Aluminum alloy casting production control method and system

Abstract

The invention discloses an aluminum alloy casting production control method and system, and relates to the technical field of industrial production control, and the method comprises the steps: obtaining the product defect index and the production efficiency of each remaining pouring scheme, calculating the product defect index and the production efficiency through an expression function to obtain the expression value of the pouring scheme, outputting a plurality of groups of pouring scheme tables, a plurality of sets of pouring scheme lists are fused to obtain a scheme list, a control system selects the pouring scheme with the maximum performance value in the scheme list to control the pouring process of a factory on the same day, and aluminum alloy products are poured according to the pouring schemes. And the control system dynamically adjusts the pouring scheme according to a comparison result of the process assignment and the gradient threshold value, and the dynamically adjusted pouring scheme is applied to subsequent pouring production of aluminum alloy products. The control method effectively guarantees the quality of the aluminum alloy product, improves the production efficiency of the aluminum alloy product and reduces the production cost.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial production control, and in particular to an aluminum alloy casting production control method and system. Background Art

[0002] Aluminum alloy casting is a common metal processing method, which is usually used to manufacture various types of parts and components. Aluminum alloy is a light, corrosion-resistant and high-strength material, so it is widely used in automobiles, ships, aerospace, construction and other fields.

[0003] In the aluminum alloy casting production process, different pouring schemes have different effects on the quality of aluminum alloy products. In the prior art, pouring is usually performed by manually setting the pouring scheme. This pouring method has the following defects:

[0004] 1. The manually set pouring scheme is highly subjective, which may result in the set pouring scheme being a general pouring scheme (i.e., not the optimal pouring scheme), which not only fails to guarantee the quality of aluminum alloy products, but also may reduce the production efficiency of aluminum alloy casting and increase production costs;

[0005] 2. When a fixed pouring scheme is selected for pouring aluminum alloy products, due to the presence of some influencing factors (generally environmental factors and equipment factors) in actual production conditions, the manually set pouring scheme cannot be combined with real-time influencing factors for dynamic pouring control, and thus cannot adapt to changes in actual production, which will further lead to a decrease in casting quality or reduced production efficiency.

[0006] Based on this, the present invention proposes an aluminum alloy casting production control method and system, which first combines the aluminum alloy product information that the factory needs to produce on the day and the factory's historical production data to automatically find the global optimal casting scheme, and then dynamically controls the casting scheme according to real-time influencing factors during the production process of the aluminum alloy products, thereby ensuring the quality of the aluminum alloy products while improving the production efficiency of the aluminum alloy products and reducing the production costs. Summary of the invention

[0007] The purpose of the present invention is to provide an aluminum alloy casting production control method and system to solve the shortcomings of the background technology.

[0008] In order to achieve the above object, the present invention provides the following technical solution: a method for controlling aluminum alloy casting production, the control method comprising the following steps:

[0009] S1: The control system randomly generates several casting schemes based on the information of aluminum alloy products that the factory needs to produce on the day and the historical production data of the factory. The historical product failure rates of all casting schemes are first obtained. After the casting schemes with product failure rates greater than the failure rate threshold are screened out, the product defect index and production efficiency of each remaining casting scheme are obtained. The performance value of the casting scheme is obtained by calculating the product defect index and production efficiency through the performance function;

[0010] S2: Select some pouring schemes according to the performance values ​​by using a virtual roulette selection method, mutate the selected pouring schemes by using a differential operation, and establish a pouring scheme table for all pouring schemes before and after the mutation;

[0011] S3: Repeat step S2 for multiple iterations. When the iteration conditions are met, output several groups of pouring plan tables, merge several groups of pouring plan tables to obtain a plan list, and the control system selects the pouring plan with the largest performance value in the plan list to control the pouring process of the factory on that day;

[0012] S4: During the pouring process of the aluminum alloy product according to the pouring plan, the control system obtains the solidification time and grain size of the aluminum alloy product after the current pouring, generates the process assignment of the aluminum alloy product after pre-processing the solidification time and grain size, and dynamically adjusts the pouring plan according to the comparison result between the process assignment and the gradient threshold, and applies the dynamically adjusted pouring plan to the subsequent pouring production of the aluminum alloy product.

[0013] In a preferred embodiment, the performance value of the casting scheme is obtained by calculating the product defect index and the production efficiency through the performance function, and the performance function expression is:

[0014] P_value=α*SCX-β*CQZ, where P_value is the performance value, SCX is the production efficiency, CQZ is the product defect index, α and β are the proportional coefficients of production efficiency and product defect index respectively, and both α and β are greater than 0.

[0015] In a preferred embodiment, a virtual roulette selection method is used to select a partial casting scheme according to the performance value, comprising the following steps:

[0016] Obtain the performance values ​​of all casting schemes, add up the performance values ​​of all casting schemes to obtain the total performance value, and obtain the round selection rate of each casting scheme by dividing the performance value by the total performance value;

[0017] Map each casting scheme to the corresponding sector area on the virtual wheel according to the round selection rate. The larger the round selection rate, the larger the sector area occupied by the casting scheme on the virtual wheel.

