A method for constructing a nodular cast iron structure and process coupling bidirectional design model

Abstract

This invention relates to the field of ductile iron technology. Specifically, it relates to a method for constructing a bidirectional design model coupling ductile iron microstructure and process. The method includes the following steps: S1, collecting historical production data of ductile iron, binding the production process, production organization, and casting products in the historical production data, and simultaneously obtaining user demand information for the casting products. This demand information is then divided into multi-dimensional demand directions, and the overall demand information is segmented and matched according to these multi-dimensional demand directions. This invention realizes the reverse derivation of ductile iron from user demand to process organization design, breaking the limitations of unidirectional design and significantly improving the accuracy and efficiency of casting design. By uniquely binding the production process, production organization, and casting products to each batch, a complete historical production data system is constructed. Combined with the division and matching of multi-dimensional demand directions, the vague user demands are transformed into quantifiable sub-indicators.

Description

Technical Field

[0001] This invention relates to the field of ductile iron technology, and more specifically, to a method for constructing a two-way design model that couples the microstructure and process of ductile iron. Background Technology

[0002] Due to its excellent mechanical properties, casting performance, and cost-effectiveness, ductile iron has become a structural material in fields such as machinery manufacturing, rail transportation, and wind power equipment. Its production process and internal structure control directly determine the performance indicators and service reliability of the casting products.

[0003] Current technology relies on manual experience to select process parameters, lacking quantitative correlation analysis between process, microstructure, and product performance. This easily leads to a disconnect between process design and actual performance requirements, excessive trial castings, and increased production raw material and time costs. At the same time, the mapping relationship between process, microstructure, and product performance is a one-way design mode, which cannot reverse-engineer the optimal process and microstructure scheme based on the user's performance requirements for the casting, making it difficult to quickly respond to personalized and customized product needs. Therefore, a method for constructing a two-way design model coupling microstructure and process for ductile iron is proposed. Summary of the Invention

[0004] The purpose of this invention is to provide a method for constructing a two-way design model that couples the microstructure and process of ductile iron, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, a method for constructing a two-way design model coupling microstructure and process of ductile iron is provided, comprising the following steps: S1. Collect historical production data of ductile iron, bind the production process and production organization in the historical production data with the casting products, and at the same time obtain the user's demand information for the casting products. Divide the demand information into multi-dimensional demand directions, and then divide and match the overall demand information according to the multi-dimensional demand directions to obtain the sub-demand information corresponding to each dimension of demand direction. S2. Calculate the similarity between the sub-item demand information of each dimension and the historical casting products, and select the corresponding number of casting products for each dimension demand direction based on the similarity. S3. Establish a design model. In the design model, summarize the requirements information of each dimension as the full score benchmark for product scoring. Based on the degree of matching between the selected casting products and the requirements information, score the product values ​​of each casting product. Then, compare and analyze the differences in product values, production processes, and production organization of different casting products to clarify the constraint relationship between production processes and production organization, the impact of process and organization changes on casting products, and establish a correlation constraint library. S4. Based on the selected casting products and demand information, and combined with the associated constraint library, perform process and organizational scope analysis. Select production processes and production organizations based on the process and organizational scope. Adjust the original production processes and production organizations corresponding to the casting products through the selected production processes and production organizations, update the product values ​​of the casting products, and obtain the highest product score for the casting products.

[0006] As a further improvement to this technical solution, in S1, the entire process data generated during the production of ductile iron is collected at the production end of the ductile iron production line, and the collected entire process data is saved as historical production data. The full-process data includes production technology, production organization, and casting products; In historical production data, the production process and organization of the same batch are linked to the various indicator data of the casting product, so that each group of casting products can be associated with its unique production process and organization data.

[0007] As a further improvement to this technical solution, in step S1, the user inputs relevant requirements for ductile iron at the production end of ductile iron, and the user-input requirements are summarized as demand information for casting products. The user's demand information for casting products is divided into multiple dimensions of demand direction, so that the demand information is divided into multiple independent and quantifiable demand dimensions. The demand dimensions include mechanical performance, process cost, service adaptability, and quality stability. Each overall requirement is assigned to a corresponding requirement dimension, so that each requirement dimension can obtain exclusive sub-requirement information, thus completing the matching of requirement information with requirement direction.

