A multi-source process parameter mapping supervision system and method based on big data model

Abstract

The present invention discloses a multi-source process parameter mapping supervision system and method based on a big data model, which relates to the technical field of process parameter analysis, collects initial process parameters in a process production line, processes the initial process parameters to obtain standardized process parameters, divides the process production line into process stages, analyzes the degree of stage product deviation in the process stages, calculates the correlation between the standardized process parameters and the degree of product deviation in the process stages, analyzes the degree of stage product deviation in the process stages, predicts the degree of product deviation in the process stages in production, predicts the product defective rate in production based on the degree of product deviation in the process stages, monitors the predicted product defective rate in real time, adjusts the standardized process parameters, and corrects the standardized process parameters. The present invention constructs a dynamic quality control closed loop by continuously predicting the product deviation in each stage and the overall product defective rate, judging the production status in real time, and improving the pertinence and effectiveness of supervision.

Description

Technical Field

[0001] The present invention relates to the technical field of process parameter analysis, and in particular to a multi-source process parameter mapping supervision system and method based on a big data model. Background Art

[0002] In industrial production, traditional process parameter monitoring methods have long faced multiple technical bottlenecks. Production process monitoring lacks a systematic hierarchical structure and often relies on fuzzy control of process parameters through manual experience, making it difficult to accurately identify the key links that impact product quality. Due to the lack of a scientifically defined process stage and quantitative analysis of parameter correlations between stages, traditional monitoring models are unable to effectively identify the specific impacts of different process stages on product deviations. This leads to blind parameter adjustments, increasing the cost of ineffective operations and hindering the precise control of product quality. This extensive management model is particularly problematic in complex production scenarios with multiple stages and parameters, severely restricting production process stability and product quality consistency. Quality control mechanisms often suffer from static principles and are unable to adapt to the real-time changes in dynamic production environments. Traditional methods typically monitor process parameters based on fixed thresholds, lacking the ability to continuously predict and respond to production processes. When product defect rates fluctuate, it is impossible to quickly identify the key process stages and parameters that cause deviations, nor is it possible to achieve an effective closed-loop control through dynamic adjustments. Furthermore, existing quality control systems often lack mechanisms for iterative corrections and anomaly alarms, preventing timely resolution of quality risks. This can easily lead to production accidents or batch-specific quality issues, further exacerbating the uncontrollability of the production process. Decision-making models rely too heavily on manual experience and lack the scientific and objective nature of data support. In traditional process management, the optimization and adjustment of process parameters are primarily based on the operator's subjective judgment and historical experience, making it difficult to uncover the underlying correlations between parameters and product quality. In complex production environments where multi-source, heterogeneous data is interwoven, the limitations of manual analysis become even more pronounced, making it easy to miss key influencing factors and leading to poor decisions. This experience-driven decision-making model not only restricts improvements in production efficiency but also struggles to meet the precision and intelligence requirements of intelligent manufacturing for process optimization, becoming a significant bottleneck in the manufacturing industry's digital transformation. Summary of the Invention

[0003] The purpose of the present invention is to provide a multi-source process parameter mapping supervision system and method based on a big data model to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solution: a multi-source process parameter mapping and supervision method based on a big data model, comprising the following steps:

[0005] S1. Collecting initial process parameters in the process production line through a sensor group, and processing the initial process parameters to obtain standardized process parameters;

[0006] S2. Based on the process flow and historical data, the process production line is divided into process stages, the degree of product deviation in each process stage is analyzed, and the correlation between the standardized process parameters and the degree of product deviation in each process stage is calculated;

[0007] S3. Analyze the correlation between the degree of product deviation and product defect rate at each process stage;

[0008] S4. Predicting the degree of product deviation in the process stage of production based on standardized process parameters, and predicting the product defect rate in production based on the degree of product deviation in the process stage;

[0009] S5. Real-time monitoring and prediction of product defect rate G, and adjustment of standardized process parameters based on the predicted product defect rate;

[0010] S6. Correct the standardized process parameters. When the standardized process parameters are judged to be abnormal, an alarm is sent to the person in charge of the process.

