Formula optimization method and device of insulating material, equipment and storage medium

Abstract

The invention discloses a formula optimization method and device of an insulating material, equipment and a storage medium, and relates to the technical field of chemical materials.The method comprises the steps that initial formula information corresponding to the current insulating material is obtained, and a physical constraint model is constructed according to the initial formula information; the physical constraint model is composed of a component compatibility constraint condition, a process feasibility constraint condition and a material performance threshold condition; generating an initial candidate formula according to a physical constraint model based on the initial formula information; performing iterative optimization on the initial candidate formula by adopting a gradient response optimization algorithm to obtain a current candidate formula; and performing physical property prediction on the current candidate formula, and determining a target optimization formula according to a physical property prediction result. The optimization process is based on a scientific model and algorithm, the limitation of traditional experience can be broken through, the relation between different performance indexes is balanced, the formula combination which may be ignored is explored, and a new possibility is provided for performance improvement of the insulating material.

Description

Technical Field

[0001] This application relates to the field of chemical materials technology, and in particular to a method, apparatus, equipment and storage medium for optimizing the formulation of insulating materials. Background Technology

[0002] With the continuous development and increasing complexity of power systems, the requirements for insulation materials in terms of heat resistance, mechanical strength, dielectric properties, and process adaptability are becoming increasingly stringent. To meet these diverse needs, the optimization of insulation material formulations has become a crucial aspect of materials science research. Traditional methods for optimizing insulation material formulations primarily rely on experimental trial and error and empirical adjustments. Engineers repeatedly adjust the material composition based on existing knowledge and experimental data, then prepare samples for performance testing to gradually improve the material's performance.

[0003] However, due to the excessive reliance on engineers' personal experience and knowledge and experimental trial-and-error adjustments in traditional methods, and the lack of quantitative optimization mechanisms, it is difficult to achieve systematic and comprehensive optimization. Consequently, it is difficult to discover innovative formulation combinations outside the scope of traditional experience, thus limiting the further improvement of insulation material performance. Summary of the Invention

[0004] The main objective of this application is to provide a method, apparatus, equipment, and storage medium for optimizing the formulation of insulating materials, aiming to solve the technical problem that existing optimization methods rely on experimental trial and error and experience-based adjustments, making it difficult to discover innovative formulation combinations beyond the scope of traditional experience.

[0005] To achieve the above objectives, this application proposes a method for optimizing the formulation of insulating materials, the method comprising:

[0006] Obtain the initial formula information corresponding to the current insulating material, and construct a physical constraint model based on the initial formula information. The physical constraint model consists of component compatibility constraints, process feasibility constraints, and material performance threshold conditions.

[0007] Based on the initial formulation information, an initial candidate formulation is generated according to the component compatibility constraints, the process feasibility constraints, and the material performance threshold conditions.

[0008] The initial candidate formulation is iteratively optimized using a gradient response optimization algorithm to obtain the current candidate formulation;

[0009] Perform physical property prediction on the current candidate formulation, and determine the target optimized formulation based on the physical property prediction results.

[0010] In one embodiment, the initial formulation information includes: formulation components, formulation process, and material properties. The step of constructing a physical constraint model based on the initial formulation information includes:

[0011] The compatibility coefficients between the formulation components are determined based on historical experience data, and a component compatibility constraint matrix is ​​established based on the compatibility coefficients.

[0012] Based on the preset process requirement information, determine the feasibility boundary function of each formulation process, and set process feasibility constraints according to the feasibility boundary function.

[0013] The lower limit threshold of the current insulating material is determined based on the material properties, and the material performance threshold conditions are set based on the lower limit threshold.

[0014] A physical constraint model is obtained by mathematical modeling based on the component compatibility constraints, the process feasibility constraints, and the material performance threshold conditions.

[0015] In one embodiment, the step of generating an initial candidate formulation based on the initial formulation information, according to the component compatibility constraints, the process feasibility constraints, and the material performance threshold conditions, includes:

[0016] The basic proportion range of each formula component is initialized based on the initial formula information, and several candidate formula points are generated based on the basic proportion range.

[0017] Among the candidate formulation points, each candidate formulation point is screened according to the component compatibility constraint, the process feasibility constraint, or the material performance threshold condition to obtain candidate formulation points that meet the conditions and determine them as initial candidate formulations.

[0018] In one embodiment, the step of iteratively optimizing the initial candidate formulation using a gradient response optimization algorithm to obtain the current candidate formulation includes:

[0019] The comprehensive performance index is determined based on the component compatibility constraints, the process feasibility constraints, and the material performance threshold conditions, and a gradient response optimization model with the comprehensive performance index as the objective function is constructed based on the initial candidate formulation.

[0020] The gradient response optimization model is solved using the quasi-Newton method to obtain the Pareto optimal frontier, and the current candidate formulation is determined based on the Pareto optimal frontier.

[0021] In one embodiment, the step of predicting the physical properties of the current candidate formulation and determining the target optimized formulation based on the prediction results includes:

[0022] The physical properties of the current candidate formulation are predicted by a preset physical property prediction model to obtain the corresponding physical property prediction results;

[0023] When the predicted physical properties meet the target performance requirements, the current candidate formulation is verified by microstructure simulation.

[0024] During the verification through microstructure simulation, the current candidate formulation is tested using prototype samples.

[0025] When the prototype sample is tested and verified, the current candidate formulation is determined as the target optimized formulation.

[0026] In one embodiment, the step of predicting the physical properties of the current candidate formulation using a preset physical property prediction model to obtain the corresponding physical property prediction result includes:

[0027] Based on the obtained molecular structure and property information of the current insulating material, a property prediction model is established, which includes: a thermal stability equation, a dielectric strength model, and a mechanical property calculator.

[0028] Based on the property prediction model, the current candidate formulation is predicted to obtain thermal stability parameter values, dielectric strength prediction values, and mechanical property quantification values.

[0029] The thermal stability parameter value, the predicted dielectric strength value, and the quantified mechanical property value are determined as the property prediction results of the current candidate formulation.

[0030] In one embodiment, after the step of determining the current candidate formulation as the target optimized formulation during prototype sample testing and verification, the method further includes:

[0031] The target optimized formula is converted into production parameters through a process compatibility interface, so that the production system to which the process compatibility structure belongs can prepare the current insulating material according to the production parameters.