[0008] As a further improvement to this technical solution, in S2, the sub-item demand information corresponding to each dimension is used as the evaluation benchmark, and various performance indicators corresponding to the casting products are extracted. The degree of matching between each casting product and the demand information of each dimension is calculated by quantitative comparison to obtain a quantitative value representing the similarity, and the quantitative value is used as the similarity. The higher the similarity score, the better the match between the casting product and the demand. Conversely, the lower the similarity score, the lower the match between the casting product and the demand.

[0009] As a further improvement to this technical solution, in step S2, casting products are selected for each dimension of demand in descending order of similarity scores, with a minimum of five products selected.

[0010] As a further improvement to this technical solution, in step S3, a computer simulation model is constructed to achieve the correlation optimization of process, organization and product, serving as a design model; In the design model, the sub-requirement information corresponding to each dimension of requirements is summarized, the target value of each dimension of requirements is determined, and the target values ​​of all dimensions are integrated into a unified product numerical scoring full score benchmark by weighted summation. Using the full score benchmark as a reference, a quantitative score is given based on the degree to which the actual performance indicators of each casting product are close to the target values ​​of each dimension. The weight of each dimension requirement is combined in the scoring process to obtain a unique product value for each casting product. The closer a product's score is to a perfect score, the better the casting meets the user's needs.

[0011] As a further improvement to this technical solution, in S3, all selected casting products are compared, the differences in product values ​​between different casting products are analyzed, and the differences in production process parameters and production organization characteristics that cause these differences are analyzed to obtain the correspondence between different production processes, production organization and product values. Using casting products as a link, we analyze the mutual constraints and influences between the production process and production organization corresponding to the same casting product, and obtain the changes in production organization caused by changes in production process parameters, as well as the differences in casting product values ​​caused by changes in production organization, and extract the correlation between process and organization. The results obtained from difference analysis and correlation analysis are structured and organized, including the reasonable value range of production processes, the qualified characteristic range of production organizations, the product value range corresponding to different combinations of processes and organizations, the correlation constraint rules between processes and organizations, and the influence law between organizations and products, thereby establishing a correlation constraint library.

[0012] As a further improvement to this technical solution, in S4, based on the original production process and production organization corresponding to the casting product selected in S2, and combined with user demand information and constraint rules in the associated constraint library, the feasible range of production process and feasible range of production organization that can meet user demand and comply with the requirements of the associated constraint library are analyzed and determined, and process and organization parameters that exceed the limits and cannot be realized are excluded. Within the defined range of process feasibility, multiple different candidate production process parameters are selected. Within the defined range of organizational feasibility, multiple different candidate production organization characteristics are selected. At the same time, during the selection process, it is ensured that the candidate processes and candidate organizations satisfy the association constraint relationships in the association constraint library.

[0013] As a further improvement to this technical solution, in S4, the original production process of the casting product is replaced with the selected candidate production process, and the original production organization is replaced with the candidate production organization. Then, according to the product value scoring standard in S3, the product value of the casting product is recalculated to complete the update of the product value and realize the synchronous adjustment of process, organization and product value. Repeatedly select and adjust the production process and organization, and update the product values, continuously iterate and optimize until the product value of the casting product no longer increases. At this point, the product value is the highest product score that the casting product can achieve, and the corresponding process and organization are the optimal process and organization for the product. During the iterative adjustment process, when the updated product value reaches the full score benchmark, the subsequent adjustment process is immediately terminated. At the same time, the various indicators of the casting product in this state, the corresponding production process parameters, and production organization characteristics are saved as the final design result.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This method for constructing a bidirectional design model coupling ductile iron microstructure and process realizes the reverse derivation from user needs to process microstructure design, breaking the limitations of unidirectional design and significantly improving the accuracy and efficiency of casting design. By uniquely binding production processes, production organization, and casting products to batches, a complete historical production data system is constructed. Combined with the division and matching of multi-dimensional demand directions, the vague user needs are transformed into quantifiable sub-indicators, ensuring that process and microstructure design always revolve around the actual needs of users. This effectively reduces meaningless process trial and error, lowers production raw material and time costs, and enables rapid response to personalized and customized casting needs in different scenarios.

