IPA production conversion time optimization method, system and device based on big data fusion algorithm

Through big data fusion algorithms and genetics and ant colony algorithms, IPA concentration and production conversion time are optimized, which solves the problem of IPA concentration control in online production conversion, improves product quality and production efficiency, and reduces time and costs.

CN118014541BActive Publication Date: 2024-12-03浙江思维特数字科技有限公司
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202410203007.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-02-23
Publication Date
2024-12-03
Estimated Expiration
2044-02-23

AI Technical Summary

Technical Problem

During the online conversion process, IPA concentration control is difficult to accurately predict, resulting in product quality not meeting standards, and existing methods rely on experience, high time costs and waste of transition materials.

Method used

Using an algorithm based on big data fusion, a material balance model is constructed by obtaining reactor data, combining genetic algorithms and ant colony algorithms, IPA concentration and production conversion time are automatically optimized, and the optimal adjustment strategy is determined.

Benefits of technology

It significantly reduces the time and workload of manual trial calculations, improves the scientificity and objectivity of the production conversion process, reduces waste, and improves production efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118014541B_ABST
    Figure CN118014541B_ABST
Patent Text Reader

Abstract

The present invention discloses an IPA conversion time optimization method, system and device based on a big data fusion algorithm, the method comprising the following steps: obtaining relevant data of each reactor; determining the flow coefficient based on the mass of each reactor, and determining the feed flow of subsequent reactors and the discharge flow of each reactor in combination with the feed flow of the first reactor; constructing a material balance model through the feed flow of each reactor, the discharge flow of each reactor, the mass of each reactor, the IPA concentration in each reactor and the IPA concentration entering the first reactor; determining the target model to be optimized of the IPA concentration based on the material balance model, and then determining the feed IPA concentration of each reactor in different time periods. The present invention combines conversion with artificial intelligence algorithms to automatically optimize the target to be optimized, can quickly find the optimal conversion time and give an IPA adjustment strategy, greatly reducing the time and workload cost of manual trial calculations, and is more scientific and objective.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence algorithm technology, and in particular to an IPA production conversion time optimization method, system and device based on a big data fusion algorithm. Background Art

[0002] IPA is a colorless volatile liquid with a slight odor. It is miscible with water and ethanol and can form an azeotrope with water. It can be used in antifreeze, quick-drying oil, etc. It can also be used as a solvent for resins and essential oils. In many cases, it can replace ethanol. IPA can also be used as a coating, rosin water, mixed resins, etc. It is colorless and transparent and can be used in pure natural products.

[0003] During the production process, the production changeover operation is a common operation for brand switching. At present, there are two ways to change production: one is to stop production first, clean the system and then change production. This method is used when continuous production equipment needs to be cleaned; the other is online production changeover, that is, switching production without stopping production. This type of production changeover method is suitable for situations where the equipment does not stop production and the product switching time is short. Online switching is conducive to switching varieties, increasing production flexibility and improving economic benefits. In the domestic market, when product needs are varied and brand switching is frequent, online production changeover is the best choice.

[0004] However, during the online conversion process, the ratio of raw and auxiliary materials will be changed, resulting in the material composition of each reactor in the production process, the semi-finished products at the pelletizer production site, and the finished products of the SSP device not meeting the quality standards of superior products, thereby reducing product quality.

[0005] In general, the feed port is only in the first reactor, and the entire process needs to go through the continuous reaction of the second and third reactors before the final concentration control result is reflected in each reactor. Therefore, this is a process with a large time lag, which poses a great challenge to process control. The current IPA concentration process control for the conversion process is based on the full backmixing assumption and the empirical formula of material balance, and will eventually be recorded in tables or data. Based on the recorded data, a certain pattern is found, that is, the outlet IPA concentration of the first reactor, the second reactor, the third reactor, the fourth reactor, and the last reactor is inferred based on parameters such as material quality and feed volume, thereby guiding the conversion process.

[0006] Therefore, the main defect at this stage is that the IPA concentration is obtained by empirical data and these data are adjusted based on the conversion results. It cannot predict at what time point and what concentration of IPA should be added to achieve the final target product concentration, and it strongly depends on the trial and experience of the process personnel. Experienced process personnel can adjust in the direction of the target according to the results of each trial, but for inexperienced process engineers, this process has too much randomness. In addition, it takes several hours to determine the entire conversion plan, which has a high time cost and produces a lot of transition materials, resulting in huge waste. Summary of the invention

[0007] In view of the problems in the prior art, the present invention provides an IPA production conversion time optimization method, system and device based on a big data fusion algorithm.

