 9. Simultaneously, operation guides and various files are outputted to a terminal machine of a sawing operation chamber 10.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application] The present invention relates to an optimal sawing method in a section steel factory.

[Conventional technology]

Saw cutting work at a shape steel factory starts with azurol material that has been rolled to about 60 to 120 m, and then cuts from 7 m to 50 m depending on the ordered length.
(Customized lengths of less than 7m are cut to a length of 7m or more due to restrictions on transportation equipment, etc., and are cut to a length that is an integral multiple of the ordered length.) However, the post-process It is possible to create a sawing plan for each billet by combining various custom lengths, taking into account the ease of work at the steel plate, the equipment capacity of the cooling bed, and the sawing pitch, while also minimizing the amount of cutoff. This is a very complex task. Moreover, after the actual rolling elongation length of the steel billet is completed and the actual rolling elongation length is known, there is only a few seconds to spare before sawing, and it is difficult to create a sawing plan manually.

Therefore, in recent years, attempts have been made to automate the above-mentioned work and control it using a computer. In other words, it is a method of creating an optimal sawing plan within a computer.
No. 84, ``Optimal Saw Cutting Method for Rolled Materials,'' etc. are known. The basic idea of this prior art is to estimate the expected cut-off amount and the combination of order lengths based on the expected rolling elongation length.
A normal distribution with the average of the unit weight of the product is considered. However, the variation in the actual rolling elongation length includes not only the variation in the unit weight of the steel billet, which is thought to follow a normal distribution, but also the amount of burning due to the length of time the billet is in the furnace, and the amount of surface scratch removal. There are many factors that cause variations that do not fit into the normal distribution, such as the size of the cutting allowance by the hot solvent equipment used for this purpose, changes in the rolling reduction of each rolling mill, and variations in the product unit weight due to roll wear. Furthermore, steel billets with several types of unit weights may be used in the same roll chance, so it is difficult to say that the actual steel billet elongation length follows a normal distribution.

This suggests that the estimate of the amount to be cut may be significantly disrupted.

In addition, when considering the combination of a single steel billet, the order length candidates are grouped into three types, but this means that the amount of cut-off is increased because it cannot fully follow the variation in the elongation of the steel billet. On the other hand, as stated in the same bulletin, if the custom length candidates are expanded to include all custom lengths, there is no possibility that the amount of cut-off for each billet will become large, but the Since the order size is generated after the order quantity has been used up, the number of types of order sizes decreases, and the degree of freedom in determining combinations decreases, which may reduce the overall yield.

[Problem that the invention seeks to solve]

The present invention was made in view of this situation, and
In order to ensure the final number of ordered lengths for each roll/chance, the number of saws required for each ordered length is determined in consideration of the order-grade yield, and the number of saws required for each ordered length is determined for each billet from among all ordered lengths. In order to leave flexibility in the arrangement just before the end of the roll chance, we introduced multiple parameters including the number of remaining saws to determine the arrangement to minimize the amount of cutting, and in addition, in the previous process, The purpose of this invention is to provide an optimal sawing method for sectioned steel that maximizes the capabilities of post-processing while ultimately producing products that satisfy the required custom size and quantity.

[Means for solving problems]

The optimal sawing method for section steel according to the present invention is achieved by: (1) when producing the ordered number of sections having a single length or different combinations of lengths given for each roll chance of the section steel; Determine the required number of saws for each ordered length for each roll and chance, taking into account the sawing yield for each length.
At some point during the opportunity, determine the degree of belonging to the FUZZY set that says, ``It is an undesirable order to cut now'' for each order, and at the same time show the ease of work for each order. Determine the linear combination parameters and calculate the expected rolling elongation length of the saw cut at equal intervals before and after the cut.
Create all possible combinations of order length and crop length for each of the seven expected rolling elongation lengths, e) Subtotal of the product of the linear combination parameter and membership function for each order length, and the crop length penalty. The nonconformity is determined by adding the penalty considering the whole to the total, and the combination of order lengths that minimizes this is determined for each of the seven expected rolling elongation lengths. When considering the order of cutting custom lengths for combinations, it is important to note that this is an unfavorable order for cutting. "r"
FUZZY set" is determined for each arrangement of order scales, and (g) considers all ways of ordering order scales for each combination of order scales, and determines the arrangement that minimizes the sum of the membership function of (This is determined as the optimal sawing plan for the applicable steel slab) When the applicable steel slab is actually rolled,
Measure the actual rolling elongation length, compare it with 7 types of optimal sawing plans, and saw according to the sawing plan with the least amount of cutting.) When all the number of saws of all ordered lengths have been cut This is characterized by ending the roll chance, and if not, returning to step (a) and repeating the steps up to (e).

