Method and system for analyzing data of purification of aluminum-silicon eutectoid bath
By constructing a data analysis model that is associated with the purification process stages of the eutectoid cell, the purification data in the aluminum-silicon alloy electrolysis production process is collected and segmented in real time. This solves the problem that existing technologies cannot effectively associate process stages, realizes refined evaluation and intelligent diagnosis, and improves the intelligence level and decision-making efficiency of purification process optimization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ORDOS MENGTAI ALUMINUM CO LTD
- Filing Date
- 2026-03-06
- Publication Date
- 2026-07-14
AI Technical Summary
In the existing aluminum-silicon alloy electrolytic production process, the data monitoring and analysis technology of the eutectoid cell purification system cannot effectively correlate with the process stages, making it difficult to accurately assess the stage-specific problems in the purification process, which affects product quality and environmental compliance.
By constructing a data analysis model that is strongly correlated with the stages of the co-exudation tank purification process, parameters from multiple monitoring points are collected in real time and divided into stages such as preheating, reaction, sedimentation, and discharge. Core characteristic parameters are extracted, process compliance scores are calculated, and weights are adjusted based on the distribution characteristics and correlation strength of historical best batch data to generate process adjustment plans.
It enables segmented and refined assessment and intelligent diagnosis of the purification process, improves the intelligence level and decision-making efficiency of purification process optimization, accurately locates abnormal stages, and generates targeted adjustment plans.
Smart Images
Figure CN122392665A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of non-ferrous metal smelting technology, and in particular to a method and system for analyzing data from the purification of an aluminum-silicon alloy eutectoid cell. Background Technology
[0002] In the electrolytic production of aluminum-silicon alloys, the eutectoid cell purification system is a crucial link in ensuring stable electrolysis processes, qualified product quality, and compliance with environmental emission standards. Existing data monitoring and analysis technologies for purification systems mainly focus on real-time acquisition and historical storage of process parameters such as temperature, pressure, and flow rate, as well as simple threshold alarms.
[0003] However, existing technologies have significant limitations in guiding process analysis and optimization. Specifically, traditional analytical methods typically treat the purification process as a black box or a single stage, performing overall statistics or trend analysis on the massive amounts of collected parameters, failing to effectively correlate data with specific process stages of the co-extrusion tank purification (such as heating, reaction, sedimentation, and discharge). This makes it difficult for technicians to accurately assess stage-specific process issues such as "whether the heating rate in the preheating stage is acceptable," "whether the pressure fluctuations in the reaction stage are within acceptable limits," and "whether the material level is stable in the sedimentation stage." When the quality of the final purified product fluctuates, operators often need to rely on experience to systematically check massive amounts of historical data, which is inefficient and makes it difficult to accurately pinpoint the specific process stage and core parameters causing the quality problem, let alone generate targeted, quantitative, and staged process adjustment plans.
[0004] Therefore, how to break down the barriers between data and process stages and achieve segmented, refined evaluation and intelligent diagnosis of the purification process is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention
[0005] This invention provides a method and system for analyzing purification data in an aluminum-silicon alloy eutectoid tank, which breaks down the barriers between data and process stages, enabling segmented and refined evaluation and intelligent diagnosis of the purification process.
[0006] On one hand, the present invention provides a method for analyzing purification data of an aluminum-silicon alloy eutectoid tank, comprising: The system acquires real-time parameters from multiple monitoring points in the purification system; these real-time parameters include temperature parameters, pressure parameters, flow parameters, material level status, and process control parameters. The acquired real-time parameters are parsed and verified to obtain purified data; According to the different stages of eutectoid tank purification, the purification data is divided into segmented data corresponding to the stages. For the segmented data of each stage, multiple core feature parameters of each stage are extracted, and the core feature parameters of each stage are compared with the standard process parameter range of the corresponding stage to calculate the process compliance score of each stage. Based on the process compliance scores at each stage, the overall pass probability of the purification process is calculated, and abnormal stages with process compliance scores below a preset score threshold are identified. Analyze the segmented data of the abnormal stage to determine the process adjustment plan; Generate an analysis report that includes stage process compliance scores, overall pass probability, and process adjustment plans.
[0007] On the other hand, the present invention also provides an aluminum-silicon alloy eutectoid tank purification data analysis system, which includes: The acquisition module is used to acquire real-time parameters from multiple monitoring points in the purification system; the real-time parameters include temperature parameters, pressure parameters, flow parameters, material level status, and process control parameters. The parsing module is used to parse and verify the acquired real-time parameters to obtain purified data; The segmentation module is used to divide the purification data into segments corresponding to different stages of the co-extrusion tank purification process. The extraction module is used to extract multiple core feature parameters for each stage of segmented data, compare the core feature parameters of each stage with the standard process parameter range of the corresponding stage, and calculate the process compliance score for each stage. The identification module is used to calculate the overall pass probability of the purification process based on the process compliance score at each stage, and to identify abnormal stages where the process compliance score is lower than a preset score threshold. The generation module is used to analyze the segmented data of the abnormal stage, determine the process adjustment plan, and generate an analysis report that includes the stage process compliance score, the overall pass probability, and the process adjustment plan.