[0032] Based on the preparation results, the process parameters of the target optimized formula are modified to obtain the target optimized scheme after process parameter modification.

[0033] Furthermore, to achieve the above objectives, this application also proposes an apparatus for optimizing the formulation of insulating materials, the apparatus comprising:

[0034] The constraint modeling module is used to obtain the initial formula information corresponding to the current insulating material and construct a physical constraint model based on the initial formula information. The physical constraint model consists of component compatibility constraints, process feasibility constraints, and material performance threshold conditions.

[0035] A two-layer optimization module is used to generate an initial candidate formulation based on the initial formulation information, according to the component compatibility constraints, the process feasibility constraints, and the material performance threshold conditions.

[0036] The two-layer optimization module is also used to iteratively optimize the initial candidate formulation using a gradient response optimization algorithm to obtain the current candidate formulation;

[0037] The property prediction module is used to predict the properties of the current candidate formulation and determine the target optimized formulation based on the property prediction results.

[0038] In addition, to achieve the above objectives, this application also proposes an insulating material formulation optimization device, the device comprising: a memory, a processor, and an insulating material formulation optimization program stored in the memory and executable on the processor, the insulating material formulation optimization program being configured to implement the steps of the insulating material formulation optimization method as described above.

[0039] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, storing a formula optimization program for an insulating material, wherein when the formula optimization program for the insulating material is executed by a processor, it implements the steps of the formula optimization method for the insulating material as described above.

[0040] This application discloses a method for optimizing the formulation of insulating materials. The method involves obtaining initial formulation information corresponding to the current insulating material and constructing a physical constraint model based on this information. The physical constraint model consists of component compatibility constraints, process feasibility constraints, and material performance threshold conditions. Based on the initial formulation information, initial candidate formulations are generated according to the component compatibility constraints, process feasibility constraints, and material performance threshold conditions. A gradient response optimization algorithm is used to iteratively optimize the initial candidate formulations to obtain the current candidate formulation. The physical properties of the current candidate formulation are predicted, and the target optimized formulation is determined based on the prediction results.

[0041] Because this application can balance the relationship between different performance indicators, and comprehensively considers multiple dimensions such as component compatibility, process feasibility, and material performance thresholds, the final optimized formulation exhibits good overall performance in practical applications. Furthermore, since the optimization process is based on scientific models and algorithms, it can overcome the limitations of traditional experience, explore potentially overlooked formulation combinations, and provide new possibilities for improving the performance of insulating materials. Attached Figure Description

[0042] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0043] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a flowchart illustrating the first embodiment of the method for optimizing the formulation of insulating materials according to this application;

[0045] Figure 2 This is a flowchart illustrating the second embodiment of the method for optimizing the formulation of insulating materials according to this application;

[0046] Figure 3 This is a flowchart illustrating the third embodiment of the method for optimizing the formulation of insulating materials according to this application;

[0047] Figure 4 This is a schematic diagram of the system architecture for optimizing the formulation of the insulating material in this application;

[0048] Figure 5 This is a schematic diagram illustrating the entire process of the insulation material formulation optimization method in this application;

[0049] Figure 6 This is a schematic diagram of the module structure of the first embodiment of the insulating material formulation optimization device of this application;

[0050] Figure 7 This is a schematic diagram of the equipment structure for optimizing the formulation of insulating materials in the embodiments of this application.

[0051] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0052] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0053] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0054] This application provides a method for optimizing the formulation of insulating materials, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the method for optimizing the formulation of insulating materials according to this application. In this embodiment, the method includes steps S10 to S40:

[0055] Step S10: Obtain the initial formula information corresponding to the current insulating material, and construct a physical constraint model based on the initial formula information. The physical constraint model consists of component compatibility constraints, process feasibility constraints, and material performance threshold conditions.

[0056] It should be noted that the executing entity in this embodiment can be a computing electronic device with data processing, network communication, and program execution capabilities, such as a mobile phone, personal computer, formula management server, or other electronic devices capable of accessing the insulation material formula optimization system. Here, we will use an optimization device (referred to as the formula optimization system) capable of accessing the insulation material formula optimization system as an example to describe this embodiment and the following embodiments.

[0057] It should be understood that the current insulating material can be an insulating material that requires formulation optimization. Formulation optimization can be achieved by optimizing the formulation components of the insulating material (e.g., by using substitute components), or by optimizing the proportions of the components (e.g., by increasing or decreasing the proportions of the original components). Furthermore, the formulation optimization process can also include optimizing the preparation processes used for different formulation components.

[0058] The initial formulation information for the current insulating material can include the names and current proportions of the formulation components used to prepare the insulating material. For example, the formulation components of a certain insulating material may be: epoxy resin (60%), silica powder (30%), and curing agent (10%).

[0059] Understandably, a physical constraint model can be a computable constraint rule obtained by mathematically modeling various physical constraints. Through a physical constraint model, the engineer's experience in the traditional way can be transformed into a quantitative optimization mechanism, which is conducive to generating candidate recipes that simultaneously satisfy multiple physical constraints.

[0060] Regarding component compatibility constraints, the interactions and compatibility relationships between the components can be determined based on initial formulation information and historical experience data. This helps identify which component combinations will cause adverse reactions and which combinations can remain stable. If a component combination at a specific ratio may produce side reactions that affect material properties, then corresponding constraints need to be set based on that specific ratio to avoid incompatible component combinations.

[0061] Specifically, historical empirical data, such as experimental data, can be used to calibrate the chemical reaction compatibility matrix (composed of several compatibility thresholds) between the ingredients in a formulation, quantifying the interactions (such as synergistic effects and inhibition of side reactions) when different ingredients are mixed. For example, for a formulation containing n ingredients, the ingredient compatibility constraint matrix can be an n×n matrix G. compat , where element g i，jThis indicates the compatibility between component i and component j. If component i and component j are compatible, the corresponding matrix element is assigned a positive value; otherwise, it is assigned a negative value.

[0062] Regarding process feasibility constraints, they can be set based on the process limitations in the actual production environment, such as the capacity of production equipment and the adjustable range of process parameters, so as to ensure that the designed formula can be produced under existing process conditions.