〔Example〕

The present invention will be specifically described below based on embodiments shown in the drawings.

FIG. 1 shows the configuration of a section steel optimal sawing system in which the optimal sawing method of the present invention is implemented.

The actual rolling elongation length of the product that has been rolled in the finish rolling mill 1 is measured using a length meter 2 and input into the computer system 3. Some V
2 E + Since there is a product type in which the rolling mill 14 is the final stand, for this product type, the actual rolling elongation length from the total length 4 is input into the computer system 3. On the other hand, the billet charging order file from the heating furnace charging bed 5 and the ordered length and number of pieces ordered for each roll chance are input to the needle counter system 3 from the input terminal 6. Create 7 types of optimal sawing plans in advance from this information, select the optimal sawing plan that best matches the actual rolling elongation length, and apply it to the hot sawing machine 7.8 and sizing machine 9. A cutting instruction is sent.At the same time, an operating guide for the saw operator and various forms are output to the terminal in the saw operator's room 10, and data input is accepted at the same time.

In the figure, 11 is the heating furnace, and 12 is the breakdown (BD).
13 is a v1 rolling mill, and 15 is a V:+E2 rolling mill.

Therefore, the optimal sawing system can be roughly divided into three steps: Calculating the required sawing amount for each ordered length to be sawed with a roll chance from the ordered length and the number of ordered pieces. Step (S)
TEP) 1. Deciding on the order length for each piece of steel...
. . . Step 2: Determination of cutting order of custom lengths in each combination . . . Step 3. The details of each step will be described below.

The purpose of step 1 is to calculate the number of required saws for each length from the number of orders for each length during the roll chance. There are two basic points to keep in mind when calculating the required number of saws. The first point is regarding the total cutting amount (total length or total weight). Figure 2 shows the relationship between the total amount of sawed products and the ordered products. It's called danyoru. That is, the planned value is set based on past actual results.

Therefore, when calculating the total cutting amount from the total order amount, it is possible to obtain the optimal value by using this order-grade yield.

Next, the second issue is due to the length structure to be ordered. A particular order size configuration (combination of length and order quantity) is determined by the standards and specifications added at the time of order. In other words, a single order length configuration table is created by grouping together items that are made of the same material and can be rolled under the same rolling conditions.
This is the ``ris'' when performing the following calculations.This ((
The data includes several types of order lengths and the number of each order. If we apply the relationship of formula (L) to each of these combinations, the relationship of formula (1) will be satisfied for the entire variety of the combination, so (1
) equation is applied to each (kuri).By the way, the relationship of equation (11) can be expressed in terms of the number of units, and if equation (2) is applied to all order quantity types P, the entire kuri (1 ) will be satisfied.

However, the relationship between long and short lengths is not equal, and the following properties must be taken into account.

(a) Order quantity long length: This length can be rounded down and non-order grade (cutting grade product 1 ordered quantity) can be applied to even shorter length.

However, if there are many defective products resulting in devalued or stratified products, we may not be able to fulfill orders. Therefore, it is safer to saw more.

Mountain) Intermediate shaku ordered: It is expected that the shaku itself will be devalued and the out-of-order occurrences will be allocated to the short shaku.

(C1 Order shortest length: It is expected that a considerable number of items will be filled from the longest length and intermediate length. Therefore, it is often possible to fill orders with fewer saw devices than the number calculated by formula (1).

In order to satisfy the above properties and obtain the number of pieces ordered for each length, set the saw cutting yield as a different order-grade yield for each length and calculate the number of pieces required to be sawed. It will be a good thing. That is, (
By changing B' (order-grade yield) in equation 2) for each order quantity (formula (3)), it is possible to set the number of pieces to be cut, taking into account the above-mentioned three properties.

The method for setting Bi will be described below.

Consider a diagrammatic model like the one shown in Figure 3 for each link.

Here, the horizontal axis represents length, and point A represents the total length of the ordered quantity during cutting.The vertical axis represents the reciprocal of the cutting yield, and
The points indicate the reciprocal of the sawing yield.