[0008] The present invention provides a data analysis method and system for the purification of aluminum-silicon alloy eutectoid tanks. By constructing a data analysis model strongly correlated with the purification process stages, the method first segments the real-time collected purification data into stages such as preheating, reaction, sedimentation, and discharge. Then, it extracts the core characteristic parameters for each stage, compares them with preset standard process intervals, and introduces a weight correction mechanism based on the distribution characteristics and correlation strength of historical best batch data to calculate the quantified process compliance score for each stage. Furthermore, it assesses the overall pass rate based on the scores of each stage and their logical relationships, and accurately locates abnormal stages. Finally, it intelligently analyzes the abnormal stages and generates process adjustment plans containing specific execution strategies, combining equipment operating status, historical adjustment effects, and parameter coupling relationships. This achieves a leap from "overall monitoring" to "segmented precise evaluation," and from "abnormal alarms" to "plan generation," improving the intelligence level and decision-making efficiency of purification process optimization. Attached Figure Description
[0009] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0010] Figure 1 This is a schematic flowchart of the data analysis method for the purification of aluminum-silicon alloy eutectoid tanks provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the aluminum-silicon alloy eutectoid tank purification data analysis system provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation
[0011] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.
[0012] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.
[0013] Figure 1 This is a schematic flowchart of the data analysis method for the purification of aluminum-silicon alloy eutectoid tanks provided in an embodiment of the present invention.
[0014] like Figure 1 As shown, the data analysis method for the purification of the aluminum-silicon alloy eutectoid tank provided in this embodiment of the invention mainly includes the following steps: 101. Obtain real-time parameters from multiple monitoring points in the purification system; In a specific implementation process, the OPC protocol data acquisition layer can communicate with the on-site DCS system and PLC controller to collect real-time data from various monitoring points of the purification system, including temperature parameters, pressure parameters, flow parameters, material level status, and process control parameters. The acquisition frequency can be set to the second level as needed to ensure the real-time performance of the parameters.
[0015] 102. The acquired real-time parameters are parsed and verified to obtain purified data; After data acquisition, the collected real-time parameters can be parsed and verified. Through standardized data formatting, outlier removal, and missing value completion, the accuracy and completeness of the purified data are ensured. The processed purified data is stored in the corresponding data table in the MySQL database. Specifically, the core characteristic parameters for the preheating stage include the heating rate, preheating endpoint temperature, and temperature uniformity; the core characteristic parameters for the reaction stage include reaction temperature stability, pressure fluctuation amplitude, reaction time, and flow rate stability; the core characteristic parameters for the sedimentation stage include sedimentation temperature, material level stabilization time, and pressure hold value; and the core characteristic parameters for the emission stage include emission flow rate uniformity, emission endpoint material level, and emission temperature.
[0016] 103. According to the different stages of eutectoid tank purification, the purification data is divided into segmented data for the corresponding stages; In a specific implementation process, based on the process sequence and characteristics of the co-extrusion tank purification, the system data analysis layer can preset the stage division rules, and automatically divide the purification data in the database into segmented data for each stage according to the time nodes and core process behaviors of the four process stages of preheating, reaction, sedimentation and discharge.
[0017] 104. For the segmented data of each stage, extract multiple core feature parameters for each stage, compare the core feature parameters of each stage with the standard process parameter range of the corresponding stage, and calculate the process compliance score of each stage. In a specific implementation process, preset core characteristic parameters can be extracted in stages, such as the heating rate and preheating endpoint temperature in the preheating stage. The core characteristic parameters of each stage are matched with the key monitoring parameters of the purification system. For the core characteristic parameters of each stage, an evaluation benchmark including an ideal target value and an allowable deviation range can be set for each core characteristic parameter based on the standard process parameter range of each core characteristic parameter. For the actual value of each core characteristic parameter, the deviation of the actual value from the evaluation benchmark is calculated. The deviation represents the degree to which the actual value deviates from the ideal process state. For each core characteristic parameter in the same stage, a weighted comprehensive calculation is performed based on the deviation of each core characteristic parameter and the evaluation weight of each core characteristic parameter to generate a value within the standardized range, which serves as the process compliance score for the same stage.
[0018] Specifically, for each core characteristic parameter of each stage of the eutectoid tank purification, an evaluation benchmark can be set in combination with its standard process parameter range. The evaluation benchmark clarifies the ideal target value of the parameter and the allowable deviation range. The ideal target value is the optimal value within the standard range, and the allowable deviation range is set based on the process requirements and actual production of aluminum-silicon alloy eutectoid tank purification. For example, the ideal target value of the preheating endpoint temperature is 580℃, and the allowable deviation range is ±5℃.
[0019] For each core characteristic parameter, calculate its deviation from the evaluation benchmark. If the actual value is between the ideal target value and the allowable deviation boundary, calculate the deviation as a proportion of the difference between the actual value and the ideal target value to the allowable deviation range. If the actual value exceeds the allowable deviation range, increase the deviation value by the excess. The larger the deviation value, the higher the degree of deviation of the actual value from the ideal process state.
[0020] The system presets evaluation weights for each core characteristic parameter within the same stage. These weights are set according to the importance of the parameter to the process objective of the stage. The system performs a weighted comprehensive calculation based on the deviation of each parameter and the evaluation weight. First, the deviation of each parameter is normalized. Then, the normalized deviation is subtracted from 1 to obtain the individual score of the parameter. The individual scores of each parameter are multiplied by the corresponding evaluation weight and summed. Finally, the calculation results are mapped to a standardized score range of 0-100 to obtain the process compliance score for that stage. The higher the score, the better the compliance of the process execution in that stage.
[0021] Furthermore, the process of obtaining the evaluation weights for each core feature parameter may include: a1. Based on the process principle of eutectoid tank purification, determine the core purification objectives of each stage, and set the basic weight of each core characteristic parameter according to the degree of influence of each core characteristic parameter on achieving the core purification objectives of the stage. Specifically, based on the process principle of aluminum-silicon alloy eutectoid tank purification, the core purification objectives of each stage can be clearly defined. For example, the core objective of the preheating stage is to achieve uniform heating of the tank to the preset temperature. According to the degree of influence of each core characteristic parameter on this objective, a basic weight is set for each parameter. The parameter with the greater influence on the core objective has a higher basic weight value. For example, the basic weight of the preheating endpoint temperature in the preheating stage is higher than that of temperature uniformity.