[0011] Furthermore, in step S1, the initial process parameters in the process production line are collected by the sensor group, and the initial process parameters are denoised, filtered and normalized to obtain standardized process parameters. The standardized process parameters are {A1, A2, ..., A i ,…,A I}, where I represents the number of standardized process parameters, A i represents the i-th standardized process parameter. The sensor group includes a temperature sensor, a pressure sensor, a flow sensor, and a speed sensor. The initial process parameters include temperature, pressure, flow, and speed. The collection and preprocessing of the initial parameters of the process production line by the sensor group can effectively improve data quality and availability. Denoising and filtering can remove noise and interference from the raw data, allowing parameters such as temperature, pressure, flow, and speed to more accurately reflect the actual state of the production line and prevent abnormal fluctuations from misleading subsequent analysis. Normalization eliminates the impact of dimensional differences between different parameters, bringing all types of data under a unified measurement standard, laying a solid foundation for subsequent process modeling, anomaly detection, parameter optimization, and other operations. The processed standardized parameters can more accurately support production line monitoring, help to promptly identify potential problems, and provide a reliable basis for quality prediction and process adjustment, thereby improving the stability and efficiency of the production process and promoting the improvement of the level of intelligent management of the production line.

[0012] Furthermore, in step S2, based on the process flow and historical data, the process production line is divided into P process stages, and the p-th process stage is analyzed, where p=1,2,…,P, and the standard stage parameters of the product of the p-th process stage are {B1,B2,…,B q ,…,B Q}, where Q represents the number of standard stage parameters of the product produced in the pth process stage, B q It represents the qth standard stage parameter of the product produced in the pth process stage. The actual stage parameter of the product in the pth process stage is {b1, b2, ..., b q ,…,b Q The standard stage parameters of the product at the pth process stage represent the preset process parameters initially input at the pth process stage. The process requirement parameters include water pressure, pipe pressure, filter temperature, water flow rate for cooling water, and engine speed. The actual stage parameters of the product at the pth process stage represent the process parameters actually collected at the pth process stage. q It represents the qth actual stage parameter of the product produced in the pth process stage, and the stage product deviation degree W of the pth process stage is obtained. p :

[0013] ;

[0014] Calculate the standardized process parameter A i Correlation coefficient w for the degree of product deviation at the pth process stage ip :

[0015] ;

[0016] Where N represents the total N production lines in the historical data, n represents the nth production line in the historical data, and A i,n represents the i-th standardized process parameter in the n-th process production line, a i W represents the average value of the i-th standardized process parameter in N production lines, p,n It represents the product deviation degree of the pth process stage in the production of the nth process line, w p It represents the average value of the product deviation degree of the pth process stage in the N-times process production line production, and then obtains the standardized process parameter A i The relevant weight W for the degree of product deviation at the pth process stage ip :

[0017] ;

[0018] When W ip When <W0, determine the standardized process parameter Ai Is not relevant to the pth process stage, otherwise, the standardized process parameter A i is the standardized process parameter related to the process stage, where W0 is the established correlation coefficient threshold, which is usually related to the number of standardized process parameters and can be set to 1 / (5*I); by dividing the process production line into multiple process stages according to the process and historical data, the standard parameters and actual parameters of each stage can be carefully compared, and the process execution effect of each stage can be accurately evaluated. This stage-by-stage analysis method can clearly locate the specific links where deviations may occur in the production process, avoid general judgments on the overall process, and improve the accuracy of problem identification. Combined with the correlation between the parameters calculated based on historical production data and the stage deviation, it can effectively screen out the key parameters that have a significant impact on each process stage, eliminate the interference of irrelevant or low-correlation parameters, and make subsequent process optimization and quality control more targeted. This method helps enterprises focus on core influencing factors, rationally allocate resources, and make timely adjustments to key parameters, thereby reducing production fluctuations, improving process stability, providing a scientific basis for the continuous improvement of product quality and optimization of production efficiency, and promoting the development of production management towards refinement and efficiency.

