An order-driven magnesium powder mixing integrated management system and mixing method
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- TANGSHAN WEIHAO MAGNESIUM POWDER
- Filing Date
- 2026-03-05
- Publication Date
- 2026-06-19
Smart Images
Figure CN122243353A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of powder material production and processing technology, and specifically refers to an order-driven integrated management system and mixing method for magnesium powder. Background Technology
[0002] In current production practices, magnesium powder blending is typically done manually. The specific process involves sales instructing blending personnel to blend magnesium powder upon receiving an order. These personnel select several semi-finished raw material containers from inventory that may meet the order's blending requirements as candidate containers. Based on the particle size distribution records of these candidate containers, they make a preliminary judgment on the blending plan. If the blended product fails to meet the target particle size distribution, the finished product must be re-screened or subjected to laser particle size analysis according to the target particle size distribution. The container selection plan or mixing parameters are then adjusted accordingly, and the process is repeated until the blending ratio within the target particle size distribution meets the customer's requirements.
[0003] However, in actual production environments, this manual mixing method will encounter the following problems: (1) Low mixing success rate. There is often an inconsistency between the screening range division method used in the inventory semi-finished raw material barrels and the target particle size range of the order. For example, the target particle size range required by the order may be "+60 mesh, 60-90 mesh, 90-120 mesh, -120 mesh", while the screening particle size range of the inventory semi-finished raw material barrels may be "+70 mesh, 70-100 mesh, 100-140 mesh, -140 mesh", or more and finer range division methods may be used. Due to differences in interval boundaries, number of intervals, or coverage, existing technologies struggle to directly obtain granularity interval data consistent with the target interval of the order before mixing. This makes it difficult to directly use the granularity information of the inventory semi-finished raw material barrels for mixing decisions. Often, manual inference of existing granularity intervals and their corresponding proportions is required to select candidate barrels. These candidate barrels are then screened according to the granularity intervals required by the customer before mixing. The lack of unified data processing and decision-making basis easily leads to a low first-time hit rate for mixing schemes, requiring multiple adjustments to the combination or proportion parameters of semi-finished raw material barrels. This increases production cycle, rework costs, and raw material loss risks, thereby significantly reducing the reliability and effectiveness of mixing.
[0004] (2) Poor inventory management. Under the existing production model, the inventory of semi-finished raw material barrels usually only records their original screening range and corresponding content data. Moreover, the range division method of raw material barrels in different batches is not uniform, and the particle size distribution varies greatly. Due to the lack of a unified processing and standardized mapping mechanism for the above screening data and particle size detection data, a large amount of existing particle size information in the inventory is difficult to be effectively integrated and utilized before mixing. At the same time, the existing inventory management methods are mostly based on static ledgers or manual registration. Information such as the quantity, weight and particle size structure of the inventory raw material barrels is difficult to be updated and centrally displayed in a timely manner. Relevant personnel cannot grasp the overall status and availability of the inventory in real time, which leads to inventory management remaining more at the weight or quantity level. It is impossible to carry out refined management and reasonable scheduling of inventory resources from the perspective of particle size structure and order matching. This easily leads to long-term accumulation, repeated screening or unreasonable use of semi-finished raw material barrels, which further affects production efficiency and resource utilization.
[0005] (3) Decentralized production process. In existing technologies, sales order placement, mixed-mix execution, and shipment management are usually completed by different positions or systems, resulting in a decentralized business process. Order information, execution status, and inventory data mainly rely on manual transmission and communication for connection. This approach not only easily leads to information lag or omissions, but also makes it difficult to achieve real-time tracking and unified management of the order execution process. When orders are changed, canceled, or mixed-mix results do not meet requirements, repeated manual coordination is often required, further affecting the overall production cycle and order delivery efficiency. Summary of the Invention
[0006] To address the technical problems existing in the prior art, the present invention provides an order-driven integrated management system and method for magnesium powder blending, the technical solution of which is as follows: On the one hand, an order-driven integrated management system for magnesium powder blending is provided. The system includes: a sales module, a blending module, a shipping module, an order management module, an inventory management module, and a management module. Each module is connected to the same system platform through a network and operates collaboratively under a unified data structure and business logic. The sales module is used for order creation and management. It supports inputting the product specifications, target granularity range and their proportion requirements corresponding to the order, and submitting the order information to the order management module for unified storage and status initialization. The mixing module is used to perform mixing decision-related operations based on the order target granularity range and inventory semi-finished raw material barrel information after the order enters the mixing stage. These operations include: mixing prediction, candidate raw material barrel screening, and mixing scheme calculation. The shipping module is used to manage the shipment of orders after the mixing is completed, and supports order outbound confirmation, shipment status update and related information feedback to the order management module. The order management module is used to manage the execution status of orders at each stage of sales, mixing and delivery, with orders as the core object, so as to realize the whole process status tracking and process constraints of orders from creation, execution to completion; The inventory management module is used to centrally manage the inventory information of semi-finished raw material barrels and finished products, supports the maintenance of inventory quantity, weight and related attribute information, and updates the inventory status in real time according to the mixing and delivery operations during the order execution process. The management module is used to uniformly manage system users and their operating permissions. It supports the division of operating permissions for different roles in the sales, mixing, and shipping ends, preventing unauthorized operations from affecting the production process and inventory data, thereby improving the security and reliability of system operation. It also centrally monitors and statistically analyzes order execution status, inventory changes, and mixing execution results, providing managers with a global operational view and decision support, enabling factory management to grasp order progress, inventory structure, and production load in real time.