[0008] In order to solve the above problems, the present invention is solved by the following technical solutions:

[0009] A method for optimizing IPA production conversion time based on big data fusion algorithm, comprising the following steps:

[0010] Obtaining data related to each reactor, wherein the data related to the reactor includes the mass of the material in each reactor, the IPA concentration in each reactor, and the IPA concentration entering the first reactor;

[0011] Determine the flow coefficient based on the mass of each reactor, and determine the feed flow of the subsequent reactors and the discharge flow of each reactor in combination with the feed flow of the first reactor;

[0012] A material balance model was constructed using the feed flow rate of each reactor, the discharge flow rate of each reactor, the mass of each reactor, the IPA concentration in each reactor, and the IPA concentration entering the first reactor;

[0013] Based on the material balance model, the target model to be optimized for IPA concentration is determined, and then the feed IPA concentration of each reactor in different time periods is determined.

[0014] As an implementable method, the flow coefficient is determined based on the mass of each reactor, and the feed flow of the first reactor is combined to determine the feed flow of the subsequent reactors and the discharge flow of each reactor, including the following steps:

[0015] Based on the mass of each reactor, the flow coefficient is obtained by combining the mass of the carboxyl group at the outlet of each reactor and the mass of the slurry IPA;

[0016] The discharge flow rate and feed flow rate of each subsequent reactor are obtained in sequence by combining the flow coefficient with the feed flow rate of the first reactor;

[0017] Wherein, the flow coefficient is expressed as follows:

[0018]

[0019] The discharge flow is expressed as follows:

[0020]

[0021] in, Indicates The feed flow rate of each reactor, Indicates The feed flow rate of each reactor, Indicates the flow coefficient.

[0022] As an implementable embodiment, the material balance model is expressed as follows:

[0023]

[0024] in, Indicates The time required for a reactor to reach the final IPA concentration from the initial state, Indicates The mass of the reactor, Indicates The final IPA concentration in each reactor is and Respectively represent The feed and discharge flow rates of each reactor, Indicates entering IPA concentration of each reactor, , indicating the The IPA concentration in the first reactor is equal to that in the IPA concentration in each reactor feed.

[0025] As an implementable embodiment, the mass of the reactor is expressed as follows:

[0026]

[0027] in, Indicates The mass of the reactor, Indicates The volume of liquid inside the reactor, Indicates The density of the liquid inside the reactor.

[0028] As an implementable method, the preset optimization target model is expressed as follows:

[0029]

[0030] in, Indicates The time required for a reactor to reach the final IPA concentration from the initial state, Indicates The mass of the reactor, and Respectively represent The feed and discharge flow rates of each reactor, Indicates The IPA concentration in each reactor, Indicates Final IPA concentration in each reactor.

[0031] As an operative embodiment, the determination of the feed IPA concentration of each reactor at different time periods comprises the following steps:

[0032] A genetic algorithm model is constructed, the feed IPA concentration is encoded as a chromosome in the genetic algorithm model, and the chromosome is selected based on the pheromone intensity in the ant colony algorithm model to perform a crossover operation to exchange genes between two chromosomes;

[0033] Randomly select chromosomes and genes that meet the preset mutation probability to determine whether they are mutated, and randomly adjust the IPA concentration at a certain time point;

[0034] Then the optimal IPA adjustment strategy and optimal production conversion time are obtained.

[0035] As an implementable method, the genetic algorithm model is constructed, the feed IPA concentration is encoded as a chromosome in the genetic algorithm model, and the chromosome is selected based on the pheromone intensity in the ant colony algorithm model to perform a crossover operation to exchange genes between two chromosomes, including the following steps:

[0036] A genetic algorithm model was constructed to encode the IPA concentration of the feed into chromosomes, and each gene on each chromosome represented the IPA concentration at a certain time point;

[0037] Each chromosome is randomly initialized, and the feed IPA concentration and the corresponding adjustment times at different time points of the production transition are preset;

[0038] Based on the fitness information of the chromosome, the pheromone intensity in the ant colony algorithm model is updated to obtain a new pheromone intensity. Based on the fitness information and the new pheromone intensity, the selection probability is obtained. The selection probability threshold is set and the chromosome is selected to obtain a suitable chromosome and retain it to the next generation.

