A method and system for real-time optimization of smelting efficiency in aluminum processing

Abstract

The present application relates to the technical field of smelting efficiency optimization, and discloses a real-time optimization method and system for the smelting efficiency of an aluminum processing process, comprising: in the smelting process of aluminum processing, constructing a parameter dependency map by analyzing the dependency relationship between parameters; combining the parameter dependency map with multiple smelting stages of the smelting process, respectively calculating the priority values ​​of the parameters corresponding to each smelting stage, sorting the priority values ​​from high to low, and obtaining the parameter priority order of each smelting stage; dynamically pruning a preset smelting control model according to the parameter priority order to obtain an optimized smelting control model; optimizing the parameters of each smelting stage through the optimized smelting control model, and controlling the smelting process to optimize the smelting efficiency in the aluminum processing process in real time; the present application can improve the parameter optimization efficiency of the smelting process and realize real-time monitoring and dynamic optimization of the smelting efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of smelting efficiency optimization, and in particular to a method and system for real-time optimization of smelting efficiency in an aluminum processing process. Background Art

[0002] In aluminum processing, the melting process is a key factor in determining product quality and production efficiency. Traditional aluminum melting control methods rely on manual experience and fixed parameter settings. Operators manually adjust parameters such as the temperature, melting time, and power of the melting equipment based on historical data and their own experience. This manual adjustment method is difficult to ensure timeliness and accuracy, and cannot quickly respond to dynamic changes in the melting process, resulting in large fluctuations in the quality of the produced aluminum products and high scrap rates.

[0003] The existing technology has the following problems: it does not take into account the differences in parameter characteristics of different smelting stages in the smelting process, and adopts a single parameter optimization method, resulting in poor parameter optimization effect; it uses a fixed model to optimize the parameters, which cannot be dynamically adjusted according to actual production conditions. The prediction errors of the long-term running model gradually accumulate, resulting in a decrease in control accuracy and the inability to achieve continuous optimization of the smelting efficiency; the redundant model structure in the model increases the system computing burden and reduces the system response speed, making it difficult to meet the real-time and high efficiency requirements of aluminum product production; in order to solve at least one of the above problems, the present invention proposes a real-time optimization method and system for the smelting efficiency of an aluminum processing process. Summary of the Invention

[0004] In view of the shortcomings of the prior art, the main purpose of the present invention is to provide a method and system for real-time optimization of the smelting efficiency of the aluminum processing process, which can effectively solve the problems in the background technology. The specific technical solutions of the present invention are as follows:

[0005] A method for real-time optimization of smelting efficiency in an aluminum processing process, comprising:

[0006] In the smelting process of aluminum processing, the dependency relationship between parameters is analyzed to construct a parameter dependency map;

[0007] Combining the parameter dependency map with multiple smelting stages of the smelting process, respectively calculating the priority values ​​of the parameters corresponding to each smelting stage, and sorting the priority values ​​from high to low to obtain the parameter priority order for each smelting stage;

[0008] Dynamically pruning the preset smelting control model according to the parameter priority order to obtain an optimized smelting control model;

[0009] The optimized smelting control model is used to optimize the parameters of each smelting stage and control the smelting process, so as to optimize the smelting efficiency in the aluminum processing process in real time.

[0010] Specifically, in the aluminum smelting process, the dependency relationship between parameters is analyzed to construct a parameter dependency graph, including:

[0011] During the smelting process of aluminum processing, each smelting stage is analyzed to obtain the parameters of each smelting stage;

[0012] Analyzing the causal relationship between the parameters to obtain the dependency relationship between the parameters;

[0013] Each parameter is taken as a node, and corresponding nodes are connected according to the dependency relationship to construct a parameter dependency graph.

[0014] Specifically, analyzing the causal relationship between the parameters to obtain the dependency relationship between the parameters includes:

[0015] Through the preset causal association model, the causal relationship between the parameters in each smelting stage is analyzed respectively to obtain the parameter dependency within the stage;

[0016] Based on the parameter dependency within the stage, connections are made between the parameters with causal relationships to obtain a parameter causal graph for each stage;

[0017] Analyzing the causal relationship of the parameters between stages according to the parameter causal diagram to obtain the cross-stage parameter dependency relationship;

[0018] The dependency relationship between parameters is obtained by combining the intra-stage parameter dependency relationship and the cross-stage parameter dependency relationship.

[0019] Specifically, the parameter dependency map and multiple smelting stages of the smelting process are combined to calculate the priority values ​​of the parameters corresponding to each smelting stage, and the priority values ​​are sorted from high to low to obtain the parameter priority order of each smelting stage, including:

[0020] Calculate the priority value of the parameters in each smelting stage by combining the connection relationship of the parameter nodes in the parameter dependency graph and the impact of each smelting stage on the manufacturing efficiency of aluminum products;

[0021] The parameters of each smelting stage are sorted from high to low according to the priority values ​​to obtain the priority order of the parameters of each smelting stage.

[0022] Specifically, the calculation of the priority value of the parameters in each smelting stage by combining the connection relationship of the parameter nodes in the parameter dependency graph and the impact of each smelting stage on the manufacturing efficiency of aluminum products includes:

[0023] According to the parameter dependency graph, the connection relationship between the parameter nodes is analyzed to obtain the parameter node graph feature vector of each smelting stage;

[0024] By analyzing the impact of each smelting stage in the smelting process on the manufacturing efficiency of aluminum products, the stage weight value of each smelting stage is calculated;

[0025] The priority value of the parameters in each smelting stage is calculated by combining the stage weight value and the parameter node graph feature vector of the corresponding smelting stage.

[0026] Specifically, the preset smelting control model is dynamically pruned according to the parameter priority order to obtain an optimized smelting control model, including:

[0027] According to the parameter priority order, the corresponding pruning strategy is matched in the preset pruning strategy library to obtain the dynamic pruning strategy;

[0028] The dynamic pruning strategy is used to dynamically prune the preset smelting control model, and the pruned model is compensated to obtain an optimized smelting control model.

