A method and system for intelligent formula optimization of TGIC-cured polyester resin
Through machine learning and particle swarm optimization algorithms, an intelligent TGIC cured polyester resin formula optimization system was established, which solved the problem of inefficiency of traditional methods, achieved rapid and accurate formulation optimization, and improved production efficiency and product quality.
Patent Information
- Application Number
- CN202410578780.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-11
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-05-11
AI Technical Summary
The traditional TGIC cured polyester resin formulation optimization method depends on laboratory tests and the experience of technicians. The process is cumbersome and inefficient, and it is impossible to accurately obtain the optimal ratio.
Using machine learning and data-driven methods, collect and analyze production data, establish an intelligent formula optimization system, quickly and accurately predict formula performance through optimization models and screening models, use particle swarm optimization algorithm to adjust the order and proportion of raw materials, optimize the curing rate and Performance metrics.
It realizes rapid and accurate formulation optimization of TGIC cured polyester resin, improves production efficiency, reduces production costs, and ensures product performance and economic benefits.
Smart Images

Figure CN118366562B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of emerging software and novel information technology service technology, and more specifically, to an intelligent formula optimization method and system for TGIC-cured polyester resin. Background Art
[0002] TGIC (melamine formate) cured polyester resin is widely used in the fields of automobiles, construction, electronics, plastic products, etc. due to its outstanding weather resistance, chemical resistance and excellent mechanical properties. In the preparation of TGIC cured polyester resin, formula optimization plays an important role in improving production efficiency, reducing production costs, and optimizing product quality. It is an important link to ensure product performance and economic benefits. It involves the precise blending of multiple components such as resin, curing agent, colorant, leveling agent, etc. However, traditional formula optimization methods rely on laboratory experiments and the experience of technicians. The design and optimization of formulas are usually carried out manually, requiring a lot of experiments and adjustments. Engineers gradually adjust the proportions of various raw materials in the formula based on their own experience and judgment until satisfactory performance and effects are achieved. Due to reliance on manual experience, multiple adjustments and optimizations are often required, and the process is cumbersome and inefficient.
[0003] For example, the patent with authorization announcement number CN103059275B provides a polyester resin for TGIC-cured transparent powder coating and a preparation method thereof. Although the patent provides a preparation method for a saturated terminal carboxyl polyester resin for a highly weather-resistant and low-temperature TGIC-cured powder coating, it still uses the traditional method, and the steps are relatively cumbersome, and it is impossible to obtain an accurate optimal ratio. Summary of the invention
[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides an intelligent formula optimization method and system for TGIC-cured polyester resin, which uses machine learning and data-driven methods to collect and analyze a large amount of production data and experimental results, and then establishes an intelligent formula optimization method and system. This system can quickly and accurately predict the performance of different formulas, and adapt to different production needs and product requirements through continuous optimization and improvement.
[0005] To achieve the above object, the present invention provides the following technical solutions:
[0006] An intelligent formula optimization method for TGIC-cured polyester resin, comprising:
[0007] The formula ratio is changed m times to obtain an original data set, where the original data set includes m groups of original data, where m is an integer greater than 1;
[0008] The optimization model was established with the original data set as the data set of the optimization model, the order of adding raw materials of TGIC-cured polyester resin as the constraint condition, and the curing rate as the optimization target;
[0009] A screening model was established with various performance indicators of TGIC-cured polyester resin as constraints and curing rate as optimization target.
[0010] Furthermore, the raw data include the amount of toluene diisocyanate added, the amount of propylene glycol added, the amount of oxidized polymer added, the activation energy of the reaction, the gas constant, the reaction temperature, the amount of colorant added and the amount of leveling agent added.
[0011] Furthermore, the method for constructing the optimization model includes:
[0012] The original data were collected, including the addition amount of toluene diisocyanate, propylene glycol, oxidized polymer, activation energy of reaction, gas constant, reaction temperature, colorant addition amount and leveling agent addition amount; the data set was divided into two parts, one for training the model and the other for verifying the model; the model was trained with the original data and the model parameters were optimized to fit the data; the validation data set was used for verification to evaluate the performance and generalization ability of the model. In the model training, the curing rate of TGIC-cured polyester resin was used as the optimization target, and the order of raw material addition in the intelligent formula of TGIC-cured polyester resin was used as the constraint condition. Finally, the model that met the preset error value requirements was selected as the final optimization model.