[0018] After all the pouring schemes are mapped, the virtual wheel is driven to rotate. When the virtual wheel stops rotating, the virtual pointer is pointed to the pouring scheme corresponding to the sector area, the selected pouring scheme sector area is deleted, and the sector areas of the remaining pouring schemes are regenerated;

[0019] When the number of selected casting schemes is equal to the quantity threshold, the casting scheme selection is completed.

[0020] In a preferred embodiment, the selected partial pouring scheme is subjected to scheme content variation through a differential operation, comprising the following steps:

[0021] By randomly selecting two different pouring schemes, the target pouring scheme is mutated to generate a variant pouring scheme. The mutation formula is: V i =X i +F*(X w -X g ), where V i is the i-th variant pouring scheme, X i is the i-th target pouring scheme, F is the scaling factor, X w , X g Two different pouring schemes were randomly selected for the random selection.

[0022] In a preferred embodiment, the logic for obtaining the solidification time is: recording the pouring start time point and the solidification completion time point of the aluminum alloy product, and obtaining the solidification time by subtracting the pouring start time point from the solidification completion time point;

[0023] The logic for obtaining the grain size is: after the aluminum alloy product is solidified, the grain size of the aluminum alloy product is obtained through a metallographic microscope.

[0024] In a preferred embodiment, the solidification time and the grain size are pre-processed to generate a process assignment for the aluminum alloy product, and the casting scheme is dynamically adjusted according to the comparison result between the process assignment and the gradient threshold, including the following steps:

[0025] The solidification time and grain size are preprocessed by a general normalization processing formula, so that the value range of the solidification time normalization value and the grain size normalization value obtained after the processing is mapped to [0,1], and the process assignment is obtained by adding the solidification time normalization value to the grain size normalization value;

[0026] The gradient threshold includes a maximum limit threshold and a minimum limit threshold, and the minimum limit threshold is smaller than the maximum limit threshold;

[0027] If the process assignment is greater than the maximum limit threshold, the control system will reduce the current pouring temperature and pouring speed by 5%;

[0028] If the process assignment is less than the minimum limit threshold, the control system increases the current pouring temperature and pouring speed by 5 percent;

[0029] If the process assignment is greater than or equal to the minimum limit threshold, and the process assignment is less than or equal to the maximum limit threshold, the control system does not dynamically adjust the casting plan.

[0030] In a preferred embodiment, the logic for obtaining the product defect index is as follows: in the actual casting process, when the absolute value of the weight deviation of the actual weight of the aluminum alloy product minus the standard weight is between 0 and the deviation threshold, it indicates that the aluminum alloy product has a slight under-casting or over-casting, and all aluminum alloy products whose absolute value of the weight deviation is between 0 and the deviation threshold when the casting scheme is used are obtained, and the absolute values ​​of the weight deviations of all aluminum alloy products are summed up to obtain the product defect index.

[0031] In a preferred embodiment, the logic for obtaining the production efficiency is: obtaining the number of aluminum alloy products cast within the monitoring time period of the casting scheme, and dividing the number of aluminum alloy products cast by the monitoring time to obtain the production efficiency.

[0032] In a preferred embodiment, the control system randomly generates several casting schemes based on the aluminum alloy product information that the factory needs to produce on the day and the factory's historical production data, including the following steps:

[0033] Obtain information about aluminum alloy products that the factory needs to produce that day, including product model, quantity, and quality requirements, and collect historical production data of the factory, including product quality and production efficiency information under different casting parameters;

[0034] According to the formulated strategy, several pouring schemes are randomly generated by a random generation tool. Each pouring scheme includes pouring temperature, pouring speed and mold temperature, and the pouring temperature, pouring speed and mold temperature need to meet constraints.

[0035] An aluminum alloy casting production control system, comprising a pouring scheme generation module, a pouring scheme update module, an iterative output module, a pouring monitoring module, and a dynamic control module;

[0036] Casting scheme generation module: randomly generates several casting schemes based on the aluminum alloy product information that the factory needs to produce on the day and the factory's historical production data;

[0037] Casting scheme update module: first obtain the historical product unqualified rate of all casting schemes, filter out the casting schemes with product unqualified rate greater than the unqualified rate threshold, calculate the performance value of each remaining casting scheme through the performance function, select some casting schemes according to the performance value through the virtual roulette selection method, mutate the content of the selected casting schemes through differential operation, and establish a casting scheme table for all casting schemes before and after the mutation;

[0038] Iterative output module: repeat the steps of the pouring scheme update module for multiple iterations. When the iteration conditions are met, output several groups of pouring scheme tables, merge several groups of pouring scheme tables to obtain a scheme list, select the pouring scheme with the largest performance value in the scheme list and send it to the dynamic control module;

[0039] Pouring monitoring module: During the pouring process of aluminum alloy products according to the pouring plan, the solidification time and grain size of the aluminum alloy products after the current pouring are obtained;

[0040] Dynamic control module: Controls the factory's pouring process of the day based on the pouring plan, generates process assignments for aluminum alloy products after pre-processing the solidification time and grain size, dynamically adjusts the pouring plan based on the comparison results between the process assignments and the gradient threshold, and applies the dynamically adjusted pouring plan to the subsequent pouring production of aluminum alloy products.