[0029] Specifically, the method of matching the corresponding pruning strategy in the preset pruning strategy library according to the parameter priority order to obtain the dynamic pruning strategy includes:

[0030] According to the parameter priority order, combined with multiple pruning strategies in the preset pruning strategy library, the matching score between each parameter and the pruning strategy is calculated respectively;

[0031] According to the matching score, voting is performed on the pruning strategies corresponding to each smelting stage parameter node to obtain the optimal pruning strategy matching each parameter node;

[0032] The optimal pruning strategies are combined to obtain a dynamic pruning strategy.

[0033] Specifically, the dynamic pruning strategy is used to dynamically prune the preset smelting control model, and the pruned model is compensated to obtain an optimized smelting control model, including:

[0034] According to the dynamic pruning strategy, combined with the priority value of the corresponding parameter, the pruning degree value of the corresponding parameter is calculated respectively;

[0035] Based on the pruning degree value at the corresponding pruning position, pruning the preset smelting control model to obtain a pruned smelting control model;

[0036] At the pruning position, pruning compensation is performed on the pruning smelting control model through a preset pruning compensation model to obtain an optimized smelting control model.

[0037] Specifically, the optimized smelting control model is used to optimize the parameters of each smelting stage and control the smelting process to optimize the smelting efficiency in the aluminum processing process in real time, including:

[0038] By optimizing the smelting control model, the parameters of each smelting stage are optimized to obtain an optimized parameter set;

[0039] generating corresponding smelting control instructions according to the optimized parameter set;

[0040] The smelting process is controlled by the smelting control instructions to optimize the smelting efficiency during the aluminum processing process in real time.

[0041] A real-time optimization system for the smelting efficiency of an aluminum processing process, used to implement the real-time optimization method for the smelting efficiency of an aluminum processing process, comprising:

[0042] The parameter dependency graph construction module constructs a parameter dependency graph by analyzing the dependency relationship between parameters during the aluminum smelting process.

[0043] A parameter priority analysis module, combining the parameter dependency graph and multiple smelting stages of the smelting process, calculates the priority values ​​of the parameters corresponding to each smelting stage, sorts the priority values ​​from high to low, and obtains the parameter priority order for each smelting stage;

[0044] A model optimization module dynamically prunes the preset smelting control model according to the parameter priority order to obtain an optimized smelting control model;

[0045] The smelting efficiency optimization module optimizes the parameters of each smelting stage through the optimized smelting control model and controls the smelting process to optimize the smelting efficiency in the aluminum processing process in real time.

[0046] Compared with the prior art, this application has the following beneficial effects:

[0047] This application constructs a parameter dependency graph based on the parameter dependencies of different smelting stages, calculates the parameter priorities, formulates a dynamic pruning strategy based on the parameter priorities, dynamically prunes the model, and dynamically prunes the model based on the dynamic associations between parameters. This can adapt to changes in the importance of parameters at different stages. Through dynamic pruning, model redundancy can be effectively removed, and while ensuring model accuracy, computing efficiency and response speed can be improved, thereby achieving real-time monitoring and dynamic optimization of smelting efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a workflow diagram of a method for real-time optimization of smelting efficiency in an aluminum processing process in Example 1 of the present invention;

[0049] Figure 2 This is a schematic diagram of the parameter dependency map constructed in Example 1 of the present invention;

[0050] Figure 3 Schematic diagram of the cross-stage parameter dependency analysis process in Example 1 of the present invention;

[0051] Figure 4 This is a structural diagram of a real-time optimization system for smelting efficiency in an aluminum processing process in Example 2 of the present invention. DETAILED DESCRIPTION

[0052] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0053] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0054] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0055] Example 1:

[0056] This embodiment provides a method for optimizing the smelting efficiency of aluminum processing in real time. Figure 1 As shown, the method for real-time optimization of smelting efficiency in an aluminum processing process comprises:

[0057] S101. During the aluminum smelting process, a parameter dependency graph is constructed by analyzing the dependency relationship between parameters.

[0058] S102. Calculating the priority values ​​of the parameters corresponding to each smelting stage based on the parameter dependency map and multiple smelting stages of the smelting process, and sorting the priority values ​​from high to low to obtain a parameter priority order for each smelting stage;

[0059] S103, dynamically pruning the preset smelting control model according to the parameter priority order to obtain an optimized smelting control model;

[0060] S104: Optimize the parameters of each smelting stage through the optimized smelting control model, and control the smelting process to optimize the smelting efficiency in the aluminum processing process in real time.

[0061] This embodiment constructs a parameter dependency map based on the parameter dependency relationship in the aluminum processing and smelting process, combines the parameter dependency map with the different smelting stages of the smelting process, calculates the parameter priority of each smelting stage, matches the corresponding dynamic pruning strategy, dynamically prunes the preset model, and uses the optimization model to perform parameter optimization and process control. Compared with the traditional single model pruning method and fixed parameter optimization method, the present application can accurately identify and adjust key parameters, so that the aluminum product smelting process can be in the best operating state at each stage, thereby improving the smelting efficiency, shortening the production cycle, and optimizing the smelting efficiency of the aluminum processing process in real time.

[0062] In this embodiment, first, the dependency relationship between parameters in the smelting process of aluminum processing is analyzed, and a parameter dependency map is constructed based on the dependency relationship. In the smelting process of aluminum products, there are many parameters that affect the smelting efficiency, including temperature, time, aluminum material composition, fuel supply and other parameters. These parameters do not exist independently, but are interrelated and influence each other. By constructing a parameter dependency map, the dependency relationship between parameters can be clarified, and the impact of parameter changes on other parameters can be analyzed, the internal logic of the smelting process can be mastered, and the adverse effects on other parameters when optimizing a certain parameter can be avoided; based on the parameter dependency map, the optimization direction can be determined more accurately, and blind adjustment of parameters can be avoided, thereby improving the efficiency and accuracy of the optimization work.