[0039] The chromosomes retained to the next generation are subjected to a crossover operation to exchange genes between the two chromosomes;

[0040] Wherein, the new pheromone intensity is expressed as follows:

[0041]

[0042] The selection probability is expressed as follows:

[0043]

[0044] represents the probability of selection, Represents chromosome The fitness information of Represents chromosome The fitness information of Represents chromosome The current pheromone strength, Represents chromosome The current pheromone strength, , Belong to 1~ Any number in Represents the total number of chromosomes.

[0045] As an implementable method, performing a crossover operation on the chromosomes retained to the next generation to exchange genes between the two chromosomes comprises the following steps:

[0046] The selection probability is modified, and two chromosomes that meet the conditions are selected for crossover to obtain a chromosome crossover pair. The selection probability is modified to:

[0047]

[0048] , Chromosomes , The fitness information of Represents chromosomes , The pheromone concentration, , It means traversing all possible chromosome pairs. , Belong to 1~ Any number in Indicates the number of all chromosomes;

[0049] The crossover point is determined based on the pheromone intensity, and the one with the highest pheromone intensity is selected as the chromosome crossover point;

[0050] Based on chromosome crossover pairs and chromosome crossover points, the chromosomes are crossover operated.

[0051] As an implementable method, the method of randomly selecting chromosomes and genes that meet a preset mutation probability to determine whether they are mutated, and randomly adjusting the IPA concentration at a certain time point, includes the following steps:

[0052] The initial mutation probability is adjusted based on the pheromone intensity to obtain a new mutation probability;

[0053] Based on the comparison between the mutation probability of each chromosome and the new mutation probability, it is determined whether to perform mutation processing, and the mutation processing is achieved by randomly adjusting the IPA concentration at a certain time point;

[0054] Wherein, the new mutation probability is expressed as follows:

[0055]

[0056] represents the initial mutation probability, represents the adjustment factor, Represents chromosome The current pheromone strength, Indicates the maximum pheromone intensity in the current population.

[0057] An IPA production conversion time optimization method system based on big data fusion algorithm, including a data acquisition module, a coefficient determination module, a model construction module and a target optimization module;

[0058] The data acquisition module is used to acquire data related to each reactor, wherein the data related to the reactor includes the mass of the material in each reactor, the IPA concentration in each reactor, and the IPA concentration entering the first reactor;

[0059] The coefficient determination module determines the flow coefficient based on the mass of each reactor, and determines the feed flow of subsequent reactors and the discharge flow of each reactor in combination with the feed flow of the first reactor;

[0060] The model building module is used to build a material balance model through the feed flow rate of each reactor, the discharge flow rate of each reactor, the mass of each reactor, the IPA concentration in each reactor and the IPA concentration entering the first reactor;

[0061] The target optimization module determines the target model to be optimized of the IPA concentration based on the material balance model, and further determines the feed IPA concentration of each reactor in different time periods.

[0062] A computer program product, characterized in that it comprises computer instructions, which, when executed by a processor, implement the steps of the method described below:

[0063] Obtaining data related to each reactor, wherein the data related to the reactor includes the mass of the material in each reactor, the IPA concentration in each reactor, and the IPA concentration entering the first reactor;

[0064] Determine the flow coefficient based on the mass of each reactor, and determine the feed flow of the subsequent reactors and the discharge flow of each reactor in combination with the feed flow of the first reactor;

[0065] A material balance model was constructed using the feed flow rate of each reactor, the discharge flow rate of each reactor, the mass of each reactor, the IPA concentration in each reactor, and the IPA concentration entering the first reactor;

[0066] Based on the material balance model, the target model to be optimized for IPA concentration is determined, and then the feed IPA concentration of each reactor in different time periods is determined.