[0041] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0042] 1. The present invention obtains the product defect index and production efficiency of each remaining casting scheme, calculates the product defect index and production efficiency through the performance function to obtain the performance value of the casting scheme, outputs several groups of casting scheme tables, merges several groups of casting scheme tables to obtain a scheme list, and the control system selects the casting scheme with the largest performance value in the scheme list to control the casting process of the factory on the day. During the casting of aluminum alloy products according to the casting scheme, the control system obtains the solidification time and grain size of the aluminum alloy product after the current casting, and generates the process assignment of the aluminum alloy product after pre-processing the solidification time and grain size. The casting scheme is dynamically adjusted according to the comparison result of the process assignment and the gradient threshold, and the dynamically adjusted casting scheme is applied to the subsequent casting production of aluminum alloy products. The control method first combines the aluminum alloy product information that the factory needs to produce on the day and the factory's historical production data to automatically find the global optimal casting scheme, and then dynamically controls the casting scheme according to real-time influencing factors during the production process of aluminum alloy products, thereby ensuring the quality of aluminum alloy products while improving the production efficiency of aluminum alloy products and reducing production costs;

[0043] 2. The present invention maximizes production efficiency and minimizes product defect index, and through comprehensive analysis of production efficiency and product defect index, can make the analysis of the casting scheme more comprehensive. Through comprehensive analysis, a trade-off and balance can be made between production efficiency and product quality, and an optimal casting scheme can be found, so that the production process can be carried out efficiently and product quality can be guaranteed;

[0044] 3. The present invention can automatically adjust the selection probability according to the size of individual fitness, and has strong adaptability. Individuals with higher fitness have a higher probability of being selected, which is conducive to retaining excellent individuals. It can adapt to different population sizes and different fitness distributions, and has high flexibility. Regardless of whether the population size is large or small, individual selection can be effectively performed. Since the selection process is based on probability, the roulette selection method helps to maintain the diversity of the population. Even individuals with lower fitness still have the possibility of being selected, which helps to maintain the diversity of the population and avoid premature convergence. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0046] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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 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 creative work are within the scope of protection of the present invention.

[0048] Example 1: Please refer to Figure 1 As shown, the aluminum alloy casting production control method described in this embodiment includes the following steps:

[0049] The control system randomly generates several casting schemes based on the information of aluminum alloy products that the factory needs to produce on that day and the historical production data of the factory. Each casting scheme includes casting temperature, casting speed and mold temperature. The historical product unqualified rate of all casting schemes is first obtained, and the casting schemes with product unqualified rate greater than the unqualified rate threshold are screened out. The product defect index and production efficiency of each remaining casting scheme are obtained. The performance value of the casting scheme is obtained by calculating the product defect index and production efficiency through the performance function. The larger the performance value, the better the comprehensive performance of the casting scheme. Some casting schemes are selected according to the performance value through the virtual roulette selection method, and the content of the selected casting schemes is mutated through the differential operation. The content of the scheme mutated through the differential operation is conducive to increasing the randomness of the casting scheme, thereby improving the control system. Robustness, avoid falling into the local optimal solution, establish a casting plan table for all casting plans before and after the mutation, repeat the above steps for multiple iterations, and generate a set of casting plan tables for each iteration. When the iteration conditions are met, output several groups of casting plan tables, merge several groups of casting plan tables to obtain a plan list, and the control system selects the casting plan with the largest performance value in the plan list to control the factory's casting process on the day. During the casting of aluminum alloy products according to the casting plan, the control system obtains the solidification time and grain size of the aluminum alloy products after the current casting, and generates the process assignment of the aluminum alloy products after pre-processing the solidification time and grain size. The casting plan is dynamically adjusted based on the comparison result of the process assignment and the gradient threshold, and the dynamically adjusted casting plan is applied to the subsequent casting production of aluminum alloy products.

[0050] This application obtains the product defect index and production efficiency of each remaining casting scheme, calculates the product defect index and production efficiency through the performance function to obtain the performance value of the casting scheme, outputs several groups of casting scheme tables, merges several groups of casting scheme tables to obtain a list of schemes, and the control system selects the casting scheme with the largest performance value in the list of schemes to control the casting process of the factory on the same day. During the casting process of the aluminum alloy product according to the casting scheme, the control system obtains the solidification time and grain size of the aluminum alloy product after the current casting, and generates the process assignment of the aluminum alloy product after pre-processing the solidification time and grain size. The casting scheme is dynamically adjusted according to the comparison result of the process assignment and the gradient threshold, and the dynamically adjusted casting scheme is applied to the subsequent casting production of aluminum alloy products. The control method first combines the aluminum alloy product information that the factory needs to produce on the same day and the factory's historical production data to automatically find the global optimal casting scheme, and then dynamically controls the casting scheme according to real-time influencing factors during the production process of the aluminum alloy product, while ensuring the quality of the aluminum alloy product, improving the production efficiency of the aluminum alloy product and reducing the production cost.