[0063] Specifically, after constructing the parameter dependency map, each smelting stage of the smelting process is analyzed based on the parameter dependency map, and the priority values ​​of the parameters in each smelting stage are calculated. The priority values ​​are sorted from high to low to obtain the parameter priority order of each smelting stage. In the smelting process of aluminum products, the key influencing parameters of different smelting stages are different. For example, in the early stage of smelting, rapid heating to a suitable smelting temperature is the key. At this time, parameters such as fuel supply and initial furnace temperature are more important. In the later stage of smelting, ensuring the uniformity of the aluminum liquid composition is the key, and parameters such as aluminum material composition and stirring time are more important. By calculating the priority values ​​of the corresponding parameters in each smelting stage and sorting them, it is possible to clearly identify the parameters that have a greater impact on the smelting efficiency in each stage, so that in the optimization process, priority can be given to the key parameters that have a greater impact on the smelting efficiency, and optimization resources can be reasonably allocated to avoid wasting too much energy on minor parameters. Parameter adjustments can be made in a targeted manner to improve the effect of smelting efficiency optimization.

[0064] Specifically, according to the parameter priority order, the preset smelting control model is dynamically pruned to obtain an optimized smelting control model. The preset smelting control model includes all parameters that affect the smelting efficiency in the smelting process and their control rules. However, in the actual smelting process, not all parameters have a significant impact on the smelting efficiency at each stage. Through the parameter priority order, the parameters that are not important at the current stage are identified, and the preset smelting control model is dynamically pruned to reduce the proportion of unimportant parameters and simplify the model structure, which can improve the model's operating efficiency and optimization effect. At the same time, after reducing unimportant parameters and rules, the model focuses more on key parameters and can more accurately reflect the actual situation of the smelting process, thereby improving the model's optimization effect.

[0065] Specifically, the parameters of each smelting stage are optimized according to the optimized smelting control model, and smelting is carried out using the optimized parameters. By optimizing the parameters of each smelting stage, the smelting parameters can be adjusted according to the characteristics and requirements of the smelting stage, so that the smelting process can be in the best state at each stage, thereby realizing real-time optimization of the smelting efficiency in the aluminum processing process.

[0066] This application constructs a parameter dependency graph based on the parameter dependencies of different smelting stages, calculates the parameter priorities, formulates a dynamic pruning strategy based on the parameter priorities, dynamically prunes the model, and dynamically prunes the model based on the dynamic associations between parameters. This can adapt to changes in the importance of parameters at different stages. Through dynamic pruning, model redundancy can be effectively removed, and while ensuring model accuracy, computing efficiency and response speed can be improved, thereby achieving real-time monitoring and dynamic optimization of smelting efficiency.

[0067] Furthermore, in the aluminum smelting process, the dependency relationship between parameters is analyzed to construct a parameter dependency map, including:

[0068] S201. During the aluminum processing smelting process, each smelting stage is analyzed to obtain parameters of each smelting stage;

[0069] S202, analyzing the causal relationship between the parameters to obtain the dependency relationship between the parameters;

[0070] S203: Take each parameter as a node, connect the corresponding nodes according to the dependency relationship, and construct a parameter dependency graph.

[0071] In this embodiment, first, each smelting stage in the smelting process of aluminum processing is analyzed, and the parameters of each smelting stage are identified. According to the industry standard process specifications for aluminum product smelting and the actual production process of the enterprise, the smelting process is divided into a preheating stage, a heating stage, a melting stage, a refining stage and a holding stage; according to the steps of the smelting process, combined with actual production experience and process knowledge, the parameters involved in each smelting stage are identified; for example, in the preheating stage, the parameters involved include preheating time, preheating temperature, ventilation volume, etc.; in the melting stage, the key parameters include furnace temperature, aluminum material input amount, stirring speed, etc.; by identifying the parameters of each smelting stage, irrelevant parameters or parameters that are not important in a specific stage are avoided from being included in the analysis range, the analysis efficiency is improved, and a data basis is provided for constructing a parameter dependency map.

[0072] Specifically, the causal relationship between parameters is analyzed to obtain the dependency relationship between parameters. There is a complex causal relationship between parameters within the same smelting stage or between different smelting stages. The change of one parameter will affect other parameters. By analyzing the causal relationship between parameters, it is clear which parameters are influencing factors and which parameters are affected factors. By grasping the parameter dependency, the parameters are adjusted during the smelting process optimization to avoid inefficiency caused by improper parameter adjustment.

[0073] like Figure 2 After analyzing the dependency relationship between parameters, each parameter is taken as a node. According to the dependency relationship between parameters, connections are established between corresponding parameter nodes. The connection line uses arrows to indicate the dependency direction, pointing from the influencing factor node to the affected factor node; for example, an arrow is drawn from the "fuel supply" node to the "furnace temperature" node to indicate the impact of fuel supply on furnace temperature; by establishing connections between all parameters with dependencies and constructing a parameter dependency graph, we can quickly understand the interaction between parameters in the smelting process. When optimizing efficiency, we can quickly locate key parameters and related influencing factors based on the parameter dependency graph, thereby improving work efficiency and accuracy.

[0074] Furthermore, the causal relationship between the parameters is analyzed to obtain the dependency relationship between the parameters, including:

[0075] S301. Analyze the causal relationship between parameters in each smelting stage using a preset causal association model to obtain intra-stage parameter dependency.

[0076] S302: Based on the parameter dependency within the stage, connect the parameters with causal relationships to obtain a parameter causal graph for each stage;

[0077] S303: Analyze the causal relationship of the parameters between stages according to the parameter causal graph to obtain the cross-stage parameter dependency relationship;

[0078] S304: Determine the dependency between parameters by combining the intra-stage parameter dependency and the cross-stage parameter dependency.

