An antioxidant raw material ratio intelligent control method and system

Abstract

The present invention relates to the technical field of intelligent control of raw material ratios, and more specifically, to an intelligent control method and system for antioxidant raw material ratios, the method comprising: obtaining a raw material ratio range for preparing an antioxidant, obtaining a solution space of a particle swarm algorithm, randomly selecting a number of particles in the solution space, iteratively updating the position of each particle, and in any updating process, if the reference weight of any position in the solution space is less than a preset threshold, then in the next updating process, the position is removed from the solution space, and the position of each particle is updated in the remaining positions until the position of each particle converges, and an optimal raw material ratio is obtained, so that when preparing an antioxidant, the raw material ratio is controlled based on the optimal raw material ratio. The present invention can improve the efficiency of determining the optimal raw material ratio while ensuring the accuracy of the optimal raw material ratio.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of raw material ratios, and more specifically, to an intelligent control method and system for antioxidant raw material ratios. Background Art

[0002] Antioxidants are of great significance in inhibiting the oxidation of materials and extending their service life, especially in the fields of plastics and rubber. Due to the different characteristics of different antioxidant raw materials, accurate raw material ratios can give full play to the advantages of each component, improve the effectiveness of antioxidants and reduce costs. With the help of intelligent control technology, the best ratio scheme for the preparation of antioxidants can be found quickly and accurately, thereby improving the quality stability of antioxidants and reducing resource waste.

[0003] In the related technology, such as the patent application document with publication number CN118707912A, a method for intelligent optimization of the production and processing technology of a sleep patch is disclosed. The process of determining the optimal ratio in this method is as follows: obtaining a sleep patch raw material database, a target product parameter set, and quality constraints, and determining the optimal ratio through an adaptive formula optimization algorithm based on the raw material database, the target product parameter set, and quality constraints to obtain the sleep patch raw material formula. Based on the optimal ratio, a continuous microfluidic mixing system is used to mix the raw materials to form a uniform sleep patch drug mixture.

[0004] However, when determining the optimal ratio in the above scheme, it is necessary to gradually adjust the ratio of each raw material through an iterative optimization method until the optimal value of the evaluation index is reached, and it is necessary to ensure that the synergistic effect of each raw material formula is maximized. When the conditions are not met, it is necessary to optimize again, which is inefficient. In addition, the use of iterative optimization may fall into a local optimal solution rather than a global optimal solution, which may reduce the efficiency and accuracy of the optimal ratio to a certain extent. Summary of the invention

[0005] In order to solve the problem of low efficiency and accuracy when determining the optimal ratio of each raw material, the present invention provides an antioxidant raw material ratio intelligent control method and system.

[0006] According to a first aspect of the present invention, there is provided a method for intelligently controlling the ratio of antioxidant raw materials, comprising:

[0007] Obtain the raw material ratio range for preparing antioxidants and obtain the solution space of the particle swarm algorithm;

[0008] A number of particles are randomly selected in the solution space, and the position of each particle is iteratively updated. In any updating process, if the reference weight of any position in the solution space is less than a preset threshold, the position is removed from the solution space in the next updating process, and the position of each particle is updated in the remaining positions until the position of each particle converges to obtain the optimal raw material ratio, so as to control the ratio of raw materials based on the optimal raw material ratio when preparing antioxidants;

[0009] The process of obtaining the reference weight is as follows: in any update process, the possibility of the existence of the optimal solution at any position in the solution space and the density of the positions searched by particles within the preset neighborhood are obtained, the product of the possibility and the density is taken as the first product, the difference between 1 and the density is calculated, the product of the preset value and the difference is taken as the second product, the sum of the first product and the second product is calculated, and the reference weight of any position in the update process is obtained.

[0010] The present invention utilizes the reference weight of each position in the solution space to remove positions with a low possibility of having an optimal solution of the raw material ratio, thereby narrowing the search range of particles, thereby improving the efficiency of particles in searching for the optimal raw material ratio to a certain extent, and avoiding premature convergence of particles and falling into a local optimal solution, thereby ensuring the accuracy of the determined optimal raw material ratio.

