Component proportion optimization method of polyurethane pressure-sensitive adhesive composition

Abstract

The invention provides a composition proportion optimization method of a polyurethane pressure-sensitive adhesive composition, and relates to the technical field of data processing. According to the method, iteration processing can be performed on a to-be-optimized component proportioning list, an initial performance value list, an initial component adjustment priority list and a target performance value list according to a component prediction model, a performance prediction model and a priority prediction model, so that a target component proportioning list can be obtained; therefore, optimization of the component proportion of the polyurethane pressure-sensitive adhesive composition is achieved, errors are gradually reduced through a closed-loop feedback mechanism, human intervention is reduced, meanwhile, the influence of the component adjustment sequence is fully considered, and improvement of optimization efficiency and accuracy is facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method for optimizing the ingredient ratio of a polyurethane pressure-sensitive adhesive composition. Background Art

[0002] The polyurethane pressure-sensitive adhesive composition is a pressure-sensitive adhesive, which is widely used in the fields of electronics, packaging, industry, etc. It combines the flexibility of polyurethane and the instant adhesion characteristics of the pressure-sensitive adhesive, and has excellent durability, weather resistance and bonding performance. The performance, cost and application effects of the polyurethane pressure-sensitive adhesive composition prepared based on different ingredient ratios are different. Therefore, it is necessary to optimize the ingredient ratio of the polyurethane pressure-sensitive adhesive composition so that the performance of the optimized polyurethane pressure-sensitive adhesive composition reaches the target.

[0003] In the prior art, the methods for optimizing the ingredient ratio of the polyurethane pressure-sensitive adhesive composition are mainly based on manual optimization or establishing a mathematical model for optimization. Among them, manual optimization adjusts the ingredient ratio through manual experience and target performance to achieve optimization; establishing a mathematical model for optimization establishes a mathematical model for describing the relationship between the input ingredient ratio and the output performance, and determines the optimal ingredient ratio through an optimization algorithm to achieve optimization.

[0004] However, the above methods also have the following technical problems:

[0005] Manual optimization mainly relies on the experience of the operator, which has great subjectivity and instability. Especially when involving multiple components, it requires repeated experiments and adjustments, which is time-consuming and laborious and prone to missing the optimal solution; when there are complex non-linear relationships or multi-variable interactions, it is difficult to establish an accurate mathematical model for describing the relationship between the input (ingredient ratio) and the output (performance), and the lack of experimental data or noise interference may also cause a large deviation between the model prediction result and the actual situation. It can be seen that relying solely on the mathematical model may lead to large prediction errors. In addition, the above methods do not consider the influence of the ingredient adjustment order on the performance, and the addition order of different ingredients may significantly affect the performance. Therefore, based on the above methods to optimize the ingredient ratio of the polyurethane pressure-sensitive adhesive composition, the optimization efficiency and accuracy are relatively low. Summary of the Invention

[0006] In view of the above technical problems, the technical solution adopted by the present invention is as follows:

[0007] A method for optimizing the ingredient ratio of a polyurethane pressure-sensitive adhesive composition, the method comprising the following steps:

[0008] S1. A list B of ingredient ratios to be optimized, an initial performance value list C corresponding to B, an initial ingredient adjustment priority list D corresponding to C, and a target performance value list V = {V1, V2,..., Vj …, V n} are input into the component prediction model to obtain the intermediate component ratio list E. B includes the component ratio to be optimized corresponding to each component identifier in the component identifier list A of the polyurethane pressure-sensitive adhesive composition. C includes R1, R2, …, R j …, R n corresponding initial performance values, and R j is the performance identifier of the j-th performance of the polyurethane pressure-sensitive adhesive composition, where j ranges from 1 to n, and n is the number of performances of the polyurethane pressure-sensitive adhesive composition. D includes the initial component adjustment priority corresponding to each component identifier in A. V j is the target performance value corresponding to R j . E includes the intermediate component ratio corresponding to each component identifier in A.

[0009] S2. Input E into the performance prediction model to obtain the intermediate performance value list F = {F1, F2, …, F j …, F n}, and F j is the intermediate performance value corresponding to R j in F.