[0079] In this embodiment, first, the causal relationship between the parameters in each smelting stage of the smelting process is analyzed to obtain the parameter dependency within the stage. In each stage of aluminum product smelting, there is an inherent causal relationship between the parameters. The change of one parameter will affect other parameters. The causal relationship of the parameters in each stage is analyzed separately through a preset causal association model. The causal association model can be a machine learning model or a statistical model. In this embodiment, the causal association model uses a Bayesian network model, and uses a large amount of historical data to learn the probabilistic causal relationship between the parameters to obtain a pre-trained Bayesian network model; the parameter data of each smelting stage are respectively input into the pre-trained Bayesian network model, and the model calculates indicators such as the correlation and influence degree between the parameters, and determines which parameters are causes and which parameters are effects, as well as the direction and intensity of the influence between the parameters; for example, the model will analyze that in the melting stage, an increase in the fuel supply will significantly increase the furnace temperature, and there is a certain time delay relationship; by analyzing the causal relationship of the parameters in each stage, the influence relationship between different parameters can be clarified.

[0080] Specifically, according to the analyzed parameter dependencies within each stage, connections are made between parameters with causal relationships to obtain parameter causal diagrams for each stage. Arrows are used as connecting lines to indicate causal directions, pointing from the cause parameter node to the effect parameter node. For example, in the melting stage, the fuel supply is the cause and the furnace temperature is the effect, and an arrow is drawn from the "fuel supply" node to the "furnace temperature" node. By constructing parameter causal diagrams for each stage separately, the degree of connection between the parameters in each smelting stage can be intuitively observed, and the interaction between the parameters can be quickly grasped.

[0081] Furthermore, based on the parameter causal diagram of each stage, the correlation between the parameter causal diagrams of different stages is analyzed, and the dependency between the parameters across stages is identified. Aluminum product smelting is a continuous process. The parameters between different stages are not independent of each other, but there is a certain causal relationship. The parameter status of the previous stage will affect the parameter performance of the subsequent stage. By analyzing the parameter causal diagram of each stage, these cross-stage parameter causal relationships are identified; by analyzing the cross-stage parameter causal relationship, it is helpful to grasp the smelting process as a whole, optimize the entire smelting process, and avoid adverse effects on subsequent stages when adjusting the parameters of a certain stage.

[0082] like Figure 3 , compare the parameter cause-effect diagrams of different stages, find the same or related parameters that appear in multiple stages, and identify the key connection points, Figure 3 Arrow A in the figure corresponds to the common parameter in stage 1 and stage 2: melting temperature, and arrow B in the figure corresponds to the common parameter in stage 1 and stage 2: melting time. In stage 1, the increase in fuel supply will increase the melting temperature. When the melting temperature increases, the chemical reaction rate will be accelerated. In stage 2, the change in melting temperature will affect other parameters. The change in melting time in stage 1 will affect the melting time in stage 2. When the melting time in stage 1 increases, in order to ensure that the quality of aluminum products is not affected, the melting time in stage 2 needs to be reduced. The same parameters in stage 1 and stage 2, namely melting temperature and melting time, are used as key connection points. Combined with the influence relationship between the key connection points, a dependency relationship is established between the parameters in stage 1 and stage 2.

[0083] For example, the furnace temperature is a common parameter in the heating stage, melting stage and refining stage, and there is a causal transmission relationship between the furnace temperature in different stages; according to the key connection points, the causal path is searched in the causal diagram of different stages to analyze how the change of the parameter in the previous stage affects other parameters in the subsequent stages; for example, in the heating stage, the increase in fuel supply causes the furnace temperature to rise, and in the melting stage, the higher furnace temperature will accelerate the melting rate of the aluminum material, thereby establishing a causal relationship between the heating stage and the melting stage through the furnace temperature; by analyzing the cross-stage parameter causal relationship, the aluminum product smelting process is analyzed from a global perspective. When optimizing efficiency, the mutual influence of each stage can be comprehensively considered to make more reasonable decisions, thereby improving the coordination and production efficiency of the entire smelting process.

[0084] After analyzing the intra-stage parameter dependencies and cross-stage parameter dependencies, duplicate relationship expressions are removed, the relationships between the same parameters are merged, the integrated parameter dependencies are logically sorted out, and the contradictory or unreasonable relationships are corrected to ensure that the entire dependency network is logically coherent and consistent, and can accurately reflect the true relationship between the parameters; the analyzed parameter dependencies can provide a reference for the aluminum product smelting optimization process, and analysis and decision-making based on the complete parameter dependencies can improve production stability, efficiency and product quality.

[0085] Furthermore, the parameter dependency map is combined with multiple smelting stages of the smelting process to calculate the priority values ​​of the parameters corresponding to each smelting stage, and the priority values ​​are sorted from high to low to obtain the parameter priority order of each smelting stage, including:

[0086] S401. Calculate the priority value of the parameters in each smelting stage based on the connection relationship of the parameter nodes in the parameter dependency graph and the impact of each smelting stage on the aluminum product manufacturing efficiency;

[0087] S402 , sorting the parameters of each smelting stage according to the priority values ​​from high to low to obtain the priority order of the parameters of each smelting stage.

[0088] In this embodiment, the parameter priority of each smelting stage is analyzed in combination with the connection relationship of the parameter nodes in the parameter dependency graph and the influence of each smelting stage on the manufacturing efficiency of aluminum products, and the priority value of the parameters in each smelting stage is calculated; in the smelting process of aluminum products, different parameters have different degrees of influence on each smelting stage, and there are mutual correlations between the parameters. The parameter dependency graph includes the connection relationship between the parameters, which reflects the scope and degree of influence of a parameter change on other parameters; and each smelting stage has different contributions and influences on the manufacturing efficiency of aluminum products. By combining the parameter association relationship and the degree of influence of the stage, the importance of each parameter in a specific smelting stage is analyzed, and the priority value is calculated. The higher the priority value, the greater the influence of the parameter on the smelting efficiency and the overall manufacturing efficiency of this stage. Parameters with high priority are given priority in the optimization process; by analyzing the importance of each parameter in the smelting stage, important parameters can be focused on when optimizing efficiency, resources can be reasonably allocated, and optimization efficiency and effect can be improved.

