A method, device and medium for optimizing the rolling performance of instant invoice ink

Abstract

This application relates to the field of optimizing the ink for instant scratch tickets, and particularly to a method, device, and medium for optimizing the winding performance of the ink for instant scratch tickets. The method includes: obtaining the initial formula set A of the ink for instant scratch tickets; traversing A to obtain the corresponding input vector, winding length, and confidence level of the winding length; training the target neural network model using the training sample set and the corresponding label set; inputting the input vector G corresponding to the target formula into the trained target neural network model to obtain the winding length L' corresponding to the target formula and the confidence level Z' of the winding length; if L' does not belong to the target length range, perform the first-round adjustment on the mass percentage of the components included in the target formula, and if the winding length corresponding to the formula after a certain adjustment in the first-round adjustment belongs to the target length range, output the formula after this adjustment in the first-round adjustment and the confidence level of the corresponding winding length. The present invention can optimize the winding performance of the ink for instant scratch tickets. i Corresponding input vector, winding length, and confidence level of the winding length; training the target neural network model using the training sample set and the corresponding label set; inputting the input vector G corresponding to the target formula into the trained target neural network model to obtain the winding length L' corresponding to the target formula and the confidence level Z' of the winding length; if L' does not belong to the target length range, perform the first-round adjustment on the mass percentage of the components included in the target formula, and if the winding length corresponding to the formula after a certain adjustment in the first-round adjustment belongs to the target length range, output the formula after this adjustment in the first-round adjustment and the confidence level of the corresponding winding length. The present invention can optimize the winding performance of the ink for instant scratch tickets.

Description

Technical Field

[0001] The invention relates to the field of optimization of instant invoice ink, and in particular to a method, equipment and medium for optimizing the roll performance of instant invoice ink. Background Art

[0002] In the production process of instant invoices, the ink coating has the function of covering important information until it is scraped off. In actual usage scenarios, the length of the ink that is scraped off in a long strip can represent the ink roll length. When the ink roll length is within the target length range, it means that the ink roll performance is good; when the ink roll length is not within the target length range, it means that the ink roll performance is poor. Different instant invoice ink formulas produce instant invoice inks with different roll performance. Some instant invoice ink formulas may produce instant invoice inks with poor roll performance. How to optimize the roll performance of instant invoice inks is an urgent problem to be solved. Summary of the invention

[0003] The object of the present invention is to provide a method, device and medium for optimizing the rolling performance of instant invoice ink, so as to optimize the rolling performance of instant invoice ink.

[0004] According to a first aspect of the present invention, a method for optimizing the rolling performance of instant invoice ink is provided, the method comprising the following steps:

[0005] S100, obtaining an initial formula set A of instant invoice ink,

[0006] ,

[0007] A i is the i-th initial formula of the instant ticket ink, the value of i ranges from 1 to n, and n is the number of initial formulas of the instant ticket ink.

[0008] S200, traverse A, get A i The corresponding input vector F i , F i Each component and the mass percentage of each component used to characterize the i-th initial formula of the instant ticket ink.

[0009] S300, traverse A, obtain A i Corresponding roll length L i and the confidence level Z of the roll length i Among them, A i The corresponding roll length is A i The length of the ink that is scraped off in long strips when making instant tickets.

[0010] S400, training the target neural network model using the training sample set and the corresponding label set to obtain a trained target neural network model; the i-th training sample in the training sample set is F i , the i-th label in the label set is .

[0011] S500, obtaining an input vector G corresponding to a target formula of instant invoice ink.

[0012] S600, input G into the trained target neural network model to obtain the roll length L' and the confidence level Z' of the roll length corresponding to the target formula of the instant invoice ink.

[0013] S700: If L' does not belong to the target length range, proceed to S800.

[0014] S800, performing a first round of adjustment on the mass percentages of the components included in the target formula of the instant invoice ink, and obtaining the roll length and the confidence level of the roll length corresponding to each adjusted formula in the first round of adjustment.

[0015] S900: If the roll length corresponding to a formula adjusted in a first round of adjustment falls within the target length range, the confidence level of the formula adjusted in the first round of adjustment and the corresponding roll length is output.