[0019] Furthermore, in step S3, a product defect rate impact analysis model is established:

[0020] ;

[0021] Among them, v p Indicates the stage product deviation degree W of the pth process stage p The coefficient of influence on product defect rate, W p,n It represents the product deviation degree of the pth process stage in the production of the nth process line, w p Y represents the average value of the product deviation degree of the pth process stage in the N-time process production line production, n represents the product defect rate in the nth process production line, y represents the average product defect rate in the Nth process production line, and then the stage product deviation degree W of the pth process stage is obtained p The relevant weight V for product defect rate p :

[0022] ;

[0023] By analyzing the relationship between the degree of product deviation and product defective rate at each process stage, the actual impact of each stage on product quality can be accurately quantified. This method uses historical production data and big data models to deeply explore the inherent connection between process deviation and quality issues, allowing companies to clearly understand the role of different process stages in the formation of product defective rates. By evaluating the weight of the impact of stage deviation on the defective rate, key process links that have a greater impact on product quality can be effectively identified, avoiding the average effort of all stages and achieving reasonable allocation and optimization of resources. Companies can focus on monitoring and adjusting stages with significant impacts, and take timely measures to reduce deviations, thereby more specifically reducing product defective rates and improving the quality control efficiency of the production process. This process provides a scientific basis for quality improvement, promotes the transformation of production management from experience-driven to data-driven, and helps companies implement precise policies and enhance product quality stability.

[0024] Furthermore, in step S4, the product deviation degree F of the pth process stage in production is predicted based on the standardized process parameters. p The product deviation degree of the pth process stage is the sum of the standardized process parameters multiplied by the corresponding weights of the product deviation degree, where the standardized process parameters A p The weight W associated with the degree of product deviation ip Correspondingly, the product defect rate in production is predicted based on the product deviation degree in the production stage, and the product defect rate in production is the sum of the product deviation degree in production and the corresponding relevant weight of the product defect rate, where the product deviation degree in production F p and product defect rate V p Corresponding; by integrating the standardized process parameters with the relevant weights of each process stage, step S4 can accurately predict the degree of product deviation at each stage and the final product defect rate based on the real-time collected production data. This method uses the association model established by historical data to convert abstract process parameters into quantifiable quality impact predictions, enabling enterprises to monitor the deviation risks of key links in real time during the production process and identify abnormal factors that may lead to defective products in advance. By focusing on parameters that are highly correlated with the deviations and defect rates at each stage, the prediction process is more targeted and scientific, avoiding the inefficiency of blind monitoring. Enterprises can adjust process parameters in a timely manner based on the prediction results, focus on controlling high-impact stages, intervene in advance in the product production process, and reduce the risk of defective product output. This method shifts quality control from post-detection to pre-prediction, improves the transparency and controllability of the production process, provides support for refined management, and helps enterprises allocate resources efficiently.

[0025] Furthermore, in step S5, during the production process of the process line, the product defect rate G is monitored and predicted. When G < G0, the system determines that the product defect rate meets the requirements, and G0 is the product defect rate threshold set by the system; when G ≥ G0, the system determines that the product defect rate does not meet the requirements, and monitors the impact value H of the pth process stage on the defect rate. p , H p =v p *W p , one by one into p = 1, 2, ..., P, and obtain the impact of P process stages on the defect rate {H1, H2, ..., H u ,…,H P}, where the maximum process stage’s impact on defect rate is H u , so the process stage corresponding to the maximum impact value is the u-th process stage, and the standardized process parameter A is obtained by analyzing the u-th process stage. i Impact value J on the degree of product deviation at the u-th process stage iu =w ip *A i , one by one into i = 1, 2, ..., I, and then get the impact value of I standardized process parameters on the product deviation degree of the u-th process stage {J 1u ,J 2u ,…,J xu ,…,J Iu}, where the maximum standardized process parameter’s impact on the product deviation degree of the u-th process stage is J xu , that is, the xth standardized process parameter has the greatest impact on the degree of product deviation at the uth process stage. The xth standardized process parameter is then corrected to the optimal xth standardized process parameter. Standardized process parameters can be obtained from historical data. By monitoring and predicting product defect rates in real time and responding dynamically, the system can quickly identify key links affecting product quality. If the predicted defect rate exceeds the set standard, the system precisely locates the process stage with the greatest impact, avoiding blind investigation of all stages and significantly improving problem diagnosis efficiency. Further, within this critical stage, the standardized process parameter with the most significant impact on product deviation is selected, allowing companies to focus on the core influencing factors and avoid wasting resources on minor parameters. By promptly correcting key parameters, the focus of quality control shifts from reactive to proactive intervention, effectively reducing the risk of defective product output. This hierarchical and progressive analysis approach, supported by historical data, enables precise control of the production process, helping companies quickly respond to quality fluctuations and improve the controllability and stability of the production process. It provides a scientific and practical solution for efficient quality optimization and process improvement, and contributes to the development of precise and intelligent production management.