[0022] a2. For each stage, select the first historical batch data with the highest purification quality level from the preset historical production database; Specifically, the system's MySQL historical production database stores a massive amount of coexisting tank purification production data, and each data entry is associated with a corresponding purification quality level evaluation result. For each process stage, the database retrieval function filters out the first historical batch data with the highest purification quality level. During the filtering process, batches with missing or abnormal data are removed to ensure the validity of the data.
[0023] a3. Based on the first historical batch data, determine the distribution concentration of the actual value of each core feature parameter and the correlation strength between each core feature parameter and the highest purification quality level at each stage. Specifically, based on the first batch of historical data after screening, the distribution concentration of the actual values of each core characteristic parameter at each stage can be calculated. The distribution concentration characterizes the stability of the parameter values during the process execution. At the same time, the correlation strength between each core characteristic parameter and the highest purification quality level can be analyzed. The correlation strength characterizes the degree of influence of the parameter on achieving the highest purification quality level.
[0024] a4. Based on the distribution concentration and the correlation strength, the basic weights are corrected to obtain the evaluation weights.
[0025] Specifically, the distribution concentration and correlation strength can be used as weight correction factors to quantify and correct the basic weights. During the correction process, the stability of the parameters and their impact on high-quality purification are taken into account. Finally, the evaluation weights of each core characteristic parameter are obtained and stored in the system's process parameter library.
[0026] Furthermore, the process of determining the distribution concentration includes: For each core feature parameter in each stage, extract all historical actual values of each core feature parameter from the historical batch data; Determine the numerical distribution center of the historical actual values, and define multiple data distribution intervals around the numerical distribution center, with the range expanding sequentially from the inside out; The statistics show the first proportion of historical actual values falling into the innermost data distribution interval, and the second proportion falling into the core data distribution interval composed of the innermost layer and the adjacent outer layer. Based on the first ratio and the second ratio, the process stability factor of each core characteristic parameter in its respective stage is evaluated as the distribution concentration. Wherein, if the first ratio is greater than a preset first standard threshold, and the difference between the first ratio and the second ratio is less than or equal to a preset difference threshold, the corresponding core feature parameter is determined to be the process stability factor of the first value. If the first ratio is between the first standard threshold and the preset second standard threshold, or if the first ratio is greater than the first standard threshold but the difference is greater than the difference threshold, the corresponding core feature parameter is determined to be the process stability factor of the second value; wherein, the first standard threshold is greater than the second standard threshold; and the first value is greater than the second value; If the first ratio is less than the second standard threshold, the corresponding core feature parameter is determined to be the process stability factor of the third value; wherein the second value is greater than the third value.
[0027] Specifically, based on the aforementioned first historical batch data, for each core characteristic parameter under each stage of coexisting tank purification, all historical actual values of the parameter are extracted from the filtered first historical batch data. During the extraction process, the timestamp and batch information of the data are retained to ensure data integrity.
[0028] The numerical distribution center is determined by calculating the arithmetic mean of historical actual values. Around this distribution center, three data distribution intervals are defined from the inside out, with the range increasing sequentially. The interval range is set according to the allowable deviation range of this parameter to ensure that the interval division conforms to the process requirements. Then, the proportion of all historical actual values falling into the innermost data distribution interval is counted, as is the proportion of the number falling into the core data distribution interval composed of the innermost and middle layers.
[0029] The system pre-sets a first standard threshold, a second standard threshold, and a difference threshold for the co-extrusion tank purification process. For example, the first standard threshold is set to 80%, the second standard threshold to 60%, and the difference threshold to 10%. The process stability factor is evaluated based on the numerical relationship between the first and second proportions: if the first proportion is >80% and the difference is ≤10%, the process stability factor for this parameter is the first value (e.g., 0.9), representing extremely high parameter distribution concentration; if the first proportion is between 60% and 80%, or the first proportion is >80% but the difference is >10%, the process stability factor is the second value (e.g., 0.7), representing moderate parameter distribution concentration; if the first proportion is <60%, the process stability factor is the third value (e.g., 0.5), representing low parameter distribution concentration. This process stability factor is used as a quantitative indicator of distribution concentration for adjustment of the basic weights.
[0030] This embodiment transforms the distribution characteristics of parameters into calculable process stability factors by extracting historical actual values, dividing distribution intervals, and statistically analyzing proportions. The quantitative results intuitively reflect the process stability of parameters in the highest quality purification batch. At the same time, the process stability factors are graded by multi-threshold condition judgment, making the assessment of distribution concentration more consistent with the actual production of coeutectoid tank purification. This provides an objective and quantitative basis for the correction of basic weights, avoids the subjectivity of stability assessment, and improves the scientific nature of weight correction.
[0031] Furthermore, the process of determining the correlation strength includes: For each core feature parameter in each stage, the optimal value range of each core feature parameter in its respective stage is determined based on the first historical batch data. Extract second historical batch data (not at the highest purification quality level) from the historical production database and locate the corresponding stage. Calculate the number of batches whose actual values deviate from the optimal value range of the stage for each core characteristic parameter, along with the average deviation. Specifically, for each core characteristic parameter of the historical batch data (not at the highest purification quality level) at the corresponding stage, determine whether its actual value is outside the optimal value range of the stage. Calculate the proportion of batches whose actual values are outside the optimal value range of the stage to the total number of batches analyzed, as the number of deviating batches. For each deviating batch, calculate the absolute difference between its actual value and the nearest boundary of the optimal value range of the stage, as the single deviation range for that batch. The average deviation is obtained by calculating the weighted average of the single deviations of all deviation batches. The degree of quality impact of each core characteristic parameter on the final purification quality is evaluated based on the number of batches deviating and the average deviation, which is used as the correlation strength; wherein, the smaller the number of batches and the smaller the average deviation, the higher the correlation strength.
