Ceramic formula optimization method and system

Abstract

The present invention relates to the technical field of ceramic production, and in particular to a ceramic formula optimization method and system, comprising: collecting formula priori knowledge data to construct a ceramic formula priori knowledge base, and collecting historical formula data and raw material data of ceramic production to construct a ceramic historical formula database; based on the priori knowledge, raw material data and historical formula data, adopting knowledge-driven and data-driven integrated learning to construct a ceramic formula performance prediction model; defining the ceramic formula raw material optimization problem as a single-objective or multi-objective optimization problem with constraints and mixed variables, and constructing an optimization model; optimizing the formula using a data-driven evolutionary optimization algorithm; performing constraint processing, robust optimization and multi-solution optimization in turn; and integrating and developing a ceramic raw material formula optimization system based on the above model and optimization algorithm; the present invention has the effect of solving the efficiency and trial-and-error cost problems of ceramic formula optimization design.

Description

Technical Field

[0001] The invention relates to the technical field of ceramic production, in particular to a ceramic formula optimization method and system. Background Art

[0002] At present, the design of ceramic raw material formulas is highly dependent on the prior knowledge of technical engineers, and there are problems such as huge trial and error costs and low efficiency in formula design.

[0003] The existing ceramic formula optimization and design mainly rely on two methods. One is to rely entirely on manual experience and obtain multiple formulas through calculation. The other method is to manually use Excel tools to design the formula, and then conduct manual experiments. The workflow is as follows: technical engineers judge the raw materials, select several raw materials, design and optimize the formula, and determine multiple ceramic formula schemes; the designed multiple formula schemes are tested through manual mixing, and the better formula scheme is selected through orthogonal experiments.

[0004] The above formula optimization design methods are all based on human subjective experience to design and judge the quality of the formula. However, considering the limited knowledge stored by humans and the limitations of problem considerations, the existing methods are not only inefficient, but also have high trial and error costs. At the same time, it is also impossible to determine whether the obtained formula is the optimal formula solution.

[0005] Therefore, it is necessary to study data-driven computational intelligence methods for optimizing ceramic raw material formulas and to build an intelligent ceramic body formula optimization system as a tool for the practicality and universality of ceramic production formula design. Summary of the invention

[0006] In view of the above-mentioned defects, the purpose of the present invention is to propose a ceramic formula optimization method and system to solve the problems of efficiency and trial-and-error cost in ceramic formula optimization design.

[0007] To achieve this object, the present invention adopts the following technical solutions:

[0008] The ceramic formula optimization method comprises the following steps:

[0009] A. Collect formula prior knowledge data, build a ceramic formula prior knowledge base, and collect historical formula data and raw material data of ceramic production to build a ceramic historical formula database;

[0010] B. Based on prior knowledge, raw material data and historical formula data, knowledge-driven and data-driven integrated learning is adopted to build a ceramic formula performance prediction model;

[0011] C. Define the ceramic formula raw material optimization problem as a single-objective or multi-objective optimization problem with constraints and mixed variables, and build an optimization model;

[0012] D. Use data-driven evolutionary optimization algorithms to optimize the formula;

[0013] E. Perform constraint processing, robust optimization and multi-solution optimization in sequence;

[0014] F. Based on the above models and optimization algorithms, an integrated ceramic raw material formula optimization system was developed.

[0015] Preferably, the step B of constructing a ceramic formulation performance prediction model includes the following contents:

[0016] b1. Based on the ceramic formula prior rules contained in the ceramic formula prior knowledge base, the ant colony mining system algorithm and genetic programming method are used to construct a decision tree and generate a decision tree model based on rule reasoning;

[0017] b2. Establish regression and classification models based on the ceramic historical formula database and machine learning methods;

[0018] b3. Through the integrated learning mechanism, the decision tree model and machine learning method are integrated to realize the ceramic formula performance prediction model that integrates data and knowledge-driven.

[0019] Preferably, the step C of constructing the optimization model includes the following contents:

[0020] c1. Set the quantity and type of raw materials in the formula;

[0021] c2. Determine the raw materials by price;

[0022] c3. Add constraints to the ceramic formula performance prediction model to form a constrained optimization mathematical model.