By the way, the cutting yield of the longest length is assumed to be a value determined based on past results, and the vertical axis indicates the position of the reciprocal of the 0 point. Here, the midpoint E of B-D in FIG. 3 is taken. Also, on the horizontal axis, the value of (order quantity length) x (order quantity), that is, the total length for each order quantity, is assigned in descending order of order quantity length. In other words, -R is the shortest length, R-3 is the second shortest length, and FA is the longest length. Next, write a straight line l passing through EIH with H being the midpoint of F-A and the intersection of C and -G.
e at the midpoint of the total length for each order quantity expressed on the horizontal axis
Find the value of and use it as the reciprocal of the sawing yield for each applicable order quantity (
Example: point ■, point J).

If you do this, for example, the shaded area A-F in Figure 3
-The area Si of P-G is Si = xiXMi X-... (4) (xl
: Since it is the length of the ordered quantity, it is the total length of sawing of the length xl.

Also, since H is the midpoint of P-G, A-F-P-G is equal to trapezoid A-F-Q-N. This relationship holds true for all order quantities: S=ΣSi ・・・・・・・・・
...(5) The total sawing amount S shown by i=1 is expressed as -M-N in the total sum A- of the trapezoid in Fig. 3, but this is because E is the midpoint of B-D. A- is equal to -B-D because of something.

Therefore, S becomes the optimal total cutting length shown in equation 11.

As proven above, two points (relationship between equations (1) and (3)) are shown as the basis of calculation by adopting each saw cutting yield (Bi) determined by the straight line l in Fig. 3. The following items will be satisfied. In an actual system, the required number of saws to be cut is required for each order quantity, so the required number of saws to be cut is determined using equation (3), and step 1 is completed. Step 2 is to ensure that the number of sawn pieces calculated in Step 1 is not too much or too small, and that the number of steel billets used is as small as possible (i.e., allocating the order amount to each billet in order to increase the sawing yield; that is, ordering Arrange the lengths.In other words, you are looking for a small one.However, if the number of types of ordered lengths P is large (nl indicating the same α).

There are many combinations of X. From among these, it is necessary to consider an arrangement that leaves a degree of freedom in arrangement during the end gap of the roll chance. Therefore, each order length belongs to the FUZZY (fuzzy) set such that "it is an undesirable order length to cut at this moment", which is a membership degree function y1.
is determined for each order length. yl The unfitness degree of 2 is defined as . Minimize this Z nl + Xl +
Let the combination of α be the optimal combination of expected billet elongation lengths.

For example, the order length xI, the number of remaining saws N i required, the degree of belonging function y1. The linear combination parameters q1 are listed in Table 1.

Table 1 (*To set 25M on a moving saw) Now, without considering cropping, if the elongation length of the steel billet is 60M, α=
Even if we consider only the combination of 0s, there are six types as shown in Table 2.

Table 2 In this example, when calculating the degree of unfitness Z for each combination, Table 2 is obtained, and in this example, the optimal combination is 25.25.
It becomes 10. However, as sawing is carried out with this arrangement, the amount of saw remaining required for the 25M scale and 10M scale decreases, and the optimal combination changes to 15°15.15 and 1°5.

As can be seen in this example, due to the nature of yl, which is the reciprocal of the number of remaining saws, if the size of α is the same, priority will be given to cutting from the ordered length with the largest number of remaining saws, so as not to reduce the number of annotation types. I have to.

Furthermore, from the relationship between the truncation amount α and qα, if qα is set to about 1, combinations with a large truncation amount α will hardly be selected, and the sawing yield can be increased. On the contrary, q
If α is made smaller, ordered lengths with a large number of remaining saws will be cut preferentially even if the yield is sacrificed to some extent.

By adopting the concept of FUZZY sets in this way, it is possible to increase the sawing yield for each billet while at the same time leaving a degree of freedom in making arrangements just before the end of the roll chance. However, it is also possible to freely change the arrangement method depending on the method of selecting each parameter.

In this way, the optimal combination is determined for the predicted elongation length of the steel billet, but in order to deal with variations in the actual elongation length of the steel billet, L-3d, L-2d, L- d
, L, L+d, L+2cl.

Seven types of combinations with L+3d and d pitches are created for one steel piece. Normally, d is set to 1 m, taking into consideration the variation in length and rolling elongation to be ordered.

In addition, the expected elongation length L of the billet is calculated by (billite weight) ÷ (unit weight per meter of product) at the start of rolling, but after the start of rolling, L is corrected according to the trend of the actual billet elongation length. , follows the variation.

Step 3 determines the optimal cutting order for the seven types of combinations created in step 2.