[0026] Furthermore, in step S6, after the xth standardized process parameter is corrected, the predicted product defect rate is monitored again. At this time, the predicted product defect rate is G1. If G1 < G0, the predicted product defect rate is monitored again after a predetermined monitoring interval. If G1 ≥ G0, the standardized process parameter is corrected again until the predicted product defect rate is less than the predetermined product defect rate threshold.

[0027] When the number of consecutive adjustments to standardized process parameters is less than or equal to the pre-set threshold, counting continues until the predicted product defect rate is less than the pre-set product defect rate threshold or the number of consecutive adjustments to standardized process parameters exceeds the pre-set threshold. When the number of consecutive adjustments to standardized process parameters exceeds the pre-set threshold, the initially set standardized process parameters are judged to be abnormal, and a parameter abnormality alarm is issued to the process personnel responsible. By establishing a revised dynamic monitoring and closed-loop adjustment mechanism, the optimization effect of key process parameters can be continuously verified, avoiding the blindness of single adjustments. If the predicted defect rate value after correction still does not meet the standard, the system can accurately locate the problematic parameter again based on historical data logic, gradually approaching the optimal production state through progressive adjustments, effectively reducing the cost of manual trial and error. The setting of the threshold for the number of consecutive adjustments not only sets reasonable boundaries for the automatic optimization of process parameters, but also triggers an abnormality alarm in a timely manner when multiple adjustments fail, prompting manual intervention to troubleshoot systemic problems and avoid long-term quality risks caused by underlying causes such as equipment failure and process design defects. This mechanism, which combines automated adjustments with manual intervention, not only ensures the stability and adaptability of the production process, but also provides double insurance for quality control in complex process scenarios. It helps companies efficiently solve immediate problems while promptly identifying potential hidden dangers, thereby promoting production management towards intelligence and reliability.

[0028] A multi-source process parameter mapping and supervision system based on a big data model, comprising: a parameter acquisition and standardization module, a process stage division and parameter correlation analysis module, a defect rate impact analysis module, a parameter deviation analysis and defect rate prediction module, a real-time monitoring and parameter correction module, and a parameter anomaly processing module;

[0029] The parameter acquisition and standardization module is used to acquire the initial process parameters in the process production line through the sensor group, and process the initial process parameters to obtain standardized process parameters;

[0030] The process stage division and parameter correlation analysis module is used to divide the process production line into process stages based on the process flow and historical data, analyze the degree of product deviation in the process stages, and calculate the correlation between the standardized process parameters and the degree of product deviation in the process stages;

[0031] The defect rate impact analysis module is used to analyze the correlation between the stage product deviation degree and the product defect rate in the process stage;

[0032] The parameter deviation analysis and defect rate prediction module is used to predict the degree of product deviation in the process stage of production based on the standardized process parameters, and to predict the product defect rate in production based on the degree of product deviation in the process stage;

[0033] The real-time monitoring and parameter correction module is used to monitor the predicted product defect rate G in real time and adjust the standardized process parameters according to the predicted product defect rate;

[0034] The parameter abnormality processing module is used to correct the standardized process parameters and alert the process responsible person when it is determined that the standardized process parameters are abnormal.

[0035] Furthermore, the sensor group includes a temperature sensor, a pressure sensor, a flow sensor and a rotation speed sensor.

[0036] Compared with the existing technology, the beneficial effects achieved by the present invention are: on the one hand, precise supervision of the production process is achieved. By scientifically dividing the process production line into stages and deeply analyzing the intrinsic relationship between product deviations at each stage and process parameters and product defect rates, the process stages and core parameters that have a key impact on product quality can be accurately located. This hierarchical correlation analysis changes the ambiguity of traditional supervision that relies on experience-based judgment, and transforms supervision work from extensive management to precise regulation of specific links, ensuring that each parameter adjustment can effectively target the key factors affecting product quality, avoiding invalid operations, and greatly improving the pertinence and effectiveness of supervision;