[0015] 2. In this method for constructing a bidirectional design model coupling ductile iron microstructure and process, similarity calculation is used to select casting product samples that match the requirements of each dimension, providing a representative reference for subsequent analysis. Then, by constructing a computer simulation design model, a unified product numerical scoring system is established, realizing the quantitative correlation between process, microstructure and product performance. At the same time, through difference analysis and correlation analysis, the coupling constraint relationship between process and microstructure and the influence law of both on product performance are clarified. The constructed process, microstructure and product correlation constraint library transforms scattered production experience into a standardized rule system, giving the selection and adjustment of process and microstructure a clear theoretical basis, effectively avoiding microstructure and performance deviations caused by fluctuations in process parameters, improving the quality stability of ductile iron casting products, and the correlation constraint library can be continuously accumulated and updated, the optimization results are reproducible, and reusable design experience can be formed, improving the design efficiency of castings of different batches and specifications. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the method for constructing a two-way design model coupling the microstructure and process of ductile iron according to the present invention. Figure 2 This is a flowchart of S1 of the present invention; Figure 3 This is a flowchart of S2 of the present invention; Figure 4 This is a flowchart of S3 of the present invention; Figure 5 This is a flowchart of S4 of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please see Figures 1-5 As shown, the purpose of this embodiment is to provide a method for constructing a two-way design model coupling the microstructure and process of ductile iron, including the following steps: S1. Collect historical production data of ductile iron, bind the production process and production organization in the historical production data with the casting products, and at the same time obtain the user's demand information for the casting products. Divide the demand information into multi-dimensional demand directions, and then divide and match the overall demand information according to the multi-dimensional demand directions to obtain the sub-demand information corresponding to each dimension of demand direction. In S1, the entire process data generated during the production of ductile iron is collected at the production end of the ductile iron production line, and the collected full process data is saved as historical production data. The full-process data includes production technology, production organization, and casting products; Through the ductile iron production end (such as production monitoring system, testing equipment), various types of data generated throughout the entire production process are collected. The entire process data only includes three types of data: production process data, production organization data, and casting product data. In historical production data, the production process and organization of the same batch are linked to the various indicator data of the casting product, so that each group of casting products can be associated with its unique production process and organization data.

[0019] In S1, the user inputs relevant requirements for ductile iron at the production end of ductile iron, and the user inputs relevant requirements are summarized as demand information for casting products. The user's demand information for casting products is divided into multiple dimensions of demand direction, so that the demand information is divided into multiple independent and quantifiable demand dimensions. The demand dimensions include mechanical performance, process cost, service adaptability, and quality stability. Each overall requirement is assigned to a corresponding requirement dimension, so that each requirement dimension can obtain exclusive sub-requirement information, thus completing the matching of requirement information with requirement direction.

[0020] S2. Calculate the similarity between the sub-item demand information of each dimension and the historical casting products, and select the corresponding number of casting products for each dimension demand direction based on the similarity. In S2, the sub-item requirement information corresponding to each dimension is used as the evaluation benchmark. Various performance indicators corresponding to the casting products are extracted. The degree of matching between each casting product and the requirement information of each dimension is calculated by quantitative comparison. The quantitative value representing the similarity is obtained and the quantitative value is used as the similarity. The higher the similarity score, the better the match between the casting product and the demand. Conversely, the lower the similarity score, the lower the match between the casting product and the requirement. The steps are as follows: Obtain the sub-requirement information corresponding to each dimension (mechanical performance, process cost, service adaptability, quality stability) from S1, clarify the target value of each performance indicator under each dimension, assign weights to each performance indicator under each dimension according to the importance of the requirements, and perform normalization processing. From the historical production database, the actual performance index values ​​of all casting products to be evaluated under the corresponding dimensions are extracted. Simultaneously, the extracted index values ​​are preprocessed, including missing value completion, outlier removal, and data standardization, to ensure data quality. Then, the weighted Euclidean distance normalization method is used to calculate the matching degree between each casting product and the requirement information of each dimension, obtaining the similarity value. The formula is as follows: ; in, Let the similarity of the casting products be defined under the j-th demand dimension. This represents the number of performance metrics under this dimension. The weight of the k-th performance metric Let k be the actual value of the k-th index in the j-th dimension of the casting product. Let the user's target value for the k-th indicator in the j-th dimension be _____. For all casting products under the same dimension, sort them from high to low according to the similarity value, and then select casting products for each dimension of demand from the sorted product list. In S2, casting products are selected for each dimension of demand in descending order of similarity scores, with a minimum of five products selected.