[0089] Specifically, after calculating the priority value of each parameter in each smelting stage, the parameters are sorted from high to low to clarify which parameters are critical and need priority attention and adjustment at this stage, and which parameters are relatively less important. In this way, in the process of efficiency optimization, important parameters can be monitored and controlled in a targeted manner to improve production efficiency and product quality.

[0090] Furthermore, the calculation of the priority value of the parameters in each smelting stage by combining the connection relationship of the parameter nodes in the parameter dependency graph and the influence of each smelting stage on the manufacturing efficiency of the aluminum product includes:

[0091] S501. Analyze the connection relationship between parameter nodes according to the parameter dependency graph to obtain the parameter node graph feature vector of each smelting stage;

[0092] S502. Calculate the stage weight value of each smelting stage by analyzing the impact of each smelting stage in the smelting process on the aluminum product manufacturing efficiency;

[0093] S503 , combining the stage weight value and the parameter node graph feature vector of the corresponding smelting stage to calculate the priority value of the parameters in each smelting stage.

[0094] In this embodiment, according to the parameter dependency graph, the connection relationship between the parameter nodes is analyzed, and the parameter node graph feature vector of each smelting stage is extracted; by analyzing the connection relationship and correlation degree between each parameter node and other parameter nodes, the degree of the node, that is, the number of edges connected to the node, is calculated respectively; betweenness centrality represents the ability of the node to control the flow of information in the network; closeness centrality represents the average shortest path length from the node to all other nodes and other graph features. The calculated graph features are combined to construct a parameter node graph feature vector. By calculating the parameter node graph feature vector, the connection relationship in the parameter dependency graph can be quantified, providing a data reference for calculating the parameter priority.

[0095] Specifically, the impact of each smelting stage in the smelting process on the efficiency of aluminum product manufacturing is analyzed, and the stage weight value of each smelting stage is calculated. During the aluminum product smelting process, different smelting stages have different degrees of influence on the manufacturing efficiency of aluminum products. Some stages are key links in determining the quality and production efficiency of aluminum products, while others play an auxiliary or subsequent processing role. By calculating the stage weight value, the importance of each smelting stage in the entire aluminum product manufacturing process can be measured. Combined with factors such as each smelting stage's position in the entire production process, its direct contribution to product quality, the length of production time, and the amount of resource consumption, each smelting stage is scored and weighted to calculate the corresponding stage weight value for each smelting stage. By calculating the stage weight value, the importance of each smelting stage in the aluminum product manufacturing process can be clarified. When calculating the parameter priority value, considering the stage weight can avoid neglecting the parameters of key stages and ensure that resources are allocated to the stages and their key parameters that have a greater impact on manufacturing efficiency, thereby improving overall production efficiency and product quality.

[0096] Specifically, a weighted summation calculation is performed by combining the stage weights and the corresponding stage's parameter node graph eigenvectors to obtain the priority values ​​of the parameters in each smelting stage. For example, in the refining stage of aluminum product smelting, the stage weight is 0.36, and the eigenvector of a certain additive dosage parameter is [3, 0.4, 0.5]. The weights assigned to the node's degree, betweenness centrality, and closeness centrality are 0.5, 0.3, and 0.2, respectively. The node's comprehensive score is first calculated as 3 × 0.5 + 0.4 × 0.3 + 0.5 × 0.2 = 1.72, and the priority value of this parameter is then calculated as 1.72 × 0.36 = 0.6192. Calculating parameter priority values ​​by combining stage weights with parameter node graph features allows for a comprehensive assessment of parameter importance from multiple dimensions, making the calculation results more consistent with actual production conditions. During production efficiency optimization, parameters with the greatest impact on manufacturing efficiency can be prioritized and optimized, improving the accuracy and effectiveness of production operations, thereby enhancing the production efficiency and quality of aluminum products.

[0097] Furthermore, the preset smelting control model is dynamically pruned according to the parameter priority order to obtain an optimized smelting control model, including:

[0098] S601: Match a corresponding pruning strategy in a preset pruning strategy library according to the parameter priority order to obtain a dynamic pruning strategy;

[0099] S602 : Dynamically prune the preset smelting control model through the dynamic pruning strategy, and compensate the pruned model to obtain an optimized smelting control model.

[0100] In this embodiment, according to the analyzed parameter priority order, corresponding pruning strategies are matched in the preset pruning strategy library to obtain a dynamic pruning strategy; the preset smelting control model includes all parameters affecting the smelting process and their control rules. However, in different stages of actual production, not all parameters have a significant impact on the smelting efficiency. The parameter priority order determines the importance of each parameter in each smelting stage. The preset pruning strategy library stores multiple pruning strategies. According to the parameter priority order, the pruning strategy library is matched to the dynamic pruning strategy that best suits the parameter importance distribution in the current stage. The smelting control model can be simplified in a targeted manner, unnecessary parameters and rules can be removed, the model operation efficiency can be improved, and the over-simplification or under-simplification problems caused by the traditional use of a single fixed pruning method can be avoided.

[0101] Specifically, after determining the dynamic pruning strategy, the preset smelting control model is dynamically pruned to remove parameters and their related control rules that are not important at the current stage, simplify the model structure, and reduce the amount of model calculation. However, directly deleting some parameters and rules will affect the integrity and accuracy of the model and affect the normal operation of the model. After pruning, the model is compensated by adjusting the weights of parameters, adding new rules, or correcting the original rules, so that the pruned model can still accurately reflect the actual situation of the smelting process and meet the needs of production control, thereby obtaining an optimized smelting control model, which provides strong support for the efficient and precise control of the aluminum product smelting process and helps to improve production efficiency and product quality.