[0016] According to a second aspect of the present invention, there is provided an electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for optimizing the rolling performance of instant invoice ink when executing the computer program.

[0017] According to a third aspect of the present invention, there is provided a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned method for optimizing the rolling performance of instant invoice ink.

[0018] Compared with the prior art, the present invention has at least the following beneficial effects:

[0019] The present invention uses each component and the mass percentage of each component that can characterize the formula of the instant invoice ink as a training sample, uses the roll length corresponding to the formula of the instant invoice ink and the confidence of the roll length as the label corresponding to the training sample, and the target neural network model trained based on the training sample and the corresponding label has the function of being able to infer the roll length and the confidence of the roll length corresponding to the formula of the instant invoice ink according to each component and the percentage of each component in the formula of the instant invoice ink; based on the trained target neural network model, the roll length and the confidence of the roll length of the target formula of the instant invoice ink are inferred, and if the inferred roll length does not belong to the target length range, indicating that when the target formula of instant invoice ink is used to make the instant invoice ink, the length of the ink in the form of a long strip that is scraped off does not meet the requirements, that is, the target formula cannot meet the requirements of the rolling performance, then the mass percentages of the components included in the target formula are adjusted, and the rolling length and confidence corresponding to the adjusted formula are obtained based on the trained target neural network model; if the rolling length corresponding to the adjusted formula belongs to the target length range, the adjusted formula and the corresponding confidence of the rolling length are output; thus, the present invention achieves the optimization of the target formula when the target formula cannot meet the requirements of the rolling performance, thereby achieving the purpose of optimizing the rolling performance of the instant invoice ink. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 This is a flow chart of a method for optimizing the rolling performance of instant invoice ink provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention. Embodiment 1

[0023] According to this embodiment, Figure 1 As shown, a method for optimizing the rolling performance of instant invoice ink is provided, and the method comprises the following steps:

[0024] S100, obtaining an initial formula set A of instant invoice ink,

[0025] ,

[0026] A i is the i-th initial formula of the instant ticket ink, the value of i ranges from 1 to n, and n is the number of initial formulas of the instant ticket ink.

[0027] In this embodiment, the initial formula set of instant invoice ink includes n formulas of instant invoice ink, any of which can be used to make instant invoice ink, and any of which includes four components, namely, resin component, solvent component, pigment component and auxiliary component, and any two different formulas do not meet the conditions that the type of each component is the same and the mass percentage of each component is the same. In addition, it should be understood that the initial formula is the formula, and the initial in the initial formula is only used to distinguish the target formula in S500.

[0028] S200, traverse A, get A i The corresponding input vector F i , F i Each component and the mass percentage of each component used to characterize the i-th initial formula of the instant ticket ink.

[0029] In this embodiment, different recipes correspond to different input vectors, and one recipe corresponds to only one input vector. As a preferred specific implementation,

[0030] ,

[0031] b i,1 , b i,2 , b i,3 and b i,4 They represent the resin component type, solvent component type, pigment component type and auxiliary component type of the i-th initial formula of the instant invoice ink, c i,1 、c i,2 、c i,3 and c i,4 represent the mass percentage of the resin component, the mass percentage of the solvent component, the mass percentage of the pigment component and the mass percentage of the auxiliary component of the i-th initial formulation of the instant invoice ink. i The i-th initial formulation of the instant ticket ink can be accurately characterized.

[0032] As a preferred specific implementation mode, b i,1 , b i,2 , b i,3 and b i,4 All are represented by one-hot encoding, b i,1 The length of b is the number of types of resin components that can be used for instant ticket ink.i,2 The length of b is the number of types of solvent components that can be used in instant ticket ink. i,3 The length of b is the number of types of pigment components that can be used in instant ticket ink. i,4 The length of b is the number of types of auxiliary ingredients that can be used in the instant ticket ink. i,1 、b i,2 、b i,3 and b i,4 The resin component type, solvent component type, pigment component type and auxiliary component type of the i-th initial formula of the instant invoice ink can be quickly determined.

[0033] S300, traverse A, obtain A i Corresponding roll length L i and the confidence level Z of the roll length i Among them, A i The corresponding roll length is A i The length of the ink that is scraped off in long strips when making instant tickets.