[0021] S3. Establish a design model. In the design model, summarize the requirements information of each dimension as the full score benchmark for product scoring. Based on the degree of matching between the selected casting products and the requirements information, score the product values ​​of each casting product. Then, compare and analyze the differences in product values, production processes, and production organization of different casting products to clarify the constraint relationship between production processes and production organization, the impact of process and organization changes on casting products, and establish a correlation constraint library. In S3, a computer simulation model is constructed to achieve the correlation optimization of process, organization and product, serving as the design model; In the design model, the sub-requirements corresponding to each dimension of requirements are summarized to determine the target values ​​for each dimension. Then, a weighted summation is used to integrate the target values ​​of all dimensions into a unified product numerical scoring benchmark. In the design model, the sub-requirement information corresponding to each dimension of requirements is summarized, the target value of each dimension of requirements is determined, and weights are assigned to each dimension according to the importance of each dimension of requirements. Then, the target values ​​of all dimensions are integrated into a unified product numerical scoring full score benchmark by weighted summation. Using the full score benchmark as a reference, a quantitative score is given based on the degree to which the actual performance indicators of each casting product are close to the target values ​​of each dimension. The weight of each dimension requirement is combined in the scoring process to obtain a unique product value for each casting product. The closer a product's score is to a perfect score, the better the casting meets the user's needs.

[0022] Using the full score benchmark as a reference, the actual performance index values ​​of each casting product under various dimensions are extracted from historical production data. Then, the extracted index values ​​are preprocessed, including missing value completion, outlier removal, and data standardization, to ensure data quality. At the same time, the scoring process combines the weights of the requirements of each dimension and uses a weighted scoring method to calculate the product value, obtaining a unique product value for each casting product. The formula is as follows: ; in, These are the product values ​​corresponding to the casting products. Let j be the weight of the j-th demand dimension. Let j be the actual performance index of the casting product in the j-th dimension. Let the target value be the value for the j-th dimension. In S3, all selected casting products are compared, and the differences in product values ​​between different casting products are analyzed. At the same time, the differences in production process parameters and production organization characteristics that cause these differences are analyzed to obtain the correspondence between different production processes, production organization and product values. By comparing all selected casting products, the differences in product values ​​between different casting products are calculated. At the same time, the differences in production process parameters and production organization characteristics that lead to these differences are analyzed. Through quantitative comparison, the source of the differences is located, and the correspondence between different production processes, production organization and product values ​​is obtained. Using casting products as a link, we analyze the mutual constraints and influences between the production process and production organization corresponding to the same casting product, and obtain the changes in production organization caused by changes in production process parameters, as well as the differences in casting product values ​​caused by changes in production organization, and extract the correlation between process and organization. Using casting products as a link, this study analyzes the mutual constraints and influences between the production process and production organization corresponding to the same casting product. Using the controlled variable method, other parameters are fixed while a single process parameter is changed, and the changes in organizational characteristics are observed to clarify the specific changes in production organization caused by changes in production process parameters. At the same time, other parameters are fixed while a single organizational characteristic is changed, and the changes in product values ​​are observed to clarify the differences in casting product values ​​caused by changes in production organization. The results are analyzed comprehensively to extract the correlation between process and organization. The results obtained from difference analysis and correlation analysis are structured and organized, including the reasonable value range of production processes, the qualified characteristic range of production organizations, the product value range corresponding to different combinations of processes and organizations, the correlation constraint rules between processes and organizations, and the influence law between organizations and products, thereby establishing a correlation constraint library.