[0007] Optionally, the mixing prediction part executed by the mixing end module includes: The particle size range data for the semi-finished raw material barrels comes from two types of testing methods: vibrating screen sieving data and laser particle size analyzer data. Vibrating screen data uses mesh size as the particle size specification unit, and its test results reflect the retention or passing ratio of powder under different screen sizes, which can directly reflect the particle size distribution of the semi-finished raw material barrels within a specific mesh size range. Laser particle size analyzer data uses micrometers as the particle size specification unit, and its output results include particle size range and corresponding range percentage, cumulative percentage, and particle size characteristic parameter D value. However, the granularity requirements of customer orders are put forward in the form of mesh size ranges, and the required proportion range for each mesh size range is clearly specified; Therefore, a particle size range mapping method based on the collaborative processing of vibrating screen data and laser particle size analyzer data is used: the data of the semi-finished product vibrating screen and the corresponding laser particle size analyzer are fitted to obtain the mapping relationship between mesh size and micrometer. When the particle size range division of the inventory semi-finished product raw material barrel is inconsistent with the target particle size range of the order, the particle size ratio of the inventory semi-finished product raw material barrel in the target range of the order is predicted according to the mapping relationship.
[0008] Optionally, the particle size interval mapping method based on the collaborative processing of vibrating screen data and laser particle size analyzer data specifically includes: First, based on the proportion value corresponding to a certain mesh size in the test results of the vibrating screen, the proportion value corresponds to the cumulative percentage of the laser particle size analyzer. Then, starting from the coarse particle size end, we look back to find the particle size in micrometers corresponding to the cumulative percentage, and establish a "mesh size - micrometer" correspondence point. Since the vibrating screen can provide up to five effective screen intervals and corresponding ratios, several discrete mesh-micron correspondence points are obtained. Based on these correspondence points, the relationship between mesh and micron is fitted by a function to construct a mapping relationship from mesh to micron. Through the mapping relationship, the micron particle size corresponding to any mesh interval can be predicted, thereby predicting the ratio value corresponding to any mesh interval. Based on this, when the target particle size range of the order is inconsistent with the original screening range of the semi-finished raw material barrel, the corresponding micron is first determined according to the mesh range of the target range of the order through the mapping relationship. Then, the interval in which the micron is located is integrated in the particle size distribution curve of the laser particle size analyzer to obtain the cumulative percentage, which is the predicted proportion of the target mesh range of the order in the semi-finished raw material barrel.
[0009] Optionally, the candidate raw material barrel screening section performed by the mixing end module includes: The system allows for configurable filtering, enabling mixing personnel to filter inventory semi-finished raw material barrels according to preset conditions based on actual production needs, or to manually select raw material barrels that must participate in mixing. The filtered or selected raw material barrels are then included in the candidate library as the input set for subsequent barrel selection and scoring. In the candidate pool, first determine if there are any semi-finished raw material barrels whose granularity interval division can completely cover the target interval of the order; if there are no perfectly matching raw material barrels, then determine if there are any raw material barrels that simultaneously satisfy the first and last intervals of the target order; if still not found, then sequentially determine if there are any raw material barrels that only satisfy the first interval of the target order or only satisfy the last interval of the target order; if none of the above situations exist, then select the raw material barrel whose first and last intervals are closest to the first and last intervals of the target order. Through this step-by-step downgrading matching strategy, avoid mixing in raw material barrels with excessively different granular structures during the barrel selection stage, and provide a stable granularity basis for subsequent mixing; After completing the interval matching and screening, a comprehensive score is made based on the inventory attributes of the semi-finished raw material barrels. Raw material barrels with earlier production dates and larger remaining weights have higher priority in the score, so as to balance inventory turnover efficiency and production stability. Based on the interval matching results and inventory attribute scores, the candidate raw material barrels are uniformly sorted, and the sorting results are output.
[0010] Optionally, the mixing scheme calculation part executed by the mixing terminal module includes: By combining linear programming solutions with least squares approximate optimization, a hierarchical solution hybrid computation strategy is formed, which prioritizes outputting hybrid schemes that satisfy all constraints when a strictly feasible solution exists, and can still output approximate hybrid schemes with clear optimization objectives and constraint boundaries when a strictly feasible solution does not exist.
[0011] Optionally, the mixing scheme calculation part executed by the mixing terminal module specifically includes: 1) Construct a mathematical model for mixing calculation In the mixing calculation process, let the candidate raw material containers participating in the mixing be...
[0012] Let the required granularity range of the order be...