[0013] The formula for calculating the curing rate is as follows:
[0014]
[0015] Where CR is the curing rate, k is the rate constant of the curing reaction, TDI represents the amount of toluene diisocyanate added, PG represents the amount of propylene glycol added, OP represents the amount of oxidized polymer added, a represents the weight coefficient of toluene diisocyanate, b represents the weight coefficient of propylene glycol added, c represents the weight coefficient of oxidized polymer added, E a represents the activation energy of the reaction, T represents the reaction temperature, CL represents the amount of colorant added, FL represents the amount of leveling agent added, d represents the coefficient of the influence of the colorant on the curing rate, e represents the coefficient of the influence of the leveling agent on the curing rate, and R represents the gas constant;
[0016] Further, the curing rate is taken as an optimization target;
[0017] Constraints:
[0018] (a) If x = 1, then
[0019] (b) If x = 0, then
[0020] (c) x and y are non-negative integers;
[0021] (d) The value range of x is {0, 1};
[0022] (e) The upper limit of CR is CR≤CR max , where CR max is the maximum acceptable cure rate;
[0023] x represents the raw material to be added first. If TDI is added, x=1; if PG is added, x=0;
[0024] CR represents the curing rate; when TDI is added first, m1 represents the weight coefficient of TDI; m2 represents the weight coefficient of PG; when PG is added first, m3 represents the weight coefficient of PG; m4 represents the weight coefficient of TDI.
[0025] Furthermore, the optimized blending formula is obtained according to the optimization model, specifically:
[0026] An initial random formula is set, and a normalization process is performed on the initial random formula to obtain a group of initial random particles;
[0027] The particle swarm optimization algorithm is used to iterate the initial particles:
[0028] The particle swarm optimization algorithm is used to iterate the initial particles:
[0029] V i+1 =w·V i +c1·r1·(pbest i -X i )+c2·r2·(gbest-X i )
[0030] X i+1 =X i +V i+1
[0031] V i+1 The particle's speed at the next moment;
[0032] V i is the current velocity of the particle;
[0033] X i+1 The position of the particle at the next moment;
[0034] X i is the current position of the particle;
[0035] pbesti is the optimal position the particle has ever encountered;
[0036] gbest: the best position encountered by the group in history, that is, the best solution for all particles;
[0037] w is the inertia weight, which helps balance the global search ability and the local search ability;
[0038] c1 and c2 are learning factors, which affect the movement of particles to the individual historical optimal and group optimal positions respectively;
[0039] r1 and r2 are randomly generated numbers in the range [0,1] to increase the randomness and diversity of the search.
[0040] Furthermore, taking the various performance indicators of TGIC-cured polyester resin as constraints and the curing rate of TGIC-cured polyester resin as the optimization target, a screening model was established:
[0041] The optimization goal is to optimize the curing rate, specifically:
[0042] argmax CR Sum
[0043] Constraints:
[0044]
[0045] Assume that the curing rate is CR, and the performance requirements include n indicators, and the corresponding weights are w1, w2, w3...w n ; Where Sum is a weighted score indicating the performance when the curing rate is CR, f i is the score of the i-th performance index when the curing rate is CR, w i is the corresponding weight.
[0046] Furthermore, the optimal blending formula is obtained according to the screening model, specifically:
[0047] An initial random formula is set, and a normalization process is performed on the initial random formula to obtain a group of initial random particles;
[0048] The particle swarm optimization algorithm is used to iterate the initial particles:
[0049] V i+1 =w·V i +c1·r1·(pbest i -X i )+c2·r2·(gbest-X i )
[0050] X i+1 =X i+V i+1
[0051] V i+1 The particle's speed at the next moment;
[0052] V i is the current velocity of the particle;
[0053] X i+1 The position of the particle at the next moment;
[0054] X i is the current position of the particle;
[0055] pbest i is the optimal position the particle has ever encountered;
[0056] gbest: the best position encountered by the group in history, that is, the best solution for all particles;
[0057] w is the inertia weight, which helps balance the global search ability and the local search ability;
[0058] c1 and c2 are learning factors, which affect the movement of particles to the individual historical optimal and group optimal positions respectively;
[0059] r1 and r2 are randomly generated numbers in the range [0,1] to increase the randomness and diversity of the search.