[0023] The results obtained from difference analysis and correlation analysis are structured and organized, including the reasonable value range of production processes, the qualified characteristic range of production organization, the product value range corresponding to different combinations of processes and organizations, the correlation constraint rules between processes and organizations, and the influence law between organizations and products. Then, the organized content is stored and managed in the form of a database or knowledge base to form a standardized process, organization, and product correlation constraint library, which is used to guide the subsequent selection and adjustment of processes and organizations.

[0024] S4. Based on the selected casting products and demand information, and combined with the associated constraint library, perform process and organizational scope analysis. Select production processes and production organizations based on the process and organizational scope. Adjust the original production processes and production organizations corresponding to the casting products through the selected production processes and production organizations, update the product values ​​of the casting products, and obtain the highest product score for the casting products.

[0025] In S4, based on the original production process and production organization corresponding to the casting products selected in S2, and combined with user demand information and constraint rules in the associated constraint library, the feasible range of production process and production organization that can meet user demand and comply with the requirements of the associated constraint library are analyzed and determined, and process and organization parameters that exceed the limits and cannot be realized are excluded. For the casting products selected in S2, extract the original production process parameters (such as melting temperature, amount of spheroidizing agent, amount of inoculant, cooling rate, etc.) and original production organization characteristics (such as graphite spheroidization rate, graphite size distribution, matrix phase ratio, grain size, etc.) corresponding to the product, as the benchmark for subsequent adjustments; Import all existing process and organizational data, as well as user requirements and S3-based related constraint libraries, into the design model to ensure that all data is accessible and analyzable. By combining user requirement information (specific requirements for each dimension and overall requirement objectives) and constraint rules in the associated constraint library (process value range, organizational characteristic interval, process and organizational relationship constraints, etc.), the compliance and adaptability analysis of the original process and organizational parameters is performed. Select the feasible range of production processes and production organization that can meet user needs and comply with the requirements of the associated restriction library, and clarify the specific value range of each process parameter (such as melting temperature 300-500℃) and each microstructure characteristic (such as graphite spheroidization rate ≥85%). Exclude process and organizational parameters that are outside the scope of the associated restriction library, cannot be achieved in production (such as temperatures that the equipment cannot reach, or processes that are too costly), or cannot meet user needs, to ensure that all parameters within the feasible range are production-feasible and demand-adaptable. Within the defined range of process feasibility, multiple different candidate production process parameters are selected. Within the defined range of organizational feasibility, multiple different candidate production organization characteristics are selected. At the same time, during the selection process, it is ensured that the candidate processes and candidate organizations satisfy the association constraint relationships in the association constraint library.