[0013] Define the following variables: : No. Mixing dosage for each raw material barrel; Total weight of the finished product required for mixing. ; : No. The mixing ratio of each raw material barrel; : No. Available inventory weight of each raw material barrel; : No. Each raw material bucket is within the particle size range Content ratio within; Order granularity range The required range of proportions; The finished product after mixing is within the particle size range The predicted proportion within is:
[0014] in, , and Use a consistent method of representing proportions; Under the constraint of a maximum number of buckets N_max, a combination search is performed on the candidate bucket set; for each combination, a feasible solution of the linear programming is solved first, and if a solution that strictly satisfies the constraints is obtained, it is output immediately; if no strictly feasible solution exists, the approximate solution with the smallest error among the combinations is selected as the output. 2) Solving rigorously feasible mixing schemes using linear programming When a mixing scheme that satisfies all constraints exists, a linear programming method is used to solve for a strictly feasible mixing ratio. The objective function of the linear programming model is set as a constant. ; The constraints are as follows; (1) Proportion normalization constraint
[0015] (2) Inventory constraints
[0016] (3) Particle size range ratio constraint
[0017] The linear programming model is used to solve strictly feasible mixing ratio schemes under the conditions of satisfying the ratio normalization constraint, inventory limit constraint and granularity range ratio constraint. When the linear programming model has a feasible solution, the obtained mixing ratio can simultaneously satisfy all granularity range constraints and raw material inventory constraints, and can be directly output as the final mixing scheme and used for actual production control. 3) Approximate mixing optimization model based on least squares criterion When the linear programming model has no feasible solution, it indicates that under the current candidate raw material barrel combination and inventory conditions, it is impossible to simultaneously satisfy the proportion constraints of all granularity intervals. Therefore, an approximate mixing optimization model based on the least squares criterion is further adopted to find the approximate mixing scheme with the minimum error under the premise of satisfying the basic physical constraints. Define granularity range The target center ratio is: ; Construct the following least squares optimization objective function: ; in, The weighting coefficients are for different granularity ranges. It is used to reflect the relative importance of different particle size ranges in specific production needs, and is set according to process requirements or empirical rules; The constraints are: , ; The approximate mixing optimization model based on the least squares criterion can minimize the overall deviation between the mixing result and the target granularity ratio, even when a strictly feasible solution cannot be obtained, provided that basic physical conditions such as ratio normalization and inventory constraints are met. In the least squares approximation stage, the granularity interval ratio constraint is no longer a hard constraint that must be met. Instead, it is achieved by minimizing the ratio between the mixing result and the target center. The deviation between the two is used to make the approximate mixture as close as possible. During the solution process, stabilization measures are combined to avoid extreme distribution of the mixing ratio, thereby improving the stability and acceptability of the approximate mixing scheme in practical engineering applications. When there are multiple approximate solutions for candidate bucket combinations, the objective function values of each approximate solution are compared, and the approximate mixing scheme with the smallest error is selected as the final output.
[0018] Optionally, when faced with an abnormal situation, the system automatically identifies the abnormal situation and provides feedback to the operator through interface prompts regarding insufficient inventory or mismatched specifications. Meanwhile, even under these abnormal circumstances, the system still provides a roughly feasible mixing solution based on the current inventory conditions to assist production personnel in making decisions or further adjusting production plans. In addition, during the order parameter input stage, if the operator fails to input certain key parameters, the system will automatically trigger a verification mechanism and prompt the system through a pop-up window indicating that the relevant information is missing.
[0019] On the other hand, a method for blending magnesium powder using the order-driven integrated magnesium powder blending management system is provided, comprising three parts: blending prediction, candidate raw material barrel screening, and blending scheme calculation.
[0020] The beneficial effects of the technical solution provided by this invention include at least the following: (1) This invention achieves a unified conversion between inventory semi-finished product particle size data and order target particle size range, improving the reliability of blending decisions. This invention establishes a mapping relationship between mesh size and micrometer by co-processing existing vibrating screen data and laser particle size analyzer data in semi-finished product raw material barrels, and predicts the corresponding proportion of any order target mesh size range in semi-finished product raw material barrels based on this mapping relationship. This enables inventory particle size data that was originally inconsistent in its range division and could not be used directly to be transformed into decision input consistent with the order target range before blending, fundamentally solving the problem in the prior art where blending decisions rely on experience due to mismatched particle size ranges.
[0021] (2) This invention transforms the mixing result from "post-event verification" to "pre-event prediction", significantly improving the first-time success rate of mixing. Before mixing, this invention can predict the proportion of the target range of the order based on the particle size distribution of the inventory semi-finished raw material barrels, thereby judging the feasibility of the mixing scheme in advance during the barrel selection and proportioning stage. This avoids the repeated adjustment problems caused by screening or testing after mixing in traditional processes, effectively reducing the number of rework, raw material waste and the risk of extended production cycle.
[0022] (3) This invention improves the stability and interpretability of the barrel selection process through a graded barrel selection and comprehensive scoring mechanism. In the barrel selection process, this invention adopts a progressively degraded interval matching strategy and combines inventory attributes such as production date and remaining weight to comprehensively score and sort the semi-finished raw material barrels. This can effectively prevent raw material barrels with excessively different particle size structures from being selected into the blending scheme at the same time, reduce the risk of blending instability, and at the same time make the barrel selection logic have clear rules and traceable basis, reducing the reliance on human experience.
