Denim dyeing process decision-making method based on knowledge graph

Abstract

The invention relates to the field of knowledge maps, in particular to a denim dyeing process decision-making method based on a knowledge map. Comprising the following steps: constructing a knowledge graph, calculating an importance score of each process parameter, setting a threshold value, and comparing the importance scores of the process parameters with the threshold value to form a key process parameter set; on the basis of the key process parameter set, an index of a dyeing effect is predicted through a dynamic semantic path enhanced dyeing effect prediction algorithm; and based on the predicted dyeing effect index, optimizing the key process parameters by using a process parameter optimization and dynamic adjustment algorithm, and generating a final process parameter vector. The problems that in a traditional dyeing process, manual key parameter screening consumes time, important factors are prone to being omitted or irrelevant parameters are prone to being included, and consequently the process effect is unstable are solved; the non-linear influence of environmental parameters on the dyeing effect is difficult to accurately capture through a traditional linear model.

Description

Technical Field

[0001] The present invention relates to the field of knowledge graphs, and in particular to a denim dyeing process decision-making method based on knowledge graphs. Background Art

[0002] Decision-making in the denim dyeing process involves multiple variables, including dye type, concentration, dyeing time, temperature, pH value, and auxiliary agent ratio. These parameters interact in complex ways. For example, excessive dye concentration can lead to color shift, while insufficient dyeing time can affect dye uptake. Furthermore, varying customer requirements (such as color depth and washing performance) and environmental regulations (such as wastewater discharge standards) further complicate process decision-making. Traditional denim dyeing process decisions often rely on empirical judgment and manual operation, resulting in low efficiency, resource waste, and environmental pollution. With the development of artificial intelligence and big data technologies, decision-making methods based on knowledge graphs offer new possibilities for optimizing denim dyeing processes. By structuredly organizing and representing domain knowledge, knowledge graphs enable intelligent recommendation, optimization, and decision support for process parameters.

[0003] Knowledge graph technology not only addresses the low efficiency and poor consistency of traditional methods, but also promotes the sustainable development of the denim industry through dynamic optimization and knowledge sharing. As the technology continues to develop, knowledge graphs will play an increasingly important role in the denim dyeing process, providing the industry with more efficient, environmentally friendly, and intelligent production methods.

[0004] However, the existing denim dyeing process decision-making methods mentioned above still have the following problems: the selection of process parameters often relies on experience or trial and error, lacks systematic data-driven decision-making; ignores the impact of environmental parameter fluctuations; and has difficulty in achieving dynamic balance between multiple objectives. Summary of the Invention

[0005] The present invention provides a denim dyeing process decision-making method based on a knowledge graph to solve the technical problems in traditional dyeing processes, such as the time-consuming manual screening of key parameters and the easy omission of important factors or the inclusion of irrelevant parameters, resulting in unstable process effects; the nonlinear influence of environmental parameters on dyeing effects is difficult to accurately capture through traditional linear models, especially in multi-hop association scenarios; it is difficult to adapt to real-time environmental fluctuations, and the production cost constraints are not fully considered, resulting in insufficient applicability of optimization results in actual production.

[0006] The present invention provides a denim dyeing process decision-making method based on a knowledge graph, which specifically includes the following technical solutions: A denim dyeing process decision-making method based on knowledge graph includes the following steps: S1. Build a knowledge graph, calculate the importance score of each process parameter, set a threshold, compare the importance score of the process parameter with the threshold, and form a set of key process parameters; S2. Based on the set of key process parameters, the dyeing effect prediction algorithm is enhanced through dynamic semantic path to predict the dyeing effect indicators; based on the predicted dyeing effect indicators, the key process parameters are optimized using the process parameter optimization and dynamic adjustment algorithm to generate the final process parameter vector.

[0007] Preferably, the S1 specifically includes: After receiving the dyeing effect input by the user, the importance score of each process parameter is calculated based on the association strength in the knowledge graph and the effect weight specified by the user; when calculating the importance score of each process parameter, each process parameter in the initial process parameter set is traversed to obtain the association strength between each process parameter and each dyeing effect, and the association strength is multiplied by the corresponding effect weight to obtain the weighted contribution. The weighted contributions of all effects are summed to obtain the importance score of the process parameter.