[0026] In S4, the original production process of the casting product is replaced with the selected candidate production process, and the original production organization is replaced with the candidate production organization. Then, according to the product value scoring standard in S3, the product value of the casting product is recalculated to complete the product value update and realize the synchronous adjustment of process, organization and product value. Within the defined process feasibility range, select multiple different candidate production process parameters evenly (at least 3 groups, to ensure coverage of different parameter ranges and avoid deviation of a single parameter). Within the defined organizational feasibility range, select multiple different candidate production organizational features evenly (at least 3 groups). During the selection process, the matching of each candidate process and candidate organization is verified one by one by referring to the process and organization association constraint rules in the association restriction library to ensure that the candidate process and candidate organization can be adapted to each other (such as a certain cooling rate corresponding to a specific graphite spheroidization rate range), and candidate combinations that do not match the process and organization are excluded. Ultimately, multiple sets of candidate processes and candidate organization combinations that meet constraints, are manufacturable, and adapt to requirements are formed, providing sufficient adjustment schemes for subsequent iterative optimization; Then, the selected candidate production process parameters replace the original production process parameters of the casting product, and the candidate production organization characteristics replace the original production organization characteristics, completing the single process and organization adjustment. According to the product value scoring standard (weighted scoring method) determined in S3, with the full score benchmark as a reference and combined with the weight of each dimension, the product value of the casting product under the new process and organization parameters is recalculated, completing the product value update and realizing the synchronous adjustment of process, organization and product value. The formula is as follows: ; in, The updated values ​​for the casting product after replacing the process or structure. Let j be the weight of the j-th demand dimension. This refers to the actual performance index value of the casting product in the j-th dimension after replacing the process or structure. Let the target value be the value for the j-th dimension. ; in, The degree of matching between candidate processes and organizations is represented by n, where n is the number of process and organization-related constraints. As an indicator function, when the k-th constraint contains candidate process parameters With candidate tissue characteristics When a match is found, γ=1; otherwise, γ=0. Record the candidate processes, candidate microstructures, and corresponding updated product values ​​after this adjustment to form an iteration record for easy comparison and analysis later. Repeatedly select and adjust the production process and organization, and update the product values, continuously iterate and optimize until the product value of the casting product no longer increases. At this point, the product value is the highest product score that the casting product can achieve, and the corresponding process and organization are the optimal process and organization for the product. During the iterative adjustment process, when the updated product value reaches the full score benchmark, the subsequent adjustment process is immediately terminated. At the same time, the various indicators of the casting product in this state, the corresponding production process parameters, and production organization characteristics are saved as the final design result.

[0027] After a normal termination and continuous iteration, if the product value after a certain update does not improve compared to the product value of the previous iteration, it is determined that the highest product score of the casting product has been reached, and the corresponding process and organization are the optimal process and optimal production organization of the product. Prioritize termination: During the iterative adjustment process, if the value of a casting product after a certain update reaches the set full score benchmark, the subsequent adjustment process will be terminated immediately without further iteration, ensuring efficiency.

[0028] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for constructing a two-way design model coupling microstructure and process of ductile iron, characterized in that: Includes the following steps: S1. Collect historical production data of ductile iron, bind the production process and production organization in the historical production data with the casting products, and at the same time obtain the user's demand information for the casting products. Divide the demand information into multi-dimensional demand directions, and then divide and match the overall demand information according to the multi-dimensional demand directions to obtain the sub-demand information corresponding to each dimension of demand direction. S2. Calculate the similarity between the sub-item demand information of each dimension and the historical casting products, and select the corresponding number of casting products for each dimension demand direction based on the similarity. S3. Establish a design model. In the design model, summarize the requirements information of each dimension as the full score benchmark for product scoring. Based on the degree of matching between the selected casting products and the requirements information, score the product values ​​of each casting product. Then, compare and analyze the differences in product values, production processes, and production organization of different casting products to clarify the constraint relationship between production processes and production organization, the impact of process and organization changes on casting products, and establish a correlation constraint library. S4. Based on the selected casting products and demand information, and combined with the associated constraint library, perform process and organizational scope analysis. Select production processes and production organizations based on the process and organizational scope. Adjust the original production processes and production organizations corresponding to the casting products through the selected production processes and production organizations, update the product values ​​of the casting products, and obtain the highest product score for the casting products.

2. The method for constructing a two-way design model coupling microstructure and process of ductile iron according to claim 1, characterized in that: In S1, the entire process data generated during the production of ductile iron is collected at the production end of the ductile iron production line, and the collected entire process data is saved as historical production data. The full-process data includes production technology, production organization, and casting products; In historical production data, the production process and organization of the same batch are linked to the various indicator data of the casting product, so that each group of casting products can be associated with its unique production process and organization data.

3. The method for constructing a two-way design model coupling microstructure and process of ductile iron according to claim 1, characterized in that: In S1, the user inputs relevant requirements for ductile iron at the production end of ductile iron, and the user-inputted requirements are summarized as demand information for casting products. The user's demand information for casting products is divided into multiple dimensions of demand direction, so that the demand information is divided into multiple independent and quantifiable demand dimensions. The demand dimensions include mechanical performance, process cost, service adaptability, and quality stability. Each overall requirement is assigned to a corresponding requirement dimension, so that each requirement dimension can obtain exclusive sub-requirement information, thus completing the matching of requirement information with requirement direction.