[0010] S3. If ∑ n j＝1 (K j × |V j - F j |) ≤ G, then take E as the target component ratio list; otherwise, input E, F, and V into the priority prediction model to obtain the intermediate component adjustment priority list H corresponding to F and enter step S4. K j is the preset importance weight corresponding to R j , G is the preset performance difference threshold, and H includes the intermediate component adjustment priority corresponding to each component identifier in A.

[0011] S4. Take E as B, take F as C, take H as D, and enter step S1.

[0012] The present invention has at least the following beneficial effects:

[0013] The present invention provides a method for optimizing the ingredient ratio of a polyurethane pressure-sensitive adhesive composition. The method inputs a list of ingredient ratios to be optimized, a list of initial performance values corresponding to the list of ingredient ratios to be optimized, a list of initial ingredient adjustment priorities corresponding to the list of initial performance values, and a list of target performance values into an ingredient prediction model to obtain an intermediate ingredient ratio list; inputs the intermediate ingredient ratio list into a performance prediction model to obtain a list of intermediate performance values corresponding to the intermediate ingredient ratio list. If the intermediate performance values in the list of intermediate performance values meet the conditions, the intermediate ingredient ratio list is used as the target ingredient ratio list. Otherwise, the intermediate ingredient ratio list, the list of intermediate performance values, and the list of target performance values are input into a priority prediction model to obtain a list of intermediate ingredient adjustment priorities corresponding to the list of intermediate performance values. The intermediate ingredient ratio list is used as the list of ingredient ratios to be optimized, the list of intermediate performance values is used as the list of initial performance values, the list of intermediate ingredient adjustment priorities is used as the list of initial ingredient adjustment priorities, and the target ingredient ratio list is obtained again. It can be seen that the present invention iteratively processes the list of ingredient ratios to be optimized, the list of initial performance values, the list of initial ingredient adjustment priorities, and the list of target performance values according to the ingredient prediction model, the performance prediction model, and the priority prediction model to obtain the target ingredient ratio list, thereby realizing the optimization of the ingredient ratio of the polyurethane pressure-sensitive adhesive composition, gradually reducing the error through a closed-loop feedback mechanism, reducing human intervention, and fully considering the influence of the ingredient adjustment order, which is beneficial to improving the optimization efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0015] Figure 1 It is a flowchart of a method for optimizing the ingredient ratio of a polyurethane pressure-sensitive adhesive composition provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0017] It should be noted that the terms "first", "second", etc. in the specification, claims and the above-mentioned drawings of the present invention are used to distinguish similar tasks and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0018] An embodiment of the present invention provides a method for optimizing the component ratio of a polyurethane pressure-sensitive adhesive composition, as Figure 1 shown, the method includes the following steps:

[0019] S1. Input the list B of component ratios to be optimized, the corresponding list C of initial performance values, the corresponding list D of initial component adjustment priorities, and the list V = {V1, V2,..., V j ,..., V n} of target performance values into the component prediction model to obtain the intermediate component ratio list E, where B includes the component ratios to be optimized corresponding to each component identifier in the component identifier list A corresponding to the polyurethane pressure-sensitive adhesive composition, C includes the initial performance values corresponding to R1, R2,..., R j ,..., R n , R j is the performance identifier of the jth performance of the polyurethane pressure-sensitive adhesive composition, the value of j ranges from 1 to n, n is the number of performances of the polyurethane pressure-sensitive adhesive composition, D includes the initial component adjustment priorities corresponding to each component identifier in A, and V j is the target performance value corresponding to R j , and E includes the intermediate component ratios corresponding to each component identifier in A.

[0020] Specifically, A includes several component identifiers of components that can be used to form the polyurethane pressure-sensitive adhesive composition.

[0021] In a specific embodiment, the component ratio to be optimized corresponding to the component identifier is the ratio of the component corresponding to the component identifier in the polyurethane pressure-sensitive adhesive composition to be optimized. The polyurethane pressure-sensitive adhesive composition to be optimized is a polyurethane pressure-sensitive adhesive composition determined by those skilled in the art from several existing polyurethane pressure-sensitive adhesive compositions according to actual needs. Among them, in two different polyurethane pressure-sensitive adhesive compositions, the ratios of the components corresponding to each component identifier are not completely the same.