[0034] In this embodiment, A i When making instant ticket ink, the ink is in the form of long strips. The length of the scraped-off ink can be determined by using A i The ink of the instant ticket is subjected to a scratch test, and the test conditions are the same for each scratch test. The length of the ink in the form of a long strip that is scratched off in each scratch test can be known, for example, it can be obtained by manual observation.

[0035] As a preferred specific implementation, obtain A i Corresponding roll length L i and the confidence level Z of the roll length i include:

[0036] S310, obtain A i The corresponding roll length test set Y i ;

[0037] ,

[0038] For use A i The length of the ink strips scraped off when the ink of the instant ticket is scratched off for the r(i)th time, r(i) ranges from 1 to R(i), R(i) is the length of the ink strips scraped off when the ink ... i The number of times the ink of the produced scratch tickets is scratched off.

[0039] S320, for Y i Perform clustering and obtain the clustering result U i ;

[0040] ,

[0041] For Y i The v(i)th cluster is obtained by clustering. The value range of v(i) is 1 to w(i), and w(i) is the value of Y i The number of clusters obtained by clustering.

[0042] As a preferred specific implementation, a density clustering method is used for clustering, such as the DBSCAN density clustering method, where each length is taken as a point, and the absolute value of the difference between two lengths is taken as the distance between the corresponding two points. Based on a preset neighborhood radius and a preset minimum number of points, clustering is performed according to the process of the DBSCAN density clustering method to obtain the final clustering result. i The lengths with smaller differences in Y are grouped into one cluster. i The lengths with larger differences are grouped in different clusters.

[0043] S330, The mean of the lengths is determined as L i ,Will The ratio of the number of included lengths to R(i) is determined as Z i ; For U i The number of clusters with the largest length is included.

[0044] In this embodiment, Z i The larger the value, the higher the L i The higher the accuracy of Z i The smaller the value, the higher the L i The lower the accuracy.

[0045] Based on S310-S330, A can be accurately obtained i Corresponding roll length L i and the confidence level Z of the roll length i .

[0046] S400, training the target neural network model using the training sample set and the corresponding label set to obtain a trained target neural network model; the i-th training sample in the training sample set is F i , the i-th label in the label set is .

[0047] In this embodiment, the number of training samples included in the training sample set is n, and the number of labels included in the label set is also n.

[0048] As an optional specific implementation, the target neural network model is a multi-layer perceptron (MLP) architecture, wherein the number of neurons in the input layer is the dimension of the input vector, and the number of neurons in the output layer is 2.

[0049] Those skilled in the art know that any neural network model training process in the prior art falls within the protection scope of the present invention. As a specific implementation, the loss function of the target neural network model is the mean square error loss, and the loss function of the target neural network model is the sum of the first part and the second part, the first part is the mean square error loss corresponding to the normalized roll length, and the second part is the mean square error loss corresponding to the confidence of the roll length.

[0050] S500, obtaining an input vector G corresponding to a target formula of instant invoice ink.

[0051] In this embodiment, the target formula of the instant invoice ink is the formula to be judged for the roll performance, and G is used to characterize each component of the target formula of the instant invoice ink and the mass percentage of each component. i The form is consistent,

[0052] when hour,

[0053] ,

[0054] and They respectively represent the resin component type, solvent component type, pigment component type and auxiliary component type of the target formula of the instant invoice ink. and They respectively represent the mass percentage of the resin component, the mass percentage of the solvent component, the mass percentage of the pigment component and the mass percentage of the auxiliary component of the target formula of the instant invoice ink. and They are also represented by one-hot encoding.

[0055] S600, input G into the trained target neural network model to obtain the roll length L' and the confidence level Z' of the roll length corresponding to the target formula of the instant invoice ink.

[0056] In this embodiment, the trained target neural network model has the function of inferring the roll length and the confidence level of the roll length of the corresponding formula based on the input vector. Therefore, after G is input into the trained target neural network model, the output of the trained target neural network model includes the roll length and the confidence level of the roll length corresponding to the target formula of the invoicing ink.

[0057] S700: If L' does not belong to the target length range, proceed to S800.