[0011] Preferably, a method for obtaining the possibility that an optimal solution exists at any position in the solution space includes:

[0012] In any updating process, the initial degree of any position in the solution space and the initial degree of each position in the neighborhood of any position are obtained, wherein the initial degree reflects the distance between the position and the position with different fitness values ​​searched by the particle, and the fitness value is the cost performance of the antioxidant prepared under the corresponding raw material ratio;

[0013] Calculate the average value of the initial degrees of all positions in the neighborhood, multiply the initial degree of any position and the difference between the initial degree of each position in the neighborhood and the average value, and use the normalized value of the average value of the calculation results as the possibility that any position in the update process has an optimal solution.

[0014] The present invention integrates various data to determine the possibility of the existence of left and right solutions at each position in the solution space, thereby ensuring the accuracy of the determined possibility.

[0015] Preferably, there is a possibility of an optimal solution at any position, satisfying the following relationship:

[0016] ;

[0017] In the formula, For the In the first update process, the solution space The possibility that there is an optimal solution at each position; For the In the first update process, the solution space The initial degree of each position; For the In the first update process, the solution space The set of all locations within the neighborhood of a location; For the In the first update process, the solution space The location in the neighborhood of The possibility that there is an optimal solution at each position; For the In the first update process, the solution space The average of the initial degrees of all positions in the neighborhood of a position; For the In the first update process, the solution space The number of locations within the neighborhood of a location; is the absolute value symbol; is the normalization function.

[0018] The present invention can limit the range of the possibility of the existence of the optimal solution at each position in the solution space to within 0-1, thereby reducing the difficulty of calculation.

[0019] Preferably, the initial degree satisfies the following relationship:

[0020] ;

[0021] In the formula, No. In the first update process, the solution space The initial degree of each position; For the During the update process, the particle searches for The fitness value of each position; For the In the first update process, the solution space The position is the same as the particle's searched The distance between the locations; is the normalization function; For the The number of positions that a particle searches during each update.

[0022] The present invention can avoid the influence of different magnitudes of distance and fitness values ​​by normalization, thereby ensuring the accuracy of the initial level.

[0023] Preferably, the method for obtaining the fitness value includes:

[0024] Obtain the average effective time of the prepared antioxidant applied to different materials under the corresponding raw material ratio at any position in the solution space, as well as the preparation cost of the prepared antioxidant;

[0025] The preset first weight and the preset second weight are used to perform weighted summation on the average effective time and the preparation cost to obtain the fitness value of any position.

[0026] Preferably, the density is a relative density, and a method for obtaining the relative density includes:

[0027] In any update process, obtain the number of positions searched by particles in the neighborhood of any position in the solution space, as well as the cumulative number of updates;

[0028] Calculate the ratio of the number to the cumulative number of updates and normalize it to get the relative density.

[0029] Preferably, the reference weight satisfies the following relationship:

[0030] ;

[0031] In the formula, For the In the first update process, the solution space The reference weight of each position; is the relative density of the positions searched by the particle within the neighborhood of the position; For the In the first update process, the solution space The possibility that there is an optimal solution at each position; is the default value.

[0032] The calculation formula of the reference weight provided by the present invention can control the reference weight to a preset value when the value of the relative density is extremely low, thereby ensuring the stability and reliability of the calculation formula.

[0033] Preferably, the method for obtaining the optimal raw material ratio includes:

[0034] When the positions of the particles converge, the raw material ratio corresponding to the position with the largest fitness value among the positions searched by the particles is taken as the optimal raw material ratio for preparing the antioxidant.

[0035] According to a second aspect of the present invention, there is provided an antioxidant raw material ratio intelligent control system, comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor executes the computer program to implement the steps of the first aspect of the present invention.