In determining the cutting order, it is necessary to consider the characteristics of the two saws and sizing machines shown in FIG. 1 and the workability in the cooling bed, which is a subsequent process. Let Aj be a factor related to the cutting order of the custom length. Aj says, ``The cutting order is such that the difference in the length of each ordered length is an integer multiple of the constant length e, that is, O, e, 2e, 3e.
, . be.

For each of these Aj, the degree of membership function fK belonging to the FUZZY set "is an unfavorable order for cutting"
Determine. For example, the above-mentioned "cutting order is such that the difference in the length of each ordered length is an integer multiple of the constant length e, that is, Q, e, 2e,
fK is defined as shown in FIG. 4 for the factor AK, ``If it is 3e, . . ., it is easy to cut.'' Here, consider all the cutting orders of the ordered lengths for each predicted steel billet elongation length determined in step 2, and consider the degree of unsuitability G regarding the cutting order for each order.

G = Σ hj The most i1
! A cutting plan will be used.

As an example, let's set the arrangement as 3 shaku: 10.12.15.
For the sake of simplicity, assuming that the only factor is the aforementioned AK (e=3), G for each cutting order is determined as shown in Table 3.

Table 3 In other words, the conditions for the cut lfr order are the same regardless of the order l. Usually this factor AK is 6 to 7N.

For the seven optimal cutting plans obtained through Steps 1, 2, and 3 above, the actual elongation length of the actually rolled steel billet was measured and converted to cold length. When there is an instruction to take post-test or size samples, subtract the required length, and compare the effective sawing lengths by subtracting cropping, and select the combination with the least amount of cutting and sawing. . This sawing performance is reflected in the number of remaining saws required and used for the next optimal sawing plan.

〔Effect of the invention〕

In principle, the optimal sawing plan is performed in the manner described above, but the cutoff length α can also be set in advance and mixed to reduce the cutoff length α.

In addition, although this method is entirely automated by a computer system, the operator at the sawing site can freely change the linear combination parameter qi of the set, the penalty qα for the cut-off amount α, the penalty qa□1 for the entire set, etc. , the contents of the arrangement can be changed at any time. It is also possible for the operator at the cutting site to specify part or all of the expected elongation length of the steel billet, including the cutting order, and the system to make arrangements for the remaining length.

The present invention makes it possible to maintain the sawing yield at a high level throughout the entire roll process, and is an invention with extremely high industrial value.

[Brief explanation of the drawing]

Fig. 1 is an explanatory diagram showing the configuration of a sawing system to which the optimal sawing method of the present invention is applied, Fig. 2 is an explanatory diagram showing the relationship between the total amount of sawing and ordered products, and Fig. 3 A diagram showing the relationship between the total length and the reciprocal of the sawing yield for explanation, and FIG. 4 is an explanatory diagram showing the relationship between the membership degree function belonging to the FUZZY set and the difference in length of the ordered length. .

Claims (1)

[Claims] When producing the ordered number of pieces for each custom length having a single length or different combinations of lengths given for each roll chance of section steel, a) a saw for each length of the custom length; Determine the required number of saws for each order length for each roll chance by taking into account the cutoff yield; a) At a certain point during the roll chance, there is a FUZZY that says ``It is not desirable to cut the order length now.'' Determine the degree of membership function belonging to the set for each ordered length, c) At the same time, determine the linear combination parameter that indicates the ease of work for each ordered length, and d) calculate the expected rolling elongation length of the saw cut. Create all possible combinations of order lengths and crop lengths for each of the seven expected rolling elongation lengths, three at equal intervals before and after e) Subtotal of the product of the linear combination parameter and the degree of membership function for each order length , the sum of the crop length penalty and the overall penalty are considered as nonconformity, and the combination of order lengths that minimizes this is determined for each of the seven expected rolling elongation lengths. When considering the cutting order of custom lengths for combinations of different types of custom lengths, the degree of membership function that belongs to the "FUZZY set", which says "This is an unfavorable order for cutting," is determined for each order of custom lengths, and the key is determined. ) For each combination of ordered lengths, consider all ways of arranging the ordered lengths, find the arrangement that minimizes the sum of the membership functions of (f), and define this as the optimal sawing plan for the applicable steel billet. When the piece is rolled, measure the actual rolling elongation length, compare it with seven optimal sawing plans, and saw it according to the sawing plan with the least amount of cutting. An optimal sawing method for shaped steel, characterized in that the roll chance is terminated when all the number of cuts have been completed, and when not, the process returns to step (a) and repeats the steps up to step (e). However, sawing yield = order filled quantity / total sawing quantity