[0058] In this embodiment, the target length range is a preset empirical value. When L' does not belong to the target length range, it is determined that the roll length corresponding to the target formula of the invoicing ink does not meet the roll performance requirements. Otherwise, it is determined that the roll length corresponding to the target formula of the invoicing ink meets the roll performance requirements.

[0059] In this embodiment, S700 further includes: if L' belongs to the target length range, outputting information and Z' indicating that the target formula of the instant invoice ink can meet the requirements of the rolling performance, and no longer executing S800-S900.

[0060] S800, performing a first round of adjustment on the mass percentages of the components included in the target formula of the instant invoice ink, and obtaining the roll length and the confidence level of the roll length corresponding to each adjusted formula in the first round of adjustment.

[0061] In this embodiment, the roll length and the confidence level of the roll length corresponding to each adjusted formula in the first round of adjustment are obtained by obtaining the input vector corresponding to each adjusted formula in the first round of adjustment and inputting each input vector into the trained neural network model. i The format is the same as that of , and will not be repeated here.

[0062] In this embodiment, when adjusting the target formula of the instant ticket ink, the type of each component in the target formula of the instant ticket ink is not changed, and only the mass percentages of different components are changed. As a preferred specific implementation, the first round of adjustment of the target formula of the instant ticket ink includes:

[0063] S810, adjusting the mass percentage of the pth component included in the target formula of the instant invoice ink to ;

[0064] in, is the mass percentage of the pth component included in the target formula of the instant ticket ink, p=1,2,3,4, t is [-h 1 ,h 1 ]A value randomly selected from the range, h 1 It is the maximum positive adjustment range corresponding to the first round of adjustment.

[0065] In this embodiment, h 1 is the preset experience value, optional, h 1 It is 3%.

[0066] S820, adjusting the mass percentage of the qth component included in the target formula of the instant invoice ink to ;

[0067] Among them, sp is the sum of the mass percentages of the components other than the pth component included in the target formula of the instant invoice ink, is the mass percentage of the qth component included in the target formula of the instant ticket ink, q=1,2,3,4 and q≠p.

[0068] Based on S810-S820, the first round of adjustment of the mass percentage of the components included in the target formula of the instant invoice ink can be achieved; the number of adjustments included in the first round of adjustment is an empirical value, and the number of adjustments included in the first round of adjustment can be increased by cyclically executing S810-S820.

[0069] S900: If the roll length corresponding to a formula adjusted in a first round of adjustment falls within the target length range, the confidence level of the formula adjusted in the first round of adjustment and the corresponding roll length is output.

[0070] In this embodiment, if there are multiple adjusted formulas in the first round of adjustment whose corresponding roll lengths all fall within the target length range, the confidence levels of the multiple adjusted formulas and corresponding roll lengths in the first round of adjustment are output.

[0071] As a preferred specific implementation, S900 further includes: if the roll length corresponding to any adjusted recipe in the first round of adjustment does not fall within the target length range, then proceed to S910.

[0072] S910, perform a second round of adjustment on the mass percentages of the components included in the target formula of the instant invoice ink, and obtain the roll length and the confidence level of the roll length corresponding to each adjusted formula in the second round of adjustment; wherein the maximum positive direction adjustment amplitude corresponding to the second round of adjustment is greater than the maximum positive direction adjustment amplitude corresponding to the first round of adjustment.

[0073] In this embodiment, the maximum positive direction adjustment range corresponding to the second round of adjustment is a preset empirical value. Optionally, the maximum positive direction adjustment range corresponding to the first round of adjustment is 3%, and the maximum positive direction adjustment range corresponding to the second round of adjustment is 5%.

[0074] S920: If the roll length corresponding to a formula adjusted in the second round of adjustment falls within the target length range, the confidence level of the formula adjusted in the second round of adjustment and the corresponding roll length is output.

[0075] In this embodiment, if there are multiple adjusted formulas in the second round of adjustment whose corresponding roll lengths all fall within the target length range, the confidence levels of the multiple adjusted formulas and corresponding roll lengths in the second round of adjustment are output.