[0060] An intelligent formula optimization system for TGIC-cured polyester resin, comprising:
[0061] Data acquisition module: change the formula ratio m times to obtain the original data set, the original data set includes m groups of original data, m>1 integer;
[0062] Optimization module: Based on the original data set, the order of adding raw materials of TGIC curing polyester resin is used as a constraint condition, and the curing rate is used as the optimization target to establish an optimization model;
[0063] Screening module: A screening model is established with various performance indicators of TGIC-cured polyester resin as constraints and curing rate as the optimization target.
[0064] An electronic device comprises a power supply, an interface, a keyboard, a memory, a central processing unit, and a computer program stored in the memory and executable on the central processing unit. When the central processing unit executes the computer program, an intelligent formula optimization method for TGIC-cured polyester resin is implemented. The interface comprises a network interface and a data interface. The network interface comprises a wired or wireless interface, and the data interface comprises an input or output interface.
[0065] A computer-readable storage medium stores a computer program, and when the computer program is executed, an intelligent formula optimization method for TGIC-cured polyester resin is implemented.
[0066] Compared with the prior art, the present invention has the following beneficial effects:
[0067] The present invention obtains the original data set, takes the original data set as the basis, takes the curing rate as the optimization target, and takes the order of adding different raw materials as the constraint condition to construct an optimization model; considers that the curing rate has different effects on multiple performance indicators of TGIC curing polyester resin, establishes a screening model, and finally obtains the optimal ratio of TGIC curing polyester resin. The present invention applies artificial intelligence technology to the technical field of TGIC curing polyester resin preparation, and solves the technical problem that the traditional formula optimization method relies on laboratory experiments and the experience of technicians. BRIEF DESCRIPTION OF THE DRAWINGS
[0068] Figure 1 It is a flow chart of an intelligent formula optimization method for TGIC-cured polyester resin in the present invention;
[0069] Figure 2 A schematic diagram of an intelligent formula optimization system for TGIC-cured polyester resin provided by one embodiment of the present invention;
[0070] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention;
[0071] Figure 4 A schematic diagram of a computer-readable storage medium in the present invention. DETAILED DESCRIPTION
[0072] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0073] Example 1
[0074] See also Figure 1 As shown, this embodiment discloses a method for optimizing an intelligent formula of a TGIC-cured polyester resin, comprising:
[0075] S10: changing the recipe ratio m times to obtain an original data set, where the original data set includes m groups of original data, where m is an integer greater than 1;
[0076] Specifically, in this example, the raw data obtained is mainly divided into two parts: first, the experimental operation is carried out in the laboratory according to the experimental design plan. This part designs the amount of raw materials to be added, the order of addition, the reaction conditions, etc. During the experiment, data is collected regularly to observe the reaction process; the collected data is analyzed; second, based on historical experience and practice, relevant data is gradually accumulated and summarized in practical production, and analyzed and sorted. This embodiment continuously changes the formula ratio and the order of raw material addition, so that the subsequent calculated data can be more accurate;
[0077] The data of each group include the amount of toluene diisocyanate added, the amount of propylene glycol added, the amount of oxidized polymer added, the reaction temperature, the amount of colorant added, and the amount of leveling agent added;
[0078] It can be understood that in the m groups of data there correspond to m amounts of toluene diisocyanate added, m amounts of propylene glycol added, m amounts of oxidized polymer added, m reaction temperatures, m amounts of colorant added, and m amounts of leveling agent added.
[0079] S20: Establishing an optimization model with the original data set as the data set of the optimization model, the order of adding raw materials of the TGIC-curable polyester resin as a constraint condition, and the curing rate as an optimization target;
[0080] The optimization model construction method includes:
[0081] The original data were collected, including the addition amount of toluene diisocyanate, propylene glycol, oxidized polymer, activation energy of reaction, gas constant, reaction temperature, colorant addition amount and leveling agent addition amount; the data set was divided into two parts, one for training the model and the other for verifying the model; the model was trained with the original data and the model parameters were optimized to fit the data; the validation data set was used for verification to evaluate the performance and generalization ability of the model. In the model training, the curing rate of TGIC-cured polyester resin was used as the optimization target, and the order of raw material addition in the intelligent formula of TGIC-cured polyester resin was used as the constraint condition. Finally, the model that met the preset error value requirements was selected as the final optimization model.