[0023] Preferably, the step b1 of using the ant colony mining system algorithm and the genetic programming method to generate a decision tree model includes the following steps:

[0024] Input the number of raw material types m, and initialize the number of cycles Nc = 0;

[0025] Set the initial raw material type k=1, select the next raw material in turn according to the ceramic historical formula database, that is, raw material type k=k+1, and modify the taboo table, and cycle until k≥m, and the cycle ends;

[0026] At the end of the above cycle, the obtained raw material data is updated to determine whether it meets the formula requirements. If so, an online inspection is performed to determine whether it is qualified. If qualified, the result is output; if it does not meet the formula requirements or the online inspection is unqualified, the number of cycles Nc=Nc+1 is set, and the above cycle is performed again.

[0027] Preferably, the robust optimization step uses a robust optimization mechanism based on probability distribution; and the multi-solution optimization uses a multi-solution optimization mechanism based on niche.

[0028] Ceramic formula optimization system, including data acquisition module, model building module and formula optimization module;

[0029] The data acquisition module is used to collect formula prior knowledge data, raw material data and historical formula data, and to build a ceramic formula prior knowledge base and a ceramic historical formula database;

[0030] The model building module is used to build a ceramic formula performance prediction model;

[0031] The recipe optimization module is used to construct an optimization model, optimize the recipe using a data-driven evolutionary optimization algorithm, and perform constraint processing, robust optimization, and multi-solution optimization;

[0032] The data acquisition module, the model building module and the recipe optimization module are electrically connected.

[0033] Preferably, the model building module is provided with a first operator module, and the first operator module generates a decision tree model using an ant colony mining system algorithm and a genetic programming method, including the following contents:

[0034] Input the number of raw material types m, and initialize the number of cycles Nc = 0;

[0035] Set the initial raw material type k=1, select the next raw material in turn according to the ceramic historical formula database, that is, raw material type k=k+1, and modify the taboo table, and cycle until k≥m, and the cycle ends;

[0036] At the end of the above cycle, the obtained raw material data is updated to determine whether it meets the formula requirements. If so, an online inspection is performed to determine whether it is qualified. If qualified, the result is output; if it does not meet the formula requirements or the online inspection is unqualified, the number of cycles Nc=Nc+1 is set, and the above cycle is performed again.

[0037] Preferably, a second operator module is provided in the recipe optimization module, and the second operator module is used to perform operations of data-driven evolutionary optimization algorithm, robust optimization based on probability distribution and multi-solution optimization based on microhabitat.

[0038] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the ceramic formula optimization method as described above when executing the program.

[0039] A non-transitory computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the ceramic formula optimization method as described above.

[0040] The above technical solution has the following beneficial effects:

[0041] Based on the currently accumulated ceramic raw material data and the prior knowledge of technical engineers, this embodiment studies a ceramic formula performance prediction model driven by data and prior knowledge, and integrates a decision tree model based on prior knowledge and a machine learning model based on historical formula data to achieve prediction and evaluation of ceramic raw material formula performance; the ceramic formula raw material optimization problem is defined as a single-objective or multi-objective optimization problem with constraints and mixed variables, and an evolutionary calculation method is designed to achieve formula optimization; in addition, in view of the inaccurate and unstable characteristics of the ceramic formula performance prediction model, and the rationality and interpretability of the generated solution of the data-driven evolutionary optimization algorithm, a robust multi-solution formula optimization algorithm is proposed, which can provide technical engineers with multiple candidate formulas with excellent prediction performance at the same time, so that technical engineers can choose a more reasonable formula for trial production, which can improve the efficiency of ceramic formula optimization design, shorten the time of formula testing, and reduce the trial and error cost of the formula, and solve the problem of the dependence of the formula design process on technical engineers. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the overall process of an embodiment of the present invention;

[0043] Figure 2 is a flow chart of constructing a ceramic formula performance prediction model in step B in an embodiment of the present invention;

[0044] Figure 3 It is a flow chart of constructing an optimization model in step C in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0046] In the description of the present invention, it is necessary to understand that the terms "longitudinal", "transverse", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, features defined as "first" or "second" may explicitly or implicitly include one or more of the features, and are used to distinguish and describe features, without distinction of order or importance.

[0047] In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0048] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0049] Combine the following Figures 1 to 3 The ceramic formula optimization method and system of the embodiment of the present invention are described as follows:

[0050] The ceramic formula optimization method comprises the following steps:

[0051] A. Collect formula prior knowledge data, build a ceramic formula prior knowledge base, and collect historical formula data and raw material data of ceramic production to build a ceramic historical formula database;

[0052] B. Based on prior knowledge, raw material data and historical formula data, knowledge-driven and data-driven integrated learning is adopted to build a ceramic formula performance prediction model;

[0053] C. Define the ceramic formula raw material optimization problem as a single-objective or multi-objective optimization problem with constraints and mixed variables, and build an optimization model;

[0054] D. Use data-driven evolutionary optimization algorithms to optimize the formula;

[0055] E. Perform constraint processing, robust optimization and multi-solution optimization in sequence;

[0056] F. Based on the above models and optimization algorithms, an integrated ceramic raw material formula optimization system was developed.