[0067] A method and device for optimizing IPA production conversion time based on a big data fusion algorithm, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, any one of the following methods is implemented:

[0068] Obtaining data related to each reactor, wherein the data related to the reactor includes the mass of the material in each reactor, the IPA concentration in each reactor, and the IPA concentration entering the first reactor;

[0069] Determine the flow coefficient based on the mass of each reactor, and determine the feed flow of the subsequent reactors and the discharge flow of each reactor in combination with the feed flow of the first reactor;

[0070] A material balance model was constructed using the feed flow rate of each reactor, the discharge flow rate of each reactor, the mass of each reactor, the IPA concentration in each reactor, and the IPA concentration entering the first reactor;

[0071] Based on the material balance model, the target model to be optimized for IPA concentration is determined, and then the feed IPA concentration of each reactor in different time periods is determined.

[0072] The present invention has significant technical effects due to the adoption of the above technical solution:

[0073] The present invention combines the production change with the artificial intelligence algorithm, establishes a material balance model based on the quality of the reactor and other related data, determines the target model to be optimized, and then obtains the optimization target and optimizes it. In the present invention, the genetic algorithm model and the ant colony algorithm model are used to automatically optimize the target to be optimized, and the optimal time for production change can be found in a relatively short time, and the adjustment strategy of IPA is given, which greatly reduces the time and workload cost of manual trial calculation, and is more scientific and objective. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative labor.

[0075] Figure 1 It is a schematic flow diagram of the method of the present invention;

[0076] Figure 2 is a flow chart of an embodiment of the present invention;

[0077] Figure 3 It is a schematic diagram of the flow chart of the crossover operation of the present invention;

[0078] Figure 4 It is a schematic diagram of a process of judging whether there is a variation and randomly adjusting the IPA concentration at a certain time point in the present invention;

[0079] Figure 5 It is a schematic diagram of the overall process of the method of the present invention;

[0080] Figure 6 It is an overall schematic diagram of the system of the present invention. DETAILED DESCRIPTION

[0081] The present invention is further described in detail below in conjunction with embodiments. The following embodiments are for explanation of the present invention but the present invention is not limited to the following embodiments.

[0082] Embodiment 1:

[0083] A method for optimizing IPA production conversion time based on big data fusion algorithm, such as Figure 1 As shown, the following steps are included:

[0084] S100, obtaining data related to each reactor, wherein the data related to the reactor includes the mass of the material in each reactor, the IPA concentration in each reactor, and the IPA concentration entering the first reactor;

[0085] S200, determining a flow coefficient based on the mass of each reactor, and combining the feed flow of the first reactor to determine the feed flow of subsequent reactors and the discharge flow of each reactor;

[0086] S300, constructing a material balance model through the feed flow rate of each reactor, the discharge flow rate of each reactor, the mass of each reactor, the IPA concentration in each reactor and the IPA concentration entering the first reactor;

[0087] S400, determining a target model to be optimized for IPA concentration based on the material balance model, and then determining the feed IPA concentration of each reactor in different time periods.

[0088] In one embodiment, the method of determining the flow coefficient based on the mass of each reactor and combining the feed flow of the first reactor to determine the feed flow of subsequent reactors and the discharge flow of each reactor includes the following steps:

[0089] Based on the mass of each reactor, the flow coefficient is obtained by combining the mass of the carboxyl group at the outlet of each reactor and the mass of the slurry IPA;

[0090] The discharge flow rate and feed flow rate of each subsequent reactor are obtained in sequence by combining the flow coefficient with the feed flow rate of the first reactor;

[0091] Wherein, the flow coefficient is expressed as follows:

[0092]

[0093] The discharge flow is expressed as follows:

[0094]

[0095] in, Indicates The feed flow rate of each reactor, Indicates The feed flow rate of each reactor, Indicates the flow coefficient.

[0096] In this embodiment, the material balance model is expressed as follows:

[0097]

[0098] in, Indicates The time required for a reactor to reach the final IPA concentration from the initial state, Indicates The mass of the reactor, Indicates The final IPA concentration in each reactor is and Respectively represent The feed and discharge flow rates of each reactor, Indicates entering IPA concentration of each reactor, , indicating the The IPA concentration in the first reactor is equal to that in the IPA concentration in each reactor feed.

[0099] The mass of the reactor is expressed as follows:

[0100]

[0101] in, Indicates The mass of the reactor, Indicates The volume of liquid inside a reactor. Here, this volume is determined based on the solutions of various relevant concentrations and liquid levels. Indicates The density of the liquid inside the reactor.