[0082] The formula for calculating the curing rate is as follows:
[0083]
[0084] Where CR is the curing rate, k is the rate constant of the curing reaction, TDI represents the amount of toluene diisocyanate added, PG represents the amount of propylene glycol added, OP represents the amount of oxidized polymer added, a represents the weight coefficient of toluene diisocyanate, b represents the weight coefficient of propylene glycol added, c represents the weight coefficient of oxidized polymer added, E arepresents the activation energy of the reaction, T represents the reaction temperature, CL represents the amount of colorant added, FL represents the amount of leveling agent added, d represents the coefficient of the influence of the colorant on the curing rate, e represents the coefficient of the influence of the leveling agent on the curing rate, and R represents the gas constant;
[0085] The curing rate is used as the optimization goal;
[0086] Constraints:
[0087] (a) If x = 1, then
[0088] (b) If x = 0, then
[0089] (c) x and y are non-negative integers;
[0090] (d) The value range of x is {0, 1};
[0091] (e) The upper limit of CR CR ≤ y max , where y max is the maximum acceptable cure rate;
[0092] x represents the raw material to be added first. If TDI is added first, x=1, and if PG is added first, x=0. When TDI is added first, m1 represents the weight coefficient of TDI; m2 represents the weight coefficient of PG; when PG is added first, m3 represents the weight coefficient of PG; and m4 represents the weight coefficient of TDI.
[0093] An initial random formula is set, and a normalization process is performed on the initial random formula to obtain a group of initial random particles;
[0094] The particle swarm optimization algorithm is used to iterate the initial particles:
[0095] V i+1 =w·V i +c1·r1·(pbest i -X i )+c2·r2·(gbest-X i )
[0096] X i+1 =X i +V i+1
[0097] V i+1 The particle's speed at the next moment;
[0098] V i is the current velocity of the particle;
[0099] X i+1The position of the particle at the next moment;
[0100] X i is the current position of the particle;
[0101] pbest i is the optimal position the particle has ever encountered;
[0102] w is the inertia weight, which helps balance the global search ability and the local search ability;
[0103] c1 and c2 are learning factors, which affect the movement of particles to the individual historical optimal and group optimal positions respectively;
[0104] r1 and r2 are randomly generated numbers in the range [0,1] to increase the randomness and diversity of the search.
[0105] Preferably, the optimization model described above is LSTM or RNN. It is worth noting that the sample data described above is mainly collected through real-time experiments and past experience and practice, and the relevant data is gradually accumulated and summarized in practical production, and analyzed and sorted; the data ratio used in the training model and the verification model here, and the preset error value are determined by those skilled in the art according to actual conditions and work experience, and this example is not limited to this;
[0106] It should be noted that the optimization model is established with the order of adding raw materials of TGIC-curing polyester resin as a constraint condition and the curing rate as the optimization target; it is mainly to optimize the impact of the addition order on the curing rate, so as to achieve the effect of optimizing the formula.
[0107] It is worth noting that the curing reaction rate directly affects the performance of the final product, such as weather resistance, chemical resistance, mechanical properties, etc. The order of addition of the raw materials also affects the curing reaction rate and is taken into account in the smart formulation of TGIC-cured polyester resin. For example, the order of addition of toluene diisocyanate (TDI) and propylene glycol (PG) may affect the initiation rate of the cross-linking reaction. If toluene diisocyanate is added first, it may react with the functional groups in the polyester resin to form active sites, thereby accelerating the addition reaction of the cross-linking agent and promoting the initiation of the cross-linking reaction. On the contrary, if propylene glycol is added first, the active sites may react with the cross-linking agent first, thereby slowing down the addition reaction of the curing agent and delaying the initiation of the curing reaction; therefore, different addition orders may cause changes in the curing rate and reaction degree;
[0108] It is worth noting that a limit is placed on the curing rate in this example, mainly because in some cases, a too fast curing rate may cause a sharp rise in temperature, or even cause safety accidents such as explosions or fires. By limiting the curing rate, the speed of the reaction can be controlled to avoid accidents. Too fast or uneven curing reaction rates may cause fluctuations or instability in product quality. Limiting the curing rate can ensure that the curing process is carried out at an appropriate speed, thereby ensuring that the product has consistent performance and quality. When preparing products such as TGIC-cured polyester resins, it is very important to set a limit on the curing rate to ensure the safety, stability and efficiency of the production process. The specific limit value is determined by those skilled in the art based on factors such as the use environment, reaction temperature, and the amount of raw materials added, and this example does not limit this;
[0109] It should be noted that an initial random recipe is set and these recipes are normalized to generate a group of initial random particles, and these initial particles are iteratively optimized using an improved particle swarm optimization algorithm until the stop condition is met and the maximum number of iterations or the optimization target range is reached; the specific values of the maximum number of iterations and the optimization target here are specifically set by the staff in this technical field, and this example does not make specific requirements.