[0063] Specifically, process feasibility constraints can be quantified by defining physical boundaries in the production process, such as temperature-viscosity curves (to ensure casting fluidity) and pressure-curing shrinkage rate relationships (to control dimensional accuracy), thus establishing a process feasibility boundary function g. proc (x (j) The input is the proportion of each ingredient in the formula (x). (j) The output is a value that represents feasibility, for example, if g proc (x (j) A value of ≥0 indicates that the process is feasible and can be carried out under existing production equipment and process conditions; otherwise, it is considered infeasible. For example, the viscosity of epoxy resin at 120 degrees Celsius needs to be below 5000 cP to be compatible with automated production lines.

[0064] Regarding material performance threshold conditions, the minimum requirements for basic performance indicators that the insulating material must meet can be determined based on the current application needs and material characteristics. These requirements may include, for example, heat resistance, mechanical strength, and dielectric properties. By setting material performance threshold conditions, it can be ensured that the optimized formulation at least meets the basic performance standards for use.

[0065] Specifically, the threshold conditions for material properties can be quantified by setting minimum standards for key performance characteristics (such as breakdown voltage ≥30kV / mm and heat distortion temperature ≥130 degrees).

[0066] It should be understood that the equipment can determine the compatibility coefficients between each formulation component based on historical experience data, and establish a component compatibility constraint matrix based on the compatibility coefficients; determine the feasibility boundary function of each formulation process based on production environment information, and set process feasibility constraints based on the feasibility boundary function; determine the current performance lower limit threshold of the insulating material based on the material characteristics, and set material performance threshold conditions based on the performance lower limit threshold; and perform mathematical modeling based on the component compatibility constraints, process feasibility constraints, and material performance threshold conditions to obtain a physical constraint model.

[0067] Step S20: Based on the initial formulation information, generate an initial candidate formulation according to the component compatibility constraints, the process feasibility constraints, and the material performance threshold conditions.

[0068] It should be understood that physical space parameterization can be performed first: the proportion range of each formulation component x i ∈[L i U i As the variable space, a uniformly distributed formula lattice is generated using orthogonal experimental design or Latin hypercube sampling. Furthermore, the formula lattice consists of N formula points, and each formula point x (j) Each formula corresponds to a specific recipe, but each recipe has different ingredients and different proportions of ingredients.

[0069] Next, the formulation points in the formulation matrix can be screened based on the aforementioned component compatibility constraints, process feasibility constraints, and material performance threshold conditions. Formulation points that do not meet the above physical constraints can be quickly eliminated to obtain an updated formulation matrix. Each formulation in this updated formulation matrix is ​​the initial candidate formulation. These initial candidate formulations meet the compatibility, process feasibility, and basic performance requirements in terms of component ratios and types, providing multiple feasible starting points for subsequent optimization.

[0070] In the specific implementation, for each recipe point x (j) We can first construct x (j) The compatibility constraint matrix between each component is set with weights based on the proportion of each component or other correlations. The compatibility score of the formulation point is calculated by weighted summation. If the score is negative, it means that the compatibility of the components at the formulation point meets the requirements, and the components in the formulation can remain stable under interaction without obvious incompatibility problems. Otherwise, it is considered that there are component incompatibility problems at the formulation point, and it can be screened out.

[0071] For each recipe point x (j) Calculate the corresponding process feasibility boundary function g respectively. proc (x (j) If g proc (x (j) )≥0 indicates that it is technically feasible; if g proc (x (j) A value less than 0 indicates that the process is not feasible and the sample can be screened out.

[0072] For each recipe point x (j) The material properties can be predicted using physical property prediction models (such as thermal stability equations, dielectric strength models, mechanical property calculators, etc.) to obtain the predicted property value P. k (x (j) Then, the predicted performance value is compared with the corresponding performance threshold P. k，minIn contrast, if all predicted performance values ​​are greater than the corresponding performance threshold, they are considered to meet the material performance threshold conditions; otherwise, they are not met and can be discarded.

[0073] Step S30: The initial candidate formulation is iteratively optimized using a gradient response optimization algorithm to obtain the current candidate formulation.

[0074] It should be understood that gradient response optimization algorithm is an optimization method based on gradient information. It can determine the search direction and step size according to the gradient of the objective function with respect to each variable (i.e., parameters such as the proportion of each component in the formula), thereby gradually adjusting the parameters of the formula to optimize the value of the objective function.

[0075] In the optimization of insulating material formulations, the objective function is usually related to the material's performance indicators, such as the comprehensive indicators of properties such as dielectric strength and mechanical strength that are desired to be improved.

[0076] It should be understood that, starting with an initial candidate formulation, the gradient response optimization algorithm continuously adjusts parameters such as the proportions of each component based on gradient information. After each adjustment, the objective function value corresponding to the new formulation is calculated and compared with the previous value. If the objective function value changes in the direction of optimization (e.g., increasing or decreasing, depending on the performance requirements), the adjustment continues in that direction; otherwise, the adjustment direction or step size is modified. Through multiple such iterative processes, the performance of the formulation is gradually optimized, resulting in a better current candidate formulation.

[0077] Specifically, this iterative process can be broken down into the following operations: repeatedly calculating sensitivity, determining the optimization direction and step size, updating the formula, and performing projection.

[0078] First, it can be based on the initial candidate formula x (j) Construct a comprehensive performance index F(x) = ∑ k w k f k (x), each performance objective f k (x) can be a quantitative expression for different properties of a material, such as strength, toughness, and corrosion resistance, with corresponding weights w. k It can be pre-configured based on the user's assessment of the importance of each factor.

[0079] Next, the comprehensive performance index F(x) is calculated with respect to x. (j) The partial derivatives of the proportions of each component in the solution, i.e., the sensitivity coefficient. It reflects the degree to which changes in the proportions of each component affect the overall performance under the current formulation. For example, if s i A positive value indicates that increasing the proportion of this component may improve overall performance; if s i A negative value may reduce overall performance.

[0080] The sensitivity coefficients are combined to form a sensitivity vector s = [s1, ..., s2]. m Then, normalize it to obtain the direction. The step size α is dynamically adjusted based on line search or a preset learning rate to ensure that all physical constraints are still met after iterative updates.

[0081] The initial candidate recipe is updated according to the following formula:

[0082]

[0083] Among them, Proj Ω Represents the projection onto the physical constraint space Ω to ensure the updated formula x (t+1) All physical constraints must be met;

[0084] Finally, the above operations of calculating sensitivity, determining optimization direction and step size, updating formula and projecting are repeated until the convergence condition is met or the maximum number of iterations is reached, and the updated formula is determined as the current candidate formula.