[0023] (4) This invention ensures the flexibility of mixing while controlling the complexity of mixing and improving production controllability. Based on the actual production and operation experience of the factory, this invention limits the number of semi-finished raw material barrels involved in mixing to no more than three, and calculates the mixing ratio based on linear programming and least squares methods. Under the premise of meeting the constraints of the target granularity range of the order, the optimal or near-optimal mixing scheme is obtained, which effectively balances the flexibility of mixing and the complexity of production operation and reduces the risk of weighing, feeding and operation errors.
[0024] (5) This invention realizes order-driven collaborative management of processes, improving production and delivery efficiency. This invention constructs an order-driven integrated factory process management system, which integrates sales ordering, mixed-mixing execution and delivery management into a unified platform, realizing centralized management and real-time synchronization of order status, inventory data and execution process, reducing the risk of information delay and omission caused by manual communication, and improving the transparency of order execution and overall delivery efficiency.
[0025] (6) This invention provides auxiliary decision-making capabilities in situations of limited inventory or mismatched specifications, thereby enhancing the robustness of the production system. In abnormal situations such as insufficient quantity of raw material barrels in stock or particle size specifications that cannot fully meet the order target requirements, this invention can automatically identify the abnormal situation and provide prompts. At the same time, it provides an approximately feasible mixing scheme based on the existing inventory conditions, providing decision-making reference for production personnel and enhancing the system's adaptability under complex production conditions. Attached Figure Description
[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0027] Figure 1 This is an overall block diagram of an order-driven integrated management system for magnesium powder blending provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of laser particle size analyzer data provided in an embodiment of the present invention; Figure 3 This is the mixing input interface provided in the embodiments of the present invention; Figure 4 This is the mixing result interface provided in the embodiments of the present invention; Figure 5 This is the interface for indicating abnormal total mixing volume provided in this embodiment of the invention; Figure 6 This is the interface for prompting abnormal situations in the total proportion provided in the embodiments of the present invention. Detailed Implementation
[0028] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.
[0029] like Figure 1 As shown in the figure, this embodiment of the invention provides an order-driven integrated management system for magnesium powder blending, the system comprising: The sales module, mixing module, shipping module, order management module, inventory management module, and management module are all connected to the same system platform via a network and operate collaboratively under a unified data structure and business logic. The sales module is used for order creation and management. It supports inputting the product specifications, target granularity range and their proportion requirements corresponding to the order, and submitting the order information to the order management module for unified storage and status initialization. The mixing module is used to perform mixing decision-related operations based on the order target granularity range and inventory semi-finished raw material barrel information after the order enters the mixing stage. These operations include: mixing prediction, candidate raw material barrel screening, and mixing scheme calculation. The shipping module is used to manage the shipment of orders after the mixing is completed, and supports order outbound confirmation, shipment status update and related information feedback to the order management module. The order management module is used to manage the execution status of orders at each stage of sales, mixing and delivery, with orders as the core object, so as to realize the whole process status tracking and process constraints of orders from creation, execution to completion; The inventory management module is used to centrally manage the inventory information of semi-finished raw material barrels and finished products, supports the maintenance of inventory quantity, weight and related attribute information, and updates the inventory status in real time according to the mixing and delivery operations during the order execution process. The management module is used to uniformly manage system users and their operating permissions. It supports the division of operating permissions for different roles in the sales, mixing, and shipping ends, preventing unauthorized operations from affecting the production process and inventory data, thereby improving the security and reliability of system operation. It also centrally monitors and statistically analyzes order execution status, inventory changes, and mixing execution results, providing managers with a global operational view and decision support, enabling factory management to grasp order progress, inventory structure, and production load in real time.
[0030] Through the above modular design, the embodiments of the present invention realize the process collaboration and data sharing of key business links such as sales order placement, mixed batch execution and delivery management under the same system platform, so that order information, execution status and inventory data can be kept consistent and real-time.
[0031] This invention addresses the issue of inconsistencies between the particle size range defined in inventory semi-finished raw material barrels and the target particle size range in orders. It proposes a method for processing data from a semi-finished product vibrating screen (as shown in Table 1) and the corresponding laser particle size analyzer data (as shown in Table 2). Figure 2 The method of fitting is used to obtain the mapping relationship between mesh size and micrometer. When the particle size range of the inventory semi-finished raw material barrel is inconsistent with the target particle size range of the order, the particle size ratio corresponding to the target range of the order can be predicted according to the mapping relationship, thus solving the problem of inconsistency between the mesh size range of the inventory semi-finished raw material barrel and the target particle size range of the order.