[0037] On the one hand, a dynamic closed loop of quality control is constructed. The method continuously predicts product deviations at each stage and the overall product defect rate by collecting and processing process parameters in real time, and judges the production status in real time according to preset standards. When the predicted defect rate exceeds the standard, it can quickly lock in the process stage that has the greatest impact on the defect rate, further screen out the process parameters with the most prominent impact in this stage, and make targeted adjustments and corrections. This dynamic monitoring and immediate response mechanism keeps the production process under control at all times, can eliminate quality risks in a timely manner, and prevent problems from escalating. At the same time, combined with iterative correction and abnormal alarm mechanisms, it ensures the system's self-regulation ability under normal fluctuations;

[0038] On the other hand, it promotes a data-driven scientific decision-making model. The method fully relies on big data models to transform historical production data into quantifiable correlation weights, using data relationships to reveal the inherent laws between process parameters and product quality, and getting rid of excessive reliance on manual experience. This data-driven analysis method can unearth parameter correlations that are difficult to detect with traditional methods, providing an objective and scientific basis for production decisions. It is particularly suitable for complex production scenarios with multiple stages and multiple parameters coupled. Through systematic modeling and analysis, the scattered process parameters are integrated into an organic supervision system, so that every link in the production process can be optimized with data support, fundamentally improving the stability of the production process and the reliability of product quality, and providing a solid methodological support for the transformation of the manufacturing industry to intelligence and digitalization. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0040] Figure 1 This is a structural diagram of a multi-source process parameter mapping and supervision system based on a big data model according to the present invention;

[0041] Figure 2 It is a flow chart of a multi-source process parameter mapping and supervision method based on a big data model of the present invention. DETAILED DESCRIPTION

[0042] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0043] See also Figure 1 and Figure 2 The present invention provides a technical solution: a multi-source process parameter mapping and supervision method based on a big data model, comprising the following steps:

[0044] S1. Collecting initial process parameters in the process production line through a sensor group, and processing the initial process parameters to obtain standardized process parameters;

[0045] S2. Based on the process flow and historical data, the process production line is divided into process stages, the degree of product deviation in each process stage is analyzed, and the correlation between the standardized process parameters and the degree of product deviation in each process stage is calculated;

[0046] S3. Analyze the correlation between the degree of product deviation and product defect rate at each process stage;

[0047] S4. Predicting the degree of product deviation in the process stage of production based on standardized process parameters, and predicting the product defect rate in production based on the degree of product deviation in the process stage;

[0048] S5. Real-time monitoring and prediction of product defect rate G, and adjustment of standardized process parameters based on the predicted product defect rate;

[0049] S6. Correct the standardized process parameters. When the standardized process parameters are judged to be abnormal, an alarm is sent to the person in charge of the process.

[0050] In step S1, the initial process parameters in the process production line are collected by the sensor group, and the initial process parameters are denoised, filtered and normalized to obtain standardized process parameters. The standardized process parameters are {A1, A2, ..., A i ,…,A I}, where I represents the number of standardized process parameters, A i represents the i-th standardized process parameter. The sensor group includes a temperature sensor, a pressure sensor, a flow sensor, and a speed sensor. The initial process parameters include temperature, pressure, flow, and speed. The collection and preprocessing of the initial parameters of the process production line by the sensor group can effectively improve data quality and availability. Denoising and filtering can remove noise and interference from the raw data, allowing parameters such as temperature, pressure, flow, and speed to more accurately reflect the actual state of the production line and prevent abnormal fluctuations from misleading subsequent analysis. Normalization eliminates the impact of dimensional differences between different parameters, bringing all types of data under a unified measurement standard, laying a solid foundation for subsequent process modeling, anomaly detection, parameter optimization, and other operations. The processed standardized parameters can more accurately support production line monitoring, help to promptly identify potential problems, and provide a reliable basis for quality prediction and process adjustment, thereby improving the stability and efficiency of the production process and promoting the improvement of the level of intelligent management of the production line.

[0051] In step S2, based on the process flow and historical data, the process production line is divided into P process stages, and the p-th process stage is analyzed, where p=1,2,…,P, and the standard stage parameters of the product in the p-th process stage are {B1,B2,…,B q ,…,B Q}, where Q represents the number of standard stage parameters of the product produced in the pth process stage, B q represents the qth standard stage parameter of the product produced in the pth process stage, and the actual stage parameter of the product in the pth process stage is {b1, b2, ..., b q ,…,b QThe standard stage parameters of the product at the pth process stage represent the preset process parameters initially input at the pth process stage. The process requirement parameters include water pressure, pipe pressure, filter temperature, water flow rate for cooling water, and engine speed. The actual stage parameters of the product at the pth process stage represent the process parameters actually collected at the pth process stage. q It represents the qth actual stage parameter of the product produced in the pth process stage, and the stage product deviation degree W of the pth process stage is obtained. p :