[0102] Furthermore, the method of matching a corresponding pruning strategy in a preset pruning strategy library according to the parameter priority order to obtain a dynamic pruning strategy includes:

[0103] S701: Calculate the matching score between each parameter and the pruning strategy based on the parameter priority order and in combination with multiple pruning strategies in a preset pruning strategy library;

[0104] S702: Voting on the pruning strategies corresponding to each smelting stage parameter node according to the matching score to obtain the optimal pruning strategy matching each parameter node;

[0105] S703: Combine the optimal pruning strategies to obtain a dynamic pruning strategy.

[0106] In this embodiment, according to the parameter priority order and in combination with multiple pruning strategies in the preset pruning strategy library, the matching score between each parameter and the pruning strategy is calculated respectively; based on historical production experience, industry knowledge and understanding of the smelting process, multiple pruning strategies are formulated. For example, when the number of high-priority parameters is small and concentrated, a strategy is formulated to delete low-priority parameters and their related rules; when high-priority parameters are scattered among different types of parameters, a strategy is formulated to retain a certain proportion of high-priority parameters according to parameter category, etc., and a pruning strategy library is constructed according to the corresponding pruning strategies.

[0107] Specifically, according to the parameter priority order, the matching score between each parameter and the pruning strategy is calculated. First, combined with the historical pruning method of the model, corresponding parameter priority characteristic conditions are formulated for each pruning strategy in the preset pruning strategy library. For example, a pruning strategy is applicable to the situation where "there are less than 3 high-priority parameters and the proportion of low-priority parameters exceeds 70%". Secondly, for the parameter priority order of each smelting stage, the priority-related characteristics of each parameter are extracted, including the priority value of the parameter, the priority ranking among all parameters in this stage, the parameter category to which it belongs, specifically including temperature, time, raw materials and other categories, as well as the degree of association with high-priority parameters of the same category or other categories. The priority characteristics of each parameter are compared with the applicable conditions of each pruning strategy in the pruning strategy library, and scored according to the degree of matching to obtain the corresponding matching score. By calculating the matching score, the degree of adaptability of each parameter to different pruning strategies can be reflected, thereby screening out the corresponding pruning strategy and improving the scientificity and accuracy of pruning strategy selection.

[0108] Specifically, after calculating the matching score of each parameter and each pruning strategy, the pruning strategies corresponding to each smelting stage parameter node are voted respectively to obtain the optimal pruning strategy matching each parameter node. Considering that the matching of a single parameter cannot directly determine the pruning strategy of the entire smelting stage, the pruning strategies corresponding to each parameter node are voted. For each parameter node, the pruning strategy with the highest matching score is selected as the voting object of the node; when there are multiple pruning strategies with the same and highest matching score, weighted voting is adopted to allocate votes according to the priority weight of the parameters, and the number of votes obtained by the pruning strategy corresponding to each parameter node is calculated. By comparing the number of votes obtained by each pruning strategy, the pruning strategy with the highest number of votes is used as the pruning strategy matched to the parameter node, and the optimal pruning strategy is matched to each parameter node; by voting to match the optimal pruning strategy, the selection of pruning strategy is more comprehensive, avoiding the irrationality caused by selecting pruning strategy based on only a few parameters.

[0109] Specifically, the optimal pruning strategies matching each parameter node are combined to obtain a complete pruning scheme, and the conflicting or repeated parts in the pruning strategies are identified. For the conflicting parts, adjustments are made by selecting a strategy that is more in line with the overall parameter priority distribution characteristics. For the repeated parts, the pruning schemes are optimized by merging or simplifying to obtain the final dynamic pruning strategy; the dynamic pruning strategy is obtained by combining the optimal pruning strategies, and the parameters of the smelting stage are considered as a whole. The priority information of each parameter is fully utilized to accurately prune the smelting control model, which not only removes unnecessary parameters and rules, but also retains key information, thereby improving the operation efficiency and accuracy of the model.

[0110] Furthermore, the dynamic pruning strategy is used to dynamically prune the preset smelting control model, and the pruned model is compensated to obtain an optimized smelting control model, including:

[0111] S801. Calculate the pruning degree values ​​of the corresponding parameters according to the dynamic pruning strategy and the priority values ​​of the corresponding parameters.

[0112] S802: Pruning the preset smelting control model at the corresponding pruning position based on the pruning degree value to obtain a pruned smelting control model;

[0113] S803 . Performing pruning compensation on the pruning smelting control model at the pruning position using a preset pruning compensation model to obtain an optimized smelting control model.

[0114] In this embodiment, first, according to the dynamic pruning strategy and the priority value of the corresponding parameter, the corresponding pruning degree value is calculated, and different parameters are pruned differently. For parameters with high priority, the pruning degree is reduced to retain their key influence on the smelting process; for parameters with low priority, the pruning degree is increased; the pruning degree value is calculated in combination with the parameter priority, specifically by basic pruning coefficient × (1-parameter priority correction coefficient × priority value / highest priority value), to calculate the pruning degree value of each parameter; the basic pruning coefficient can be determined according to the type of dynamic pruning strategy to reflect the overall pruning intensity; the parameter priority correction coefficient can be determined based on experience The corresponding value is used to adjust the influence of priority on the degree of pruning. For example, if a parameter priority value is 8, the highest priority value is 10, the basic pruning coefficient is 0.8, and the parameter priority correction coefficient is 0.6, then the pruning degree of the parameter is 0.8×(1-0.6×8 / 10)=0.416. All parameters in each smelting stage are calculated separately to obtain the pruning degree value corresponding to each parameter. By calculating the pruning degree value, different parameters can be pruned to different degrees, which not only ensures the effective simplification of unimportant parameters and reduces the amount of model calculation, but also protects the influence of key parameters on the smelting process and maintains the control accuracy of the model.