[0022] Specifically, the proportion of components is used to represent the ratio of components.

[0023] Optionally, the proportion of components can be expressed as a percentage; for example: in the XX composition composed of Component 1, Component 2, Component 3, and Component 4, the proportion of Component 1 is 70%, the proportion of Component 2 is 10%, the proportion of Component 3 is 4%, and the proportion of Component 4 is 6%.

[0024] Specifically, the initial performance value is the performance value of the polyurethane pressure-sensitive adhesive composition prepared based on each proportion of components to be optimized in the list of proportions of components to be optimized; it can be understood as: the performance value of the performance of the polyurethane pressure-sensitive adhesive composition to be optimized.

[0025] Specifically, the performance of the polyurethane pressure-sensitive adhesive composition includes adhesion performance, flexibility, cohesive strength, heat resistance, low temperature resistance, aging resistance, and light transmittance.

[0026] Specifically, the initial component adjustment priority is used to represent the adjustment order of the proportion of the component corresponding to the component identifier in the process of adjusting the performance value corresponding to each performance of the polyurethane pressure-sensitive adhesive composition from the initial performance value to the target performance value.

[0027] Furthermore, preferentially adjust the proportion of the component corresponding to the component identifier with a small initial component adjustment priority; it can be understood that: the smaller the initial component adjustment priority, the earlier the adjustment order of the proportion of the component corresponding to the component identifier with the initial component adjustment priority, for example: the initial component adjustment priority corresponding to component identifier 3 is 1, and the initial component adjustment priorities corresponding to other component identifiers are all greater than 1, which means that in the process of adjusting the performance value corresponding to each performance of the polyurethane pressure-sensitive adhesive composition from the initial performance value to the target performance value, first adjust the proportion of the component corresponding to component identifier 3.

[0028] In a specific embodiment, the initial component adjustment priority is determined by those skilled in the art according to the cost required to adjust each component (including material cost, process complexity, and process difficulty); the lower the comprehensive cost required to adjust the component corresponding to the component identifier with the initial component adjustment priority, the smaller the initial component adjustment priority, and the comprehensive cost can be calculated from the quantified material cost, process complexity, and process difficulty. Preferentially adjusting the component with a small comprehensive cost is beneficial to cost savings.

[0029] Specifically, the intermediate component proportion is the proportion of the component output by the component prediction model; it can be understood as: the proportion of each component in the polyurethane pressure-sensitive adhesive composition corresponding to the list of predicted target performance values.

[0030] Optionally, the component prediction model is a model obtained by a person skilled in the art through training a multi-layer perceptron (MLP) for the component ratio prediction task. Among them, the MLP is a feedforward artificial neural network model that can map multiple input data sets to a single output data set and can be used to process complex non-linear relationships. Through a large amount of training data, the MLP model can learn the relationship between the existing component ratios, performance values, component adjustment priorities and the target performance value, so as to accurately predict the ratios of each component and output them.

[0031] S2. Input E into the performance prediction model to obtain the corresponding intermediate performance value list F = {F1, F2,..., F j , …, F n}, where F j is the intermediate performance value corresponding to R j in F.

[0032] Specifically, the intermediate performance value is the performance value output by the performance prediction model; it can be understood as: the performance value of predicting the performance of the polyurethane pressure-sensitive adhesive composition prepared based on each intermediate component ratio in the intermediate component ratio list.

[0033] Optionally, the performance prediction model is a model obtained by a person skilled in the art through training the support vector regression (SVR) algorithm for the performance prediction task. SVR is a regression method of support vector machine (SVM) with good generalization ability, which can effectively process small sample data, high-dimensional data and non-linear relationships. SVR can fit the data points by constructing an optimal hyperplane and allow a certain error range, so as to accurately predict the performance values corresponding to each performance according to the ratios of all components output by the component prediction model.

[0034] S3. If ∑ n j＝1 (K j ×|V j -F j |) ≤ G, then take E as the target component ratio list; otherwise, input E, F and V into the priority prediction model to obtain the corresponding intermediate component adjustment priority list H of F and enter step S4. K j is the preset importance weight corresponding to R j , and G is the preset performance difference threshold. Among them, H includes the intermediate component adjustment priorities corresponding to each component identifier in A.

