Multi-Mode Steel Bar Sleeve Optimization Method Applicable to Ultra-Large Scale

Abstract

The present invention discloses a multi-mode steel bar set material optimization method applicable to ultra-large scale, including: S1, scenario description; S2, estimation of the scale of steel bar set materials; S3, comparing scale and scale_threshold. If scale > scale_threshold, the greedy algorithm is used to optimize the cutting plan of each length of raw material in turn. After determining the cutting plan, calculate the estimated value scale of the scale of the remaining steel bars to be cut. If scale > scale_threshold, continue to use the greedy algorithm for iterative optimization until scale ≤ scale_threshold, and then use the global optimal algorithm to optimize the cutting plan of the remaining steel bars to be cut. In view of the problem that in the case of ultra-large scale set materials with too many single-time set material steel bar sizes or too many short materials, the calculation time of the general set material calculation method is too long or even memory overflow occurs and the computer cannot calculate, the present invention utilizes the characteristics of fast calculation speed and linear growth of calculation scale of the greedy algorithm to reduce the calculation scale of steel bar set materials, and at the same time combines the global optimization algorithm to basically achieve the optimal material yield while achieving efficient calculation, meeting the requirements of ultra-large scale steel bar set material optimization.

Description

Technical Field

[0001] The present invention relates to the technical field of steel bar processing. More specifically, the present invention relates to a multi-mode steel bar set optimization method applicable to ultra-large scale. Background Art

[0002] Nowadays, steel bar processing equipment is becoming more and more advanced, and the level of automation and intelligence is getting higher and higher. However, in actual production, the bar sets of steel bars usually rely on the experience of workers, and a large amount of steel bar waste is often caused during the blanking process. At present, some universities and steel bar processing equipment manufacturers are researching steel bar set optimization technologies. These technologies can achieve a relatively high material yield in small-scale blanking scenarios. As the blanking scale continues to increase, the amount of calculation will show a geometric growth, resulting in too long calculation time or even unable to calculate the result, or even if the result is calculated, the material yield will be lost. Summary of the Invention

[0003] Another object of the present invention is to provide a multi-mode steel bar set optimization method applicable to ultra-large scale to meet the optimization requirements of ultra-large scale steel bar sets.

[0004] In order to achieve these objects and other advantages of the present invention, there is provided a multi-mode steel bar set optimization method applicable to ultra-large scale, including:

[0005] S1. Scenario description

[0006] The raw material information is represented by an array a, a = [a1 a2 … a i …], where a i represents the length of the i-th raw material;

[0007] The information of the steel bars to be blanked is represented by an array b and an array c, b = [b1 b2 … b j …], c = [c1 c2 … c j …], where b j and c j represent the length and quantity of the j-th steel bar to be blanked respectively; the steel bar set scale threshold is scale_threshold;

[0008] S2. Estimation of steel bar set scale

[0009] Select the longest raw material a max in the raw material array a;

[0010] Calculate the estimated value scale of the set scale, and the calculation formula is as follows:

[0011] In the formula, len(b) is the length of the steel bar array b to be blanked, that is, the number of elements in b, and [] represents rounding down;

[0012] S3. Compare scale and scale_threshold. If scale > scale_threshold, use the greedy algorithm to optimize the cutting plan for each length of raw material in turn. After determining the cutting plan and fully cutting the materials, calculate the predicted value scale of the remaining steel bars to be cut again. If scale > scale_threshold, continue to use the greedy algorithm for iterative optimization until scale ≤ scale_threshold, and then use the global optimal algorithm to optimize the cutting plan for the remaining steel bars to be cut.

[0013] Preferably, the steps of using the greedy algorithm to optimize the cutting plan for each length of raw material in step S3 include the following steps:

[0014] A. For each length of raw material a i , establish an integer programming mathematical model with the highest single-pass yield rate as the goal. The integer programming mathematical model is as follows:

[0015] max f = b·x T / a i

[0016]

[0017] In the formula, x is the decision variable, x = [x1 x2 … x j …], x j is the cutting quantity of the jth length of steel bar in the array b;

[0018] B. Use the branch and bound method to obtain the optimal solution x, and calculate the highest single-pass yield rate f according to x max = b·x T / a i ;

[0019] C. Take the largest number of single-pass cut steel bars as the new constraint condition, and establish a new integer programming mathematical model. The new integer programming mathematical model is as follows:

[0020]

[0021]

[0022] D. Use the branch and bound method to obtain the optimal solution x, and calculate the highest single-pass yield rate f according to x max = b·x T / a i .