[0115] Specifically, after calculating the pruning degree value of each parameter, the corresponding pruning position is found in the preset smelting control model according to the corresponding pruning degree value, and the model is pruned. The pruning position corresponds to the definition, calculation logic, constraint conditions and other parts related to the parameters in the model; according to the determined pruning operation type, the model is pruned at the corresponding pruning position, including deletion, modification or simplification. After completing the corresponding pruning operation for all parameters in each smelting stage, a pruned smelting control model is obtained; by pruning the preset smelting control model, unnecessary parts of the model can be removed, the complexity and calculation amount of the model can be reduced, and differentiated operations can be performed according to the pruning degree value. While simplifying the model, the core functions and key control logic of the model are retained to the maximum extent, ensuring that the model can still effectively control the smelting process after pruning.

[0116] Specifically, after the model is pruned, the pruned smelting control model is pruned and compensated at the pruning position to compensate for the problem of missing or inaccurate logical relationships in the model caused by the pruning operation. The pruned smelting control model is compensated at the pruning position through a preset pruning compensation model to repair the logical loopholes caused by pruning and adjust the relationship between parameters so that the model can still accurately reflect the actual situation of the smelting process after simplification and meet the needs of production control, thereby obtaining an optimized smelting control model; the preset pruning compensation model contains a variety of compensation strategies for different pruning effects, including a parameter substitution strategy, which calculates by replacing the deleted parameters with other relevant parameters; a rule correction strategy, which re-formulates or modifies the affected control rules; and a relationship adjustment strategy, which adjusts the correlation between parameters; at the pruning position, the pruned smelting control model is compensated by a matching compensation strategy to obtain an optimized smelting control model, and the output accuracy and effectiveness of the optimized smelting control model are verified. By performing pruning compensation on the pruned smelting control model, the impact of the pruning operation can be compensated, the integrity and accuracy of the model can be guaranteed, and the production quality and efficiency of aluminum products can be ensured.

[0117] Furthermore, the optimized smelting control model is used to optimize the parameters of each smelting stage and control the smelting process to optimize the smelting efficiency in the aluminum processing process in real time, including:

[0118] S901, optimizing the parameters of each smelting stage by optimizing the smelting control model to obtain an optimized parameter set;

[0119] S902. Generate corresponding smelting control instructions according to the optimization parameter set;

[0120] S903: Control the smelting process through the smelting control instruction to optimize the smelting efficiency during the aluminum processing process in real time.

[0121] In this embodiment, after obtaining the optimized smelting control model, the parameters of each smelting stage are optimized respectively through the optimized smelting control model to obtain an optimized parameter set. In each smelting stage, parameter data is collected in real time by sensors, and the real-time data is input into the optimized smelting control model of the corresponding stage. The model calculates the optimal value of each parameter. By calculating the optimal value of the parameters of each smelting stage, the optimized parameter set is obtained. The optimal parameters are calculated by associating the stage characteristics and the parameters, and the smelting process can be adjusted in real time.

[0122] Specifically, the optimized parameter set is converted into a melting control instruction executable by the equipment, and a mapping relationship between the parameter optimization value and the equipment operation is established; for example, the temperature optimization value corresponds to the fuel valve opening, and the stirring time corresponds to the motor running time, and the corresponding parameter values ​​are converted into equipment status control instructions; the melting process is controlled according to the melting control instructions, and the equipment executes the instructions and provides real-time feedback of status data, thereby optimizing and controlling the melting process in real time; through real-time control and optimization, the processing time of aluminum products can be shortened and the efficiency of the melting process can be improved.

[0123] Example 2

[0124] In this embodiment, if Figure 4 , provides a real-time optimization system for the smelting efficiency of an aluminum processing process, which is used to implement the real-time optimization method for the smelting efficiency of an aluminum processing process, comprising:

[0125] The parameter dependency graph construction module constructs a parameter dependency graph by analyzing the dependency relationship between parameters during the aluminum smelting process.

[0126] A parameter priority analysis module, combining the parameter dependency graph and multiple smelting stages of the smelting process, calculates the priority values ​​of the parameters corresponding to each smelting stage, sorts the priority values ​​from high to low, and obtains the parameter priority order for each smelting stage;

[0127] A model optimization module dynamically prunes the preset smelting control model according to the parameter priority order to obtain an optimized smelting control model;

[0128] The smelting efficiency optimization module optimizes the parameters of each smelting stage through the optimized smelting control model and controls the smelting process to optimize the smelting efficiency in the aluminum processing process in real time.

[0129] In this embodiment, the parameter dependency map construction module analyzes the dependency relationship between parameters during the smelting process of aluminum processing, and constructs a parameter dependency map based on the intrinsic connection between the parameters, which can reflect the interaction between the parameters and provide a reference for parameter optimization and model optimization; the parameter priority analysis module combines the parameter dependency map with multiple stages of the aluminum product smelting process, analyzes the importance of the parameters in each stage, assigns a priority value to the parameters of each stage, and sorts them from high to low according to the priority value, determines the priority order of the parameters in each stage, identifies key parameters, and can prioritize optimization of key parameters during the optimization process to avoid wasting time and resources on secondary parameters, improve optimization efficiency, and help improve the efficiency of the entire smelting process and product quality.

[0130] Specifically, the model optimization module prunes and optimizes the preset smelting control model according to the parameter priority order obtained by the parameter priority analysis module. Through dynamic pruning, it removes parameters and related rules that are not important in the model at the current stage, simplifies the model structure, and compensates the pruned model to ensure the accuracy and effectiveness of the model. Finally, an optimized smelting control model is obtained, redundant information in the smelting control model is removed, and the operation speed and response efficiency of the model are improved. The smelting efficiency optimization module uses the optimized smelting control model obtained by the model optimization module to perform real-time parameter optimization and smelting control on each stage of the aluminum product smelting process. By continuously adjusting the parameters, the smelting process is always in the best operating state, realizing real-time optimization of the smelting efficiency in the aluminum processing process, improving production efficiency, reducing production costs, and improving product quality.