[0052] ;

[0053] Calculate the standardized process parameter A i Correlation coefficient w for the degree of product deviation at the pth process stage ip :

[0054] ;

[0055] Where N represents the total N production lines in the historical data, n represents the nth production line in the historical data, and A i,n represents the i-th standardized process parameter in the n-th process production line, a i W represents the average value of the i-th standardized process parameter in N production lines, p,n It represents the product deviation degree of the pth process stage in the production of the nth process line, w p It represents the average value of the product deviation degree of the pth process stage in the N-times process production line production, and then obtains the standardized process parameter A i The relevant weight W for the degree of product deviation at the pth process stage ip :

[0056] ;

[0057] When W ip When <W0, determine the standardized process parameter A i Is not relevant to the pth process stage, otherwise, the standardized process parameter A iis the standardized process parameter related to the process stage, where W0 is the established correlation coefficient threshold, which is usually related to the number of standardized process parameters and can be set to 1 / (5*I); by dividing the process production line into multiple process stages according to the process and historical data, the standard parameters and actual parameters of each stage can be carefully compared, and the process execution effect of each stage can be accurately evaluated. This stage-by-stage analysis method can clearly locate the specific links where deviations may occur in the production process, avoid general judgments on the overall process, and improve the accuracy of problem identification. Combined with the correlation between the parameters calculated based on historical production data and the stage deviation, it can effectively screen out the key parameters that have a significant impact on each process stage, eliminate the interference of irrelevant or low-correlation parameters, and make subsequent process optimization and quality control more targeted. This method helps enterprises focus on core influencing factors, rationally allocate resources, and make timely adjustments to key parameters, thereby reducing production fluctuations, improving process stability, providing a scientific basis for the continuous improvement of product quality and optimization of production efficiency, and promoting the development of production management towards refinement and efficiency.

[0058] In step S3, a product defect rate impact analysis model is established:

[0059] ;

[0060] Among them, v p Indicates the stage product deviation degree W of the pth process stage p The coefficient of influence on product defect rate, W p,n It represents the product deviation degree of the pth process stage in the production of the nth process line, w p Y represents the average value of the product deviation degree of the pth process stage in the N-time process production line production, n represents the product defect rate in the nth process production line, y represents the average product defect rate in the Nth process production line, and then the stage product deviation degree W of the pth process stage is obtained p The relevant weight V for product defect rate p :

[0061] ;

[0062] By analyzing the relationship between the degree of product deviation and product defective rate at each process stage, the actual impact of each stage on product quality can be accurately quantified. This method uses historical production data and big data models to deeply explore the inherent connection between process deviation and quality issues, allowing companies to clearly understand the role of different process stages in the formation of product defective rates. By evaluating the weight of the impact of stage deviation on the defective rate, key process links that have a greater impact on product quality can be effectively identified, avoiding the average effort of all stages and achieving reasonable allocation and optimization of resources. Companies can focus on monitoring and adjusting stages with significant impacts, and take timely measures to reduce deviations, thereby more specifically reducing product defective rates and improving the quality control efficiency of the production process. This process provides a scientific basis for quality improvement, promotes the transformation of production management from experience-driven to data-driven, and helps companies implement precise policies and enhance product quality stability.

[0063] In step S4, the product deviation degree F of the pth process stage in production is predicted based on the standardized process parameters. p The product deviation degree of the pth process stage is the sum of the standardized process parameters multiplied by the corresponding weights of the product deviation degree, where the standardized process parameters A p The weight W associated with the degree of product deviation ip Correspondingly, the product defect rate in production is predicted based on the product deviation degree in the production stage, and the product defect rate in production is the sum of the product deviation degree in production and the corresponding relevant weight of the product defect rate, where the product deviation degree in production F p and product defect rate V p By integrating standardized process parameters with the relevant weights of each process stage, step S4 accurately predicts the degree of product deviation at each stage and the final product defect rate based on real-time production data. This approach leverages a correlation model established with historical data to transform abstract process parameters into quantifiable quality impact predictions. This allows companies to monitor deviation risks at key stages in real time during the production process and proactively identify abnormal factors that may lead to defective products. By focusing on parameters that are highly correlated with deviation and defect rates at each stage, the prediction process becomes more targeted and scientific, avoiding the inefficiency of blind monitoring. Based on the prediction results, companies can promptly adjust process parameters, focus on high-impact stages, and intervene proactively during the production process to reduce the risk of defective product output. This approach shifts quality control from post-testing to pre-testing, improving the transparency and controllability of the production process, supporting refined management, and helping companies allocate resources efficiently.