[0085] Step S40: Perform property prediction on the current candidate formulation and determine the target optimized formulation based on the property prediction results.

[0086] It should be noted that a pre-established property prediction model can be used to predict various performance indicators, such as thermal stability, dielectric strength, and mechanical properties, based on the components and proportions in the current candidate formulation.

[0087] The aforementioned property prediction models are usually established based on the molecular structure of materials, the principle of component interaction, and existing experimental data. They can quickly and relatively accurately provide estimated values ​​of formulation performance, avoiding the need for a large number of actual experimental tests each time and greatly improving optimization efficiency.

[0088] In practice, the predicted performance indicators can be compared with the pre-set target performance requirements. If the predicted physical properties of the current candidate formulation meet or are close to the target requirements, then the current candidate formulation can be determined as the target optimized formulation. If not, further optimization iterations can be performed based on the current candidate formulation until a target optimized formulation that meets the requirements is found.

[0089] For example, the target optimized formulation obtained by optimizing the formulation based on the aforementioned initial formulation: epoxy resin (60%), silica powder (30%), and curing agent (10%) can be: epoxy resin (55%), micron-sized silica powder (20%), nano-sized silica powder (15%), and modified curing agent (10%).

[0090] Compared to the initial formulation, this optimized formulation can achieve improved dielectric strength (32kV / mm-38kV / mm), reduced processing temperature (150°C-135°C), and reduced material costs (by 12%).

[0091] This embodiment systematically considers various constraints and constructs a comprehensive physical constraint model, thereby ensuring that the generated initial candidate formulations have high feasibility and performance potential. Through iterative optimization using the gradient response optimization algorithm, it can efficiently search the formulation space, gradually improving the various performance indicators of the candidate formulations, avoiding the blindness and inefficiency of traditional trial-and-error methods. Simultaneously, by using property prediction to guide the optimization process, the number and cost of actual experiments are reduced, accelerating the R&D process and enabling the acquisition of target optimized formulations that meet multiple requirements in a shorter time.

[0092] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 , Figure 2 This is a flowchart illustrating the second embodiment of the method for optimizing the formulation of insulating materials according to this application.

[0093] In this embodiment, to specifically illustrate how to generate candidate formulations, step S20 includes: steps S201 to S202:

[0094] Step S201: Initialize the basic proportion range of each formula component according to the initial formula information, and generate several candidate formula points based on the basic proportion range.

[0095] It should be understood that, based on the proportions of each component in the initial formula, and combined with past experience and material characteristics, a reasonable proportion range, or basic proportion range, can be set for each component. For example, if a certain component accounts for 10% in the initial formula, but considering the fluctuations in actual demand and the necessity of optimization, its basic proportion range may be set between 5% and 15%.

[0096] It should also be noted that the interactions between different components can be taken into account when setting the basic ratio range. For example, two components need to maintain a certain ratio to form a stable network structure, so the basic ratio range can be adjusted based on this interaction.

[0097] Next, the physical space can be parameterized based on the set basic ratio range, and a formula matrix can be generated within the basic ratio range of each component. This allows each candidate formula point in the formula matrix to cover combinations of different component ratios, which is beneficial for comprehensively exploring the formula space and increases the chance of finding candidate formulas that meet subsequent physical constraints.

[0098] Step S202: Among the candidate formulation points, each candidate formulation point is screened according to the component compatibility constraint, the process feasibility constraint, or the material performance threshold condition to obtain the candidate formulation points that meet the conditions and determine them as the initial candidate formulations.

[0099] It should be understood that for each candidate formulation point, it can be screened in sequence according to the component compatibility constraint, the process feasibility constraint, or the material performance threshold condition, so that the obtained initial candidate formulation can simultaneously meet the requirements of the above physical constraints.

[0100] Specifically, when screening based on component compatibility constraints, it is possible to check whether the proportions of each component in each candidate formulation meet the previously determined compatibility requirements. For example, if the sum of the proportions of two components is less than a pre-set compatibility threshold, then the candidate formulation that does not meet the compatibility requirements is eliminated.

[0101] When screening based on process feasibility constraints, it's important to consider process requirements to determine whether candidate formulations can be implemented in actual production. For example, if certain component proportions might lead to excessively high material viscosity, preventing smooth passage through the production equipment's pipes, candidate formulations that do not meet the process requirements should be eliminated.

[0102] When screening based on material performance thresholds, existing empirical formulas, databases, or simple performance prediction models can be used to preliminarily determine whether candidate formulations meet basic performance requirements. For example, the dielectric strength of the material can be estimated based on the component ratio to determine if it meets the corresponding minimum requirement; if not, candidate formulations that do not meet the minimum requirement can be eliminated.

[0103] It should be noted that after the above screening based on the physical constraint model, an initial candidate formulation can be obtained. This initial candidate formulation is the starting point for subsequent optimization and can provide a feasible basis for subsequent iterative optimization.

[0104] Furthermore, in order to achieve a comprehensive improvement in the performance indicators of the formulation based on the initial candidate formulation, step S30 specifically includes: steps S301 to S302:

[0105] Step S301: Determine the comprehensive performance index based on the component compatibility constraints, the process feasibility constraints, and the material performance threshold conditions, and construct a gradient response optimization model with the comprehensive performance index as the objective function based on the initial candidate formulation.

[0106] It should be understood that the comprehensive performance index can be a multivariate function that can comprehensively reflect the key properties of the material, such as dielectric strength, mechanical strength, and thermal stability, while also taking into account the requirements of the production process, such as production cost and production efficiency.

[0107] Specifically, the quantitative values ​​corresponding to each performance target in the component compatibility constraint, the process feasibility constraint, and the material performance threshold condition can be weighted and summed to form a comprehensive performance index F(x) = ∑ k w k f k (x), each performance objective f k The weight w corresponding to (x) k It can be pre-configured based on the user's assessment of the importance of each factor.

[0108] Next, using the determined comprehensive performance index as the objective function, the proportion of each component as the variable, and the component compatibility constraint, process feasibility constraint, and material performance threshold condition as constraints, a gradient response optimization model is constructed based on the initial candidate formulation. The gradient direction indicates the optimal path for adjusting the component proportion.