4. The method for constructing a two-way design model coupling microstructure and process of ductile iron according to claim 1, characterized in that: In S2, the sub-item demand information corresponding to each dimension is used as the evaluation benchmark. Various performance indicators corresponding to the casting products are extracted. The degree of matching between each casting product and the demand information of each dimension is calculated by quantitative comparison. A quantitative value representing the similarity is obtained, and the quantitative value is used as the similarity. The higher the similarity score, the better the match between the casting product and the demand. Conversely, the lower the similarity score, the lower the match between the casting product and the demand.

5. The method for constructing a two-way design model coupling microstructure and process of ductile iron according to claim 1, characterized in that: In step S2, casting products are selected for each dimension of demand in descending order of similarity scores, with a minimum of five products selected.

6. The method for constructing a two-way design model coupling microstructure and process of ductile iron according to claim 1, characterized in that: In step S3, a computer simulation model is constructed to achieve the optimization of the correlation between process, organization and product, serving as the design model; In the design model, the sub-requirement information corresponding to each dimension of requirements is summarized, the target value of each dimension of requirements is determined, and the target values ​​of all dimensions are integrated into a unified product numerical scoring full score benchmark by weighted summation. Using the full score benchmark as a reference, a quantitative score is given based on the degree to which the actual performance indicators of each casting product are close to the target values ​​of each dimension. The weight of each dimension requirement is combined in the scoring process to obtain a unique product value for each casting product. The closer a product's score is to a perfect score, the better the casting meets the user's needs.

7. The method for constructing a two-way design model coupling microstructure and process of ductile iron according to claim 1, characterized in that: In S3, all selected casting products are compared, and the differences in product values ​​between different casting products are analyzed. At the same time, the differences in production process parameters and production organization characteristics that cause these differences are analyzed to obtain the correspondence between different production processes, production organization and product values. Using casting products as a link, we analyze the mutual constraints and influences between the production process and production organization corresponding to the same casting product, and obtain the changes in production organization caused by changes in production process parameters, as well as the differences in casting product values ​​caused by changes in production organization, and extract the correlation between process and organization. The results obtained from difference analysis and correlation analysis are structured and organized, including the reasonable value range of production processes, the qualified characteristic range of production organizations, the product value range corresponding to different combinations of processes and organizations, the correlation constraint rules between processes and organizations, and the influence law between organizations and products, thereby establishing a correlation constraint library.

8. The method for constructing a two-way design model coupling microstructure and process of ductile iron according to claim 1, characterized in that: In S4, based on the original production process and production organization corresponding to the casting product selected in S2, and combined with user demand information and constraint rules in the associated constraint library, the feasible range of production process and production organization that can meet user demand and comply with the requirements of the associated constraint library are analyzed and determined, and process and organization parameters that exceed the limits and cannot be realized are excluded. Within the defined range of process feasibility, multiple different candidate production process parameters are selected. Within the defined range of organizational feasibility, multiple different candidate production organization characteristics are selected. At the same time, during the selection process, it is ensured that the candidate processes and candidate organizations satisfy the association constraint relationships in the association constraint library.

9. The method for constructing a two-way design model coupling microstructure and process of ductile iron according to claim 1, characterized in that: In S4, the original production process of the casting product is replaced with the selected candidate production process, and the original production organization is replaced with the candidate production organization. Then, according to the product value scoring standard in S3, the product value of the casting product is recalculated to complete the product value update and realize the synchronous adjustment of process, organization and product value. Repeatedly select and adjust the production process and organization, and update the product values, continuously iterate and optimize until the product value of the casting product no longer increases. At this point, the product value is the highest product score that the casting product can achieve, and the corresponding process and organization are the optimal process and organization for the product. During the iterative adjustment process, when the updated product value reaches the full score benchmark, the subsequent adjustment process is immediately terminated. At the same time, the various indicators of the casting product in this state, the corresponding production process parameters, and production organization characteristics are saved as the final design result.

Images