[0035] Specifically, the larger the preset importance weight, the higher the importance degree of the performance corresponding to the performance identifier corresponding to the preset importance weight.

[0036] In a specific embodiment, the preset importance weight is determined by those skilled in the art according to the importance degree of the properties of the polyurethane pressure-sensitive adhesive composition, which will not be elaborated here.

[0037] In a specific embodiment, the usage scenario information corresponding to the polyurethane pressure-sensitive adhesive composition is input into the weight acquisition model to obtain the preset importance weight corresponding to each performance identifier of the polyurethane pressure-sensitive adhesive composition. The weight acquisition model is a neural network model trained by those skilled in the art for the weight acquisition task, which will not be elaborated here.

[0038] Specifically, the usage scenario information corresponding to the polyurethane pressure-sensitive adhesive composition is information related to the scenario where the polyurethane pressure-sensitive adhesive composition is used; for example: the temperature, humidity, area, etc. of the scenario where the polyurethane pressure-sensitive adhesive composition is used.

[0039] Specifically, the preset performance difference threshold is determined by those skilled in the art according to the importance degree of the properties of the polyurethane pressure-sensitive adhesive composition and the allowable error corresponding to the performance value of the target performance of the polyurethane pressure-sensitive adhesive composition, which will not be elaborated here.

[0040] Through the above steps, it is determined whether to use the intermediate component ratio list as the target component ratio list according to the preset importance weight, intermediate performance value, target performance value, and preset performance difference threshold corresponding to the performance identifier. The greater the preset importance weight, the higher the importance degree of the performance corresponding to the performance identifier corresponding to the preset importance weight, fully considering the importance degree of the performance. The importance degree of the performance is also different in different usage scenarios. By flexibly adjusting the preset importance weight according to different usage scenarios, the requirements of different usage scenarios can be met, which is beneficial to improving the accuracy of the obtained target component ratio list.

[0041] Specifically, the target component ratio list includes the target component ratio corresponding to each component identifier in A. The target component ratio corresponding to the component identifier can be understood as: in the intermediate component ratio list corresponding to the target component ratio list, the intermediate component ratio corresponding to the component identifier.

[0042] Specifically, the intermediate component adjustment priority is the component adjustment priority output by the priority prediction model; it can be understood as: the adjustment order of the ratio of the component corresponding to the component identifier in the process of predicting the adjustment of the performance values corresponding to the various properties of the polyurethane pressure-sensitive adhesive composition from the intermediate performance value to the target performance value.

[0043] Further, preferentially adjust the ratio of the component corresponding to the component identifier corresponding to the adjustment priority of the small intermediate component; it can be understood that: the smaller the adjustment priority of the intermediate component, the earlier the adjustment order of the ratio of the component corresponding to the component identifier corresponding to the adjustment priority of the intermediate component; for example: the adjustment priority of the intermediate component corresponding to component identifier 3 is 1, and the adjustment priorities of the intermediate components corresponding to other component identifiers are all greater than 1, which means that in the process of adjusting the performance values corresponding to the respective performances of the polyurethane pressure-sensitive adhesive composition from the intermediate performance values to the target performance values, first adjust the ratio of the component corresponding to component identifier 3.

[0044] Optionally, the priority prediction model is a model obtained by a person skilled in the art through training a decision tree for the task of obtaining the component adjustment priority. The decision tree is a rule-based learning algorithm that can perform rule learning and decision-making based on various input conditions (such as the predicted ratios of the respective components, the predicted performance values of the respective performances, and the ultimately required performance values of the respective performances), so as to output a reasonable component adjustment priority.

[0045] In a specific embodiment, input B, C, and V into the priority prediction model to obtain D, and use the priority prediction model to obtain the initial component adjustment priority list, which reduces human intervention and improves the optimization efficiency.

[0046] S4. Take E as B, take F as C, take H as D, and enter step S1.