[0036] The present invention has the following effects:

[0037] The present invention calculates the reference weight of each position in the solution space, sets a threshold, and screens the positions that meet the conditions, so as to accurately and effectively narrow the search range of the particles, thereby improving the efficiency of the particles in searching for the optimal raw material ratio in the solution space; and when determining the reference weight, the possibility of the existence of the optimal solution at the corresponding position and the density of the positions searched by the particles in the neighborhood are combined, so as to ensure the accuracy of the determined reference weight; and it can avoid the particles from converging too early and falling into the local optimal solution, so as to ensure the accuracy of the determined optimal raw material ratio. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0039] Figure 1 The present invention is a schematic diagram of the steps of an intelligent control method for the ratio of antioxidant raw materials according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] 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 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.

[0041] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0042] Reference Figure 1 , an antioxidant raw material ratio intelligent control method, comprising steps S1-S2, specifically as follows:

[0043] S1: Obtain the raw material ratio range for preparing antioxidants and obtain the solution space of the particle swarm algorithm.

[0044] It should be noted that, in order to refine the search for the optimal value of the raw material ratio and clarify the solution space range of the particle swarm algorithm, the present invention expresses each raw material ratio for preparing the antioxidant in the form of 4 decimal places, and the sum of all raw material ratios for preparing the antioxidant is 1.

[0045] Specifically, the value range of the raw material ratio for preparing antioxidants can be used as the solution space of the particle swarm algorithm, and then the sum of all raw material ratios is 1 as a constraint condition to discretize the solution space so that the minimum adjustment unit of each position in the solution space is 0.0001, thereby obtaining a finite number of positions in the solution space.

[0046] The solution space is a professional term in the particle swarm algorithm, which is not described in detail in this embodiment. Each position in the solution space corresponds to a value of the raw material ratio for preparing the antioxidant.

[0047] S2: Randomly select a number of particles in the solution space, and iteratively update the position of each particle. During any update process, if the reference weight of any position in the solution space is less than a preset threshold, then in the next update process, the position is removed from the solution space, and the position of each particle is updated in the remaining positions until the position of each particle converges to obtain the optimal raw material ratio, so that the raw material ratio can be controlled based on the optimal raw material ratio when preparing the antioxidant.

[0048] It should be noted that, due to the large scope of the solution space, the process of searching for the best solution in the solution space using the traditional particle swarm algorithm needs to traverse each position in the solution space, and when iteratively updating the position of the particle, it depends on the cost-effectiveness of the antioxidant prepared under the raw material ratio corresponding to the position of the particle, that is, each time the position of the particle is updated, the antioxidant needs to be prepared once, the experimental cost is large, and the efficiency is low. Therefore, the present invention improves the traditional particle swarm algorithm, reduces the search range of each particle when iteratively updating the position of the particle, thereby reducing the number of times the antioxidant is prepared.

[0049] Optionally, by randomly selecting a number of particles in the solution space, the particle swarm algorithm can be prevented from falling into a local optimal solution too early, ensuring that the particles can effectively search the entire solution space. The number of particles selected in this embodiment is 300, and this embodiment does not specifically limit the number of particles.

[0050] The reference weight refers to the importance of each position in the solution space when evaluating the iterative update of the particle position, so as to remove positions with less importance to shorten the search range of each particle. For example, in any update process, if the reference weight of any position in the solution space is less than a preset threshold, such as 0.3, then in the next update process, the position can be removed, that is, each particle does not need to search at the position, thereby narrowing the search range of each particle and improving the search efficiency. This embodiment does not specifically limit the setting of the threshold.