[0023] Preferably, in step S3, after optimizing the cutting plan of each length of raw material in turn using the greedy algorithm, the nesting combination corresponding to the optimal solution x with the largest number of single-cutting steel bars and the highest material utilization rate is used for sufficient cutting, and the remaining array c of the quantity of steel bars to be cut is updated.

[0024] Preferably, optimizing the cutting plan of the remaining steel bars to be cut using the global optimization algorithm includes the following steps:

[0025] a. Calculation of the combination of steel bars to be cut

[0026] Exhaustively list all cutting combinations corresponding to each length a of the raw material. Each cutting combination includes all possible cases that the raw material of this length can be cut. The remaining combinations of steel bars to be cut corresponding to each length of raw material are spliced to form the cutting combination T of all raw materials. i t t mn represents the quantity of the m-th steel bar to be cut in the n-th combination. Calculate the length of the raw material used for each combination, and represent it with the array y, y = [y1 y2 … y i …], y i represents the length of the raw material corresponding to the i-th combination.

[0027] b. Establishing and solving the global optimization mathematical model

[0028] Establish an integer programming model with the goal of minimizing the total length of the consumed raw material. Among them, the decision variable x is the usage times of each column combination in T, x = [x1 x2 … x i …], x i is the usage times of the i-th column cutting combination in T. The constraint condition is that the quantity of each length of steel bar to be cut corresponding to x is equal to the cutting requirement quantity in the updated c. The integer programming model is as follows:

[0029] min x·y T

[0030]

[0031] Use the branch and bound method to obtain the optimal solution x, and cut the remaining steel bars to be cut according to the cutting combination corresponding to x to complete the cutting of all steel bar nests.

[0032] The present invention has at least the following beneficial effects: Aiming at the problem that in the case of ultra-large-scale nesting where the number of single-nesting steel bar sizes is too large or the short materials are too many, the calculation time of the general nesting calculation method is too long or even the computer cannot calculate due to memory overflow, the present invention utilizes the characteristics of fast calculation speed and linear growth of the calculation scale of the greedy algorithm to reduce the calculation scale of the steel bar nesting, and at the same time combines the global optimization algorithm to basically achieve the optimal material utilization rate while achieving efficient calculation, meeting the requirements of optimizing ultra-large-scale steel bar nesting.

[0033] Other advantages, objects and features of the present invention will be partly reflected by the following description and partly understood by those skilled in the art through the research and practice of the present invention. Description of the Drawings

[0034] Figure 1 It is a flowchart of an optimization method for ultra-large-scale multi-mode steel bar sets applicable to the present invention. Detailed Description of the Invention

[0035] The following further describes the present invention in detail with reference to the drawings, so that those skilled in the art can implement it according to the description in the specification.

[0036] It should be noted that in the description of the present invention, the orientation or positional relationship indicated by the terms "transverse", "longitudinal", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention.

[0037] As Figure 1 shown, the present invention provides an optimization method for ultra-large-scale multi-mode steel bar sets, including:

[0038] S1. Scenario description

[0039] For the optimization cutting scenario of a certain specification of steel bars, the information includes the length of the raw material of this specification, the length and corresponding quantity of the steel bars to be cut. Generally, the raw material supply is sufficient and can be regarded as unlimited supply.

[0040] The raw material information is represented by an array a, a = [a1 a2 … a i …], where a i represents the length of the i-th raw material;

[0041] The information of the steel bars to be cut is represented by an array b and an array c, b = [b1 b2 … b j …], c = [c1 c2 … c j .], where b j and c j represent the length and quantity of the j-th cut steel bar respectively; the scale threshold scale_threshold of the steel bar set is set to 10 20 .