[0110] S30: Establishing a screening model with various performance indicators of TGIC-cured polyester resin as constraints and curing rate as optimization target;
[0111] The screening model is constructed as follows:
[0112] The optimization goal is to optimize the curing rate, specifically:
[0113] argmax CR Sum
[0114] Constraints:
[0115]
[0116] Assume that the curing rate is CR, and the performance requirements include n indicators, and the corresponding weights are w1, w2, w3...w n ; Where Sum is a weighted score indicating the performance when the curing rate is CR, f i is the score of the i-th performance index when the curing rate is CR, w i is the corresponding weight.
[0117] The initial formula is set as the optimized formula obtained after the optimization model is optimized, and the initial formula is normalized to obtain a group of initial random particles;
[0118] The particle swarm optimization algorithm is used to iterate the initial particles:
[0119] V i+1 =w·V i +c1·r1·(pbest i -X i )+c2·r2·(gbest-X i )
[0120] X i+1 =X i +V i+1
[0121] V i+1 The particle's speed at the next moment;
[0122] V i is the current velocity of the particle;
[0123] X i+1 The position of the particle at the next moment;
[0124] X i is the current position of the particle;
[0125] pbest i is the optimal position the particle has ever encountered;
[0126] gbest: the best position encountered by the group in history, that is, the best solution for all particles;
[0127] w is the inertia weight, which helps balance the global search ability and the local search ability;
[0128] c1 and c2 are learning factors, which affect the movement of particles to the individual historical optimal and group optimal positions respectively;
[0129] r1 and r2 are randomly generated numbers in the range [0,1] to increase the randomness and diversity of the search.
[0130] It should be noted that the screening model is established on the basis of the optimization model with the various performance indicators of TGIC-cured polyester resin as constraints and the curing rate as the optimization target; the main function of this module is to screen the various parameters of the formula after the optimization model to obtain the optimal curing rate, so that the various properties of the prepared TGIC-cured polyester resin can reach the most balanced state;
[0131] It is worth noting that the high and low curing rates also have different effects on the performance indicators of TGIC-cured polyester resins. For example, in the preparation process of TGIC-cured polyester resins, a high curing rate can often form a denser and more stable network structure, so it has better weather resistance and can better resist the erosion of natural environmental factors such as ultraviolet rays, oxidation, and humidity; secondly, the network structure formed by TGIC-cured polyester resins is denser, so it usually has higher hardness, wear resistance and scratch resistance; TGIC-cured polyester resins with moderate curing reaction rates usually have good leveling properties and can form a smooth and uniform surface after construction; and when the curing rate is low, the curing speed can be slowed down, reducing material waste during the curing process, especially for large production batches or situations that require particularly precise construction. This is particularly important; therefore, in the preparation process of TGIC-cured polyester resins, a suitable curing rate can make the performance of the prepared products optimal; therefore, it is necessary to comprehensively consider the various performance indicators and actual application scenarios of TGIC-cured polyester resins to obtain the optimal curing rate, thereby obtaining the optimal ratio.
[0132] It should be noted that an initial random recipe is set and these recipes are normalized to generate a group of initial random particles, and these initial particles are iteratively optimized using an improved particle swarm optimization algorithm until the stop condition is met and the maximum number of iterations or the optimization target range is reached; the specific values of the maximum number of iterations and the optimization target here are specifically set by the staff in this technical field, and this example does not make specific requirements.
[0133] Example 2
[0134] See also Figure 2 As shown, this embodiment discloses an intelligent formula optimization system for TGIC-cured polyester resin; in addition, this embodiment is based on Example 1, and the same parts as Example 1 are not repeated here;
[0135] Data acquisition module: change the formula ratio m times to obtain the original data set, the original data set includes m groups of original data, m>1 integer;
[0136] Optimization module: The optimization model is established with the original data set as the data set of the optimization model, the order of adding raw materials of TGIC curing polyester resin as the constraint condition, and the curing rate as the optimization target;
[0137] Screening module: A screening model is established with various performance indicators of TGIC-cured polyester resin as constraints and curing rate as the optimization target.