[0109] Finally, by adjusting the component ratios, the overall performance index can be optimized while satisfying all constraints, thus obtaining the optimal solution for the gradient response optimization model.

[0110] Step S302: Solve the gradient response optimization model using the quasi-Newton method to obtain the Pareto optimal frontier, and determine the current candidate formulation based on the Pareto optimal frontier.

[0111] It should be understood that the quasi-Newton method is a commonly used optimization algorithm that determines the search direction and step size by approximating the second derivative information of the objective function. When solving gradient response optimization models, the quasi-Newton method can gradually adjust the proportions of each component in the initial candidate formulation, causing the overall performance index to change in the direction of optimization.

[0112] The Pareto optimal front represents a set of optimal solutions that balance various performance objectives, meaning it is impossible to improve one performance objective without worsening others. This Pareto optimal front can be obtained using a quasi-Newton method, where each formulation on the Pareto optimal front represents a possible solution that balances performance indicators.

[0113] In practice, when the optimization process reaches the convergence condition, the formulation currently on the Pareto optimal front can be determined as the current candidate formulation.

[0114] This embodiment employs a two-layer optimization mechanism. The first layer explores the formulation space within the boundaries of the physical constraint model based on physical constraints to obtain initial candidate formulations. The second layer uses a gradient response optimization algorithm to comprehensively improve performance indicators. This approach effectively searches the formulation space while satisfying physical constraints, finding formulation combinations with superior performance. Because the optimization process in this embodiment is based on scientific models and algorithms, it can overcome the limitations of traditional experience and explore potentially overlooked formulation combinations, providing new possibilities for improving the performance of insulating materials.

[0115] Based on the first and second embodiments of this application, in the third embodiment of this application, the content that is the same as or similar to that in embodiments one and two above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 3 , Figure 3 This is a flowchart illustrating the third embodiment of the method for optimizing the formulation of insulating materials according to this application.

[0116] In this embodiment, to specifically illustrate how to perform multi-level verification on the current candidate formulation to finally determine the target optimized formulation, step S40 specifically includes: steps S401 to S404:

[0117] Step S401: Perform property prediction on the current candidate formulation using a preset property prediction model to obtain the corresponding property prediction results.

[0118] It should be understood that this pre-defined property prediction model can be established based on the molecular structure characteristics of current insulating materials. The molecular structure of current insulating materials can include properties such as the length of molecular chains, the degree of branching, and the types and numbers of functional groups. For example, the number of specific functional groups in a molecular chain may affect the dielectric strength of the material, and the length and branching of the molecular chains may be closely related to mechanical properties. Based on this molecular structure characteristic information, a property prediction model that reflects the relationship between the macroscopic properties and microstructure of the material can be established.

[0119] The preset property prediction model may include: a thermal stability equation, a dielectric strength model, and a mechanical property calculator. To further illustrate the rapid property prediction process for the current candidate formulation, step S401 also includes: steps S4011 to S4013.

[0120] Step S4011: Establish a property prediction model based on the obtained molecular structure and characteristic information of the current insulating material.

[0121] It should be noted that the thermal stability equation is an equation describing the thermal stability of a material by studying its thermal degradation behavior at different temperatures. This equation typically considers key thermal performance parameters such as the glass transition temperature and thermal decomposition temperature, as well as the relationship between temperature, time, and the degree of thermal degradation. For example, it can predict the performance change trend of a material under long-term high-temperature conditions.

[0122] A dielectric strength model is a mathematical model constructed by analyzing the influence of factors such as the molecular polarization characteristics and electronic structure of a material on its dielectric strength. This model can predict the dielectric strength of a material under different electric field strengths and frequencies based on its composition and microstructure, providing a basis for evaluating the insulation performance of materials.

[0123] The mechanical calculator can be a model that calculates the mechanical properties (such as tensile strength and elastic modulus) of a material, taking into account microstructural factors such as intermolecular forces and crosslinking density. By inputting the material's composition and structural information, the calculator can predict the material's mechanical properties, providing support for mechanical performance evaluation in practical applications.

[0124] Step S4012: Perform property prediction on the current candidate formulation according to the property prediction model to obtain thermal stability parameter values, dielectric strength prediction values, and mechanical property quantification values.

[0125] Step S4013: Determine the thermal stability parameter value, the predicted dielectric strength value, and the quantified mechanical property value as the property prediction results of the current candidate formulation.

[0126] It should be understood that the types and proportions of each component in the current candidate formulation, as well as relevant molecular structure parameters (such as molecular weight, functional group density, etc.), can be input into the property prediction model. Based on these input data and combined with the pre-constructed relationship between molecular structure and performance, the property prediction model performs corresponding calculations and predictions.

[0127] In practical implementation, the thermal stability equation can be used, combined with the molecular structure characteristics of each component, to calculate the thermal stability parameters of the current candidate formulation, such as the glass transition temperature (Tg) and thermal decomposition temperature (Td). Based on the dielectric strength model, considering the polarization and conductivity characteristics of each component, the predicted dielectric strength value can be calculated. For example, the predicted dielectric strength of the candidate formulation is 35 kV / mm. Using a mechanical property calculator, based on the molecular chain structure and interactions of each component, the quantitative values ​​of mechanical properties can be evaluated, such as a tensile strength of 80 MPa and an elastic modulus of 3 GPa.

[0128] Step S402: When the property prediction results meet the target performance requirements, perform microstructure simulation verification on the current candidate formulation.

[0129] It should be noted that, based on the actual application requirements of the insulating material, the target performance requirements such as thermal stability, dielectric strength, and mechanical properties that it should achieve can be clearly defined, and corresponding quantification thresholds can be set. Then, the predicted physical properties (thermal stability parameter values, predicted dielectric strength values, and quantified mechanical property values) can be compared one by one with the quantification thresholds corresponding to the target performance requirements.

[0130] If the predicted values ​​of thermal stability parameters, dielectric strength, and mechanical properties all meet or exceed the corresponding target performance requirements, then the current candidate formulation theoretically meets the target performance requirements.