[0051] Embodiment 2: The control system randomly generates several casting plans based on the information of aluminum alloy products that the factory needs to produce on the day and the historical production data of the factory. Each casting plan includes casting temperature, casting speed and mold temperature, and includes the following steps:

[0052] Obtain information about aluminum alloy products that the factory needs to produce that day, including product model, quantity, quality requirements, etc. Collect historical production data of the factory, including product quality and production efficiency under different casting parameters;

[0053] According to the established strategy, a number of pouring schemes are randomly generated through the random generation tool. Each pouring scheme includes parameters such as pouring temperature, pouring speed, and mold temperature. When generating a pouring scheme, factors such as production demand, product quality requirements, and factory historical production data need to be considered to ensure that the generated scheme meets the actual production situation and quality requirements;

[0054] For example, the factory needs to produce aluminum alloy products on the same day, including 100 products of model A1 and 50 products of model B2. The quality requirement of product A1 is that the density is not less than 2.7g / cm 3 The quality requirement of product B2 is that the density is not less than 2.8g / cm 3 In addition, according to the factory's historical production data, the pouring temperature is between 650℃ and 750℃, the pouring speed is between 1m / s and 2m / s, and the mold temperature is between 200℃ and 300℃;

[0055] The casting temperature is between 650℃ and 750℃, the casting speed is between 1m / s and 2m / s, and the mold temperature is between 200℃ and 300℃ as constraints, that is, the casting temperature, casting speed and mold temperature in the casting plan need to meet the constraints;

[0056] Assume that according to the established strategy, three random pouring schemes are generated by the random generation tool as follows:

[0057] Casting scheme 1: casting temperature 700℃; casting speed 1.5m / s; mold temperature 250℃;

[0058] Casting scheme 2: casting temperature 680℃; casting speed 1.2m / s; mold temperature 220℃;

[0059] Casting scheme 3: casting temperature 720℃; casting speed 1.8m / s; mold temperature 280℃.

[0060] When generating these pouring plans, the product model, quantity and quality requirements, as well as the range of pouring parameters in the factory's historical production data are taken into account. The generated plan ensures that the pouring temperature, pouring speed and mold temperature are within a reasonable range and meet the product quality requirements.

[0061] Obtaining the historical product failure rates of all casting schemes and filtering out casting schemes whose product failure rates are greater than the failure rate threshold includes the following steps:

[0062] Obtain the number of all aluminum alloy products cast during the historical use of the casting scheme and the number of unqualified aluminum alloy products, and obtain the product unqualified rate by dividing the number of unqualified aluminum alloy products by the number of all aluminum alloy products cast. If the product unqualified rate is greater than the unqualified rate threshold, it indicates that the aluminum alloy casting yield of the casting scheme is low, so the casting scheme is not selected and needs to be screened out;

[0063] In the present application, after the aluminum alloy product is formed, the actual weight of the aluminum alloy product is measured, and the absolute value of the weight deviation is obtained by subtracting the standard weight from the actual weight. If the absolute value of the weight deviation is greater than the deviation threshold, it indicates that there are a large number of holes inside the aluminum alloy product or a large number of surface depressions. Therefore, the aluminum alloy product with an absolute value of the weight deviation greater than the deviation threshold is marked as an unqualified product, wherein the deviation threshold is equal to the standard weight multiplied by 5%. When the absolute value of the weight deviation is greater than the deviation threshold, it indicates that the aluminum alloy product is under-poured or over-poured. Under-poured means that the aluminum alloy product has holes or surface defects, and over-poured means that the aluminum alloy product is over-poured at the pouring port. First, it will lead to a waste of aluminum alloy resources. Second, it is necessary to polish the over-poured areas of the aluminum alloy product, thereby increasing the processing cost of the aluminum alloy product.

[0064] Obtain the product defect index and production efficiency of each remaining casting scheme, calculate the product defect index and production efficiency through the performance function to obtain the performance value of the casting scheme. The larger the performance value, the better the comprehensive performance of the casting scheme, including the following steps:

[0065] The logic for obtaining the product defect index is as follows: in the actual pouring process, when the absolute value of the weight deviation of the actual weight of the aluminum alloy product minus the standard weight is between 0 and the deviation threshold, although the aluminum alloy product cannot be judged as an unqualified product, the aluminum alloy product still has slight under-pouring or over-pouring. Therefore, all aluminum alloy products with absolute values ​​of weight deviation between 0 and the deviation threshold when the pouring scheme is used are obtained, and the absolute values ​​of the weight deviations of all aluminum alloy products are summed to obtain the product defect index, that is, the product defect index is expressed as the sum of the absolute values ​​of the weight deviations of all aluminum alloy products with slight under-pouring or over-pouring.

[0066] It should be noted that the present application obtains aluminum alloy products with slight under-casting or over-casting, because generally speaking, it is okay if the number of aluminum alloy products with slight under-casting or over-casting is small. If the number of aluminum alloy products with slight under-casting or over-casting is too large, it will still reduce customer satisfaction with the overall aluminum alloy products. Therefore, the present application considers the product defect index.