[0032] Specifically, based on the first batch of historical data, for each core characteristic parameter in each stage, the upper and lower quartiles of the historical actual value of the parameter in the first batch of historical data can be calculated. The quartile interval is determined as the optimal value range of the parameter in the stage. This value range represents the optimal operating range of the parameter to achieve the highest purification quality.
[0033] Extract the second historical batch data (not the highest purification quality level) from the MySQL historical production database. After removing invalid data, locate the data to the corresponding purification process stage. For each core characteristic parameter, determine whether the actual value of the parameter in the second historical batch data exceeds the optimal value range of the stage. Calculate the proportion of batches exceeding the value range to the total number of second historical batches analyzed, as the deviation batch ratio. For each deviation batch, calculate the absolute difference between the actual value of the parameter and the nearest boundary of the optimal value range, as the single deviation magnitude.
[0034] The deviation magnitude is assigned a weight based on the production output ratio of each deviating batch. A weighted average of all deviation magnitudes is calculated to obtain the average deviation magnitude. The correlation strength is assessed by combining the proportion of deviating batches and the average deviation magnitude. The lower the proportion of deviating batches and the smaller the average deviation magnitude, the smaller the impact on purification quality when the parameter deviates from the optimal value range. Conversely, the greater the impact, the higher the proportion of deviating batches and the larger the average deviation magnitude. The degree of quality impact is quantified and graded, and converted into a quantitative value between 0 and 1 as a correlation strength index. The higher the value, the stronger the correlation, which can be used to adjust the basic weights.
[0035] This embodiment achieves a quantitative assessment of the correlation strength between core characteristic parameters and the highest cleanliness level. By determining the optimal value range for a given stage, the parameter values are directly correlated with the highest cleanliness level. Furthermore, by analyzing the parameter deviations of non-highest quality batches, the proportion of batches with deviations and the average deviation magnitude are statistically analyzed, objectively reflecting the degree of impact of parameter deviations from the optimal state on cleanliness quality. This provides a basis for adjusting the basic weights in line with actual production, enabling the adjusted assessment weights to accurately reflect the actual impact of parameters on cleanliness quality and improving the accuracy of subsequent process compliance scoring.
[0036] Further, based on the distribution concentration and the correlation strength, the basic weights are corrected to obtain the evaluation weights, including: Based on the distribution concentration, a stability adjustment coefficient for each core feature parameter is determined; wherein, the higher the distribution concentration, the larger the corresponding stability adjustment coefficient. Based on the correlation strength, the quality influence coefficient of each core feature parameter is determined; wherein, the higher the correlation strength, the greater the corresponding quality influence coefficient. For each core feature parameter, the product of the basic weight, the stability adjustment coefficient, and the quality influence coefficient is calculated to obtain the intermediate weight of each core feature parameter. The intermediate weights of all core feature parameters within the same stage are normalized so that the sum of the weights of each parameter is 1, thus obtaining the evaluation weights.
[0037] Specifically, the stability adjustment coefficient is first determined by dividing the process stability factor of the distribution concentration into three levels and setting different stability adjustment coefficient ranges accordingly. When the process stability factor is the first value (high concentration), the stability adjustment coefficient is 1.1-1.2; when the process stability factor is the second value (medium concentration), the stability adjustment coefficient is 0.9-1.0; and when the process stability factor is the third value (low concentration), the stability adjustment coefficient is 0.7-0.8. At the same time, the adjustment coefficient is fine-tuned in combination with the process stability requirements of each stage of the eutectoid tank purification. For example, the reaction stage has higher requirements for parameter stability, so the adjustment coefficient of the high concentration parameter in this stage is increased by 0.05. Finally, the stability adjustment coefficient of each core characteristic parameter is obtained, realizing a precise match between the distribution concentration and the adjustment coefficient.
[0038] Secondly, the quality influence coefficient is determined. The quantitative values of the correlation strength are divided into three levels: low, medium, and high, with values of 0-0.3, 0.3-0.7, and 0.7-1.0. Corresponding quality influence coefficients are set: low correlation strength coefficient is 0.8-0.9, medium correlation strength coefficient is 0.9-1.1, and high correlation strength coefficient is 1.1-1.3. For the core parameters of key process stages in the co-extrusion tank purification (such as sedimentation and discharge), the coefficient is increased by 0.1 to ensure that the coefficient is close to the key points of the process. The higher the correlation strength, the greater the quality influence coefficient.
[0039] For each core characteristic parameter, the set base weight is multiplied by the stability adjustment coefficient and the quality influence coefficient mentioned above to obtain the intermediate weight of the parameter. The intermediate weights of all core characteristic parameters in the same process stage are summed, and then the intermediate weight of each parameter is divided by the sum to complete the normalization process, so that the sum of the evaluation weights of all parameters in the same stage is 1. The final evaluation weight is stored in the system process parameter library.
[0040] This embodiment combines the stability adjustment coefficient and the quality influence coefficient with the basic weight through a dual-coefficient correction mechanism. This allows the evaluation weight to simultaneously consider the process stability of core characteristic parameters and their actual impact on purification quality, solving the problem of traditional weight settings that only consider process importance. Furthermore, by setting coefficient levels and fine-tuning the differences between process stages, the coefficients are made more closely aligned with the process characteristics of aluminum-silicon alloy eutectoid tank purification. Normalization ensures the standardization and calculability of the evaluation weights. The corrected evaluation weights are more scientific and accurate, providing a reliable weight basis for subsequent process compliance scoring and effectively improving the objectivity and reference value of the scoring results.