[0047] Specifically, after obtaining the target component ratio list, prepare the polyurethane pressure-sensitive adhesive composition based on the respective target component ratios in the target component ratio list to achieve the optimization of the component ratio of the polyurethane pressure-sensitive adhesive composition.

[0048] Through the above steps, the three models of the component prediction model, the performance prediction model, and the priority prediction model are used to perform iterative processing on the list of component ratios to be optimized, the list of initial performance values, the list of initial component adjustment priorities, and the list of target performance values, obtain the list of target component ratios, thereby realizing the optimization of the component ratio of the polyurethane pressure-sensitive adhesive composition, gradually reducing the error through the closed-loop feedback mechanism, reducing human intervention, and fully considering the influence of the component adjustment order, which is beneficial to improving the optimization efficiency and accuracy.

[0049] In a specific embodiment, before step S1, the following step S01 is further included:

[0050] S01. Obtain the count value count, and the initial value of count is 0.

[0051] After step S2, the following steps S21-S25 are included:

[0052] S21, if count < L, insert F into the intermediate performance value list set and proceed to step S24, where L is the preset cycle number threshold; if count ≥ L, insert M in the current intermediate performance value list set M p-L+1 , M p-L+2 , …, M p as a candidate performance value list to obtain a candidate performance value list set N = {N1, N2, ..., N y ,…,N q},N y = {N y1 , N y2 ,…,N yj ,…,N yn}, where M = {M1, M2, ..., M x , …, M p},M x is the xth intermediate performance value list in M, x ranges from 1 to p, p is the number of intermediate performance value lists in M, M p-L+1 is the list of the p-L+1th intermediate performance values ​​in M, M p-L+2 is the list of the p-L+2th intermediate performance values ​​in M, N y is the yth candidate performance value list, y ranges from 1 to q, q is the number of candidate performance value lists, N yj N y Medium R j Corresponding candidate performance values, among which, those skilled in the art know that the preset cycle number threshold is a value pre-set by those skilled in the art according to actual needs, for example: 3, 4, 5, which will not be repeated here.

[0053] Specifically, the intermediate performance value list set is initially NULL.

[0054] Specifically, q=L.

[0055] Specifically, N y Medium R j The corresponding candidate performance value can be understood as y The corresponding intermediate performance value list R j The corresponding intermediate performance value.

[0056] S22. Get N y The corresponding target performance difference T y , T y Meet the following conditions:

[0057] T y =∑ n j＝1 (K j ×|V j -N yj |).

[0058] S23. If T1 ≤ G, T2 ≤ G, …, T y ≤ G, …, T p ≤ G, then take the intermediate component ratio list corresponding to the intermediate performance value list of min(T1, T2, …, T y , …, T p ) as the target component ratio list; otherwise, go to step S24. min() is the minimum value acquisition function.

[0059] S24. Input E, F, and V into the priority prediction model to obtain H.

[0060] S25. Let count = count + 1, take E as B, take F as C, take H as D, and go to step S2.

[0061] Through the above steps, set the count value. According to the continuously updated count value and the preset loop count threshold, obtain the set of intermediate performance value lists. When the count value is not less than the preset loop count threshold, extract the last L intermediate performance value lists from the set of intermediate performance value lists as the candidate performance value lists, which ensures that only the most recently obtained intermediate performance value lists are considered in the optimization process, facilitating the improvement of optimization efficiency and accuracy; obtain the target performance difference corresponding to each candidate performance value list. If there is a target performance difference greater than the preset performance difference threshold, it indicates that there is an intermediate component ratio list corresponding to the intermediate performance value lists obtained in the last L times that cannot be used as the target component ratio list. At this time, update the count value to update the set of intermediate performance value lists and re-obtain the target component ratio list. If all the target performance differences are not greater than the preset performance difference threshold, it indicates that the intermediate component ratio lists corresponding to the intermediate performance value lists obtained in the last L times can all be used as the target component ratio lists, ensuring that the final result is optimal or near-optimal in multiple performance indicators. At this time, take the smallest target performance difference as the intermediate component ratio list corresponding to the corresponding intermediate performance value list as the target component ratio list, further improving the optimization accuracy.