[0067] The logic for obtaining production efficiency is: obtain the number of aluminum alloy products cast by the casting scheme within the monitoring time period, and obtain the production efficiency by dividing the number of aluminum alloy products cast by the previous monitoring time. The higher the production efficiency, the more aluminum alloy products can be cast by the casting scheme per unit time, thereby improving the overall production efficiency of aluminum alloy products.

[0068] In summary, in this application, it is necessary to maximize the production efficiency and minimize the product defect index. By comprehensively analyzing the production efficiency and the product defect index, the analysis of the casting scheme can be more comprehensive. The performance value of the casting scheme is obtained by calculating the product defect index and the production efficiency through the performance function. The performance function expression is:

[0069] P_value=α*SCX-β*CQZ, where P_value is the performance value, SCX is the production efficiency, CQZ is the product defect index, α and β are the proportional coefficients of production efficiency and product defect index respectively, and α and β are both greater than 0. The larger the performance value, the better the comprehensive performance of the casting scheme.

[0070] This application maximizes production efficiency and minimizes product defect index. By comprehensively analyzing production efficiency and product defect index, the analysis of the casting scheme can be more comprehensive. Through comprehensive analysis, a trade-off and balance can be made between production efficiency and product quality to find an optimal casting scheme so that the production process can be carried out efficiently while ensuring product quality.

[0071] The virtual roulette selection method is used to select some casting solutions according to the performance value, including the following steps:

[0072] Obtain the performance values ​​of all casting schemes, add up the performance values ​​of all casting schemes to obtain the total performance value, obtain the round selection rate of each casting scheme by dividing the performance value by the total performance value, map each casting scheme to the corresponding sector area on the virtual roulette according to the round selection rate, the larger the round selection rate, the larger the sector area occupied by the casting scheme on the virtual roulette, after all casting schemes are mapped, drive the virtual roulette to rotate, when the virtual roulette stops rotating, point the virtual pointer to the casting scheme corresponding to the sector area, select it, delete the selected casting scheme sector area, and regenerate the sector area of ​​the remaining casting schemes, when the number of selected casting schemes is equal to the quantity threshold, the casting scheme selection is completed;

[0073] This application uses a virtual roulette selection method to select a casting scheme, which has the following effects:

[0074] It can automatically adjust the selection probability according to the size of individual fitness, and has strong adaptability. Individuals with higher fitness have a higher probability of being selected, which is conducive to retaining excellent individuals. It can adapt to different population sizes and different fitness distributions, and has high flexibility. Whether the population size is large or small, individual selection can be carried out effectively. Since the selection process is based on probability, the roulette selection method helps to maintain the diversity of the population. Even individuals with lower fitness still have the possibility of being selected, which helps to maintain the diversity of the population and avoid premature convergence.

[0075] The selected casting schemes are mutated through differential operation. The mutated casting scheme content is beneficial to increase the randomness of the casting scheme, thereby improving the robustness of the control system and avoiding falling into the local optimal solution. A casting scheme table is established for all casting schemes before and after the mutation, including the following steps:

[0076] By randomly selecting two different pouring schemes, the target pouring scheme is mutated to generate a variant pouring scheme. The mutation formula is: V i =X i +F*(X w -X g ), where V i is the i-th variant pouring scheme, X i is the i-th target pouring scheme, F is the scaling factor, X w , X g Two different pouring schemes were randomly selected for the random selection;

[0077] Assume that the selected partial pouring plan includes X 1 , X 2 , X 3 , X 1 , X 2 , X 3 The pouring schemes are:

[0078] X 1 =(A700℃, B250℃, C1.5m / s);

[0079] X 2 =(A680℃, B220℃, C1.2m / s);

[0080] X 3 =(A720℃, B280℃, C1.8m / s);

[0081] A, B, and C are pouring temperature, mold temperature, and pouring speed, respectively;

[0082] In this example, the scaling factor F is 0.5, then X 1 , X2 , X 3 Variations of the pouring scheme The pouring scheme can be:

[0083] V 1 =X 1 +0.5*(X 2 -X 3 )=(A680℃, B220℃, C1.2m / s);

[0084] V 2 =X 2 +0.5*(X 3 -X 1 )=(A690℃, B235℃, C1.35m / s);

[0085] V 3 =X 3 +0.5*(X 1 -X 2 )=(A730℃, B295℃, C1.95m / s);

[0086] Through differential mutation operation, the pouring scheme X 1 , X 2 , X 3 Variant casting schemes V were generated respectively 1 、V 2 、V 3 .

[0087] Repeat the selection and mutation steps for multiple iterations. Each iteration generates a set of pouring plan tables. When the iteration conditions are met, several sets of pouring plan tables are output. Several sets of pouring plan tables are merged to obtain a plan list. The control system selects the pouring plan with the largest performance value in the plan list to control the pouring process of the factory on that day, including the following steps:

[0088] Repeat the selection and mutation steps for multiple iterations, each of which generates a set of casting plan tables. When the number of iterations is equal to the number threshold, the iteration is completed and several sets of casting plan tables are output.