[0076] In this embodiment, if the roll length corresponding to any adjusted formula in the second round of adjustment does not fall within the target length range, the mass percentage of the components included in the target formula of the instant invoice ink is adjusted for the third round, and the roll length and the confidence of the roll length corresponding to each adjusted formula in the third round of adjustment are obtained. If the roll length corresponding to the formula adjusted in a certain round of adjustment falls within the target length range, the confidence of the adjusted formula and the corresponding roll length in the third round of adjustment is output; if the roll length corresponding to any adjusted formula in the third round of adjustment does not fall within the target length range, the mass percentage of the components included in the target formula of the instant invoice ink is adjusted for the fourth round, and so on, until the roll length corresponding to a certain adjusted formula in a certain round of adjustment falls within the target length range, then the confidence of the adjusted formula and the corresponding roll length in the adjustment round is output. Among them, the maximum positive direction adjustment amplitude corresponding to the third round of adjustment is greater than the maximum positive direction adjustment amplitude corresponding to the second round of adjustment, and the maximum positive direction adjustment amplitude corresponding to the fourth round of adjustment is greater than the maximum positive direction adjustment amplitude corresponding to the third round of adjustment.

[0077] In this embodiment, each component and the mass percentage of each component that can characterize the formula of the instant invoice ink are used as training samples, and the roll length corresponding to the formula of the instant invoice ink and the confidence of the roll length are used as labels corresponding to the training samples. The target neural network model trained based on the training samples and the corresponding labels has the function of inferring the roll length and the confidence of the roll length corresponding to the formula of the instant invoice ink according to each component and the percentage of each component of the formula of the instant invoice ink; the roll length and the confidence of the roll length of the target formula of the instant invoice ink are inferred based on the trained target neural network model. If the inferred roll length does not belong to the target length range, indicating that when the target formula of instant invoice ink is used to make the instant invoice ink, the length of the ink in the form of a long strip that is scraped off does not meet the requirements, that is, the target formula cannot meet the requirements of the rolling performance, then the mass percentages of the components included in the target formula are adjusted, and the rolling length and confidence corresponding to the adjusted formula are obtained based on the trained target neural network model; if the rolling length corresponding to the adjusted formula belongs to the target length range, the adjusted formula and the corresponding confidence of the rolling length are output; therefore, this embodiment optimizes the target formula when the target formula cannot meet the requirements of the rolling performance, thereby achieving the purpose of optimizing the rolling performance of the instant invoice ink. Embodiment 2

[0078] According to this embodiment, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for optimizing the rolling performance of the instant invoice ink in the first embodiment when executing the computer program. Embodiment 3

[0079] According to this embodiment, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for optimizing the rolling performance of the instant invoice ink in the first embodiment described above is implemented.

[0080] Although some specific embodiments of the present invention have been described in detail by way of example, it will be appreciated by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It will also be appreciated by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.

Claims

1. A method for optimizing the rolling performance of instant invoice ink, characterized in that: The method comprises the following steps: S100, obtaining an initial formula set A of instant invoice ink, ， A i is the i-th initial formula of the instant ticket ink, i ranges from 1 to n, and n is the number of initial formulas of the instant ticket ink; S200, traverse A, get A i The corresponding input vector F i , F i Each component and the mass percentage of each component used to characterize the i-th initial formula of the instant ticket ink; S300, traverse A, obtain A i Corresponding roll length L i and the confidence level Z of the roll length i Among them, A i The corresponding roll length is A i The length of the ink that is scraped off in a long strip when making the instant ticket; S400, training the target neural network model using the training sample set and the corresponding label set to obtain a trained target neural network model; the i-th training sample in the training sample set is F i , the i-th label in the label set is ; S500, obtaining an input vector G corresponding to a target formula of instant invoice ink; S600, input G into the trained target neural network model to obtain the roll length L' and the confidence level Z' of the roll length corresponding to the target formula of the instant invoice ink; the trained target neural network model has the function of inferring the roll length and the confidence level of the roll length of the corresponding formula according to the input vector; S700, if L' does not belong to the target length range, enter S800; S800, performing a first round of adjustment on the mass percentages of the components included in the target formula of the instant invoice ink, and obtaining the roll length and the confidence level of the roll length corresponding to each adjusted formula in the first round of adjustment; S900, if the roll length corresponding to a certain adjusted formula in the first round of adjustment falls within the target length range, output the confidence level of the adjusted formula and the corresponding roll length in the first round of adjustment; Get A i Corresponding roll length L i and the confidence level Z of the roll length i include: S310, obtain A i The corresponding roll length test set Y i ; ， For use A i The length of the ink strips scraped off when the ink of the instant ticket is scratched off for the r(i)th time, r(i) ranges from 1 to R(i), R(i) is the length of the ink strips scraped off when the ink ... i The number of times the ink of the produced instant tickets is scratched off and tested; S320, for Y i Perform clustering and obtain the clustering result U i ; ， For Y i The v(i)th cluster is obtained by clustering. The value range of v(i) is 1 to w(i), and w(i) is the value of Y i The number of clusters obtained by clustering; S330, The mean of the lengths is determined as L i ,Will The ratio of the number of included lengths to R(i) is determined as Z i ; For U i The number of clusters with the largest length is included.