[0051] Specifically, in any updating process, the steps of obtaining the reference weights of each position in the solution space include:

[0052] Step 1: In any update process, obtain the possibility that the optimal solution exists at any position in the solution space;

[0053] In an exemplary embodiment of the present invention, the possibility of the existence of an optimal solution at any position in the solution space can be determined by the following steps:

[0054] (1) In any updating process, the initial degree of any position in the solution space and the initial degree of each position within the preset neighborhood of any position are obtained, wherein the initial degree reflects the distance between the position and the positions with different fitness values ​​searched by the particle, and the fitness value is the cost-effectiveness of the antioxidant prepared under the corresponding raw material ratio;

[0055] Among them, the initial degree refers to the data of the distance between each position in the solution space and the position with a larger fitness value searched by the particle, which is used to preliminarily evaluate the possibility that each position in the solution space has an optimal solution. If any position in the solution space is relatively close to the position with a larger fitness value searched by the particle during any update process, it means that the possibility of the optimal solution at this position is high. It should be noted that the fitness value is a professional term for the particle swarm algorithm, and this embodiment will not be described in detail here. This embodiment uses the cost-effectiveness of the antioxidant prepared under the corresponding raw material ratio at each position in the solution space as the fitness value of the corresponding position.

[0056] Optionally, the preset neighborhood range of any position in the solution space is a range with the position as the center and a neighborhood radius of 0.2. Of course, a suitable neighborhood range can also be set according to specific circumstances. This embodiment does not specifically limit the size of the neighborhood range.

[0057] Specifically, during any update process, the initial degree of any position in the solution space satisfies the following relationship:

[0058] ;

[0059] In the formula, No. In the first update process, the solution space The initial degree of each position; For the During the update process, the particle searches for The fitness value of each position; For the In the first update process, the solution space The position is the same as the particle's searched The distance between the locations; is the normalization function; For the The number of positions that a particle searches during each update.

[0060] Optionally, the distance between any position in the solution space and any position searched by the particle may be a Euclidean distance or a Manhattan distance. This embodiment does not specifically limit the type of distance.

[0061] It should be noted that since the distance between different positions in high-dimensional space is much smaller than the fitness value in magnitude, when obtaining the initial degree of any position in the solution space during each update process, it is necessary to linearly normalize the distance between any position in the solution space and any position searched by the particle, as well as the fitness value of any position searched by the particle, so as to ensure the accuracy of the initial degree.

[0062] In an exemplary embodiment of the present invention, the determination of the fitness value of each particle can be achieved by the following steps:

[0063] The average effective time of the prepared antioxidant applied to different materials and the preparation cost of the prepared antioxidant under the corresponding raw material ratio at any position in the solution space are obtained; the average effective time and the preparation cost are weightedly summed using the preset first weight and the preset second weight to obtain the fitness value of any position.

[0064] It should be noted that antioxidants are substances that can inhibit or slow down the auto-oxidation reaction rate of polymer materials. After adding antioxidants, the longer the service life of the material, the better the antioxidant effect of the antioxidant. In addition, when preparing antioxidants, the preparation cost should be reduced while maintaining a good antioxidant effect. Therefore, determining the fitness value in combination with this feature can ensure the accuracy of the optimal raw material ratio.

[0065] Optionally, the first weight can be set to 2 and the second weight can be set to -1, so that the first weight and the second weight are used to weight the average effective time of the antioxidant prepared in different materials under the corresponding raw material ratio at any position in the solution space, and the preparation cost, so as to obtain the fitness value of the position. This embodiment does not specifically limit the size of the first weight and the second weight, as long as the sum of the first weight and the second weight is 1.

[0066] (2) Calculate the average value of the initial degrees of all positions in the neighborhood, multiply the initial degree of any position and the difference between the initial degree of each position in the neighborhood and the average value, and use the normalized value of the average value of the calculation results as the possibility that any position in the update process has an optimal solution.

[0067] Specifically, in any update process, there is a possibility that an optimal solution exists at any position in the solution space, satisfying the following relationship:

[0068] ;

[0069] In the formula, For the In the first update process, the solution space The possibility that there is an optimal solution at each position; For the In the first update process, the solution space The initial degree of each position; For the In the first update process, the solution space The set of all locations within the neighborhood of a location; For the In the first update process, the solution space The location in the neighborhood of The possibility that there is an optimal solution at each position; For the In the first update process, the solution space The average of the initial degrees of all positions in the neighborhood of a position; For the In the first update process, the solution space The number of locations within the neighborhood of a location; is the absolute value symbol; is the normalization function.