[0008] Preferably, the S2 specifically includes: The dynamic semantic path enhanced dyeing effect prediction algorithm predicts the indicators of the dyeing effect in the denim dyeing process through multi-hop semantic path analysis of the knowledge graph, combined with process parameters and real-time environmental data.

[0009] Preferably, the S2 specifically includes: In the implementation process of the dynamic semantic path enhanced coloring effect prediction algorithm, a path weight is assigned to each path. For each path, the nonlinear influence of the environmental parameters on the coloring effect is calculated. The difference between the environmental parameters and the environmental reference value is divided by the environmental reference value and the absolute value is taken. The result is multiplied by the contribution coefficient of the environmental parameters in the path as the input of the hyperbolic tangent function.

[0010] Preferably, the S2 specifically includes: In the implementation of the dynamic semantic path enhanced coloring effect prediction algorithm, the output of the hyperbolic tangent function is multiplied by the path weight and summed over all paths to obtain the total contribution of the environmental parameters to the coloring effect.

[0011] Preferably, the S2 specifically includes: In the implementation process of the dynamic semantic path enhanced dyeing effect prediction algorithm, the contribution of process parameters is calculated through the regression model. Each process parameter value is multiplied by the corresponding regression coefficient and summed up. The regression model intercept is added to the sum to generate a linear prediction term. The total contribution of environmental parameters to the dyeing effect is combined with the linear prediction term to obtain the dyeing effect index.

[0012] Preferably, the S2 specifically includes: In the process of implementing the process parameter optimization and dynamic adjustment algorithm, the difference between the dyeing effect index and the target value of the dyeing effect index is calculated, the difference between the two is squared and divided by the square of the reference deviation of the dyeing effect index to generate an effect deviation term, each effect deviation term is multiplied by the priority of the dyeing effect index, and the indicators of all dyeing effects are summed to obtain the weighted effect deviation.

[0013] Preferably, the S2 specifically includes: In the implementation of the process parameter optimization and dynamic adjustment algorithm, the process cost under the process parameter vector is divided by the reference value of the process cost to generate a cost item, and then the cost item is multiplied by the cost weight to obtain a weighted cost item; the weighted effect deviation and the weighted cost item are added to form the objective function value, and the gradient descent method is used to minimize the objective function to obtain the final process parameter vector.

[0014] The beneficial effects of the technical solution of the present invention are: 1. By constructing a structured knowledge graph, integrating historical process databases, experimental records and literature information, and systematically storing dyeing-related entities (such as dye types and fabric characteristics) and relationships (such as the relationship between dye concentration and color depth), the correlation between process parameters and dyeing effects is mined, significantly improving query efficiency and accuracy.

[0015] 2. By combining the association strength in the knowledge graph and the effect weight set by the user, the importance score of each process parameter is innovatively calculated, and the key process parameter set is screened based on the user-defined threshold. This achieves accurate screening of process parameters, avoids interference from irrelevant parameters, and focuses on parameters that significantly affect the dyeing effect, thereby improving the pertinence and efficiency of process design and adapting to the needs of different process scenarios.

[0016] 3. By utilizing multi-hop semantic path analysis of the knowledge graph and combining process parameters with real-time environmental parameters, we innovatively predict dyeing effect indicators (such as color depth and uniformity). By analyzing the nonlinear impact of environmental parameter deviations on dyeing effects and combining path weights and regression models, we achieve high-precision effect prediction, which can dynamically adapt to environmental fluctuations, ensure that the predicted dyeing effect indicators are close to actual production conditions, and improve the controllability of the dyeing effect.

[0017] 4. Through optimization algorithms (such as gradient descent or genetic algorithms), dyeing effect deviation and production cost are comprehensively considered, key process parameters are dynamically adjusted, and the final process parameter vector is generated, achieving a balance between effect goals and cost constraints, significantly improving resource utilization and production efficiency, while ensuring that the dyeing effect is close to user needs and enhancing the adaptability and economy of the process. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of the denim dyeing process decision-making method based on knowledge graph described in the present invention. DETAILED DESCRIPTION

[0019] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0020] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0021] The following describes in detail a specific scheme of a denim dyeing process decision method based on a knowledge graph provided by the present invention with reference to the accompanying drawings.