[0089] During the pouring process of the aluminum alloy product according to the pouring scheme, the control system obtains the solidification time and grain size of the aluminum alloy product after the current pouring, including the following steps:

[0090] The logic for obtaining the solidification time is: record the time when the aluminum alloy product starts pouring and the time when the solidification is completed, and obtain the solidification time by subtracting the time when the pouring starts from the time when the solidification is completed. Too fast or too slow solidification time will affect the quality of the aluminum alloy product, as follows:

[0091] 1) The effect of too fast solidification time is: when the solidification time is too fast, it means that the solidification time is short. In the faster solidification process, the unsolidified metal liquid may be trapped under the solidified surface, forming pores or holes;

[0092] 2) The effects of too slow solidification time are: long solidification time will increase the shrinkage of the material, easily produce internal shrinkage holes or surface shrinkage cracks, resulting in reduced product quality or even making the product unusable. Long solidification time will extend the production cycle, reduce production efficiency and increase production costs.

[0093] The logic for obtaining the grain size is as follows: After the aluminum alloy product solidifies, the grain size of the aluminum alloy product is obtained through a metallographic microscope. A grain size that is too large or too small will affect the quality of the aluminum alloy product, as follows:

[0094] 1) Large grain size: Large grain size will lead to a reduction in grain boundary area, thereby reducing the strength and toughness of the material. The grain boundary energy inside large grains is low, which easily leads to grain slip and yield, making the material easy to deform and fracture. Large grain size will increase the grain boundary length of the material, making local stress concentration and crack propagation at the grain boundary easier, reducing the fatigue resistance of aluminum alloy products. Large grain size will reduce the plastic deformation performance of the material, causing the material to be prone to brittle fracture, thereby affecting the processing and forming properties of aluminum alloy products.

[0095] 2) Small grain size: Small grain size will increase the grain boundary area, increase the dislocation and dislocation density at the grain boundary, thereby increasing the strength and hardness of the material, making the aluminum alloy product have better tensile and compressive properties. Although small grain size can improve the strength of the material, too small grains may cause the material's toughness to decrease, prone to brittle fracture, and affect the impact resistance and toughness of the aluminum alloy product. Small grain size will increase the dislocation and dislocation density at the grain boundary, making it easier to crack at the grain boundary, thereby reducing the material's fatigue resistance and affecting the service life of the aluminum alloy product.

[0096] To sum up, for aluminum alloy products, the solidification time should be within the standard solidification time range, and the grain size should also be within the standard grain size range, so as to ensure the quality of aluminum alloy products.

[0097] After pre-processing the solidification time and the grain size, a process assignment of the aluminum alloy product is generated, and the casting scheme is dynamically adjusted according to the comparison result between the process assignment and the gradient threshold, and the dynamically adjusted casting scheme is applied in the subsequent casting production of the aluminum alloy product, including the following steps:

[0098] The solidification time and grain size are preprocessed by a general normalization processing formula, so that the value range of the solidification time normalization value and the grain size normalization value obtained after the processing is mapped to [0,1], and the process assignment is obtained by adding the solidification time normalization value to the grain size normalization value;

[0099] The general normalization calculation formula is: In the formula, x norn is the normalized value of the parameter, x is the original parameter, max(x) is the maximum value of the original parameter, min(x) is the minimum value of the original parameter. The normalization method for solidification time and grain size is the same as above and will not be repeated here.

[0100] The gradient threshold includes a maximum limit threshold and a minimum limit threshold, and the minimum limit threshold is smaller than the maximum limit threshold;

[0101] If the process assignment is greater than the maximum limit threshold, it indicates that the solidification time is too slow, the grain size is too large, or the solidification time is slow and the grain size is large. The control system dynamically adjusts the pouring temperature and pouring speed in the pouring plan by reducing the current pouring temperature and pouring speed by 5%.

[0102] During the processing of aluminum alloy products, by lowering the pouring temperature, the transition speed from the liquid state to the solid state of the aluminum alloy can be slowed down and the solidification time can be prolonged. The main reason is the heat release during the solidification process. When the aluminum alloy changes from liquid to solid, a certain amount of heat will be released, and the heat release rate depends on the temperature difference of the solidification interface. A higher pouring temperature will lead to a smaller temperature difference at the solidification interface, thereby faster heat release and a relatively shorter solidification time; while a lower pouring temperature will increase the temperature difference at the solidification interface, slow down the heat release rate, and relatively prolong the solidification time.

[0103] For example, suppose there are two aluminum alloy products J1 and J2, and their pouring temperatures are 750℃ and 650℃ respectively. Under the same solidification conditions, the pouring temperature of product J1 is higher, resulting in a smaller temperature difference at the solidification interface and a faster heat release rate, so its solidification time is relatively short, while the pouring temperature of product J2 is lower, the temperature difference at the solidification interface is larger, and the heat release rate is slower, so its solidification time is relatively long. Therefore, by lowering the pouring temperature, the heat release rate of the aluminum alloy during the solidification process can be slowed down, and the solidification time can be relatively extended, which can provide more sufficient time for the formation of grains, which is beneficial to the control of grain size, thereby improving the quality and performance of aluminum alloy products.

[0104] If the process assignment is less than the minimum limit threshold, it indicates that the solidification time is too fast, the grain size is too small, or the solidification time is fast and the grain size is small. The control system dynamically adjusts the pouring temperature and pouring speed in the pouring plan by increasing the current pouring temperature and pouring speed by 5%.