[0057] Specifically, this embodiment, based on the currently accumulated ceramic raw material data and the prior knowledge of technical engineers, studies a ceramic formula performance prediction model driven by data and prior knowledge, and integrates a decision tree model based on prior knowledge and a machine learning model based on historical formula data to achieve prediction and evaluation of ceramic raw material formula performance; the ceramic formula raw material optimization problem is defined as a single-objective or multi-objective optimization problem with constraints and mixed variables, and an evolutionary calculation method is designed to achieve formula optimization; in addition, in view of the inaccurate and unstable characteristics of the ceramic formula performance prediction model, and the rationality and interpretability of the generated solution of the data-driven evolutionary optimization algorithm, a robust multi-solution formula optimization algorithm is proposed, which can provide technical engineers with multiple candidate formulas with excellent prediction performance at the same time, so that technical engineers can choose a more reasonable formula for trial production, which can improve the efficiency of ceramic formula optimization design, shorten the time of formula testing, and reduce the trial and error cost of the formula, and solve the problem of the dependence of the formula design process on technical engineers.

[0058] In this embodiment, the ceramic formula prior knowledge base is obtained by connecting with the company's ceramic production technical engineers to understand the specific background prior knowledge of the ceramic raw material formula, and constructing a ceramic formula prior knowledge base through knowledge expression to form a series of ceramic formula prior rules in the form of IF Then. In addition, historical formula data of ceramic production, including various successful and failed formulas, are collected and annotated to form a ceramic historical formula database.

[0059] In step D, the recipe is optimized using the latest proposed hierarchical particle swarm learning optimization algorithm assisted by classification models and other related algorithms.

[0060] Preferably, the step B of constructing a ceramic formulation performance prediction model includes the following contents:

[0061] b1. Based on the ceramic formula prior rules contained in the ceramic formula prior knowledge base, the ant colony mining system algorithm and genetic programming method are used to construct a decision tree and generate a decision tree model based on rule reasoning;

[0062] b2. Establish regression and classification models based on the ceramic historical formula database and machine learning methods;

[0063] b3. Through the integrated learning mechanism, the decision tree model and machine learning method are integrated to realize the ceramic formula performance prediction model that integrates data and knowledge-driven.

[0064] Specifically, the ceramic formula performance prediction model is driven by the ceramic historical formula database, and must also conform to the various ceramic formula prior knowledge and principles accumulated by ceramic production technical engineers to ensure that the constructed model is interpretable and reasonable. Therefore, a knowledge-driven and data-driven integrated learning model must be adopted, and ultimately the ceramic formula performance prediction model must be tested in the test set and newly collected data, and optimized with technical engineers to verify the effectiveness of this ceramic formula performance prediction model.

[0065] Specifically, the data from the ceramic formula prior knowledge base and the ceramic historical formula database, including the chemical composition and physical data of the ceramic formula, the types of raw materials, the proportion of each raw material, etc., are used as model training and rule benchmarks; combined with the above database and ant colony mining and genetic programming methods, each optimal raw material is found one by one to combine into the optimal formula system.

[0066] After multiple raw materials are combined, they are matched with the chemical and physical data of the formula in the ceramic historical formula database to determine whether they meet the requirements. If not, the cycle is repeated to find the best combination of multiple raw materials to form a formula. If they do, the online verification test is performed to determine whether they meet the technical requirements through various tests. If they do, the results are output. If not, the cycle is repeated again to find the best combination of raw materials to form a formula.

[0067] Preferably, the step C of constructing the optimization model includes the following contents:

[0068] c1. Set the quantity and type of raw materials in the formula;

[0069] c2. Determine the raw materials by price;

[0070] c3. Add constraints to the ceramic formula performance prediction model to form a constrained optimization mathematical model.

[0071] Specifically, based on the ceramic formula performance prediction model, combined with production cost, product category, quality requirements and other conditions, a constrained optimization mathematical model of the ceramic formula is constructed; wherein the constraints are production cost, chemical composition constraint, dosage constraint and classification dosage constraint, etc.; by defining the ceramic formula raw material optimization problem as a single-objective or multi-objective optimization problem with constraints and mixed variables, an evolutionary calculation method is designed to achieve formula optimization.