[0131] Next, computer simulation techniques, such as molecular dynamics simulations and quantum chemical calculations, can be used to simulate the microstructure of the current candidate formulations. By simulating the motion and interactions of material molecules, the microstructure can be further verified to see if it supports the predicted macroscopic properties. For example, it can be observed whether there are effective cross-linking structures between molecular chains to enhance mechanical properties, or whether there are suitable molecular arrangements to improve dielectric strength.

[0132] Finally, it can be determined whether the microstructure simulation results show a performance trend consistent with the predicted physical properties, and whether there are any obvious structural defects or microscopic features that contradict the expected performance. If so, the current candidate formulation is deemed to have passed the microstructure simulation verification. For example, if the simulation shows a stable cross-linking network between molecular chains and that the electronic structure can effectively hinder charge migration, which meets the expectations for the microstructure of a good insulating material, then the verification is passed.

[0133] Step S403: When verifying through microstructure simulation, the current candidate formulation is tested with prototype samples.

[0134] Step S404: When verifying through prototype sample testing, the current candidate formulation is determined as the target optimized formulation.

[0135] It should be understood that prototype test verification can be a process of preparing actual prototype samples based on the current candidate formulation and then using a series of experimental test methods to verify their performance.

[0136] The testing can include thermal stability testing (such as thermogravimetric analysis), dielectric strength testing (such as withstand voltage breakdown testing), and mechanical property testing (such as tensile testing). By comparing the experimental test results with the predicted physical properties and target performance requirements, the actual performance of the current candidate formulation in practical applications can be verified.

[0137] It should be noted that when the experimental test results of the prototype sample meet the target performance requirements and are basically consistent with the predicted physical properties, the current candidate formulation is considered to have passed the prototype sample test verification, and the current candidate formulation can be determined as the target optimized formulation.

[0138] It should be understood that the optimized formulation has undergone rigorous verification, from property prediction to microstructure simulation and then to actual prototype testing, proving that it can meet the performance requirements of insulating materials both theoretically and practically, and is a formulation that can be tested in actual production and application.

[0139] Furthermore, in order to apply the target optimized formulation to the actual production process for verification, and to make process corrections to the target optimized formulation based on process feedback during production, after step S404, the process further includes: steps S501 to S504:

[0140] Step S501: Convert the target optimized formula into production parameters through the process compatibility interface, so that the production system to which the process compatibility structure belongs can prepare the current insulating material according to the production parameters.

[0141] It should be understood that the process compatibility interface is a crucial link connecting the formulation optimization system and the actual production system. It can convert information such as component ratios and process conditions in the target optimized formulation into production parameters that the production system can recognize and execute.

[0142] For example, the proportions of epoxy resin, filler, and curing agent in the target formulation can be converted into specific ingredient weights or volumes, and process conditions such as curing temperature and time can be converted into parameters for the temperature settings and time controllers of the heating equipment on the production line.

[0143] It should also be noted that different production systems have different equipment and technical requirements. Process compatibility interfaces need to be adapted to the specific production equipment and process flow. For example, some production equipment may have specific requirements for material flowability; the interface will take this into account when changing production parameters, adjusting the ingredient ratios or adding appropriate additives to ensure the materials can be processed smoothly in the equipment.

[0144] In the specific implementation, data such as the proportions of each component and performance requirements in the target optimized formula are input into the process compatibility interface. The interface parses this data, extracts parameters related to the production process, and maps the extracted parameters to corresponding production parameters according to the characteristics of the production system and process requirements. Finally, the process compatibility interface generates a file or data package containing all necessary production parameters, which can be directly read and executed by the production system.

[0145] Step S502: Determine whether the target optimized formulation meets the requirements of actual process feasibility based on the preparation results.

[0146] It should be understood that the formula optimization system can collect relevant production data, such as actual ingredient weights, temperature profiles, and mixing times, during the preparation of insulating materials based on generated production parameters in the production system. Simultaneously, it records any anomalies that occur during production, such as equipment malfunctions or material flowability issues.

[0147] In addition, the prepared insulating material can be subjected to performance tests, including tests on key indicators such as thermal stability, dielectric strength, and mechanical properties. The test results are compared with the target performance requirements to assess whether the material meets the design requirements. For example, tests may reveal that the dielectric strength of the prepared material is slightly lower than the target value, or the glass transition temperature is lower than expected.

[0148] It should be noted that a set of process feasibility standards can be developed in advance based on production experience and industry standards, including material flowability, controllability of the curing reaction, and production efficiency. For example, the material flowability during the production process should meet the processing requirements of the equipment, the curing reaction should be completed within a set time, and the production efficiency should reach a certain level.

[0149] In practice, the preparation results can be compared and analyzed with process feasibility standards. For example, if the prepared material exhibits good fluidity during production, complete curing reaction, and meets production efficiency requirements, then the target optimized formulation is considered to meet the actual process feasibility requirements. Conversely, if fluidity issues or incomplete curing reaction occur during production, the formulation is considered not to meet the process feasibility requirements.

[0150] Step S503: If so, the target optimized formulation shall be used as the actual application formulation.

[0151] It should be understood that if the target optimized formulation passes the evaluation of the preparation results and the verification of process feasibility, it can be used as a formulation for practical application. This practical application formulation has undergone comprehensive verification from theoretical optimization to actual production and can be promoted and used in actual production.

[0152] Furthermore, detailed records and archives can be maintained for formulations identified for practical application, including formulation components, proportions, process parameters, key data from the preparation process, and performance test results. This facilitates subsequent production and quality control and provides a reference for future formulation optimization.

[0153] Step S504: If not, update the physical constraint model based on the judgment result and return to the step of generating the initial candidate formulation until the regenerated target optimized formulation meets the actual process feasibility requirements.

[0154] It should be understood that if the target optimized formulation fails the process feasibility verification, the physical constraint model needs to be updated to return to step S20 and regenerate the initial candidate formulation. The physical constraint model can be updated in the following ways:

[0155] Adjusting component compatibility constraints: If certain components are found to be incompatible or cause process problems in actual production, their compatibility can be reassessed and adjusted in the physical constraint model. For example, if a certain filler is found to cause poor material flowability at a specific proportion, the upper limit of that filler's proportion can be lowered.

[0156] Revise process feasibility constraints: Process feasibility constraints can be updated based on process problems encountered in actual production. For example, if it is found that excessive curing time affects production efficiency, the upper limit of curing time can be appropriately shortened.