[0102] Therefore, the preset optimization target model is expressed as follows:

[0103]

[0104] in, Indicates The time required for a reactor to reach the final IPA concentration from the initial state, Indicates The mass of the reactor, and Respectively represent The feed and discharge flow rates of each reactor, Indicates The IPA concentration in each reactor, Indicates Final IPA concentration in each reactor.

[0105] In step S400, the feed IPA concentration of each reactor in different time periods is further determined, including the following steps:

[0106] S410, constructing a genetic algorithm model, encoding the feed IPA concentration as a chromosome in the genetic algorithm model, and selecting the chromosome based on the pheromone intensity in the ant colony algorithm model to perform a crossover operation to exchange genes between two chromosomes;

[0107] S420, randomly selecting chromosomes and genes that meet the preset mutation probability to determine whether they are mutated, and randomly adjusting the IPA concentration at a certain time point;

[0108] S430, and then obtain the optimal IPA adjustment strategy and the optimal production changeover time.

[0109] In this field, the optimization process is to find the most appropriate random adjustment of the IPA concentration at a certain time point, and then obtain the optimal IPA adjustment strategy and the optimal conversion time. Here, the adjustment strategy can be understood as adjusting the time and IPA concentration. Only by adjusting the strategy can the optimal conversion time be obtained. This whole set of logic can ensure the highest efficiency and the best time control of the conversion. The existing genetic algorithm is a search algorithm used to solve optimization in computational mathematics, and is a type of evolutionary algorithm. There is a good solution to the optimization problem, that is, to evolve a population of abstract representations (called chromosomes) of a certain number of candidate solutions (called individuals) to a better solution. Generally, binary representation (that is, a string of 0 and 1) is used, and evolution starts from a population of completely random individuals, and then occurs generation after generation. In each generation, the fitness of the entire population is evaluated, and multiple individuals are randomly selected from the current population (based on their fitness), and a new life population is generated through natural selection and mutation. This population becomes the current population in the next iteration of the algorithm. So, how can the IPA concentration in this field and the genetic algorithm be combined?

[0110] First, the decision variable of the optimization problem (i.e., the IPA concentration of the feed) is encoded into a chromosome, where binary encoding or real number encoding can be used, and real number encoding is applied in this application;

[0111] Design chromosome structure: Each chromosome represents a potential solution. The chromosome can be composed of a series of genes representing the operating parameters at different time points. In this field, a gene can represent the IPA concentration at a specific time point.

[0112] Generate an initial population: The initial population consists of multiple randomly generated chromosomes, each of which is randomly initialized to cover the possible range of the decision variable. For example, if you consider adjusting the IPA concentration every 10 minutes within an hour, each chromosome will contain 6 genes, each representing the IPA concentration at a 10-minute time point. A possible chromosome (i.e., a potential solution) can be: [0.35, 0.5, 1.2, 1.8, 2.0, 2.2] This represents the IPA concentration value planned to be achieved at different time points in the conversion process;

[0113] Chromosome selection strategy: The selection process determines which chromosomes are retained to the next generation. In this step, potential excellent solutions are selected to enter the next generation based on the fitness of the chromosome. In the ant colony algorithm, pheromones can be used to indicate the quality of the path. In addition, in the genetic algorithm, fitness is used to select chromosomes for crossover and mutation. Determine the crossover probability and then judge whether the chromosome participates in the crossover;

[0114] Formulate mutation strategy: Mutation is the process of introducing random changes in the genes of chromosomes to increase the diversity of the population. The mutation probability is set and chromosomes and genes are randomly selected for mutation.