[0032] Table 1 Vibrating Screen Data
[0033] Optionally, the mixing prediction part executed by the mixing end module includes: The particle size range data for the semi-finished raw material barrels comes from two types of testing methods: vibrating screen sieving data and laser particle size analyzer data. Vibrating screen data uses mesh size as the particle size unit, and its results reflect the proportion of powder retained or passing through different screen sizes (as shown in Table 1, +20 mesh corresponds to a value of 3.3, indicating that 3.3% of the powder remains on a 20-mesh screen; +40 mesh corresponds to 27.1, indicating that 27.1% of the powder remains on a 40-mesh screen; -100 mesh corresponds to 0.6, indicating that 0.6% of the powder passes through a 100-mesh screen). This directly reflects the particle size distribution of the semi-finished raw material barrels within a specific mesh size range. Laser particle size analyzer data uses micrometers (μm) as the particle size unit, and its output includes the particle size range and the corresponding range percentage, cumulative percentage, and the particle size characteristic parameter D value. The D value is used to characterize the statistical properties of powder particle size distribution. For example, D50 = 280.7 μm means that 50% of the particles in the semi-finished raw material container have a particle size smaller than 280.7 μm. The particle size range and the corresponding range percentage and cumulative percentage reflect the distribution of powder content within each micron particle size range, providing more refined particle size distribution information.
[0034] However, the two types of data mentioned above have significant differences and limitations in practical applications. On the one hand, vibrating screens use "mesh" as the unit, and due to the equipment structure, a maximum of five screens can usually be placed in one screening, corresponding to a maximum of six particle size intervals, which is relatively coarse. On the other hand, although laser particle size analyzers can provide fine distribution data for particle size intervals of up to tens or even hundreds of micrometers, their unit is micrometers, which is inconsistent with the mesh unit of vibrating screens, and there is no unified and clear correspondence between mesh and micrometers in the existing technology.
[0035] However, the granularity requirements of customer orders are put forward in the form of mesh size ranges, and the required proportion range for each mesh size range is clearly specified; Therefore, how to effectively convert the micron-sized particle size distribution information obtained by the laser particle size analyzer into a mesh size range ratio consistent with the order target has become a key technical problem for realizing mixing prediction and barrel selection decision-making.
[0036] Therefore, a particle size range mapping method based on the collaborative processing of vibrating screen data and laser particle size analyzer data is used: the data of the semi-finished product vibrating screen and the corresponding laser particle size analyzer are fitted to obtain the mapping relationship between mesh size and micrometer. When the particle size range division of the inventory semi-finished product raw material barrel is inconsistent with the target particle size range of the order, the particle size ratio of the inventory semi-finished product raw material barrel in the target range of the order is predicted according to the mapping relationship.
[0037] Optionally, the particle size interval mapping method based on the collaborative processing of vibrating screen data and laser particle size analyzer data specifically includes: First, based on the proportion value corresponding to a certain mesh size in the test results of the vibrating screen (for example, +20 mesh corresponds to 3.3%), the proportion value corresponds to the cumulative percentage of the laser particle size analyzer. Then, starting from the coarse particle size end, we look back to find the particle size in micrometers corresponding to the cumulative percentage to establish a "mesh size - micrometer" correspondence point. Since the vibrating screen can provide up to five effective screen intervals and corresponding ratios, several discrete mesh-micron correspondence points are obtained. Based on these correspondence points, the relationship between mesh and micron is fitted by a function to construct a mapping relationship from mesh to micron. Through the mapping relationship, the micron particle size corresponding to any mesh interval can be predicted, thereby predicting the ratio value corresponding to any mesh interval. Based on this, when the target particle size range of the order is inconsistent with the original screening range of the semi-finished raw material barrel, the corresponding micron is first determined according to the mesh range of the target range of the order through the mapping relationship. Then, the interval in which the micron is located is integrated in the particle size distribution curve of the laser particle size analyzer to obtain the cumulative percentage, which is the predicted proportion of the target mesh range of the order in the semi-finished raw material barrel.
[0038] Based on this prediction result, the embodiments of the present invention can make decisions on the selection and proportion of raw material barrels in the inventory before mixing, providing reliable data for subsequent mixing and realizing the pre-prediction and optimization of the mixing process.
[0039] Optionally, the candidate raw material barrel screening section performed by the mixing end module includes: The system allows for configurable filtering, enabling mixing personnel to filter inventory semi-finished raw material barrels according to preset conditions (such as production date, powder specifications, remaining weight in the barrel, etc.) based on actual production needs, or to manually select raw material barrels that must participate in mixing. The filtered or selected raw material barrels are then included in the candidate library as the input set for subsequent barrel selection and scoring. In the candidate pool, first determine if there are any semi-finished raw material barrels whose granularity interval division can completely cover the target interval of the order; if there are no perfectly matching raw material barrels, then determine if there are any raw material barrels that simultaneously satisfy the first and last intervals of the target order; if still not found, then sequentially determine if there are any raw material barrels that only satisfy the first interval of the target order or only satisfy the last interval of the target order; if none of the above situations exist, then select the raw material barrel whose first and last intervals are closest to the first and last intervals of the target order. Through this step-by-step downgrading matching strategy, avoid mixing in raw material barrels with excessively different granular structures during the barrel selection stage, and provide a stable granularity basis for subsequent mixing; After completing the interval matching and screening, a comprehensive score is made based on the inventory attributes of the semi-finished raw material barrels (such as production date and remaining weight). Raw material barrels with earlier production dates and larger remaining weights have higher priority in the scoring, so as to balance inventory turnover efficiency and production stability. Based on the interval matching results and inventory attribute scores, the candidate raw material barrels are uniformly sorted, and the sorting results are output.