[0157] Adjusting material performance threshold conditions: If the properties of the prepared material do not meet expectations, the material performance threshold conditions need to be re-evaluated. For example, if the dielectric strength does not reach the target value, the dielectric strength threshold requirement can be appropriately increased.

[0158] In practice, after updating the physical constraint model, the process returns to the step of generating the initial candidate formulation and restarts the optimization process. This includes re-initializing the basic proportion ranges of each formulation component based on the updated physical constraint model, generating new candidate formulation points, and then screening and optimizing them. Through this iterative process, the formulation is gradually improved until the regenerated target optimized formulation meets the requirements of practical process feasibility.

[0159] This embodiment achieves multi-level verification of the current candidate formulation through rapid property prediction, microstructure simulation, and prototype sample testing. The target optimized formulation, determined through multi-level verification, is then converted into process parameters based on process compatibility. Process feasibility is verified during production until a practically feasible target optimized formulation is obtained. This ensures that the target optimized formulation not only possesses excellent theoretical performance but also demonstrates feasibility and reliability in actual production. Furthermore, this embodiment, based on iterative optimization using actual process feedback, effectively improves the optimization quality and efficiency of insulating material formulations, providing strong support for industrial production.

[0160] Furthermore, this can be combined with Figure 4 The system architecture of the method for optimizing the formulation of insulating materials in this application is described. Figure 4 This is a schematic diagram of the system architecture for optimizing the formulation of the insulating material in this application.

[0161] Depend on Figure 4It can be seen that the system architecture for optimizing the formulation of insulating materials can be divided into a physical constraint modeling part, a multi-objective optimization engine, a physical property prediction model, and a process compatibility interface part.

[0162] In the physical constraint modeling part, a component compatibility constraint matrix can be established to ensure chemical stability (component compatibility constraint condition), a process feasibility boundary function can be constructed to ensure manufacturability (process feasibility constraint condition), and a physical lower limit threshold of material properties can be defined to ensure basic functionality (material property threshold condition). In this way, a physical constraint model can be constructed and connected to a multi-objective optimization engine.

[0163] In the multi-objective optimization engine, the initial formulation information of the current insulating material can be obtained from the initial formulation library. Then, a two-layer optimization mechanism is adopted in the multi-objective optimization engine: the first layer explores the formulation space based on physical constraints within the boundary of the physical constraint model to obtain the initial candidate formulation; the second layer uses the gradient response optimization algorithm to comprehensively improve the performance indicators, which can effectively search in the formulation space and determine the current candidate formulation while satisfying the physical constraints.

[0164] Next, a pre-established property prediction model can be used to predict various performance indicators, such as thermal stability, dielectric strength, and mechanical properties, based on the components and proportions in the current candidate formulation. This property prediction model can be a simplified property prediction equation based on the molecular structure and interaction principles of materials. Furthermore, after property prediction, microstructure simulation and prototype sample testing can be performed to obtain the target optimized formulation that has passed multiple levels of verification.

[0165] Finally, the optimized formula is input into the process compatibility interface and converted into production parameters. This allows the insulation material to be prepared in the production system based on the production parameters, and continuous feedback optimization of the multi-objective optimization engine is achieved based on process constraints.

[0166] Accordingly, you can also refer to this section. Figure 5 This application provides a complete description of the formulation optimization method for the insulating material. Figure 5 This is a schematic diagram of the entire process of the formulation optimization method for the insulating material in this application.

[0167] Depend on Figure 5 As can be seen, the physical constraint space can be initialized first, and then constraint modeling can be performed based on the composition compatibility constraint, process feasibility constraint, and material performance threshold condition to obtain the physical constraint model.

[0168] After generating the initial candidate formulation based on the physical constraint model, the gradient response optimization algorithm is used to iteratively optimize the initial candidate formulation until the convergence condition corresponding to the objective function is met, and the current candidate formulation is obtained.

[0169] The current candidate formulation is subjected to rapid prediction of physical properties, simulation of microstructure, and testing of prototype samples in sequence, and the target optimized formulation is output.

[0170] The target optimized formula is converted into process parameters based on process compatibility, and the process feasibility of the target optimized scheme is judged. If the process feasibility is met, the output is given; otherwise, the physical preset model is adjusted and optimized again.

[0171] The method described in this application is based on iterative optimization using feedback from actual processes, which can effectively improve the optimization quality and efficiency of insulation material formulations and provide strong support for industrial production.

[0172] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the formulation optimization method of the insulating material of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0173] This application also provides a device for optimizing the formulation of insulating materials. Please refer to... Figure 6 , Figure 6 This is a schematic diagram of the module structure of the insulating material formulation optimization device of this application. The device includes:

[0174] The constraint modeling module 601 is used to obtain the initial formula information corresponding to the current insulating material and construct a physical constraint model based on the initial formula information. The physical constraint model consists of component compatibility constraints, process feasibility constraints, and material performance threshold conditions.

[0175] The dual-layer optimization module 602 is used to generate an initial candidate formulation based on the initial formulation information, according to the component compatibility constraints, the process feasibility constraints, and the material performance threshold conditions.

[0176] The dual-layer optimization module 602 is further used to iteratively optimize the initial candidate formula using a gradient response optimization algorithm to obtain the current candidate formula;

[0177] The property prediction module 603 is used to predict the properties of the current candidate formulation and determine the target optimized formulation based on the property prediction results.

[0178] This embodiment systematically considers various constraints and constructs a comprehensive physical constraint model, thereby ensuring that the generated initial candidate formulations have high feasibility and performance potential. Through iterative optimization using the gradient response optimization algorithm, it can efficiently search the formulation space, gradually improving the various performance indicators of the candidate formulations, avoiding the blindness and inefficiency of traditional trial-and-error methods. Simultaneously, by using property prediction to guide the optimization process, the number and cost of actual experiments are reduced, accelerating the R&D process and enabling the acquisition of target optimized formulations that meet multiple requirements in a shorter time.

[0179] This application also provides an apparatus for optimizing the formulation of insulating materials, the apparatus comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method for optimizing the formulation of insulating materials in the first embodiment described above.