[0115] In step S410, a genetic algorithm model is constructed, the feed IPA concentration is encoded as a chromosome in the genetic algorithm model, and the chromosome is selected based on the pheromone intensity in the ant colony algorithm model to perform a crossover operation to exchange genes between two chromosomes, such as Figure 2 As shown, the following steps are included:

[0116] S411, constructing a genetic algorithm model, encoding the IPA concentration of the feed into chromosomes, and each gene on each chromosome represents the IPA concentration at a certain time point;

[0117] S412, randomly initialize each chromosome, preset the feed IPA concentration and the corresponding adjustment times at different time points of the production conversion;

[0118] S413, based on the fitness information of the chromosome, the pheromone strength in the ant colony algorithm model is updated to obtain a new pheromone strength, based on the fitness information and the new pheromone strength, a selection probability is obtained, a selection probability threshold is set and the chromosome is selected to obtain a suitable chromosome and retain it to the next generation;

[0119] S414, performing a crossover operation on the chromosomes retained to the next generation to exchange genes between the two chromosomes;

[0120] Wherein, the new pheromone intensity is expressed as follows:

[0121]

[0122] The selection probability is expressed as follows:

[0123]

[0124] represents the probability of selection, Represents chromosome The fitness information of Represents chromosome The fitness information of Represents chromosome The current pheromone strength, Represents chromosome The current pheromone strength, , Belong to 1~ Any number in Represents the total number of chromosomes.

[0125] In step S414, a crossover operation is performed on the chromosomes retained to the next generation to exchange genes between the two chromosomes, such as Figure 3 As shown, the following steps are included:

[0126] S4141, modify the selection probability, select two chromosomes that meet the conditions for crossover, and obtain a chromosome crossover pair, and the selection probability is modified to:

[0127]

[0128] , Chromosomes , The fitness information of Represents chromosomes , The pheromone concentration, , It means traversing all possible chromosome pairs. , Belong to 1~ Any number in Indicates the number of all chromosomes;

[0129] S4142, determining the crossover point based on the pheromone intensity, and selecting the highest pheromone intensity as the chromosome crossover point;

[0130] S4143. Based on chromosome crossover pairs and chromosome crossover points, perform a crossover operation on the chromosomes.

[0131] In step S420, chromosomes and genes that meet the preset mutation probability are randomly selected to determine whether they are mutated, and the IPA concentration at a certain time point is randomly adjusted, such as Figure 4 As shown, the following steps are included:

[0132] S421, adjusting the initial mutation probability based on the pheromone intensity to obtain a new mutation probability;

[0133] S422, based on the comparison between the mutation probability of each chromosome and the new mutation probability, determine whether to perform mutation processing, wherein the mutation processing is achieved by randomly adjusting the IPA concentration at a certain time point;

[0134] Wherein, the new mutation probability is expressed as follows:

[0135]

[0136] represents the initial mutation probability, represents the adjustment factor, Represents chromosome The current pheromone strength, Indicates the maximum pheromone intensity in the current population.

[0137] In actual operation, the whole method can be found in the attached Figure 5 As shown, Figure 5 This is a complete optimization process of the present invention during actual operation.

[0138] Embodiment 2:

[0139] An IPA production conversion time optimization method system based on big data fusion algorithm, such as Figure 6 As shown, it includes a data acquisition module 100, a coefficient determination module 200, a model construction module 300 and a target optimization module 400;

[0140] The data acquisition module 100 is used to acquire data related to each reactor, wherein the data related to the reactor includes the mass of the material in each reactor, the IPA concentration in each reactor, and the IPA concentration entering the first reactor;

[0141] The coefficient determination module 200 determines the flow coefficient based on the mass of each reactor, and determines the feed flow of subsequent reactors and the discharge flow of each reactor in combination with the feed flow of the first reactor;

[0142] The model building module 300 is used to build a material balance model according to the feed flow rate of each reactor, the discharge flow rate of each reactor, the mass of each reactor, the IPA concentration in each reactor and the IPA concentration entering the first reactor;

[0143] The target optimization module 400 determines the target model to be optimized of the IPA concentration based on the material balance model, and further determines the feed IPA concentration of each reactor in different time periods.

[0144] Various changes and modifications can be made without departing from the spirit and scope of the present invention, and all equivalent technical solutions also belong to the scope of the present invention.

[0145] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0146] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0147] The present invention is described with reference to the flowcharts and / or block diagrams of the method, terminal device (system), and computer program product according to the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0148] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0149] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0150] It should be noted that:

[0151] The "one embodiment" or "embodiment" mentioned in the specification means that the specific features, structures or characteristics described in conjunction with the embodiment are included in at least one embodiment of the present invention. Therefore, the phrases "one embodiment" or "embodiment" appearing in various places throughout the specification do not necessarily refer to the same embodiment.