[0042] S2. Estimation of the scale of the steel bar set

[0043] The estimated value scale of the set material scale represents the evaluation of the scale of the steel bar set material to be cut currently. The scale of the steel bar set material is the number of all possible matching combinations of steel bars. Its actual value needs to be obtained by calculating all combinations. For a super-large-scale steel bar set material, calculating all steel bar combinations is extremely time-consuming and may even cause memory overflow and be unable to calculate. Therefore, it is estimated, and the estimation method is as follows:

[0044] Select the longest raw material a from the raw materials a max ;

[0045] Calculate the estimated value scale of the set material scale. The calculation formula is as follows:

[0046] In the formula, len(b) is the length of the steel bar array b to be cut, that is, the number of elements in b, and [] represents rounding down.

[0047] S3. Compare scale and scale_threshold. If scale ≤ scale_threshold, directly use the global optimal algorithm to optimize and obtain the cutting plan for all steel bars to be cut; if scale > scale_threshold, use the greedy algorithm to optimize the cutting plan for each length of raw material in turn. After determining the cutting plan and fully cutting, calculate the estimated value scale of the set material scale of the remaining steel bars to be cut again. If scale > scale_threshold, continue to use the greedy algorithm for iterative optimization until scale ≤ scale_threshold, and then use the global optimal algorithm to optimize and obtain the cutting plan for the remaining steel bars to be cut;

[0048] Among them, optimizing the cutting plan for each length of raw material using the greedy algorithm includes the following steps:

[0049] A. For each length of raw material a i , establish an integer programming mathematical model with the highest single-pass yield rate as the goal. The integer programming mathematical model is as follows:

[0050] max f = b·x T / a i

[0051]

[0052] In the formula, x is the decision variable, x = [x1 x2 … x j …], x j is the cutting quantity of the jth size steel bar in b;

[0053] B. Use the branch and bound method to obtain the optimal solution x, and calculate the highest single-pass yield rate f according to x max = b·xT / a i ;

[0054] C. While achieving the highest single-time yield, it is also necessary to make as many steel bars as possible in a single batch. Therefore, for each length of raw material a i , taking the maximum number of steel bars cut in a single batch as the new constraint condition, a new integer programming mathematical model is established. The new integer programming mathematical model is as follows:

[0055]

[0056]

[0057] D. Use branch and bound method to find the optimal solution x, and calculate the highest single yield f based on x. max =b·x T / a i .

[0058] After optimizing the cutting plan for each length of raw material using a greedy algorithm, the set of materials corresponding to the optimal solution x that maximizes the number of rebars cut in a single cut and the highest yield rate is used for full cutting, and the array c of the remaining rebars to be cut is updated. To improve cutting efficiency and ensure the yield rate, the cutting plan for each length of raw material is optimized using a greedy algorithm. The cutting plan corresponding to the length with the highest yield rate is selected. Based on the cutting requirements, this combination is used for multiple cuttings as much as possible, and then the array c of the remaining rebars to be cut is updated.

[0059] The global optimal algorithm generates all the cutting schemes for the remaining steel bars to be cut at one time, rather than just one combination at a time as in the greedy algorithm. The global optimal algorithm is used to optimize the cutting schemes for the remaining steel bars to be cut, which includes the following steps:

[0060] a. Calculation of steel bar combination to be cut

[0061] The combination of steel bars to be cut is all possible length combinations of the remaining steel bars to be cut. First, the length a of each raw material is exhaustively listed. i All corresponding cutting combinations, including all situations in which the raw material of this length can be cut, are combined to form the cutting combination T of all raw materials by splicing the remaining steel bars to be cut corresponding to each length of raw material. t mn Indicates the number of steel bars to be cut in the mth combination. Calculate the length of raw materials used for each combination, represented by array y, y = [y1 y2 … y i …], y i Indicates the length of raw materials used for the i-th combination;

[0062] b. Establish and solve the global optimization mathematical model

[0063] An integer programming model is established with the goal of minimizing the total length of the consumed raw materials. Among them, the decision variable x is the number of times each column combination in T is used, x = [x1 x2 … x i …], x i is the number of times the i-th column cutting combination in T is used. The constraint condition is that the quantity of each length of the steel bars to be cut corresponding to x is equal to the cutting requirement quantity in the updated c. The integer programming model is as follows:

[0064] min x·y T

[0065]

[0066] The optimal solution x is obtained by using the branch and bound method. The remaining steel bars to be cut are cut according to the cutting combination corresponding to x, and the cutting of all steel bar sets is completed.