[0062] In a specific embodiment, before step S1, the following steps are further included to obtain B:

[0063] S001. Obtain the original component ratio list set Q and the original performance value list set U corresponding to Q, where Q = {Q1, Q2, …, Q e , …, Q f}, Q e is the e-th original component ratio list, the value of e ranges from 1 to f, f is the number of original component ratio lists, U = {U1, U2, …, U e , …, U f}, Ue For Q e The corresponding list of original performance values.

[0064] Specifically, the original ingredient ratio list is the ingredient ratio list corresponding to one of several existing polyurethane pressure-sensitive adhesive compositions, and includes the original ingredient ratios corresponding to each ingredient identifier in A.

[0065] Specifically, the original ingredient ratio corresponding to the ingredient identifier is the ratio of the ingredient corresponding to the ingredient identifier in the polyurethane pressure-sensitive adhesive composition corresponding to its corresponding original ingredient ratio list.

[0066] Specifically, two different polyurethane pressure-sensitive adhesive compositions corresponding to different original ingredient ratio lists.

[0067] Specifically, the original performance value list includes R1, R2, …, R j , …, R n The corresponding original performance values.

[0068] Specifically, the original performance value in the original performance value list is the performance value of the polyurethane pressure-sensitive adhesive composition prepared based on each original ingredient ratio in the original ingredient ratio list corresponding to its corresponding original performance value list. It can be understood as: the performance value of the polyurethane pressure-sensitive adhesive composition corresponding to the original ingredient ratio list corresponding to its corresponding original performance value list.

[0069] S002. Obtain sample data and construct a sample data set X, where (Q e , U e , U r ) is used as the input feature in the sample data, and Z er is used as the output value corresponding to the input feature (Q e , U e , U r ) in the sample data. U r is the list of original performance values corresponding to Q r , Q r is the r-th original ingredient ratio list, where r ranges from 1 to f and r ≠ e. Z er is the cost value required to adjust the original ingredient ratios in Q e so that Q e is the same as Q r , and thus U e is the same as U r .

[0070] Specifically, the cost value is calculated by quantifying the material cost, process complexity, and process difficulty involved in adjusting the ratios of each ingredient.

[0071] Specifically, if Z er is a positive number, it indicates that the original component ratio in Q e is adjusted so that Q e is the same as Q r , and further so that U e is the same as U r . In this case, an additional cost corresponding to Z er needs to be paid (including material cost, process complexity, and process difficulty); if Z er is 0, it indicates that the original component ratio in Q e is adjusted so that Q e is the same as Q r , and further so that U e is the same as U r . In this case, no additional cost needs to be paid; if Z er is a negative number, it indicates that the original component ratio in Q e is adjusted so that Q e is the same as Q r , and further so that U e is the same as U r . In this case, the cost corresponding to Z er needs to be reduced.

[0072] Specifically, X includes f×(f - 1) sample data.

[0073] Specifically, one sample data includes an input feature and the output value corresponding to the input feature.

[0074] S003. Train a neural network model according to X to obtain a cost prediction model, and the cost prediction model can predict the target cost according to the input feature.

[0075] S004. Use (Q e , U e , V) as the target input feature corresponding to Q e and input it into the cost prediction model to obtain the target cost DJ e corresponding to Q e .

[0076] S005. Use the original component ratio list corresponding to min(Q1, Q2, …, Q e , …, Q f ) as B.

[0077] Through the above steps, a sample data set is constructed based on the original ingredient ratio list set, the original performance value list set, and the cost value calculated by considering the material cost, process complexity, and process difficulty involved in adjusting the ratios of various ingredients after quantization. This fully takes into account the material cost, process complexity, and process difficulty of the ingredients and enables the cost value to be objectively measured. The neural network model is trained based on the sample data set to obtain a cost value prediction model, enabling the neural network model to capture complex input-output relationships, thereby improving the accuracy of the target cost value prediction. Based on the original ingredient ratio list, the original performance value list corresponding to the original ingredient ratio list, the target performance value list, and the cost value prediction model, the target cost value corresponding to the original ingredient ratio list is obtained. The target cost value can represent the cost required to adjust the original performance value list corresponding to the original ingredient ratio list to the target performance value list. Therefore, the original ingredient ratio list corresponding to the minimum target cost value is used as the ingredient ratio list to be optimized. Selecting the original ingredient ratio list that requires the least cost as the ingredient ratio list to be optimized can save costs and avoid unnecessary resource consumption.