[0040] Optionally, the mixing scheme calculation part executed by the mixing terminal module includes: By combining linear programming with least squares approximate optimization, a hierarchical solution strategy for mixing is formed. This strategy prioritizes outputting mixing schemes that satisfy all constraints when a strictly feasible solution exists. Even when a strictly feasible solution does not exist, it can still output approximate mixing schemes with clear optimization objectives and constraint boundaries. This avoids the problem of mixing calculations being unable to proceed due to constraint conflicts, and significantly improves the robustness, continuity, and engineering applicability of mixing decisions under complex inventory conditions and multiple constraints.
[0041] Optionally, the mixing scheme calculation part executed by the mixing terminal module specifically includes: 1) Construct a mathematical model for mixing calculation In the mixing calculation process, let the candidate raw material containers participating in the mixing be...
[0042] Let the required granularity range of the order be...
[0043] Define the following variables: : No. Mixing amount (kg) for each raw material barrel; Total weight of the finished product (kg) required. ; : No. The mixing ratio of each raw material barrel; : No. Available inventory weight (kg) of each raw material drum; : No. Each raw material bucket is within the particle size range Content ratio within; Order granularity range The required range of proportions; The finished product after mixing is within the particle size range The predicted proportion within is:
[0044] in, , and Use a consistent method of representing proportions; Under the constraint of a maximum number of buckets N_max, a combination search is performed on the candidate bucket set; for each combination, a feasible solution of the linear programming is solved first, and if a solution that strictly satisfies the constraints is obtained, it is output immediately; if no strictly feasible solution exists, the approximate solution with the smallest error among the combinations is selected as the output. 2) Solving rigorously feasible mixing schemes using linear programming When a mixing scheme that satisfies all constraints exists, a linear programming method is used to solve for a strictly feasible mixing ratio. The objective function of the linear programming model is set as a constant. ; The constraints are as follows; (1) Proportion normalization constraint
[0045] (2) Inventory constraints
[0046] (3) Particle size range ratio constraint
[0047] The linear programming model is used to solve strictly feasible mixing ratio schemes under the conditions of satisfying the ratio normalization constraint, inventory limit constraint and granularity range ratio constraint. When the linear programming model has a feasible solution, the obtained mixing ratio can simultaneously satisfy all granularity range constraints and raw material inventory constraints, and can be directly output as the final mixing scheme and used for actual production control. 3) Approximate mixing optimization model based on least squares criterion When the linear programming model has no feasible solution, it indicates that under the current candidate raw material barrel combination and inventory conditions, it is impossible to simultaneously satisfy the proportion constraints of all granularity intervals. Therefore, an approximate mixing optimization model based on the least squares criterion is further adopted to find the approximate mixing scheme with the minimum error under the premise of satisfying the basic physical constraints. Define granularity range The target center ratio is: ; Construct the following least squares optimization objective function: ; in, The weighting coefficients are for different granularity ranges. It is used to reflect the relative importance of different particle size ranges in specific production needs, and is set according to process requirements or empirical rules; The constraints are: , ; The approximate mixing optimization model based on the least squares criterion can minimize the overall deviation between the mixing result and the target granularity ratio, even when a strictly feasible solution cannot be obtained, provided that basic physical conditions such as ratio normalization and inventory constraints are met. In the least squares approximation stage, the granularity interval ratio constraint is no longer a hard constraint that must be met. Instead, it is achieved by minimizing the ratio between the mixing result and the target center. The deviation between the two is used to make the approximate mixture as close as possible. During the solution process, stabilization measures are combined to avoid extreme distribution of the mixing ratio, thereby improving the stability and acceptability of the approximate mixing scheme in practical engineering applications. When there are multiple approximate solutions for candidate bucket combinations, the objective function values of each approximate solution are compared, and the approximate mixing scheme with the smallest error is selected as the final output.
[0048] Therefore, the overall process of the mixing end in this embodiment of the invention is as follows: the mixing personnel input the order target requirements, such as... Figure 3 The mixing input interface shown yields the mixing scheme, as follows: Figure 4 The interface showing the mixing results is shown.
[0049] Optionally, when faced with abnormal situations (such as insufficient quantity of semi-finished raw material barrels in stock, or the particle size of existing raw material barrels being unable to meet the target particle size range requirements of the order), the system automatically identifies the abnormal situation and provides feedback to the operator through interface prompts regarding insufficient inventory or mismatched specifications. Meanwhile, even under these abnormal circumstances, the system still provides a roughly feasible mixing solution based on the current inventory conditions to assist production personnel in making decisions or further adjusting production plans. Furthermore, during the order parameter input phase, if the operator fails to input certain key parameters (such as total mixing quantity, target granularity range, or target ratio), the system will automatically trigger a verification mechanism and display a pop-up message indicating the missing information (e.g., the total mixing quantity was not entered). Figure 5 The pop-up window shown indicates that when the entered target proportion does not meet reasonable constraints, such as the sum of the proportions in each target interval not being equal to or exceeding 100%, the system will also provide a prompt through the interface pop-up window, such as... Figure 6 As shown, invalid orders are prevented from entering the mixing calculation process, thereby ensuring the effectiveness of the mixing decision and the reliability of the calculation results.