[0041] 105. Based on the process compliance scores at each stage, calculate the overall pass probability of the purification process, and identify abnormal stages where the process compliance scores are lower than the preset score threshold. In a specific implementation process, the influence weight of the process compliance score on the final purification result can be determined according to the sequential logical relationship and action transmission path of each stage. Among them, the stage located at the end of the process chain or that plays a decisive role in the final quality is given a higher influence weight. Based on the influence weight, the process compliance scores of each stage are merged to obtain a comprehensive process evaluation value. The comprehensive process evaluation value is mapped to a preset probability transformation model to calculate the overall pass probability, and at the same time, the stage with a score lower than the preset threshold of the system is marked as an abnormal stage.
[0042] Specifically, based on the process compliance scores at each stage, the logical relationship and process effect transmission path of the four stages of eutectoid tank purification—preheating, reaction, sedimentation, and discharge—are first analyzed. According to the process characteristics of aluminum-silicon alloy eutectoid tank purification, the sedimentation and discharge stages at the end of the process chain have a stronger decisive influence on the final purification quality. Therefore, higher influence weights are assigned to the sedimentation and discharge stages, such as a weight of 0.15 for the preheating stage, 0.25 for the reaction stage, 0.3 for the sedimentation stage, and 0.3 for the discharge stage. The sum of the influence weights of each stage is 1. The weight settings can be dynamically adjusted according to the process optimization results of actual production.
[0043] The process compliance score (0-100 points) at each stage is multiplied by its corresponding influence weight, and the products are summed to obtain the comprehensive evaluation value of the purification process. The comprehensive evaluation value reflects the overall level of process compliance at each stage. A probability transformation model is pre-built in the system. This model is trained based on massive historical production data of eutectoid tank purification. The comprehensive evaluation value of the process is used as the model input. The model maps the comprehensive evaluation value of 0-100 to the overall pass probability of 0-100% through piecewise linear mapping. For example, a comprehensive evaluation value of 90 points or above corresponds to an overall pass probability of 99%, and a comprehensive evaluation value below 60 points corresponds to an overall pass probability below 50%. The mapping relationship closely matches the correlation between the comprehensive score and the purification pass result in actual production.
[0044] This embodiment sets differentiated influence weights for each stage based on the process chain characteristics of the eutectoid tank purification process. The comprehensive process evaluation value obtained by weighted fusion can fully reflect the overall process execution level of the purification process. Combined with the preset probability transformation model, the comprehensive evaluation value is mapped to the overall qualification probability, making the qualification assessment of the purification process more intuitive and quantitative, and providing accurate quantitative basis for quality control of eutectoid tank purification production.
[0045] 106. Analyze the segmented data of the abnormal stage to determine the process adjustment plan; In a specific implementation process, based on the process compliance score of the abnormal stage and the comparison results of each core characteristic parameter, at least one deviation parameter causing the low score can be located; the deviation direction and magnitude of the actual value of each deviation parameter from the standard process parameter range can be analyzed, and preliminary adjustment suggestions for each deviation parameter can be obtained based on a preset process parameter adjustment knowledge base; combined with the current operating status and historical adjustment effect data of the purification equipment corresponding to the abnormal stage, the preliminary adjustment suggestions can be optimized and conflict resolved to generate the process adjustment plan.
[0046] Specifically, based on the results of abnormal stage identification and the overall pass probability calculation, the process compliance score details and the comparison results of each core characteristic parameter of the abnormal stage are retrieved first. The core characteristic parameter with the largest deviation from the standard process parameter range and the lowest contribution to the stage score is located as the deviation parameter. Multiple interrelated deviation parameters can be located at the same time.
[0047] For each deviation parameter, the deviation direction (e.g., too high, too low) and specific deviation magnitude (e.g., temperature too high, pressure too low) of the actual value are analyzed. The system has a preset process parameter adjustment knowledge base, which stores the deviation adjustment rules of each core parameter in each stage of the aluminum-silicon alloy eutectoid tank purification. The rules are formulated based on process principles and historical adjustment experience. According to the deviation direction and magnitude, the corresponding preliminary adjustment suggestions are matched from the knowledge base. For example, if the preheating end temperature is 5°C low, the matched preliminary suggestion is to increase the heating power by 10%.
[0048] Subsequently, the system collects the current operating status of the corresponding purification equipment during the abnormal phase, including information such as equipment load, operating speed, and component wear. At the same time, it retrieves historical adjustment effect data of the deviation parameter from the MySQL historical database to analyze the matching between the preliminary adjustment suggestions and the current operating status of the equipment. If the preliminary suggestion exceeds the equipment operating load, the adjustment range is reduced. If there is a process conflict between the preliminary suggestions of multiple deviation parameters (such as one suggestion to increase the flow rate and another suggestion to decrease the flow rate), the conflict is resolved according to the importance of the parameter to the process, and the adjustment suggestions with a greater impact on the core process objectives are retained first. Finally, the preliminary adjustment suggestions are optimized and integrated to generate a targeted process adjustment plan.
[0049] This embodiment can accurately locate the deviation parameters that cause abnormal stages, avoiding indiscriminate parameter checking and improving the efficiency of problem localization. Based on the matching of deviation direction and magnitude, the initial adjustment suggestions are made more targeted. The initial suggestions are optimized and conflict-resolved by combining the current operating status of the equipment and historical adjustment effect data, which effectively avoids the problems of mismatch between adjustment suggestions and actual equipment and conflict of multiple parameter adjustments. This makes the generated process adjustment plan more in line with the actual production and has executability. At the same time, the application of the process parameter adjustment knowledge base enables rapid matching of adjustment suggestions and improves the efficiency of generating process adjustment plans.