[0070] Optionally, when the initial degrees of the positions within the neighborhood of any position in the solution space are small and the overall difference of the positions within the neighborhood is small, it means that there is a high possibility that there is no optimal solution at that position, that is, the possibility that there is an optimal solution at that position is low.

[0071] Step 2: During any update process, obtain the density of the positions searched by the particle within the neighborhood of any position;

[0072] In an exemplary embodiment of the present invention, the density of positions searched by particles within the neighborhood of any position is a relative density, and the relative density can be determined by the following steps:

[0073] In any update process, the number of positions searched by particles in the neighborhood of any position in the solution space and the cumulative number of updates are obtained; the ratio of the number to the cumulative number of updates is calculated and normalized to obtain the relative density.

[0074] Specifically, in any update process, the relative density of the positions searched by particles in the neighborhood of any position satisfies the following relationship:

[0075] ;

[0076] In the formula, For the In the first update process, the solution space The relative density of positions searched by particles within the neighborhood of a position; The number of positions that the particle has searched within the neighborhood of this position; is the normalization function.

[0077] Step 3: Take the product of the possibility and the density as the first product, calculate the difference between 1 and the density, take the product of the preset value and the difference as the second product, calculate the sum of the first product and the second product, and obtain the reference weight of any position in the update process.

[0078] Specifically, during any update process, the reference weight of any position satisfies the following relationship:

[0079] ;

[0080] In the formula, For the In the first update process, the solution space The reference weight of each position; is the relative density of the positions searched by the particle within the neighborhood of the position; For the In the first update process, the solution space The possibility that there is an optimal solution at each position; is a preset value. .

[0081] Optionally, during any updating process, if the relative density of positions searched by particles in the neighborhood of any position in the solution space is low, and the possibility of the existence of an optimal solution at that position is low, then the reference weight of that position is low, and when the reference weight of that position is lower than a certain value, then that position can be removed during the next updating process, thereby narrowing the search range of the particles and improving the efficiency of determining the optimal raw material ratio for preparing antioxidants.

[0082] Furthermore, the positions with low reference weights in the solution space can be removed during each update process by setting a threshold. For example, during any update process, the reference weight calculation formula can be used to calculate the reference weights of each position in the solution space during the update process, and then a threshold can be set, such as 0.3. When the reference weight of any position is less than 0.3, the position is removed, so that the range of the screened positions can be used as the range that the particles need to search during the update process to narrow the search range of the particles.

[0083] In an exemplary embodiment of the present invention, the determination of the optimal raw material ratio can be achieved by the following steps:

[0084] When the positions of the particles converge, the raw material ratio corresponding to the position with the largest fitness value among the positions searched by the particles is taken as the optimal raw material ratio for preparing the antioxidant.

[0085] Optionally, the present invention does not reduce the number of particles when iteratively updating the positions of particles, which can avoid premature convergence of particles while reducing the number of experiments, thereby improving the efficiency of determining the optimal raw material ratio while ensuring the accuracy of the determined optimal raw material ratio.

[0086] Optionally, after the optimal raw material ratio for preparing the antioxidant is determined, the data can be sent to the control module in real time, thereby realizing intelligent control of the antioxidant raw material ratio.

[0087] The present invention also provides an antioxidant raw material ratio intelligent control system, the system includes a memory and a processor, and a computer program is stored in the memory, the computer program integrates the function of an antioxidant raw material ratio intelligent control method, when the computer program is executed, the antioxidant raw material ratio intelligent control method can improve the efficiency of determining the optimal raw material ratio while ensuring the accuracy of the optimal raw material ratio.