[0138] Example 3
[0139] Figure 3 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Figure 3 As shown, according to another aspect of the present application, an electronic device 500 is also provided. The electronic device 500 may include one or more processors and one or more memories. The memories store computer readable codes, and when the computer readable codes are executed by one or more processors, the path planning method described above may be executed.
[0140] The method or system according to the embodiment of the present application can also be used by Figure 3 The electronic device architecture shown in FIG. Figure 3 As shown, the electronic device 500 may include a bus 501, one or more CPUs 502, a read-only memory (ROM) 503, a random access memory (RAM) 504, a communication port 505 connected to a network, an input / output component 506, a hard disk 507, etc. The storage device in the electronic device 500, such as the ROM 503 or the hard disk 507, may store a smart formula optimization method for TGIC-cured polyester resin provided in the present application. A smart formula optimization method for TGIC-cured polyester resin may, for example, include: changing the formula ratio m times to obtain an original data set, the original data set includes m groups of original data, and m>1 integer; based on the original data set, with the order of adding raw materials of the TGIC-cured polyester resin as a constraint condition, and with the curing rate as an optimization target, an optimization model is established; with the various performance indicators of the TGIC-cured polyester resin as constraints, and with the curing rate as an optimization target, a screening model is established.
[0141] Furthermore, the electronic device 500 may also include a user interface 508. Figure 3 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 3 One or more components of an electronic device are shown.
[0142] Example 4
[0143] See also Figure 4 As shown, this embodiment discloses a computer-readable storage medium, on which a computer program is stored. When the computer program is executed, the above-mentioned intelligent formula optimization method for TGIC-cured polyester resin is implemented.
[0144] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters, weights and thresholds in the formula are set by technicians in this field according to actual conditions.
[0145] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through a wired network or a wireless network. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD) or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0146] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0147] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0148] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0149] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0150] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0151] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
[0152] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. An intelligent formula optimization method for TGIC-cured polyester resin, characterized in that: include: S10: changing the recipe ratio m times to obtain an original data set, where the original data set includes m groups of original data, where m is an integer greater than 1; The raw data include: the amount of toluene diisocyanate added, the amount of propylene glycol added, the amount of oxidized polymer added, the activation energy of the reaction, the gas constant, the reaction temperature, the amount of colorant added and the amount of leveling agent added; S20: Establishing an optimization model with the original data set as the data set of the optimization model, the order of adding raw materials of the TGIC-curable polyester resin as a constraint condition, and the curing rate as an optimization target; The order of adding raw materials of TGIC curing polyester resin is used as a constraint condition, and the curing rate is used as an optimization target as follows: The curing rate is used as the optimization goal; Constraints: (a) If x = 1, then (b) If x = 0, then (c) x and y are non-negative integers; (d) The value range of x is {0, 1}; (e) The upper limit of CR is CR≤CR max ; x represents the raw material to be added first. If TDI is added, x=1; if PG is added, x=0; CR represents the curing rate; when TDI is added first, m1 represents the weight coefficient of TDI; m2 represents the weight coefficient of PG; when PG is added first, m3 represents the weight coefficient of PG; m4 represents the weight coefficient of TDI; CR max is the maximum acceptable cure rate; S30: Taking the various performance indicators of TGIC-cured polyester resin as constraints and the curing rate as the optimization target, a screening model is established. The above-mentioned, taking the various performance indicators of TGIC-cured polyester resin as constraints, taking the curing rate as the optimization target, and taking the curing rate as the optimization target, specifically: argmax CR Sum Constraints: Assume that the curing rate is CR, and the performance requirements include n indicators, and the corresponding weights are w1, w2, w3...w n ; Where Sum is a weighted score indicating the performance when the curing rate is CR, f i is the score of the i-th performance index when the curing rate is CR, w i is the corresponding weight.