[0050] Furthermore, by combining the current operating status of the purification system with historical adjustment effect data, the preliminary adjustment suggestions are optimized and conflict resolution is achieved to generate the process adjustment plan, including: The current operating status is analyzed to identify the degree of approximation of equipment load, operational stability, and key safety parameters, which serve as constraints for implementing adjustments. Assess the impact of each preliminary adjustment proposal on the overall system energy consumption, production efficiency, and upstream and downstream process segments, and calculate the comprehensive cost and expected benefits of the adjustment. Identify the process coupling relationship between various deviation parameters, predict the fluctuation of associated deviation parameters caused by the adjustment of a single deviation parameter, and assess the degree of parameter coupling influence of the fluctuation based on adjustment records under similar process conditions in historical adjustment effect data. Based on the constraints, the comprehensive cost and expected benefit assessment results, and the degree of parameter coupling influence, the preliminary adjustment suggestions are screened, and the screened target adjustment suggestions are prioritized. The process adjustment plan is generated by integrating the target adjustment suggestions and referring to the specific execution steps of successful cases in historical adjustment effect data; wherein, the process adjustment plan includes specific execution equipment, adjustment parameters, adjustment amount and adjustment priority.
[0051] Specifically, the current operating status of the purification system is first comprehensively analyzed. The proportion of the actual load of the equipment to the rated load, the stability indicators such as vibration and temperature of the equipment, and the difference between the key safety parameters such as pressure and temperature and the safety threshold are extracted. The constraints for implementing the adjustment are that the equipment load does not exceed 90% of the rated load, the operating stability indicators are within the qualified range, and the difference between the key safety parameters and the threshold is greater than the preset safety margin. The execution boundary of the adjustment suggestion is then clarified.
[0052] For each preliminary adjustment suggestion, assess its impact on the overall system energy consumption, the purification production efficiency, and the degree of fluctuation in upstream and downstream process parameters. For example, the adjustment suggestion to increase heating power will increase energy consumption and may also affect the temperature parameters in the reaction stage. Calculate the comprehensive cost of the adjustment based on energy consumption cost and equipment wear cost, and calculate the expected benefits based on the improvement in purification quality and the improvement in production efficiency to obtain the benefit-cost ratio of each preliminary adjustment suggestion.
[0053] By analyzing the process principle of co-extrusion tank purification, the process coupling relationship between various deviation parameters is identified. For example, there is a positive correlation between gas flow rate and pressure. The fluctuation trend and amplitude of related deviation parameters caused by the adjustment of a single deviation parameter are predicted. Then, similar adjustment records under similar process conditions are retrieved from historical adjustment effect data. The actual fluctuation amplitude and process impact of related parameters are statistically analyzed and quantified into a value of 0-1 as the degree of parameter coupling influence. The higher the value, the greater the impact of coupling fluctuation.
[0054] Based on the aforementioned constraints, benefit-cost ratio, and parameter coupling influence, the preliminary adjustment suggestions were screened from multiple dimensions: First, adjustment suggestions that violated the constraints were eliminated, and then suggestions with a benefit-cost ratio greater than 1 and a parameter coupling influence less than 0.5 were retained to obtain the target adjustment suggestions; subsequently, the target adjustment suggestions were prioritized and a ranking evaluation system was constructed. The benefit-cost ratio (weight 0.4), parameter coupling influence (weight 0.3), and adjustment execution difficulty (weight 0.3) were selected as evaluation indicators, and each target adjustment suggestion was quantitatively scored. The higher the score, the higher the priority. The adjustment execution difficulty was set according to the equipment operation complexity and adjustment time.
[0055] Finally, after integrating and sorting the target adjustment suggestions, successful adjustment cases under the same or similar process conditions are retrieved from historical adjustment effect data. Referring to the specific execution steps in the cases, the specific equipment to be executed, the parameters to be adjusted, the specific adjustment amount, and the adjustment priority of the process adjustment plan are clarified. For example, the heating power of the reactor is adjusted first (first priority), and then the gas inlet flow rate is adjusted (second priority), forming a complete and executable process adjustment plan.
[0056] 107. Generate an analysis report that includes stage process compliance scores, overall pass probability, and process adjustment plans.
[0057] In a specific implementation process, an analysis report can be generated, which includes process compliance scores for each stage, overall pass probability values, and specific process adjustment plans. The report can be viewed on both web and mobile devices.
[0058] The aluminum-silicon alloy eutectoid tank purification data analysis method in this embodiment constructs a data analysis model strongly correlated with the purification process stages of the eutectoid tank. First, the real-time collected purification data is segmented into stages such as preheating, reaction, sedimentation, and discharge. Then, the core characteristic parameters of each stage are extracted and compared with preset standard process intervals. A weight correction mechanism based on the distribution characteristics and correlation strength of historical best batch data is introduced to calculate the quantified process compliance score for each stage. Furthermore, the overall pass probability is evaluated based on the scores of each stage and their logical relationships, and abnormal stages are accurately located. Finally, combined with the equipment operating status, historical adjustment effects, and parameter coupling relationships, the abnormal stages are intelligently analyzed and a process adjustment plan containing specific execution strategies is generated. This achieves a leap from "overall monitoring" to "segmented precise evaluation" and from "abnormal alarm" to "plan generation," improving the intelligence level and decision-making efficiency of purification process optimization.
[0059] Based on the same general inventive concept, this invention also protects an aluminum-silicon alloy eutectoid tank purification data analysis system. The aluminum-silicon alloy eutectoid tank purification data analysis system provided by this invention will be described below. The aluminum-silicon alloy eutectoid tank purification data analysis system described below can be referred to in correspondence with the aluminum-silicon alloy eutectoid tank purification data analysis method described above.