[0072] Preferably, the step b1 of using the ant colony mining system algorithm and the genetic programming method to generate a decision tree model includes the following steps:

[0073] Input the number of raw material types m, and initialize the number of cycles Nc = 0;

[0074] Set the initial raw material type k=1, select the next raw material in turn according to the ceramic historical formula database, that is, raw material type k=k+1, and modify the taboo table, and cycle until k≥m, and the cycle ends;

[0075] At the end of the above cycle, the obtained raw material data is updated to determine whether it meets the formula requirements. If so, an online inspection is performed to determine whether it is qualified. If qualified, the result is output; if it does not meet the formula requirements or the online inspection is unqualified, the number of cycles Nc=Nc+1 is set, and the above cycle is performed again.

[0076] Preferably, the robust optimization step uses a robust optimization mechanism based on probability distribution; and the multi-solution optimization uses a multi-solution optimization mechanism based on niche.

[0077] Specifically, in practical applications, the above evolutionary optimization algorithm needs to be further tuned. The complexity of the actual ceramic production process puts forward new requirements for the design of the optimization algorithm. For example, due to the huge cost of trial and error, offline data-driven optimization must be used as much as possible in the optimization process to reduce unnecessary trial and error times; the formula must meet many constraints such as cost and raw material quality; due to the uncertainty of the ceramic formula performance prediction model, the solution generated by the evolutionary optimization algorithm needs to have high robustness, and at the same time must be reasonable and interpretable, and conform to the industry's technical cognition. In view of the above complex characteristics, it is necessary to adopt mechanisms such as robust optimization based on probability distribution and multi-solution optimization based on microhabitats, so as to provide enterprises with multiple optimization solutions at the same time in one algorithm operation, so that technical engineers can select solutions with high rationality and interpretability for trial production, thereby effectively reducing the cost of trial and error and improving production efficiency.

[0078] Specifically, a robust optimization mechanism based on probability distribution is used to improve the optimization performance of the algorithm: according to the distribution of the current solution in the search space, candidate solutions are selected according to probability, which can make full use of the information obtained from evolution and search around the effective information. It also maintains the algorithm's ability to explore areas with less information, ensures the diversity of the population, and avoids falling into local optimality. The multi-solution optimization mechanism based on microhabitats can prevent all individuals from converging to one area, thereby providing solutions that meet the constraints with greater differences, and ultimately obtain multiple optimized solutions for technical engineers to choose from.

[0079] This embodiment also discloses a ceramic formula optimization system, including a data acquisition module, a model building module and a formula optimization module;

[0080] The data acquisition module is used to collect formula prior knowledge data, raw material data and historical formula data, and to build a ceramic formula prior knowledge base and a ceramic historical formula database;

[0081] The model building module is used to build a ceramic formula performance prediction model;

[0082] The recipe optimization module is used to construct an optimization model, optimize the recipe using a data-driven evolutionary optimization algorithm, and perform constraint processing, robust optimization, and multi-solution optimization;

[0083] The data acquisition module, the model building module and the recipe optimization module are electrically connected.

[0084] Preferably, the model building module is provided with a first operator module, and the first operator module generates a decision tree model using an ant colony mining system algorithm and a genetic programming method, including the following contents:

[0085] Input the number of raw material types m, and initialize the number of cycles Nc = 0;

[0086] Set the initial raw material type k=1, select the next raw material in turn according to the ceramic historical formula database, that is, raw material type k=k+1, and modify the taboo table, and cycle until k≥m, and the cycle ends;

[0087] At the end of the above cycle, the obtained raw material data is updated to determine whether it meets the formula requirements. If so, an online inspection is performed to determine whether it is qualified. If qualified, the result is output; if it does not meet the formula requirements or the online inspection is unqualified, the number of cycles Nc=Nc+1 is set, and the above cycle is performed again.

[0088] Preferably, a second operator module is provided in the recipe optimization module, and the second operator module is used to perform operations of data-driven evolutionary optimization algorithm, robust optimization based on probability distribution and multi-solution optimization based on microhabitat.

[0089] This embodiment also discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the ceramic formula optimization method described above when executing the program.

[0090] This embodiment also discloses a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the ceramic formula optimization method described above are implemented.

[0091] Other components and operations of the ceramic formula optimization method and system according to the embodiment of the present invention are well known to those skilled in the art and will not be described in detail here.