[0022] Refer to the attached Figure 1 , which shows a flow chart of a denim dyeing process decision method based on a knowledge graph provided by one embodiment of the present invention, the method comprising the following steps: S1. Build a knowledge graph, calculate the importance score of each process parameter, set a threshold, compare the importance score of the process parameter with the threshold, and form a set of key process parameters; Data related to the denim dyeing process is extracted from historical process databases, experimental records, and literature, including process parameter values, environmental parameters, and corresponding dyeing effects (such as color depth and uniformity). A structured knowledge graph is constructed to store entities and relationships related to the dyeing process. These entities include dye types (such as indigo dyes and sulfur dyes), fabric properties (such as cotton fiber density and fabric thickness), equipment parameters (such as dye vat pressure and stirring speed), and environmental parameters (such as humidity and temperature). Relationships are represented as triples, such as "dye concentration - effect - color depth", and stored in a graph database (such as Neo4j) for convenient and efficient querying. After receiving the dyeing effect input by the user, in order to evaluate the contribution of each process parameter to the dyeing effect, the importance score of each process parameter is calculated based on the association strength in the knowledge graph and the effect weight specified by the user. When calculating the importance score of each process parameter, each process parameter in the initial process parameter set is traversed to obtain the association strength between each process parameter and each dyeing effect. The association strength is multiplied by the corresponding effect weight to obtain the weighted contribution. The weighted contributions of all effects are summed to obtain the importance score of the process parameter. The calculation formula for the importance score of process parameters is:

[0023] in, Indicates the Importance score of each process parameter; Indicates that comprehensive consideration Process parameter pairs The contribution of each dyeing effect (such as color depth, uniformity); Indicates the The weight of each dyeing effect reflects the user's priority for different dyeing targets. For example, if the depth of color is more important, a higher weight will be given. It is specified by the user according to the process requirements. ; Indicates the first The process parameters and The correlation strength of each dyeing effect is extracted from the knowledge graph, such as the edge weight of the triple (dye concentration-effect-color depth), which is assigned through historical data statistics or expert knowledge. ; Users directly set thresholds based on process requirements and retain process parameters whose importance scores are greater than or equal to the thresholds to form a set of key process parameters. Specific process parameter values ​​can be extracted from the historical process database. If there is no historical process data, the default recommended values ​​in the knowledge graph, such as industry standard values, are used. The threshold screening method ensures a strong correlation between process parameters and dyeing targets, and dynamically adjusts the threshold to adapt to different process scenarios, avoiding the omission of key process parameters or the inclusion of irrelevant process parameters; S2. Based on the set of key process parameters, the dyeing effect prediction algorithm is enhanced by a dynamic semantic path to predict the dyeing effect index; based on the predicted dyeing effect index, the key process parameters are optimized using a process parameter optimization and dynamic adjustment algorithm to generate a final process parameter vector; Based on the key process parameter set, the dyeing effect prediction algorithm is enhanced by dynamic semantic path to predict the dyeing effect index; The dynamic semantic path enhanced dyeing effect prediction algorithm aims to predict the indicators of dyeing effect in the denim dyeing process, such as color depth and uniformity, by combining multi-hop semantic path analysis of the knowledge graph with process parameters and real-time environmental data. The innovation lies in utilizing the topological structure of the knowledge graph to analyze the multi-hop semantic paths from environmental parameters to dyeing effects in the knowledge graph. The path set is generated by the existing knowledge graph query algorithm (such as depth-first search) to ensure that all valid paths from environmental parameters to dyeing effects are covered. Specifically, a path weight is assigned to each path. , calculate the nonlinear influence of environmental parameters on dyeing effect, divide the difference between environmental parameters and environmental reference values ​​by the environmental reference value and take the absolute value, further multiply it by the contribution coefficient of environmental parameters in the path as the input of hyperbolic tangent function, when the environmental parameter is equal to the environmental reference value, the difference between the environmental parameter and the environmental reference value is 0, the hyperbolic tangent function output is 0, and the environmental contribution is 0. When the deviation between the environmental parameter and the environmental reference value increases, the output of the hyperbolic tangent function increases, multiply the output of the hyperbolic tangent function by the path weight, and sum all paths to obtain the total contribution of the environmental parameters to the dyeing effect; Furthermore, the contribution of process parameters is calculated through the regression model. Each process parameter value is multiplied by the corresponding regression coefficient and the sum is added. The regression model intercept is added to the sum to generate a linear prediction term. The linear contribution of process parameters and the influence of environmental parameters are combined to predict the index of dyeing effect. Combining linear regression and nonlinear path analysis of knowledge graph, the calculation formula of the dyeing effect index is:

[0024] in, The predicted An indicator of dyeing effect; Represents the intercept of the regression model, provides a benchmark for prediction, and is obtained by training the linear regression model with historical data; Express The contributions of each process parameter are summed up; Indicates the The regression coefficient of the process parameter reflects the The process parameters have an impact on the The linear influence strength of each dyeing effect is obtained by training the linear regression model with historical data, and the value range is ; Indicates the process parameters; Indicates the knowledge graph from environmental parameters to Sum up all multi-hop semantic paths of the coloring effect; Indicates the path The weight of the path For the first The potential influence of each staining effect, ; represents the hyperbolic tangent function, which is used to simulate the amplification effect of environmental deviation on the dyeing effect; Indicates that all The deviation contributions of each environmental parameter are summed up; Indicates the path Medium environmental parameters The contribution coefficient of ; Indicates the Environmental parameter values; Indicates environmental reference values, i.e. ideal or standard environmental conditions, such as temperature and humidity recommended by the process; It represents the total contribution of environmental parameters to the dyeing effect; represents the linear prediction term, which is obtained based on the existing regression model; The above formula adds the linear prediction term and the total contribution of environmental parameters to the dyeing effect. It describes that the dyeing effect index is the result of the combined effect of process parameters and environmental factors. The linear contribution of process parameters represents the direct impact of controllable operating variables on the dyeing effect, while the environmental impact captures the nonlinear perturbations of uncontrollable or external conditions. The combination of the two can comprehensively describe the physical and chemical processes in the dyeing process, close to actual production scenarios. Based on the predicted dyeing effect indicators, the process parameter optimization and dynamic adjustment algorithm is used to optimize the key process parameters to approach the user-specified target dyeing effect while meeting the cost constraints, and generate the final process parameter vector that adapts to real-time environmental fluctuations; The process parameter optimization and dynamic adjustment algorithm calculates the difference between the dyeing effect index and the target value of the dyeing effect index, squares the difference between the two and divides it by the square of the dyeing effect index reference deviation to generate an effect deviation term, the dyeing effect index reference deviation is the maximum deviation of the target range, multiplies each effect deviation term by the priority of the dyeing effect index, and sums all the dyeing effect indices to obtain a weighted effect deviation, further, divides the process cost under the process parameter vector by the reference value of the process cost to generate a cost term, and then multiplies the cost term by the cost weight to obtain a weighted cost term; the weighted effect deviation and the weighted cost term are added to form an objective function value, and the objective function is minimized using the gradient descent method, which is a technical means well known to those skilled in the art and will not be described in detail here; The calculation formula of the final process parameter vector is:

[0025] in, Represents the final process parameter vector, which is used as the final output to guide the actual dyeing process and is solved by an optimization algorithm (such as gradient descent); Represents the minimization operation, finding the process parameter vector that minimizes the objective function, and searching for the optimal solution through optimization algorithms (such as gradient descent); Indicates that all The sum of the indicators of the dyeing effect is calculated; represents the effect deviation term; Indicates the The priority of each dyeing effect index is used to adjust the contribution of different dyeing effect indicators to the objective function. and ; Indicates the user-specified The target value of the dyeing effect index is used as the optimization target; Indicates the The reference deviation of each dyeing effect index is used for normalization and is determined by the process standard; represents the cost weight, reflecting the importance of cost in optimization, ; represents the production cost under the process parameter vector; A reference value representing the process cost, such as a budget ceiling, specified by the user; It represents the weighted effect deviation, reflecting the degree of deviation of the process parameters on a specific dyeing effect. The significance of multiplying it by the priority of the dyeing effect index is to adjust the contribution of each dyeing effect deviation to the overall objective function according to the user's preference for the importance of different dyeing effects, reflecting the quantification of the relative importance of different dyeing effects. represents the weighted cost item; The above formula incorporates the concept of multi-objective optimization. The weighted effect deviation in the formula represents the deviation between the predicted dyeing effect index and the target value of the dyeing effect index, reflecting the comprehensive performance of the process parameters in meeting the user's dyeing effect target. The weighted cost term reflects the economic efficiency of resource consumption. The two are added together to form the objective function. Taking into account the degree of dyeing effect and the constraints of production costs, finding a balance between dyeing effect and cost is crucial in actual production. It ensures that while meeting dyeing quality requirements, production costs are controlled to improve economic benefits. On the premise of meeting the user's dyeing effect goals and cost constraints, the final process parameter vector is generated, which improves resource utilization and production efficiency.