[0060] Figure 2 This is a schematic diagram of the aluminum-silicon alloy eutectoid tank purification data analysis system provided in an embodiment of the present invention, as shown below. Figure 2 As shown, the aluminum-silicon alloy eutectoid tank purification data analysis system of this embodiment includes an acquisition module 21, an analysis module 22, a division module 23, an extraction module 24, an identification module 25, and a generation module 26.
[0061] The acquisition module 21 is used to acquire real-time parameters from multiple monitoring points of the purification system; the real-time parameters include temperature parameters, pressure parameters, flow parameters, material level status, and process control parameters. The parsing module 22 is used to parse and verify the acquired real-time parameters to obtain purified data; The segmentation module 23 is used to divide the purification data into segmented data corresponding to different stages of the co-extrusion tank purification process. Extraction module 24 is used to extract multiple core feature parameters for each stage of segmented data, compare the core feature parameters of each stage with the standard process parameter range of the corresponding stage, and calculate the process compliance score of each stage. The identification module 25 is used to calculate the overall pass probability of the purification process based on the process compliance score of each stage, and at the same time identify abnormal stages where the process compliance score is lower than the preset score threshold. The generation module 26 is used to analyze the segmented data of the abnormal stage, determine the process adjustment plan, and generate an analysis report that includes the stage process compliance score, the overall pass probability, and the process adjustment plan.
[0062] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. The electronic device may include: a processor 310, a communication interface 320, a memory 330, and a communication bus 340. The processor 310, communication interface 320, and memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions stored in the memory 330 to execute a data analysis method for cleaning aluminum-silicon alloy eutectoid cells.
[0063] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0064] It should be noted that all relevant information that may be involved in the various embodiments of the present invention is processed in strict accordance with the requirements of laws and regulations, following the principles of legality, legitimacy, and necessity, based on the reasonable purpose of the business scenario, and is information that users actively provide or generate during the use of the product / service, as well as information obtained with user authorization.
[0065] The information processed by this invention may vary depending on the specific product / service scenario and should be based on the specific scenario in which the user uses the product / service. This may involve user account information, device information, or other related information. This invention will treat the relevant information and its processing with the utmost diligence.
[0066] This invention places great emphasis on the security of relevant information and has adopted reasonable and feasible security protection measures that comply with industry standards to protect user information and prevent unauthorized access, public disclosure, use, modification, damage or loss of relevant information.
[0067] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0068] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.
[0069] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for analyzing purification data in an aluminum-silicon alloy eutectoid tank, characterized in that, include: The system acquires real-time parameters from multiple monitoring points in the purification system; these real-time parameters include temperature parameters, pressure parameters, flow parameters, material level status, and process control parameters. The acquired real-time parameters are parsed and verified to obtain purified data; According to the different stages of eutectoid tank purification, the purification data is divided into segmented data corresponding to the stages. For the segmented data of each stage, multiple core feature parameters of each stage are extracted, and the core feature parameters of each stage are compared with the standard process parameter range of the corresponding stage to calculate the process compliance score of each stage. Based on the process compliance scores at each stage, the overall pass probability of the purification process is calculated, and abnormal stages with process compliance scores below a preset score threshold are identified. Analyze the segmented data of the abnormal stage to determine the process adjustment plan; Generate an analysis report that includes stage process compliance scores, overall pass probability, and process adjustment plans.
2. The method for analyzing purification data in an aluminum-silicon alloy eutectoid tank according to claim 1, characterized in that, By comparing the core characteristic parameters of each stage with the corresponding standard process parameter ranges, the process compliance score for each stage is calculated, including: Based on the standard process parameter range for each core characteristic parameter, an evaluation benchmark is set for each core characteristic parameter, including the ideal target value and the allowable deviation range. For each core characteristic parameter, the deviation of the actual value from the evaluation benchmark is calculated; the deviation represents the degree to which the actual value deviates from the ideal process state. For each core feature parameter within the same stage, a weighted comprehensive calculation is performed based on the deviation of each core feature parameter and the evaluation weight of each core feature parameter to generate a value within the standardized range, which serves as the process compliance score within the same stage.
3. The method for analyzing purification data in an aluminum-silicon alloy eutectoid tank according to claim 2, characterized in that, The process of obtaining the evaluation weights for each core feature parameter includes: Based on the process principle of eutectoid tank purification, the core purification objectives of each stage are determined, and the basic weight of each core characteristic parameter is set according to the degree of influence of each core characteristic parameter on achieving the core purification objective of its respective stage. For each stage, the first historical batch data with the highest purification quality level is selected from the pre-set historical production database; Based on the first historical batch data, determine the distribution concentration of the actual values of each core feature parameter and the correlation strength between each core feature parameter and the highest purification quality level at each stage. The basic weights are adjusted based on the distribution concentration and the correlation strength to obtain the evaluation weights.
4. The method for analyzing purification data in an aluminum-silicon alloy eutectoid tank according to claim 3, characterized in that, The process of determining the distribution concentration includes: For each core feature parameter in each stage, extract all historical actual values of each core feature parameter from the historical batch data; Determine the numerical distribution center of the historical actual values, and define multiple data distribution intervals around the numerical distribution center, with the range expanding sequentially from the inside out; The statistics show the first proportion of historical actual values falling into the innermost data distribution interval, and the second proportion falling into the core data distribution interval composed of the innermost layer and the adjacent outer layer. Based on the first ratio and the second ratio, the process stability factor of each core characteristic parameter in its respective stage is evaluated as the distribution concentration. Wherein, if the first ratio is greater than a preset first standard threshold, and the difference between the first ratio and the second ratio is less than or equal to a preset difference threshold, the corresponding core feature parameter is determined to be the process stability factor of the first value. If the first ratio is between the first standard threshold and the preset second standard threshold, or if the first ratio is greater than the first standard threshold but the difference is greater than the difference threshold, the corresponding core feature parameter is determined to be the process stability factor of the second value; wherein, the first standard threshold is greater than the second standard threshold; and the first value is greater than the second value; If the first ratio is less than the second standard threshold, the corresponding core feature parameter is determined to be the process stability factor of the third value; wherein the second value is greater than the third value.