[0078] In step S005, it further includes: taking the polyurethane pressure-sensitive adhesive composition corresponding to the original ingredient ratio list of min(Q1, Q2,..., Q e ,..., Q f ) as the polyurethane pressure-sensitive adhesive composition to be optimized.

[0079] The present invention provides a method for optimizing the composition ratio of a polyurethane pressure-sensitive adhesive composition. The method inputs a list of composition ratios to be optimized, a list of initial performance values corresponding to the list of composition ratios to be optimized, a list of initial component adjustment priorities corresponding to the list of initial performance values, and a list of target performance values into a component prediction model to obtain an intermediate composition ratio list; inputs the intermediate composition ratio list into a performance prediction model to obtain a list of intermediate performance values corresponding to the intermediate composition ratio list. If the intermediate performance values in the list of intermediate performance values meet the conditions, the intermediate composition ratio list is used as the target composition ratio list. Otherwise, the intermediate composition ratio list, the list of intermediate performance values, and the list of target performance values are input into a priority prediction model to obtain a list of intermediate component adjustment priorities corresponding to the list of intermediate performance values. The intermediate composition ratio list is used as the list of composition ratios to be optimized, the list of intermediate performance values is used as the list of initial performance values, and the list of intermediate component adjustment priorities is used as the list of initial component adjustment priorities and the target composition ratio list is obtained again. It can be seen that the present invention iteratively processes the list of composition ratios to be optimized, the list of initial performance values, the list of initial component adjustment priorities, and the list of target performance values according to the component prediction model, the performance prediction model, and the priority prediction model to obtain the target composition ratio list, thereby realizing the optimization of the composition ratio of the polyurethane pressure-sensitive adhesive composition, gradually reducing the error through a closed-loop feedback mechanism, reducing human intervention, and fully considering the influence of the component adjustment order, which is beneficial to improving the optimization efficiency and accuracy.

[0080] An embodiment of the present invention further provides a non-transitory computer-readable storage medium, which can be set in an electronic device to store a computer program related to a method for implementing a method in the method embodiment. The computer program is loaded and executed by the processor to implement the method provided in the above embodiment.

[0081] An embodiment of the present invention further provides an electronic device, including: a processor, a memory, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method provided in the above embodiment is implemented.

[0082] An embodiment of the present invention further provides a computer program product, which includes program code. When the program product runs on an electronic device, the program code is used to cause the electronic device to execute the steps in the method according to various exemplary embodiments of the present invention described above in this specification.

[0083] Although some specific embodiments of the present invention have been described in detail by way of examples, those skilled in the art should understand that the above examples are only for illustration and not for limiting the scope of the present invention. Those skilled in the art should also understand that various modifications can be made to the embodiments without departing from the scope and spirit of the present invention.

Claims

1. A method for optimizing the composition ratio of a polyurethane pressure-sensitive adhesive composition, characterized in that: The method comprises the following steps: S1, the list of components to be optimized, the list of initial performance values, C, the list of initial component adjustment priorities, D, and the list of target performance values, V = {V1, V2, ..., V j , …, V n } is input into the component prediction model to obtain the intermediate component ratio list E, B includes the component ratio to be optimized corresponding to each component identifier in the component identifier list A corresponding to the polyurethane pressure-sensitive adhesive composition, and C includes R1, R2, ..., R j , …, R n The corresponding initial performance value, R j is the performance identifier of the jth performance of the polyurethane pressure-sensitive adhesive composition, j ranges from 1 to n, n is the number of the performances of the polyurethane pressure-sensitive adhesive composition, D includes the initial component adjustment priority corresponding to each component identifier in A, V j For R j The corresponding target performance value, E includes the intermediate component ratio corresponding to each component identification in A; S2. Input E into the performance prediction model to obtain the intermediate performance value list F corresponding to E = {F1, F2, ..., F j , …, F n }, F j R in F j The corresponding intermediate performance value; S3, if n j＝1 (K j ×|V j -F j |)≤G, then E is used as the target component ratio list, otherwise, E, F and V are input into the priority prediction model to obtain the intermediate component adjustment priority list H corresponding to F and enter step S4, K j For R j The corresponding preset importance weight, G is the preset performance difference threshold, and H includes the intermediate component adjustment priority corresponding to each component identifier in A; S4. Set E as B, F as C, H as D and go to step S1.