[0026] In summary, a denim dyeing process decision-making method based on knowledge graph was completed.

[0027] The order in which the embodiments of the invention are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0028] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0029] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A denim dyeing process decision method based on knowledge graph, characterized in that: The following steps are involved: S1. Build a knowledge graph, calculate the importance score of each process parameter, set a threshold, compare the importance score of the process parameter with the threshold, and form a set of key process parameters; S2. Based on the set of key process parameters, the dyeing effect prediction algorithm is enhanced through dynamic semantic path to predict the dyeing effect indicators; based on the predicted dyeing effect indicators, the key process parameters are optimized using the process parameter optimization and dynamic adjustment algorithm to generate the final process parameter vector.

2. The denim dyeing process decision method based on knowledge graph according to claim 1, characterized in that: Said S1 specifically includes: After receiving the dyeing effect input by the user, the importance score of each process parameter is calculated based on the association strength in the knowledge graph and the effect weight specified by the user; when calculating the importance score of each process parameter, each process parameter in the initial process parameter set is traversed to obtain the association strength between each process parameter and each dyeing effect, and the association strength is multiplied by the corresponding effect weight to obtain the weighted contribution. The weighted contributions of all effects are summed to obtain the importance score of the process parameter.

3. The denim dyeing process decision method based on knowledge graph according to claim 1, characterized in that: Said S2 specifically includes: The dynamic semantic path enhanced dyeing effect prediction algorithm predicts the indicators of the dyeing effect in the denim dyeing process through multi-hop semantic path analysis of the knowledge graph, combined with process parameters and real-time environmental data.

4. The denim dyeing process decision method based on knowledge graph according to claim 3, characterized in that: Said S2 specifically includes: In the implementation process of the dynamic semantic path enhanced coloring effect prediction algorithm, a path weight is assigned to each path. For each path, the nonlinear influence of the environmental parameters on the coloring effect is calculated. The difference between the environmental parameters and the environmental reference value is divided by the environmental reference value and the absolute value is taken. The result is multiplied by the contribution coefficient of the environmental parameters in the path as the input of the hyperbolic tangent function.

5. The denim dyeing process decision method based on knowledge graph according to claim 4, characterized in that: Said S2 specifically includes: In the implementation of the dynamic semantic path enhanced coloring effect prediction algorithm, the output of the hyperbolic tangent function is multiplied by the path weight and summed over all paths to obtain the total contribution of the environmental parameters to the coloring effect.

6. The denim dyeing process decision method based on knowledge graph according to claim 5, characterized in that: Said S2 specifically includes: In the implementation process of the dynamic semantic path enhanced dyeing effect prediction algorithm, the contribution of process parameters is calculated through the regression model. Each process parameter value is multiplied by the corresponding regression coefficient and summed up. The regression model intercept is added to the sum to generate a linear prediction term. The total contribution of environmental parameters to the dyeing effect is combined with the linear prediction term to obtain the dyeing effect index.

7. The denim dyeing process decision method based on knowledge graph according to claim 1, characterized in that: Said S2 specifically includes: In the process of implementing the process parameter optimization and dynamic adjustment algorithm, the difference between the dyeing effect index and the target value of the dyeing effect index is calculated, the difference between the two is squared and divided by the square of the reference deviation of the dyeing effect index to generate an effect deviation term, each effect deviation term is multiplied by the priority of the dyeing effect index, and the indicators of all dyeing effects are summed to obtain the weighted effect deviation.

8. The denim dyeing process decision method based on knowledge graph according to claim 7, characterized in that: Said S2 specifically includes: In the implementation of the process parameter optimization and dynamic adjustment algorithm, the process cost under the process parameter vector is divided by the reference value of the process cost to generate a cost item, and then the cost item is multiplied by the cost weight to obtain a weighted cost item; the weighted effect deviation and the weighted cost item are added to form the objective function value, and the gradient descent method is used to minimize the objective function to obtain the final process parameter vector.

Images