[0152] In addition, it should be noted that the shapes and names of the parts and components of the specific embodiments described in this specification may be different. Any equivalent or simple changes made based on the structure, features and principles described in the patent concept of the present invention are included in the protection scope of the patent of the present invention. The technicians in the technical field of the present invention can make various modifications or supplements to the specific embodiments described or replace them in a similar manner, as long as they do not deviate from the structure of the present invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. A method for optimizing IPA production conversion time based on big data fusion algorithm, characterized in that: The following steps are involved: Obtaining data related to each reactor, including the mass of the material in each reactor, the IPA concentration in each reactor, and the IPA concentration entering the first reactor; Determine the flow coefficient based on the mass of each reactor, and combine the feed flow of the first reactor to determine the feed flow of subsequent reactors and the discharge flow of each reactor; A material balance model was constructed using the feed flow rate of each reactor, the discharge flow rate of each reactor, the mass of each reactor, the IPA concentration in each reactor, and the IPA concentration entering the first reactor. Based on the material balance model, a target model for IPA concentration to be optimized is determined, and then the feed IPA concentration of each reactor at different time periods is determined. Specifically, a genetic algorithm model is constructed, the feed IPA concentration is encoded as a chromosome in the genetic algorithm model, and chromosomes are selected based on the pheromone intensity in the ant colony algorithm model to perform a crossover operation to exchange genes between two chromosomes; chromosomes and genes that meet a preset mutation probability are randomly selected to determine whether they have mutated, and the IPA concentration at a certain time point is randomly adjusted; thereby, an optimal IPA adjustment strategy and an optimal production changeover time are obtained; The target model to be optimized is expressed as follows: in, Indicates the The time required for a reactor to reach the final IPA concentration from the initial state, Indicates the The mass of the reactor, and Respectively represent The feed and discharge flow rates of each reactor, Indicates the The IPA concentration in each reactor, Indicates the The final IPA concentration in each reactor.

2. The IPA production conversion time optimization method based on big data fusion algorithm according to claim 1 is characterized in that: The method of determining the flow coefficient based on the mass of each reactor and combining the feed flow of the first reactor to determine the feed flow of subsequent reactors and the discharge flow of each reactor includes the following steps: Based on the mass of each reactor, the flow coefficient is obtained by combining the mass of the carboxyl group at the outlet of each reactor and the mass of the slurry IPA; The discharge flow rate and feed flow rate of each subsequent reactor are obtained in sequence by combining the flow coefficient with the feed flow rate of the first reactor; Wherein, the flow coefficient is expressed as follows: The discharge flow rate is expressed as follows: in, Indicates the The feed flow rate of each reactor, Indicates the The feed flow rate of each reactor, Indicates the flow coefficient.

3. The IPA production conversion time optimization method based on big data fusion algorithm according to claim 1 is characterized in that: The material balance model is expressed as follows: in, Indicates the The time required for a reactor to reach the final IPA concentration from the initial state, Indicates the The mass of the reactor, Indicates the The final IPA concentration in each reactor is and Respectively represent The feed and discharge flow rates of each reactor, Indicates entering IPA concentration of each reactor, , indicating the The IPA concentration of the first reactor is equal to that of the IPA concentration in each reactor feed.

4. The IPA production conversion time optimization method based on big data fusion algorithm according to claim 1 is characterized in that: The mass of the reactor is expressed as follows: in, Indicates the The mass of the reactor, Indicates the The volume of liquid inside the reactor, Indicates the The density of the liquid inside the reactor.

5. The IPA production conversion time optimization method based on big data fusion algorithm according to claim 1 is characterized in that: The genetic algorithm model is constructed, the feed IPA concentration is encoded as a chromosome in the genetic algorithm model, and the chromosome is selected based on the pheromone intensity in the ant colony algorithm model to perform a crossover operation to exchange genes between two chromosomes, including the following steps: A genetic algorithm model was constructed to encode the feed IPA concentration into chromosomes, where each gene on each chromosome represents the IPA concentration at a certain time point. Each chromosome is randomly initialized, and the feed IPA concentration and the corresponding adjustment times at different time points of the production transition are preset; Based on the fitness information of the chromosome, the pheromone intensity in the ant colony algorithm model is updated to obtain a new pheromone intensity. Based on the fitness information and the new pheromone intensity, the selection probability is obtained. The selection probability threshold is set and the chromosome is selected to obtain a suitable chromosome and retain it to the next generation. Perform a crossover operation on the chromosomes that are retained in the next generation to exchange genes between the two chromosomes; Wherein, the new pheromone intensity is expressed as follows: The selection probability is expressed as follows: represents the probability of selection, Represents chromosome The fitness information of Represents chromosome The fitness information of Represents chromosome Current pheromone strength, Represents chromosome Current pheromone strength, , Belong to 1~ Any number in Indicates the total number of chromosomes.