2. The method for optimizing the composition ratio of the polyurethane pressure-sensitive adhesive composition according to claim 1, characterized in that: A includes several component identifiers of components that can be used to form the polyurethane pressure-sensitive adhesive composition.

3. The method for optimizing the composition ratio of the polyurethane pressure-sensitive adhesive composition according to claim 2, characterized in that: The component ratio to be optimized corresponding to the component identification is the ratio of the component corresponding to the component identification in the polyurethane pressure-sensitive adhesive composition to be optimized.

4. The method for optimizing the composition ratio of the polyurethane pressure-sensitive adhesive composition according to claim 1, characterized in that: The initial performance value is a performance value of the polyurethane pressure-sensitive adhesive composition prepared based on each component ratio to be optimized in the list of component ratios to be optimized.

5. The method for optimizing the composition ratio of the polyurethane pressure-sensitive adhesive composition according to claim 2, characterized in that: The initial component adjustment priority is used to indicate the order of adjusting the proportions of components corresponding to the component identifiers in the process of adjusting the performance values ​​corresponding to various properties of the polyurethane pressure-sensitive adhesive composition from initial performance values ​​to target performance values.

6. The method for optimizing the composition ratio of the polyurethane pressure-sensitive adhesive composition according to claim 5, characterized in that: The smaller the initial component adjustment priority is, the earlier the order of adjusting the proportion of the component corresponding to the component identifier corresponding to the initial component adjustment priority is.

7. The method for optimizing the composition ratio of the polyurethane pressure-sensitive adhesive composition according to claim 1, characterized in that: The larger the preset importance weight is, the higher the importance of the performance corresponding to the performance identifier corresponding to the preset importance weight is.

8. The method for optimizing the composition ratio of the polyurethane pressure-sensitive adhesive composition according to claim 1, characterized in that: Before step S1, the method further includes the following step S01: S01. Obtain a count value count, where count is initially 0.

9. The method for optimizing the composition ratio of the polyurethane pressure-sensitive adhesive composition according to claim 8, characterized in that: S2 then includes the following steps S21-S25: S21, if count < L, insert F into the intermediate performance value list set and proceed to step S24, where L is the preset cycle number threshold; if count ≥ L, insert M in the current intermediate performance value list set M p-L+1 , M p-L+2 , …, M p as a candidate performance value list to obtain a candidate performance value list set N = {N1, N2, ..., N y ,…,N q },N y = {N y1 , N y2 ,…,N yj ,…,N yn }, where M = {M1, M2, ..., M x , …, M p },M x is the xth intermediate performance value list in M, x ranges from 1 to p, p is the number of intermediate performance value lists in M, M p-L+1 is the list of the p-L+1th intermediate performance values ​​in M, M p-L+2 is the list of the p-L+2th intermediate performance values ​​in M, N y is the yth candidate performance value list, y ranges from 1 to q, q is the number of candidate performance value lists, N yj N y Medium R j The corresponding candidate performance values; S22. Get N y The corresponding target performance difference T y , T y Meet the following conditions: T y =∑ n j＝1 (K j ×|V j -N yj |); S23. If T1≤G, T2≤G, ..., T y ≤G,…,T p ≤G, then min(T1, T2, …, T y ,…,T p ) The intermediate component ratio list corresponding to the intermediate performance value list corresponding to the intermediate performance value list is used as the target component ratio list, otherwise, enter step S24, min() is the minimum value acquisition function; S24, input E, F, and V into the priority prediction model to obtain H; S25. Let count=count+1, take E as B, F as C, H as D and go to step S2.

10. The method for optimizing the composition ratio of the polyurethane pressure-sensitive adhesive composition according to claim 9, characterized in that: The intermediate performance value list set is initially NULL.

Images