[0088] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0089] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. An antioxidant raw material ratio intelligent control method, characterized in that: include: Obtain the raw material ratio range for preparing antioxidants and obtain the solution space of the particle swarm algorithm; A number of particles are randomly selected in the solution space, and the position of each particle is iteratively updated. In any updating process, if the reference weight of any position in the solution space is less than a preset threshold, the position is removed from the solution space in the next updating process, and the position of each particle is updated in the remaining positions until the position of each particle converges to obtain the optimal raw material ratio, so as to control the ratio of raw materials based on the optimal raw material ratio when preparing antioxidants; The process of obtaining the reference weight is as follows: in any updating process, obtaining the possibility that any position in the solution space has an optimal solution includes: in any updating process, obtaining the initial degree of any position in the solution space, and the initial degree of each position in the neighborhood of any position, wherein the initial degree reflects the distance between the position and the position with different fitness values ​​searched by the particle, and the fitness value is the cost performance of the antioxidant prepared under the corresponding raw material ratio; The average value of the initial degrees of all positions within the neighborhood is calculated, and the initial degree of any position and the difference between the initial degree of each position within the neighborhood and the average value are multiplied, and the normalized value of the average value of the obtained calculation result is used as the possibility that the optimal solution exists at any position in the update process, and the density of the positions searched by particles within the preset neighborhood. The density is a relative density. The method for obtaining the relative density includes: in any update process, obtaining the number of positions searched by particles within the neighborhood of any position in the solution space and the cumulative number of updates; calculating the ratio of the number to the cumulative number of updates, and normalizing it to obtain the relative density, taking the product of the possibility and the density as the first product, calculating the difference between 1 and the density, taking the product of the preset value and the difference as the second product, calculating the sum of the first product and the second product, and obtaining the reference weight of any position in the update process.

2. The method for intelligently controlling the ratio of antioxidant raw materials according to claim 1, characterized in that: The possibility that any of the above positions has an optimal solution satisfies the following relationship: ； In the formula, For the In the first update process, the solution space The possibility that there is an optimal solution at each position; For the In the first update process, the solution space The initial degree of each position; For the In the first update process, the solution space The set of all locations within the neighborhood of a location; For the In the first update process, the solution space The location in the neighborhood of The possibility that there is an optimal solution at each position; For the In the first update process, the solution space The average of the initial degrees of all positions in the neighborhood of a position; For the In the first update process, the solution space The number of locations within the neighborhood of a location; is the absolute value symbol; is the normalization function.

3. The method for intelligently controlling the ratio of antioxidant raw materials according to claim 1, characterized in that: The initial degree satisfies the following relationship: ； In the formula, No. In the first update process, the solution space The initial degree of each position; For the During the update process, the particle searches for The fitness value of each position; For the In the first update process, the solution space The position is the same as the particle's searched The distance between the locations; is the normalization function; For the The number of positions that a particle searches during each update.

4. The method for intelligently controlling the ratio of antioxidant raw materials according to claim 3, characterized in that: The method for obtaining the fitness value includes: Obtaining the average effective time of the prepared antioxidant applied to different materials under the corresponding raw material ratio at any position in the solution space, and the preparation cost of the prepared antioxidant; Using a preset first weight and a preset second weight, weighted sum is performed on the average effective time and the preparation cost to obtain a fitness value of any position; The first weight is a positive number, the second weight is a negative number, and the sum of the first weight and the second weight is 1.

5. The method for intelligently controlling the ratio of antioxidant raw materials according to claim 1 or 2, characterized in that: The reference weight satisfies the following relationship: ； In the formula, For the In the first update process, the solution space The reference weight of each position; is the relative density of the positions searched by the particle within the neighborhood of the position; For the In the first update process, the solution space The possibility that there is an optimal solution at each position; is the default value.

6. The method for intelligently controlling the ratio of antioxidant raw materials according to claim 1, characterized in that: The method for obtaining the optimal raw material ratio comprises: When the positions of the particles converge, the raw material ratio corresponding to the position with the largest fitness value among the positions searched by the particles is used as the optimal raw material ratio for preparing the antioxidant.

7. An intelligent control system for antioxidant raw material ratio, characterized in that: The antioxidant raw material ratio intelligent control system comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 6.

Images