6. The IPA production conversion time optimization method based on big data fusion algorithm according to claim 5 is characterized in that: The crossover operation is performed on the chromosomes retained to the next generation to exchange genes between the two chromosomes, comprising the following steps: The selection probability is modified to select two chromosomes that meet the conditions for crossover to obtain a chromosome crossover pair. The selection probability is modified to: , Chromosomes , The fitness information of Represents chromosomes , The pheromone concentration, , Indicates traversing all possible chromosome pairs, , Belong to 1~ Any number in Indicates the number of all chromosomes; The crossover point is determined based on the pheromone intensity, and the one with the highest pheromone intensity is selected as the chromosome crossover point; Based on chromosome crossover pairs and chromosome crossover points, the chromosomes are crossover operated.

7. The IPA production conversion time optimization method based on big data fusion algorithm according to claim 5 is characterized in that: The method of randomly selecting chromosomes and genes that meet a preset mutation probability to determine whether they are mutated and randomly adjusting the IPA concentration at a certain time point includes the following steps: The initial mutation probability is adjusted based on the pheromone intensity to obtain a new mutation probability; Based on the comparison between the mutation probability of each chromosome and the new mutation probability, it is determined whether to perform mutation processing, and the mutation processing is achieved by randomly adjusting the IPA concentration at a certain time point; The new mutation probability is expressed as follows: represents the initial mutation probability, represents the adjustment factor, Represents chromosome Current pheromone strength, Indicates the maximum pheromone intensity in the current population.

8. An IPA production conversion time optimization method system based on big data fusion algorithm, characterized in that: It includes data acquisition module, coefficient determination module, model construction module and target optimization module; The data acquisition module is used to obtain data related to each reactor, and the data related to the reactor include the mass of the material in each reactor, the IPA concentration in each reactor, and the IPA concentration entering the first reactor; The coefficient determination module determines the flow coefficient based on the mass of each reactor, and combines the feed flow of the first reactor to determine the feed flow of subsequent reactors and the discharge flow of each reactor; The model building module is used to build a material balance model based on the feed flow rate of each reactor, the discharge flow rate of each reactor, the mass of each reactor, the IPA concentration in each reactor, and the IPA concentration entering the first reactor; The target optimization module determines the target model to be optimized for IPA concentration based on the material balance model, and further determines the feed IPA concentration of each reactor at different time periods. Specifically, the module constructs a genetic algorithm model, encodes the feed IPA concentration as a chromosome in the genetic algorithm model, selects chromosomes based on the pheromone intensity in the ant colony algorithm model, and then performs a crossover operation to exchange genes between two chromosomes; randomly selects chromosomes and genes that meet a preset mutation probability to determine whether they have mutated, and randomly adjusts the IPA concentration at a certain time point; thereby obtaining the optimal IPA adjustment strategy and the optimal production changeover time; The target model to be optimized is expressed as follows: in, Indicates the The time required for a reactor to reach the final IPA concentration from the initial state, Indicates the The mass of the reactor, and Respectively represent The feed and discharge flow rates of each reactor, Indicates the The IPA concentration in each reactor, Indicates the The final IPA concentration in each reactor.

9. A computer program product, characterized in that The method comprises computer instructions, which implement the method according to any one of claims 1 to 7 when executed by a processor.

10. An IPA production changeover time optimization device based on a big data fusion algorithm, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

Citation Information

Patent Citations

  • Genetic ant colony algorithm based determination method for water cool storage air-conditioning system energy-saving optimal control strategy

    CN106940068A

  • Database query optimization method based on ant colony genetic dynamic fusion algorithm

    CN115391385A

  • Genetic algorithm optimization method

    WO2002082371A2

  • Task scheduling method, device and computer storage medium

    WO2018036282A1