[0064] In step S5, during the production process of the process line, the product defect rate G is monitored and predicted. When G < G0, the system determines that the product defect rate meets the requirements, and G0 is the product defect rate threshold set by the system; when G ≥ G0, the system determines that the product defect rate does not meet the requirements, and monitors the impact value H of the pth process stage on the defect rate. p , H p =v p *W p , one by one into p = 1, 2, ..., P, and obtain the impact of P process stages on the defect rate {H1, H2, ..., H u ,…,H P}, where the maximum process stage’s impact on defect rate is H u , so the process stage corresponding to the maximum impact value is the u-th process stage, and the standardized process parameter A is obtained by analyzing the u-th process stage. i Impact value J on the degree of product deviation at the u-th process stage iu =w ip *A i , one by one into i = 1, 2, ..., I, and then get the impact value of I standardized process parameters on the product deviation degree of the u-th process stage {J 1u ,J 2u ,…,J xu ,…,J Iu}, where the maximum standardized process parameter’s impact on the product deviation degree of the u-th process stage is J xu , that is, the xth standardized process parameter has the greatest impact on the degree of product deviation at the uth process stage. The xth standardized process parameter is then corrected to the optimal xth standardized process parameter. Standardized process parameters can be obtained from historical data. By monitoring and predicting product defect rates in real time and responding dynamically, the system can quickly identify key links affecting product quality. If the predicted defect rate exceeds the set standard, the system precisely locates the process stage with the greatest impact, avoiding blind investigation of all stages and significantly improving problem diagnosis efficiency. Further, within this critical stage, the standardized process parameter with the most significant impact on product deviation is selected, allowing companies to focus on the core influencing factors and avoid wasting resources on minor parameters. By promptly correcting key parameters, the focus of quality control shifts from reactive to proactive intervention, effectively reducing the risk of defective product output. This hierarchical and progressive analysis approach, supported by historical data, enables precise control of the production process, helping companies quickly respond to quality fluctuations and improve the controllability and stability of the production process. It provides a scientific and practical solution for efficient quality optimization and process improvement, and contributes to the development of precise and intelligent production management.

[0065] In step S6, after the xth standardized process parameter is corrected, the predicted product defect rate is monitored again. The predicted product defect rate at this time is G1. If G1 < G0, the predicted product defect rate is monitored again after a pre-set monitoring interval. If G1 ≥ G0, the standardized process parameter is corrected again until the predicted product defect rate is less than the pre-set product defect rate threshold.

[0066] When the number of consecutive adjustments to standardized process parameters is less than or equal to the pre-set threshold, counting continues until the predicted product defect rate is less than the pre-set product defect rate threshold or the number of consecutive adjustments to standardized process parameters exceeds the pre-set threshold. When the number of consecutive adjustments to standardized process parameters exceeds the pre-set threshold, the initially set standardized process parameters are judged to be abnormal, and a parameter abnormality alarm is issued to the process personnel responsible. By establishing a revised dynamic monitoring and closed-loop adjustment mechanism, the optimization effect of key process parameters can be continuously verified, avoiding the blindness of single adjustments. If the predicted defect rate value after correction still does not meet the standard, the system can accurately locate the problematic parameter again based on historical data logic, gradually approaching the optimal production state through progressive adjustments, effectively reducing the cost of manual trial and error. The setting of the threshold for the number of consecutive adjustments not only sets reasonable boundaries for the automatic optimization of process parameters, but also triggers an abnormality alarm in a timely manner when multiple adjustments fail, prompting manual intervention to troubleshoot systemic problems and avoid long-term quality risks caused by underlying causes such as equipment failure and process design defects. This mechanism, which combines automated adjustments with manual intervention, not only ensures the stability and adaptability of the production process, but also provides double insurance for quality control in complex process scenarios. It helps companies efficiently solve immediate problems while promptly identifying potential hidden dangers, thereby promoting production management towards intelligence and reliability.