[0180] The following is for reference. Figure 7 , Figure 7 This is a schematic diagram of the structure of the insulation material formulation optimization device of this application. The insulation material formulation optimization device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), etc., as well as fixed terminals such as digital TVs, desktop computers, etc. Figure 7 The apparatus for optimizing the formulation of insulating materials shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0181] like Figure 7As shown, the insulation material formulation optimization device may include a processor 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the insulation material formulation optimization device. The processor 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. The communication device 1009 allows the insulation material formulation optimization device to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows insulation material formulation optimization devices with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0182] The insulating material formulation optimization device provided in this application, employing the insulating material formulation optimization method in the above embodiments, can solve the technical problem of insulating material formulation optimization. Compared with the prior art, the beneficial effects of the insulating material formulation optimization device provided in this application are the same as the beneficial effects of the insulating material formulation optimization method provided in the above embodiments, and other technical features in the insulating material formulation optimization device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0183] This application also provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the method for optimizing the formulation of insulating materials in the above embodiments.

[0184] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0185] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described method for optimizing the formulation of insulating materials, thereby solving the technical problem of optimizing the formulation of insulating materials. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the method for optimizing the formulation of insulating materials provided in the above embodiments, and will not be repeated here.

[0186] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other elements in the process, method, article, or system that includes that element.

[0187] The sequence numbers of the above embodiments of the present invention are merely for description and do not represent the superiority or inferiority of the embodiments. They are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the content of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included within the patent protection scope of this application.

Claims

1. A method for optimizing the formulation of an insulating material, characterized in that, The method comprises: obtaining initial formula information corresponding to the current insulation material, and constructing a physical constraint model according to the initial formula information, wherein the physical constraint model is composed of component compatibility constraint conditions, process feasibility constraint conditions and material performance threshold conditions; generating an initial candidate formula based on the initial formula information, the component compatibility constraint conditions, the process feasibility constraint conditions and the material performance threshold conditions; iteratively optimizing the initial candidate formula by using a gradient response optimization algorithm to obtain a current candidate formula; performing physical property prediction on the current candidate formula, and determining a target optimized formula according to the physical property prediction result.

2. The method of claim 1, wherein, The initial formula information includes formula components, formula processes and material characteristics. The step of constructing a physical constraint model according to the initial formula information comprises: determining compatibility coefficients between each of the formula components based on historical experience data, and establishing a component compatibility constraint matrix according to the compatibility coefficients; determining feasibility boundary functions of each formula process based on preset process requirement information, and setting process feasibility constraint conditions according to the feasibility boundary functions; determining a performance lower limit threshold of the current insulation material according to the material characteristics, and setting a material performance threshold condition according to the performance lower limit threshold; mathematically modeling according to the component compatibility constraint conditions, the process feasibility constraint conditions and the material performance threshold conditions to obtain a physical constraint model.

3. The method of claim 1, wherein, The step of generating an initial candidate formula based on the initial formula information, the component compatibility constraint conditions, the process feasibility constraint conditions and the material performance threshold conditions comprises: initializing basic proportion ranges of each formula component according to the initial formula information, and generating a plurality of candidate formula points based on the basic proportion ranges; screening each of the candidate formula points according to the component compatibility constraint conditions, the process feasibility constraint conditions or the material performance threshold conditions to obtain a candidate formula point that meets the conditions, and determining the candidate formula point as an initial candidate formula.

4. The method of claim 1, wherein, The step of iteratively optimizing the initial candidate formula by using a gradient response optimization algorithm to obtain a current candidate formula comprises: determining a comprehensive performance index according to the component compatibility constraint conditions, the process feasibility constraint conditions and the material performance threshold conditions, and constructing a gradient response optimization model with the comprehensive performance index as an objective function based on the initial candidate formula; solving the gradient response optimization model by using a quasi-Newton method to obtain a Pareto optimal frontier, and determining a current candidate formula according to the Pareto optimal frontier.

5. The method of claim 1, wherein, The step of performing physical property prediction on the current candidate formula, and determining a target optimized formula according to the physical property prediction result comprises: performing physical property prediction on the current candidate formula by using a preset physical property prediction model to obtain a corresponding physical property prediction result; when the physical property prediction result meets a target performance requirement, performing microstructure simulation verification on the current candidate formula; when the microstructure simulation verification is passed, performing prototype sample test verification on the current candidate formula; The current candidate formula is determined as the target optimization formula when verified by prototype sample testing.

6. The method of claim 5, wherein, The step of predicting the physical properties of the current candidate formula by the preset physical property prediction model comprises: A physical property prediction model is established according to the obtained molecular structure characteristic information of the current insulating material, and the physical property prediction model comprises a thermal stability equation, a dielectric strength model and a mechanical performance calculator; The thermal stability parameter value, the dielectric strength prediction value and the mechanical performance quantitative value are determined as the physical property prediction results of the current candidate formula. The step of determining the current candidate formula as the target optimization formula when verified by prototype sample testing further comprises:

7. The method of claim 5, wherein, The target optimization formula is converted into production parameters through a process compatibility interface, so that a production system to which the process compatibility structure belongs prepares the current insulating material according to the production parameters; The target optimization formula is corrected in process parameters according to the preparation result, and a target optimization scheme corrected in process parameters is obtained. The device comprises:

8. An insulation material formulation optimization apparatus, characterized by, A constraint modeling module is configured to obtain initial formula information corresponding to a current insulating material, and to construct a physical constraint model according to the initial formula information, wherein the physical constraint model is composed of component compatibility constraint conditions, process feasibility constraint conditions and material performance threshold conditions; A double-layer optimization module is configured to generate an initial candidate formula based on the initial formula information according to the component compatibility constraint conditions, the process feasibility constraint conditions and the material performance threshold conditions; The double-layer optimization module is further configured to obtain a current candidate formula by iteratively optimizing the initial candidate formula using a gradient response optimization algorithm; A physical property prediction module is configured to predict the physical properties of the current candidate formula, and to determine a target optimization formula according to the physical property prediction results. The device comprises a memory, a processor and an insulating material formula optimization program stored on the memory and executable on the processor, and the insulating material formula optimization program, when executed by the processor, implements the steps of the insulating material formula optimization method according to any one of claims 1 to 7.

9. An insulation material formulation optimization apparatus, characterized by, The storage medium stores an insulating material formula optimization program, and the insulating material formula optimization program, when executed by a processor, implements the steps of the insulating material formula optimization method according to any one of claims 1 to 7.

10. A storage medium, characterized by ​

Images