[0131] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A real-time optimization method for aluminum processing smelting efficiency, characterized in that: include: In the smelting process of aluminum processing, the dependency relationship between parameters is analyzed to construct a parameter dependency map; Combining the parameter dependency map with multiple smelting stages of the smelting process, respectively calculating the priority values ​​of the parameters corresponding to each smelting stage, and sorting the priority values ​​from high to low to obtain the parameter priority order for each smelting stage; Dynamically pruning the preset smelting control model according to the parameter priority order to obtain an optimized smelting control model; Optimizing the parameters of each smelting stage and controlling the smelting process by the optimized smelting control model to optimize the smelting efficiency in the aluminum processing process in real time; In the aluminum smelting process, the dependency relationship between parameters is analyzed to construct a parameter dependency graph, including: During the smelting process of aluminum processing, each smelting stage is analyzed to obtain the parameters of each smelting stage; Analyzing the causal relationship between the parameters to obtain the dependency relationship between the parameters; Taking each parameter as a node, connecting the corresponding nodes according to the dependency relationship to construct a parameter dependency graph; The parameter dependency graph and the multiple smelting stages of the smelting process are combined to calculate the priority values ​​of the parameters corresponding to each smelting stage, and the priority values ​​are sorted from high to low to obtain the parameter priority order of each smelting stage, including: Calculate the priority value of the parameters in each smelting stage by combining the connection relationship of the parameter nodes in the parameter dependency graph and the impact of each smelting stage on the manufacturing efficiency of aluminum products; The parameters of each smelting stage are sorted from high to low according to the priority values ​​to obtain the priority order of the parameters of each smelting stage.

2. The method for real-time optimization of smelting efficiency in an aluminum processing process according to claim 1, characterized in that: The analyzing the causal relationship between the parameters to obtain the dependency relationship between the parameters includes: Through the preset causal association model, the causal relationship between the parameters in each smelting stage is analyzed respectively to obtain the parameter dependency within the stage; Based on the parameter dependency within the stage, connections are made between the parameters with causal relationships to obtain a parameter causal graph for each stage; Analyzing the causal relationship of the parameters between stages according to the parameter causal diagram to obtain the cross-stage parameter dependency relationship; The dependency relationship between parameters is obtained by combining the intra-stage parameter dependency relationship and the cross-stage parameter dependency relationship.

3. The method for real-time optimization of smelting efficiency in an aluminum processing process according to claim 1, characterized in that: The calculation of the priority value of the parameters in each smelting stage by combining the connection relationship of the parameter nodes in the parameter dependency graph and the influence of each smelting stage on the manufacturing efficiency of the aluminum product includes: According to the parameter dependency graph, the connection relationship between the parameter nodes is analyzed to obtain the parameter node graph feature vector of each smelting stage; By analyzing the impact of each smelting stage in the smelting process on the manufacturing efficiency of aluminum products, the stage weight value of each smelting stage is calculated; The priority value of the parameters in each smelting stage is calculated by combining the stage weight value and the parameter node graph feature vector of the corresponding smelting stage.

4. The method for real-time optimization of smelting efficiency in aluminum processing according to claim 1, characterized in that: The method of dynamically pruning the preset smelting control model according to the parameter priority order to obtain an optimized smelting control model includes: According to the parameter priority order, the corresponding pruning strategy is matched in the preset pruning strategy library to obtain the dynamic pruning strategy; The dynamic pruning strategy is used to dynamically prune the preset smelting control model, and the pruned model is compensated to obtain an optimized smelting control model.

5. The method for real-time optimization of smelting efficiency in aluminum processing according to claim 4, characterized in that: The method of matching the corresponding pruning strategy in the preset pruning strategy library according to the parameter priority order to obtain the dynamic pruning strategy includes: According to the parameter priority order, combined with multiple pruning strategies in the preset pruning strategy library, the matching score between each parameter and the pruning strategy is calculated respectively; According to the matching score, voting is performed on the pruning strategies corresponding to each smelting stage parameter node to obtain the optimal pruning strategy matching each parameter node; The optimal pruning strategies are combined to obtain a dynamic pruning strategy.

6. The method for real-time optimization of smelting efficiency in aluminum processing according to claim 4, characterized in that: The dynamic pruning strategy is used to dynamically prune the preset smelting control model, and the pruned model is compensated to obtain an optimized smelting control model, including: According to the dynamic pruning strategy, combined with the priority value of the corresponding parameter, the pruning degree value of the corresponding parameter is calculated respectively; Based on the pruning degree value at the corresponding pruning position, pruning the preset smelting control model to obtain a pruned smelting control model; At the pruning position, pruning compensation is performed on the pruning smelting control model through a preset pruning compensation model to obtain an optimized smelting control model.

7. The method for real-time optimization of smelting efficiency in aluminum processing according to claim 1, characterized in that: The optimized smelting control model is used to optimize the parameters of each smelting stage and control the smelting process to optimize the smelting efficiency in the aluminum processing process in real time, including: By optimizing the smelting control model, the parameters of each smelting stage are optimized to obtain an optimized parameter set; generating corresponding smelting control instructions according to the optimized parameter set; The smelting process is controlled by the smelting control instructions to optimize the smelting efficiency in the aluminum processing process in real time.

8. A real-time optimization system for aluminum processing smelting efficiency, characterized by: A method for realizing a real-time optimization method for the melting efficiency of an aluminum processing process as claimed in any one of claims 1 to 7, comprising: The parameter dependency graph construction module constructs a parameter dependency graph by analyzing the dependency relationship between parameters during the aluminum smelting process. A parameter priority analysis module, combining the parameter dependency graph and multiple smelting stages of the smelting process, calculates the priority values ​​of the parameters corresponding to each smelting stage, sorts the priority values ​​from high to low, and obtains the parameter priority order for each smelting stage; A model optimization module dynamically prunes the preset smelting control model according to the parameter priority order to obtain an optimized smelting control model; The smelting efficiency optimization module optimizes the parameters of each smelting stage through the optimized smelting control model and controls the smelting process to optimize the smelting efficiency in the aluminum processing process in real time.

Images