2. The method for optimizing the rolling performance of instant invoice ink according to claim 1, characterized in that: The first round of adjustments to the target formulation of the instant ticket ink included: S810, adjusting the mass percentage of the pth component included in the target formula of the instant invoice ink to ; in, is the mass percentage of the pth component included in the target formula of the instant ticket ink, p=1,2,3,4, t is a value randomly selected from the range of [-h1,h1], and h1 is the maximum positive adjustment amplitude corresponding to the first round of adjustment; S820, adjusting the mass percentage of the qth component included in the target formula of the instant invoice ink to ; Among them, s p is the sum of the mass percentages of the components other than the pth component included in the target formula of the instant invoice ink, is the mass percentage of the qth component included in the target formula of the instant ticket ink, q=1,2,3,4 and q≠p.

3. The method for optimizing the rolling performance of instant invoice ink according to claim 2, characterized in that: S900 also includes: if the roll length corresponding to any adjusted recipe in the first round of adjustment does not fall within the target length range, then proceed to S910; S910, performing a second round of adjustment on the mass percentages of the components included in the target formula of the instant invoice ink, and obtaining the roll length and the confidence level of the roll length corresponding to each adjusted formula in the second round of adjustment; wherein the maximum positive direction adjustment amplitude corresponding to the second round of adjustment is greater than the maximum positive direction adjustment amplitude corresponding to the first round of adjustment; S920: If the roll length corresponding to a formula adjusted in the second round of adjustment falls within the target length range, the confidence level of the formula adjusted in the second round of adjustment and the corresponding roll length is output.

4. The method for optimizing the rolling performance of instant invoice ink according to claim 1, characterized in that: ， b i,1 、b i,2 、b i,3 and b i,4 They represent the resin component type, solvent component type, pigment component type and auxiliary component type of the i-th initial formula of the instant invoice ink, c i,1 、c i,2 、c i,3 and c i,4 They respectively represent the mass percentage of the resin component, the mass percentage of the solvent component, the mass percentage of the pigment component and the mass percentage of the auxiliary agent component of the i-th initial formula of the instant invoice ink.

5. The method for optimizing the rolling performance of instant invoice ink according to claim 4, characterized in that: b i,1 、b i,2 、b i,3 and b i,4 All are represented by one-hot encoding, b i,1 The length of b is the number of types of resin components that can be used for instant ticket ink. i,2 The length of b is the number of types of solvent components that can be used in instant ticket ink. i,3 The length of b is the number of types of pigment components that can be used in instant ticket ink. i,4 The length of is the number of preset types of auxiliary ingredients that can be used in instant ticket ink.

6. The method for optimizing the rolling performance of instant invoice ink according to claim 1, characterized in that: S700 also includes: if L' belongs to the target length range, outputting information and Z' indicating that the target formula of the instant invoice ink can meet the requirements of the rolling performance, and no longer executing S800-S900.

7. The method for optimizing the rolling performance of instant invoice ink according to claim 1, characterized in that: The loss function of the target neural network model is the mean square error loss.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for optimizing the rolling performance of instant invoice ink as described in any one of claims 1 to 7 is implemented.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for optimizing the rolling performance of instant invoice ink as described in any one of claims 1 to 7 is implemented.

Images