[0105] If the process assignment is greater than or equal to the minimum limit threshold, and the process assignment is less than or equal to the maximum limit threshold, it indicates that the solidification time and grain size are within the standard range. The control system does not dynamically adjust the casting plan and continues to use the current casting plan to cast aluminum alloy products.

[0106] Embodiment 3: The aluminum alloy casting production control system described in this embodiment includes a pouring scheme generation module, a pouring scheme update module, an iterative output module, a pouring monitoring module, and a dynamic control module;

[0107] Casting scheme generation module: Combined with the aluminum alloy product information that the factory needs to produce on the day and the factory's historical production data, several casting schemes are randomly generated. Each casting scheme includes casting temperature, casting speed and mold temperature. Several casting schemes are sent to the casting scheme update module;

[0108] Casting scheme update module: first obtain the historical product unqualified rate of all casting schemes, screen out the casting schemes with product unqualified rate greater than the unqualified rate threshold, obtain the product defect index and production efficiency of each remaining casting scheme, calculate the product defect index and production efficiency through the performance function to get the performance value of the casting scheme. The larger the performance value, the better the comprehensive performance of the casting scheme. Select some casting schemes according to the performance value through the virtual roulette selection method, and perform scheme content mutation on the selected casting schemes through differential operation. Performing scheme content mutation through differential operation is conducive to increasing the randomness of the casting scheme, thereby improving the robustness and avoiding falling into the local optimal solution. A casting scheme table is established for all casting schemes before and after the mutation, and the casting scheme table is sent to the iterative output module;

[0109] Iterative output module: repeat the steps of the pouring scheme update module for multiple iterations. Each iteration generates a set of pouring scheme tables. When the iteration conditions are met, several sets of pouring scheme tables are output. Several sets of pouring scheme tables are merged to obtain a scheme list. The pouring scheme with the largest performance value in the scheme list is selected and sent to the dynamic control module.

[0110] Pouring monitoring module: During the pouring process of aluminum alloy products according to the pouring plan, the solidification time and grain size of the aluminum alloy products after the current pouring are obtained, and the solidification time and grain size are sent to the dynamic control module;

[0111] Dynamic control module: Controls the factory's pouring process of the day based on the pouring plan, generates process assignments for aluminum alloy products after pre-processing the solidification time and grain size, dynamically adjusts the pouring plan based on the comparison results between the process assignments and the gradient threshold, and applies the dynamically adjusted pouring plan to the subsequent pouring production of aluminum alloy products.

[0112] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0113] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0114] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0115] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0116] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for controlling aluminum alloy casting production, characterized in that: The control method comprises the following steps: S1: The control system randomly generates several casting schemes based on the information of aluminum alloy products that the factory needs to produce on the day and the historical production data of the factory. The historical product failure rates of all casting schemes are first obtained. After the casting schemes with product failure rates greater than the failure rate threshold are screened out, the product defect index and production efficiency of each remaining casting scheme are obtained. The performance value of the casting scheme is obtained by calculating the product defect index and production efficiency through the performance function; S2: Select some pouring schemes according to the performance values ​​by using a virtual roulette selection method, mutate the selected pouring schemes by using a differential operation, and establish a pouring scheme table for all pouring schemes before and after the mutation; S3: Repeat step S2 for multiple iterations. When the iteration conditions are met, output several groups of pouring plan tables, merge several groups of pouring plan tables to obtain a plan list, and the control system selects the pouring plan with the largest performance value in the plan list to control the pouring process of the factory on that day; S4: During the pouring process of the aluminum alloy product according to the pouring plan, the control system obtains the solidification time and grain size of the aluminum alloy product after the current pouring, generates the process assignment of the aluminum alloy product after pre-processing the solidification time and grain size, and dynamically adjusts the pouring plan according to the comparison result between the process assignment and the gradient threshold, and applies the dynamically adjusted pouring plan to the subsequent pouring production of the aluminum alloy product.

2. The aluminum alloy casting production control method according to claim 1, characterized in that: The performance value of the casting scheme is obtained by calculating the product defect index and production efficiency through the performance function. The performance function expression is: P_value=α*SCX-β*CQZ, where P_value is the performance value, SCX is the production efficiency, CQZ is the product defect index, α and β are the proportional coefficients of production efficiency and product defect index respectively, and both α and β are greater than 0.

3. The aluminum alloy casting production control method according to claim 2, characterized in that: The virtual roulette selection method is used to select some casting solutions according to the performance value, including the following steps: Obtain the performance values ​​of all casting schemes, add up the performance values ​​of all casting schemes to obtain the total performance value, and obtain the round selection rate of each casting scheme by dividing the performance value by the total performance value; Map each casting scheme to the corresponding sector area on the virtual wheel according to the round selection rate. The larger the round selection rate, the larger the sector area occupied by the casting scheme on the virtual wheel. After all the pouring schemes are mapped, the virtual wheel is driven to rotate. When the virtual wheel stops rotating, the virtual pointer is pointed to the pouring scheme corresponding to the sector area, the selected pouring scheme sector area is deleted, and the sector areas of the remaining pouring schemes are regenerated; When the number of selected casting schemes is equal to the quantity threshold, the casting scheme selection is completed.