[0067] The present invention combines the greedy algorithm and the global optimization algorithm to realize the optimization of multi-mode steel bar sets. First, the calculation scale is estimated according to the bill of materials. If the calculation amount exceeds the set scale threshold, it switches to the greedy algorithm mode. As the cutting progresses and some steel bars of certain sizes are cut, the set scale decreases accordingly. If it decreases to not exceed the set scale threshold, it switches to the global optimization algorithm mode; the characteristics of fast calculation speed and linear growth of the calculation scale of the greedy algorithm are used to reduce the calculation scale of the steel bar sets, and at the same time, combined with the global optimization algorithm, while achieving efficient calculation, the optimal material utilization rate is basically realized, meeting the requirements of ultra-large-scale steel bar set optimization, and solving the problem that in the case of ultra-large-scale sets with too many steel bar sizes or too many short materials in a single set, the existing set calculation methods take too long to calculate or even cause memory overflow and the computer cannot calculate.

[0068] 1) Case scenario

[0069] A certain steel bar processing factory has sufficient raw materials of Φ28mm steel bars with a length specification of 9000mm. Now it is necessary to cut a batch of Φ28mm steel bars, including 48 pieces of 2100mm steel bars, 40 pieces of 1100mm steel bars, 15 pieces of 500mm steel bars, and 20 pieces of 300mm steel bars. In this scenario, a =

[9000] , b = [2100 1100 500 300], c = [48 40 15 20], and scale_threshold can be set customarily. The present invention recommends setting it to 10 20 , but if set to this value, the current case needs to list a bill of materials of extremely large scale. For the convenience of explanation, here scale_threshold is set to 10000.

[0070] 2) Estimation of the scale of steel bar sets

[0071] First, select the longest raw material 9000 from raw material a, and then calculate the estimated value of the nested material scale according to the formula.

[0072] 3) Use the greedy algorithm for nesting and reduce the scale.

[0073] Since the scale > 10000, the steel bar nesting is evaluated as an extra-large scale. First, use the greedy algorithm for nesting. The process is as follows:

[0074] For raw material 9000, establish an integer programming mathematical model with the highest single - pass material yield rate as the goal as follows:

[0075] max f = b·x T / 9000 = [2100 1100 500 300]·x T / 9000

[0076]

[0077] Use the branch - and - bound method to obtain the optimal solution x = [0 0 15 5], and calculate the highest single - pass material yield rate f according to x. max = b·x T / 9000 = [2100 1100 500 300]·[0 0 15 5] T / 9000 = 100%

[0078] Then, with the largest number of single - pass cut steel bars as the new constraint condition, establish a new integer programming mathematical model as follows:

[0079]

[0080]

[0081] Use the branch - and - bound method to obtain the optimal solution x = [0 3 0 19].

[0082] Use the nesting combination of the optimal solution x = [0 3 0 19] for cutting. According to the cutting requirements, this combination can only be used once. After this cutting, update the remaining array c of the steel bars to be cut as [48 37 15 1]. Recalculate the scale of the remaining steel bars to be cut, which is still 17280, and scale > 10000. Continue to use the greedy algorithm to obtain the optimal solution x = [0 7 2 1]. According to the cutting requirements, this combination can only be used once. After this cutting, update the remaining array c of the steel bars to be cut as [48 30 13 0].

[0083] 4) Use the global optimization algorithm to optimize the nesting of the remaining steel bars

[0084] Recalculate the scale of the remaining steel bars to be cut = [9000 / 2100]×[9000 / 1100]×[9000 / 500] = 4×8×18 = 576. Since scale ≤ 10000, the global optimization algorithm is used to optimize the cutting plan of the remaining steel bars to be cut. The process is as follows:

[0085] First, calculate all the cutting combinations corresponding to the 9000mm raw material In this case, the number of combinations is 178, which takes up a large amount of space and is omitted here for description.

[0086] Then, establish an integer programming model with the goal of minimizing the total length of the consumed raw materials as follows:

[0087] min x·[9000 9000 … 9000] T

[0088]

[0089] Finally, use the branch and bound method to obtain the optimal solution x = [3 11 1 1], and the corresponding nesting plan The number of uses is [3 11 1 1]. Cut the remaining steel bars to be cut according to this plan to complete the nesting of all the steel bars.