5. The method for analyzing purification data in an aluminum-silicon alloy eutectoid tank according to claim 3, characterized in that, The process of determining the correlation strength includes: For each core feature parameter in each stage, the optimal value range of each core feature parameter in its respective stage is determined based on the first historical batch data. Extract the second historical batch data (not the highest purification quality level) from the historical production database, locate the corresponding stage, and count the number of batches whose actual values of each core characteristic parameter deviate from the optimal value range of the stage and the average deviation. The degree of quality impact of each core characteristic parameter on the final purification quality is evaluated based on the number of batches deviating and the average deviation, which is used as the correlation strength; wherein, the smaller the number of batches and the smaller the average deviation, the higher the correlation strength.
6. The method for analyzing purification data in an aluminum-silicon alloy eutectoid tank according to claim 3, characterized in that, The evaluation weights are obtained by adjusting the basic weights based on the distribution concentration and the correlation strength, including: Based on the distribution concentration, a stability adjustment coefficient for each core feature parameter is determined; wherein, the higher the distribution concentration, the larger the corresponding stability adjustment coefficient. Based on the correlation strength, the quality influence coefficient of each core feature parameter is determined; wherein, the higher the correlation strength, the greater the corresponding quality influence coefficient. For each core feature parameter, the product of the basic weight, the stability adjustment coefficient, and the quality influence coefficient is calculated to obtain the intermediate weight of each core feature parameter. The intermediate weights of all core feature parameters within the same stage are normalized so that the sum of the weights of each parameter is 1, thus obtaining the evaluation weights.
7. The method for analyzing purification data in an aluminum-silicon alloy eutectoid tank according to claim 1, characterized in that, Based on the process compliance scores at each stage, the overall pass probability of the cleanroom process is calculated, including: Based on the sequential logical relationship and the transmission path of each stage, the influence weight of the process compliance score on the final purification result is determined at different stages; among them, the stages located at the end of the process chain or that play a decisive role in the final quality are given higher influence weights. Based on the aforementioned influence weights, the process compliance scores of each stage are integrated to obtain a comprehensive process evaluation value. The overall process evaluation value is mapped to a preset probability transformation model to calculate the overall pass probability.
8. The method for analyzing purification data in an aluminum-silicon alloy eutectoid tank according to claim 7, characterized in that, Analyze the segmented data of the abnormal phase to determine the process adjustment plan, including: Based on the process compliance score of the abnormal stage and the comparison results of each core feature parameter, locate at least one deviation parameter that caused the low score; The actual values of each deviation parameter are analyzed to determine the direction and magnitude of the deviation from the standard process parameter range. Based on a pre-set process parameter adjustment knowledge base, preliminary adjustment suggestions are obtained for each deviation parameter. By combining the current operating status of the purification equipment corresponding to the abnormal stage with historical adjustment effect data, the preliminary adjustment suggestions are optimized and conflict resolution is achieved, thereby generating the process adjustment plan.
9. The method for analyzing purification data in an aluminum-silicon alloy eutectoid tank according to claim 8, characterized in that, Based on the current operating status of the purification system and historical adjustment effect data, the preliminary adjustment suggestions are optimized and conflict resolution is performed to generate the process adjustment plan, including: The current operating status is analyzed to identify the degree of approximation of equipment load, operational stability, and key safety parameters, which serve as constraints for implementing adjustments. Assess the impact of each preliminary adjustment proposal on the overall system energy consumption, production efficiency, and upstream and downstream process segments, and calculate the comprehensive cost and expected benefits of the adjustment. Identify the process coupling relationship between various deviation parameters, predict the fluctuation of associated deviation parameters caused by the adjustment of a single deviation parameter, and assess the degree of parameter coupling influence of the fluctuation based on adjustment records under similar process conditions in historical adjustment effect data. Based on the constraints, the comprehensive cost and expected benefit assessment results, and the degree of parameter coupling influence, the preliminary adjustment suggestions are screened, and the screened target adjustment suggestions are prioritized. The process adjustment plan is generated by integrating the target adjustment suggestions and referring to the specific execution steps of successful cases in historical adjustment effect data; wherein, the process adjustment plan includes specific execution equipment, adjustment parameters, adjustment amount and adjustment priority.
10. A data analysis system for the purification of an aluminum-silicon alloy eutectoid tank, characterized in that, include: The acquisition module is used to acquire real-time parameters from multiple monitoring points in the purification system; the real-time parameters include temperature parameters, pressure parameters, flow parameters, material level status, and process control parameters. The parsing module is used to parse and verify the acquired real-time parameters to obtain purified data; The segmentation module is used to divide the purification data into segments corresponding to different stages of the co-extrusion tank purification process. The extraction module is used to extract multiple core feature parameters for each stage of segmented data, compare the core feature parameters of each stage with the standard process parameter range of the corresponding stage, and calculate the process compliance score for each stage. The identification module is used to calculate the overall pass probability of the purification process based on the process compliance score at each stage, and to identify abnormal stages where the process compliance score is lower than a preset score threshold. The generation module is used to analyze the segmented data of the abnormal stage, determine the process adjustment plan, and generate an analysis report that includes the stage process compliance score, the overall pass probability, and the process adjustment plan.