4. The aluminum alloy casting production control method according to claim 3, characterized in that: The selected partial pouring schemes are subjected to scheme content variation through differential operation, including the following steps: By randomly selecting two different pouring schemes, the target pouring scheme is mutated to generate a variant pouring scheme. The mutation formula is: V i =X i +F*(X w -X g ), where V i is the i-th variant pouring scheme, X i is the i-th target pouring scheme, F is the scaling factor, X w , X g Two different pouring schemes were randomly selected for the random selection.

5. The aluminum alloy casting production control method according to claim 4, characterized in that: The logic for obtaining the solidification time is as follows: recording the pouring start time and solidification completion time of the aluminum alloy product, and obtaining the solidification time by subtracting the pouring start time from the solidification completion time; The logic for obtaining the grain size is: after the aluminum alloy product is solidified, the grain size of the aluminum alloy product is obtained through a metallographic microscope.

6. The aluminum alloy casting production control method according to claim 5, characterized in that: After pre-processing the solidification time and grain size, a process value of the aluminum alloy product is generated, and the casting scheme is dynamically adjusted according to the comparison result between the process value and the gradient threshold, including the following steps: The solidification time and grain size are preprocessed by a general normalization processing formula, so that the value range of the solidification time normalization value and the grain size normalization value obtained after the processing is mapped to [0,1], and the process assignment is obtained by adding the solidification time normalization value to the grain size normalization value; The gradient threshold includes a maximum limit threshold and a minimum limit threshold, and the minimum limit threshold is smaller than the maximum limit threshold; If the process assignment is greater than the maximum limit threshold, the control system will reduce the current pouring temperature and pouring speed by 5%; If the process assignment is less than the minimum limit threshold, the control system increases the current pouring temperature and pouring speed by 5 percent; If the process assignment is greater than or equal to the minimum limit threshold, and the process assignment is less than or equal to the maximum limit threshold, the control system does not dynamically adjust the casting plan.

7. The aluminum alloy casting production control method according to claim 6, characterized in that: The logic for obtaining the product defect index is as follows: in the actual pouring process, when the absolute value of the weight deviation of the actual weight of the aluminum alloy product minus the standard weight is between 0 and the deviation threshold, it indicates that the aluminum alloy product has a slight under-pouring or over-pouring, and all aluminum alloy products whose absolute values ​​of weight deviation are between 0 and the deviation threshold when the pouring scheme is used are obtained, and the absolute values ​​of the weight deviations of all aluminum alloy products are summed up to obtain the product defect index.

8. The aluminum alloy casting production control method according to claim 7, characterized in that: The logic for obtaining the production efficiency is: obtaining the number of aluminum alloy products cast within the monitoring time period of the casting scheme, and dividing the number of aluminum alloy products cast by the monitoring time to obtain the production efficiency.

9. The aluminum alloy casting production control method according to claim 8, characterized in that: The control system randomly generates several casting plans based on the aluminum alloy product information that the factory needs to produce on the day and the factory's historical production data, including the following steps: Obtain information about aluminum alloy products that the factory needs to produce that day, including product model, quantity, and quality requirements, and collect historical production data of the factory, including product quality and production efficiency information under different casting parameters; According to the formulated strategy, several pouring schemes are randomly generated by a random generation tool. Each pouring scheme includes pouring temperature, pouring speed and mold temperature, and the pouring temperature, pouring speed and mold temperature need to meet constraints.

10. An aluminum alloy casting production control system, used to implement the control method according to any one of claims 1 to 9, characterized in that: It includes pouring scheme generation module, pouring scheme update module, iterative output module, pouring monitoring module and dynamic control module; Casting scheme generation module: randomly generates several casting schemes based on the aluminum alloy product information that the factory needs to produce on the day and the factory's historical production data; Casting scheme update module: first obtain the historical product unqualified rate of all casting schemes, filter out the casting schemes with product unqualified rate greater than the unqualified rate threshold, calculate the performance value of each remaining casting scheme through the performance function, select some casting schemes according to the performance value through the virtual roulette selection method, mutate the content of the selected casting schemes through differential operation, and establish a casting scheme table for all casting schemes before and after the mutation; Iterative output module: repeat the steps of the pouring scheme update module for multiple iterations. When the iteration conditions are met, output several groups of pouring scheme tables, merge several groups of pouring scheme tables to obtain a scheme list, select the pouring scheme with the largest performance value in the scheme list and send it to the dynamic control module; Pouring monitoring module: During the pouring process of aluminum alloy products according to the pouring plan, the solidification time and grain size of the aluminum alloy products after the current pouring are obtained; Dynamic control module: Controls the factory's pouring process of the day based on the pouring plan, generates process assignments for aluminum alloy products after pre-processing the solidification time and grain size, dynamically adjusts the pouring plan based on the comparison results between the process assignments and the gradient threshold, and applies the dynamically adjusted pouring plan to the subsequent pouring production of aluminum alloy products.

Images