2. The intelligent formula optimization method for TGIC-cured polyester resin according to claim 1, characterized in that: The method for constructing the optimization model includes: The original data were collected, including the addition amount of toluene diisocyanate, propylene glycol, oxidized polymer, activation energy of reaction, gas constant, reaction temperature, colorant addition amount and leveling agent addition amount; the data set was divided into two parts, one for training the model and the other for verifying the model; the model was trained with the original data and the model parameters were optimized to fit the data; the validation data set was used for verification to evaluate the performance and generalization ability of the model. In the model training, the curing rate of TGIC-cured polyester resin was used as the optimization target, and the order of raw material addition in the intelligent formula of TGIC-cured polyester resin was used as the constraint condition. Finally, the model that met the preset error value requirements was selected as the final optimization model. The formula for calculating the curing rate is as follows: Where CR is the curing rate, k is the rate constant of the curing reaction, TDI represents the amount of toluene diisocyanate added, PG represents the amount of propylene glycol added, OP represents the amount of oxidized polymer added, a represents the weight coefficient of toluene diisocyanate, b represents the weight coefficient of propylene glycol added, c represents the weight coefficient of oxidized polymer added, E a It represents the activation energy of the reaction, T represents the reaction temperature, CL represents the amount of colorant added, FL represents the amount of leveling agent added, d represents the coefficient of the influence of the colorant on the curing rate, e represents the coefficient of the influence of the leveling agent on the curing rate, and R represents the gas constant.
3. The intelligent formula optimization method for TGIC-cured polyester resin according to claim 1, characterized in that: The optimized blending formula is obtained according to the optimization model, specifically: An initial random formula is set, and a normalization process is performed on the initial random formula to obtain a group of initial random particles; The particle swarm optimization algorithm is used to iterate the initial particles: V i+1 =w·V i +c1·r1·(pbest i -X i )+c2·r2·(gbest-X i ) X i+1 =X i +V i+1 V i+1 The particle's speed at the next moment; V i is the current velocity of the particle; X i+1 The position of the particle at the next moment; X i is the current position of the particle; pbest i is the optimal position the particle has ever encountered; w is the inertia weight, which helps balance the global search ability and the local search ability; c1 and c2 are learning factors, which affect the movement of particles to the individual historical optimal and group optimal positions respectively; r1 and r2 are randomly generated numbers in the range [0,1] to increase the randomness and diversity of the search.
4. The intelligent formula optimization method for TGIC-cured polyester resin according to claim 1, characterized in that: The optimal blending formula is obtained according to the screening, specifically: The initial formula is set as the optimized formula obtained after the optimization model is optimized, and the initial formula is normalized to obtain a group of initial random particles; The particle swarm optimization algorithm is used to iterate the initial particles: V i+1 =w·V i +c1·r1·(pbest i -X i )+c2·r2·(gbest-X i ) X i+1 =X i +V i+1 V i+1 The particle's speed at the next moment; V i is the current velocity of the particle; X i+1 The position of the particle at the next moment; X i is the current position of the particle; pbest i is the optimal position the particle has ever encountered; gbest: the best position encountered by the group in history, that is, the best solution for all particles; w is the inertia weight, which helps balance the global search ability and the local search ability; c1 and c2 are learning factors, which affect the movement of particles to the individual historical optimal and group optimal positions respectively; r1 and r2 are randomly generated numbers in the range [0,1] to increase the randomness and diversity of the search.
5. An intelligent formula optimization system for TGIC-cured polyester resin, used to implement the intelligent formula optimization method for TGIC-cured polyester resin according to any one of claims 1 to 4, characterized in that: include: Data acquisition module: change the formula ratio m times to obtain the original data set, the original data set includes m groups of original data, m>1 integer; Optimization module: Based on the original data set, the order of adding raw materials of TGIC curing polyester resin is used as a constraint condition, and the curing rate is used as the optimization target to establish an optimization model; Screening module: A screening model is established with various performance indicators of TGIC-cured polyester resin as constraints and curing rate as the optimization target.
6. An electronic device comprising a power supply, an interface, a keyboard, a memory, a central processing unit, and a computer program stored in the memory and executable on the central processing unit, characterized in that: When the central processing unit executes the computer program, the intelligent formula optimization method for TGIC-cured polyester resin according to any one of claims 1 to 4 is implemented, and the interface includes a network interface and a data interface, the network interface includes a wired or wireless interface, and the data interface includes an input or output interface.
7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the intelligent formula optimization method for TGIC-cured polyester resin according to any one of claims 1 to 4 is implemented.