[0050] This invention also provides a method for blending magnesium powder using the order-driven integrated magnesium powder blending management system, comprising three parts: blending prediction, candidate raw material barrel screening, and blending scheme calculation.
[0051] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. An order-driven integrated management system for magnesium powder blending, characterized in that, The system includes: a sales module, a mixing module, a shipping module, an order management module, an inventory management module, and a management module. Each module is connected to the same system platform via a network and operates collaboratively under a unified data structure and business logic. The sales module is used for order creation and management. It supports inputting the product specifications, target granularity range and their proportion requirements corresponding to the order, and submitting the order information to the order management module for unified storage and status initialization. The mixing module is used to perform mixing decision-related operations based on the order target granularity range and inventory semi-finished raw material barrel information after the order enters the mixing stage. These operations include: mixing prediction, candidate raw material barrel screening, and mixing scheme calculation. The shipping module is used to manage the shipment of orders after the mixing is completed, and supports order outbound confirmation, shipment status update and related information feedback to the order management module. The order management module is used to manage the execution status of orders at each stage of sales, mixing and delivery, with orders as the core object, so as to realize the whole process status tracking and process constraints of orders from creation, execution to completion; The inventory management module is used to centrally manage the inventory information of semi-finished raw material barrels and finished products, supports the maintenance of inventory quantity, weight and related attribute information, and updates the inventory status in real time according to the mixing and delivery operations during the order execution process. The management module is used to uniformly manage system users and their operating permissions. It supports the division of operating permissions for different roles in the sales, mixing, and shipping ends, preventing unauthorized operations from affecting the production process and inventory data, thereby improving the security and reliability of system operation. It also centrally monitors and statistically analyzes order execution status, inventory changes, and mixing execution results, providing managers with a global operational view and decision support, enabling factory management to grasp order progress, inventory structure, and production load in real time.
2. The system according to claim 1, characterized in that, The mixing prediction component executed by the mixing module includes: The particle size range data for the semi-finished raw material barrels comes from two types of testing methods: vibrating screen sieving data and laser particle size analyzer data. Vibrating screen data uses mesh size as the particle size specification unit, and its test results reflect the retention or passing ratio of powder under different screen sizes, which can directly reflect the particle size distribution of the semi-finished raw material barrels within a specific mesh size range. Laser particle size analyzer data uses micrometers as the particle size specification unit, and its output results include particle size range and corresponding range percentage, cumulative percentage, and particle size characteristic parameter D value. However, the granularity requirements of customer orders are put forward in the form of mesh size ranges, and the required proportion range for each mesh size range is clearly specified; Therefore, a particle size range mapping method based on the collaborative processing of vibrating screen data and laser particle size analyzer data is used: the data of the semi-finished product vibrating screen and the corresponding laser particle size analyzer are fitted to obtain the mapping relationship between mesh size and micrometer. When the particle size range division of the inventory semi-finished product raw material barrel is inconsistent with the target particle size range of the order, the particle size ratio of the inventory semi-finished product raw material barrel in the target range of the order is predicted according to the mapping relationship.
3. The system according to claim 2, characterized in that, The particle size interval mapping method based on the collaborative processing of vibrating screen data and laser particle size analyzer data specifically includes: First, based on the proportion value corresponding to a certain mesh size in the test results of the vibrating screen, which corresponds to the cumulative percentage of the laser particle size analyzer, we look back from the coarse particle size end to find the particle size in micrometers corresponding to the cumulative percentage, and establish a "mesh size - micrometer" correspondence point. Since the vibrating screen can provide up to five effective screen intervals and corresponding ratios, several discrete mesh-micron correspondence points are obtained. Based on these correspondence points, the relationship between mesh and micron is fitted by a function to construct a mapping relationship from mesh to micron. Through the mapping relationship, the micron particle size corresponding to any mesh interval can be predicted, thereby predicting the ratio value corresponding to any mesh interval. Based on this, when the target particle size range of the order is inconsistent with the original screening range of the semi-finished raw material barrel, the corresponding micron is first determined according to the mesh range of the target range of the order through the mapping relationship. Then, the interval in which the micron is located is integrated in the particle size distribution curve of the laser particle size analyzer to obtain the cumulative percentage, which is the predicted proportion of the target mesh range of the order in the semi-finished raw material barrel.