[0090] Organize the nesting plans calculated in 3) and 4). The total material yield is 97.716%. The complete nesting plan is as follows:

[0091]

[0092] Although the implementation embodiments of the present invention have been disclosed as above, they are not limited to only the applications listed in the specification and the embodiments. It can be fully applied to various fields suitable for the present invention. For those familiar with the field, additional modifications can be easily made. Therefore, without departing from the general concept defined by the claims and the equivalent scope, the present invention is not limited to specific details and the illustrated examples here.

Claims

1. A multi-mode steel bar sleeve material optimization method applicable to ultra-large scale, characterized in that, Including: S1. Scene description The raw material information is represented by an array a, a = [a1 a2 … a i …], where a i represents the length of the i-th raw material; The information of the steel bars to be cut is represented by array b and array c, b = [b1 b2 … b j …], c = [c1 c2 … c j …], where b j and c j respectively represent the length and quantity of the j-th type of steel bar to be cut; the scale threshold of the steel bar nesting scale is scale_threshold; S2. Estimation of the scale of steel bar sets Select the longest raw material a from the raw material array a max ; Calculate the estimated value scale of the set material scale, and the calculation formula is as follows: Where len(b) is the length of the reinforcing bar array b to be cut, that is, the number of elements in b, and [] represents rounding down. S3. Compare scale with scale_threshold. If scale > scale_threshold, use the greedy algorithm to optimize the cutting plan of each length of raw material in turn. After determining the cutting plan and fully cutting, calculate the estimated value scale of the set material of the remaining steel bars to be cut again. If scale > scale_threshold, continue to use the greedy algorithm for iterative optimization until scale ≤ scale_threshold, and then use the global optimal algorithm to optimize the cutting plan of the remaining steel bars to be cut.

2. The multi-mode steel bar sleeve material optimization method applicable to ultra-large scale according to claim 1, characterized in that, The steps of using the greedy algorithm to optimize the cutting plan of each length of raw material in step S3 include the following steps: A. For each length of raw material a i , an integer programming mathematical model is established with the goal of maximizing the single-pass yield. The integer programming mathematical model is as follows: max f = b·x T / a i where x is the decision variable, x = [x1 x2 … x j …], x j is the cutting quantity of the j-th length of steel bar in the array b; B. Obtain the optimal solution x using the branch and bound method, and calculate the highest single - pass yield f based on x max = b·x T / a i ; C. Taking the maximum number of steel bars cut in a single time as a new constraint condition, establish a new integer programming mathematical model, and the new integer programming mathematical model is as follows: D. Obtain the optimal solution x using the branch and bound method, and calculate the highest single - pass yield f according to x max = b·x T / a i .

3. The multi-mode steel bar sleeve material optimization method applicable to ultra-large scale according to claim 2, characterized in that, In step S3, after using the greedy algorithm to optimize the cutting plan of each length of raw material in turn, use the set material combination corresponding to the optimal solution x with the maximum number of steel bars cut in a single time and the highest material yield rate for full cutting, and update the array c of the remaining number of steel bars to be cut.

4. The multi-mode steel bar sleeve material optimization method applicable to ultra-large scale according to claim 3, characterized in that The steps of using the global optimal algorithm to optimize the cutting plan of the remaining steel bars to be cut include the following steps: a. Calculation of the combination of steel bars to be cut Exhaustively list the length a of each raw material i For all corresponding cutting combinations, each cutting combination includes all the cases that the raw material of this length can be cut. Splice the remaining steel bars to be cut combinations corresponding to each length of raw material to form the cutting combination T of all raw materials t mn represents the quantity of the m-th steel bar to be cut in the n-th combination. Calculate the length of the raw material used for each combination, and represent it with the array y, y = [y1 y2 … y i …], y i represents the length of the raw material corresponding to the i-th combination b. Establish and solve the global optimization mathematical model An integer programming model is established with the goal of minimizing the total length of the consumed raw materials. Among them, the decision variable x is the number of times each column combination in T is used, x = [x1 x2 … x i …], x i is the number of times the i-th cutting combination in T is used. The constraint condition is that the quantity of each length of the steel bars to be cut corresponding to x is equal to the cutting requirement quantity in the updated c. The integer programming model is as follows: min x·y T Use the branch and bound method to obtain the optimal solution x, and cut the remaining steel bars to be cut according to the cutting combination corresponding to x to complete the cutting of all steel bar sets.

Images