[0092] Each module in the above ceramic formula optimization system can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in an electronic device in the form of hardware, or can be stored in a memory of an electronic device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0093] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0094] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0095] The above description of the specific embodiments of the present invention is only for illustrating the technical route and features of the present invention, and its purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly, but the present invention is not limited to the above specific implementation methods. Any changes or modifications made within the scope of the claims of the present invention should be included in the protection scope of the present invention.

Claims

1. A ceramic formula optimization method, characterized in that: The steps include: A. Collect formula prior knowledge data, build a ceramic formula prior knowledge base, and collect historical formula data and raw material data of ceramic production to build a ceramic historical formula database; B. Based on prior knowledge, raw material data and historical formula data, knowledge-driven and data-driven integrated learning is adopted to build a ceramic formula performance prediction model; C. Define the ceramic formula raw material optimization problem as a single-objective or multi-objective optimization problem with constraints and mixed variables, and build an optimization model; D. Use data-driven evolutionary optimization algorithms to optimize the formula; E. Perform constraint processing, robust optimization and multi-solution optimization in sequence; F. Based on the above models and optimization algorithms, a ceramic raw material formula optimization system was developed; The step B of constructing a ceramic formulation performance prediction model includes the following contents: b1. Based on the ceramic formula prior rules contained in the ceramic formula prior knowledge base, the ant colony mining system algorithm and genetic programming method are used to construct a decision tree and generate a decision tree model based on rule reasoning; b2. Establish regression and classification models based on the ceramic historical formula database and machine learning methods; b3. Through the integrated learning mechanism, the decision tree model and machine learning method are integrated to realize the ceramic formula performance prediction model that integrates data and knowledge-driven.

2. The ceramic formula optimization method according to claim 1, characterized in that: The C step of constructing the optimization model includes the following contents: c1. Set the quantity and type of raw materials in the formula; c2. Determine the raw materials by price; c3. Add constraints to the ceramic formula performance prediction model to form a constrained optimization mathematical model.

3. The ceramic formula optimization method according to claim 1, characterized in that: The step b1 in which the ant colony mining system algorithm and the genetic programming method are used to generate a decision tree model includes the following steps: Input the number of raw material types m, and initialize the number of cycles Nc=0; Set the initial raw material type k=1, select the next raw material in turn according to the ceramic historical formula database, that is, raw material type k=k+1, and modify the taboo table, and loop until k≥m, and the loop ends; At the end of the above cycle, the obtained raw material data is updated to determine whether it meets the formula requirements. If so, an online inspection is performed to determine whether it is qualified. If qualified, the result is output; if it does not meet the formula requirements or the online inspection is unqualified, the number of cycles Nc=Nc+1 is set, and the above cycle is performed again.

4. The ceramic formula optimization method according to claim 1, characterized in that: The robust optimization step uses a robust optimization mechanism based on probability distribution; the multi-solution optimization uses a multi-solution optimization mechanism based on niche.

5. Ceramic formula optimization system, characterized by: It includes data acquisition module, model building module and recipe optimization module; The data acquisition module is used to collect formula prior knowledge data, raw material data and historical formula data, and to build a ceramic formula prior knowledge base and a ceramic historical formula database; The model building module is used to build a ceramic formula performance prediction model; The recipe optimization module is used to construct an optimization model, optimize the recipe using a data-driven evolutionary optimization algorithm, and perform constraint processing, robust optimization, and multi-solution optimization; The data acquisition module, the model building module and the recipe optimization module are electrically connected; The model building module is provided with a first operator module, which generates a decision tree model using an ant colony mining system algorithm and a genetic programming method, including the following contents: Input the number of raw material types m, and initialize the number of cycles Nc=0; Set the initial raw material type k=1, select the next raw material in turn according to the ceramic historical formula database, that is, raw material type k=k+1, and modify the taboo table, and loop until k≥m, and the loop ends; At the end of the above cycle, the obtained raw material data is updated to determine whether it meets the formula requirements. If so, an online inspection is performed to determine whether it is qualified. If qualified, the result is output; If the formula requirements are not met or the online inspection is unqualified, the number of cycles Nc=Nc+1 is set and the above cycle is repeated.

6. The ceramic formula optimization system according to claim 5, characterized in that: The recipe optimization module is provided with a second operator module, and the second operator module is used to perform operations of data-driven evolutionary optimization algorithm, robust optimization based on probability distribution and multi-solution optimization based on microhabitat.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the ceramic formula optimization method according to any one of claims 1 to 4 are implemented.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the ceramic formula optimization method according to any one of claims 1 to 4 are implemented.

Images