4. The system according to claim 1, characterized in that, The candidate raw material bin screening process performed by the mixing module includes: The system allows for configurable filtering, enabling mixing personnel to filter inventory semi-finished raw material barrels according to preset conditions based on actual production needs, or to manually select raw material barrels that must participate in mixing. The filtered or selected raw material barrels are then included in the candidate library as the input set for subsequent barrel selection and scoring. In the candidate pool, first determine if there are any semi-finished raw material barrels whose granularity interval division can completely cover the target interval of the order; if there are no perfectly matching raw material barrels, then determine if there are any raw material barrels that simultaneously satisfy the first and last intervals of the target order; if still not found, then sequentially determine if there are any raw material barrels that only satisfy the first interval of the target order or only satisfy the last interval of the target order; if none of the above situations exist, then select the raw material barrel whose first and last intervals are closest to the first and last intervals of the target order. Through this step-by-step downgrading matching strategy, avoid mixing in raw material barrels with excessively different granular structures during the barrel selection stage, and provide a stable granularity basis for subsequent mixing; After completing the interval matching and screening, a comprehensive score is made based on the inventory attributes of the semi-finished raw material barrels. Raw material barrels with earlier production dates and larger remaining weights have higher priority in the score, so as to balance inventory turnover efficiency and production stability. Based on the interval matching results and inventory attribute scores, the candidate raw material barrels are uniformly sorted, and the sorting results are output.
5. The system according to claim 1, characterized in that, The mixing scheme calculation part executed by the mixing terminal module includes: By combining linear programming solutions with least squares approximate optimization, a hierarchical solution hybrid computation strategy is formed, which prioritizes outputting hybrid schemes that satisfy all constraints when a strictly feasible solution exists, and can still output approximate hybrid schemes with clear optimization objectives and constraint boundaries when a strictly feasible solution does not exist.
6. The system according to claim 5, characterized in that, The mixing scheme calculation part executed by the mixing terminal module specifically includes: 1) Construct a mathematical model for mixing calculation In the mixing calculation process, let the candidate raw material containers participating in the mixing be... Let the required granularity range of the order be... Define the following variables: : No. Mixing dosage for each raw material barrel; Total weight of the finished product required for mixing. ; : No. The mixing ratio of each raw material barrel; : No. Available inventory weight of each raw material barrel; : No. Each raw material bucket is within the particle size range Content ratio within; Order granularity range The required range of proportions; The finished product after mixing is within the particle size range The predicted proportion within is: in, , and Use a consistent method of representing proportions; Under the constraint of a maximum number of buckets N_max, a combination search is performed on the candidate bucket set; for each combination, a feasible solution of the linear programming is solved first, and if a solution that strictly satisfies the constraints is obtained, it is output immediately; if no strictly feasible solution exists, the approximate solution with the smallest error among the combinations is selected as the output. 2) Solving rigorously feasible mixing schemes using linear programming When a mixing scheme that satisfies all constraints exists, a linear programming method is used to solve for a strictly feasible mixing ratio. The objective function of the linear programming model is set as a constant. ; The constraints are as follows; (1) Proportion normalization constraint (2) Inventory constraints (3) Particle size range ratio constraint The linear programming model is used to solve strictly feasible mixing ratio schemes under the conditions of satisfying the ratio normalization constraint, inventory limit constraint and granularity range ratio constraint. When the linear programming model has a feasible solution, the obtained mixing ratio can simultaneously satisfy all granularity range constraints and raw material inventory constraints, and can be directly output as the final mixing scheme and used for actual production control. 3) Approximate mixing optimization model based on least squares criterion When the linear programming model has no feasible solution, it indicates that under the current candidate raw material barrel combination and inventory conditions, it is impossible to simultaneously satisfy the proportion constraints of all granularity intervals. Therefore, an approximate mixing optimization model based on the least squares criterion is further adopted to find the approximate mixing scheme with the minimum error under the premise of satisfying the basic physical constraints. Define granularity range The target center ratio is: ; Construct the following least squares optimization objective function: ； in, The weighting coefficients are for different granularity ranges. It is used to reflect the relative importance of different particle size ranges in specific production needs, and is set according to process requirements or empirical rules; The constraints are: , ; The approximate mixing optimization model based on the least squares criterion can minimize the overall deviation between the mixing result and the target granularity ratio, even when a strictly feasible solution cannot be obtained, provided that basic physical conditions such as ratio normalization and inventory constraints are met. In the least squares approximation stage, the granularity interval ratio constraint is no longer a hard constraint that must be met. Instead, it is achieved by minimizing the ratio between the mixing result and the target center. The deviation between the two is used to make the approximate mixture as close as possible. During the solution process, stabilization measures are combined to avoid extreme distribution of the mixing ratio, thereby improving the stability and acceptability of the approximate mixing scheme in practical engineering applications. When there are multiple approximate solutions for candidate bucket combinations, the objective function values of each approximate solution are compared, and the approximate mixing scheme with the smallest error is selected as the final output.
7. The system according to claim 1, characterized in that, When faced with an abnormal situation, the system automatically identifies the abnormal situation and provides feedback to the operator through interface prompts, indicating that the inventory is insufficient or the specifications do not match. Meanwhile, even under these abnormal circumstances, the system still provides a roughly feasible mixing solution based on the current inventory conditions to assist production personnel in making decisions or further adjusting production plans. In addition, during the order parameter input stage, if the operator fails to input certain key parameters, the system will automatically trigger a verification mechanism and prompt the system through a pop-up window indicating that the relevant information is missing.
8. A method for blending magnesium powder using the order-driven integrated magnesium powder blending management system described in any one of claims 1-7, characterized in that, include: The system consists of three parts: blending prediction, candidate raw material barrel screening, and blending scheme calculation.