[0067] A multi-source process parameter mapping and supervision system based on a big data model, comprising: a parameter acquisition and standardization module, a process stage division and parameter correlation analysis module, a defect rate impact analysis module, a parameter deviation analysis and defect rate prediction module, a real-time monitoring and parameter correction module, and a parameter anomaly processing module;

[0068] The parameter acquisition and standardization module is used to collect the initial process parameters in the process production line through the sensor group, and process the initial process parameters to obtain standardized process parameters;

[0069] The process stage division and parameter correlation analysis module is used to divide the process production line into process stages based on the process flow and historical data, analyze the degree of product deviation in each process stage, and calculate the correlation between the standardized process parameters and the degree of product deviation in each process stage;

[0070] The defect rate impact analysis module is used to analyze the correlation between the degree of product deviation and product defect rate at each process stage;

[0071] The parameter deviation analysis and defect rate prediction module is used to predict the degree of product deviation in the process stage of production based on standardized process parameters, and to predict the product defect rate in production based on the degree of product deviation in the process stage;

[0072] The real-time monitoring and parameter correction module is used to monitor and predict the product defect rate G in real time and adjust the standardized process parameters according to the predicted product defect rate;

[0073] The parameter anomaly processing module is used to correct the standardized process parameters and to alert the process responsible personnel when the standardized process parameters are judged to be abnormal.

[0074] Example 1: An electronic component production line manufactures precision circuit boards. The production process includes circuit printing, component placement, reflow soldering, and quality inspection. During daily production, the system first collects initial process parameters such as temperature, pressure, flow rate, and rotational speed in real time using sensors deployed at each process step. These parameters undergo data cleaning to remove abnormal fluctuations and interference signals, and are then normalized according to unified standards to form a standardized parameter set for analysis.

[0075] Next, the system divides the entire production line into several key process stages based on historical production data and process flow charts. Taking the "component placement" stage as an example, standard parameters for this stage include placement head positioning accuracy, nozzle pressure, and component pickup time. The system compares the actual parameters for this stage in current production with the standard parameters to calculate the degree of deviation in product dimensions, solder joint quality, and other aspects. Furthermore, by analyzing historical data, the system calculates the correlation between each standardized process parameter and product deviation at that stage, identifying key parameters with significant impact on deviation. For example, it has been found that fluctuations in placement head temperature are highly correlated with solder joint defects.

[0076] During real-time production, the system uses currently collected process parameters and historical correlation weights to predict product deviations at each stage and further estimate the overall product defect rate. If a monitoring session reveals a defect rate exceeding a preset standard, the system automatically traces back to the process stage with the greatest impact. For example, by calculating the contribution of each stage to the defect rate, the system identified the "reflow soldering" stage as the primary cause of the current increase in defect rates. The system then conducts an in-depth analysis of the relevant process parameters for that stage and discovers that abnormal fluctuations in conveyor belt speed have the greatest impact on soldering uniformity.

[0077] The system prioritizes adjusting this key parameter and continuously monitors the production results. If the defect rate doesn't decrease significantly after the initial adjustment, the system will gradually optimize the parameter according to pre-set rules. If the number of consecutive adjustments reaches a predetermined limit and the target is still not met, the system will determine that there may be problems such as improper initial parameter settings or equipment anomalies, and immediately alert the process engineer, prompting manual intervention for troubleshooting.

[0078] Through this hierarchical correlation analysis and dynamic adjustment mechanism, the company's production line can quickly locate key parameters that affect product quality, while reducing manual intervention and significantly improving the stability of the production process and the efficiency of defective product warning.

[0079] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A multi-source process parameter mapping and supervision method based on a big data model, characterized by: The method comprises the following steps: S1. Collecting initial process parameters in the process production line through a sensor group, and processing the initial process parameters to obtain standardized process parameters; S2. Based on the process flow and historical data, the process production line is divided into process stages, the degree of product deviation in each process stage is analyzed, and the correlation between the standardized process parameters and the degree of product deviation in each process stage is calculated; S3. Analyze the correlation between the degree of product deviation and product defect rate at each process stage; S4. Predicting the degree of product deviation in the process stage of production based on standardized process parameters, and predicting the product defect rate in production based on the degree of product deviation in the process stage; S5. Real-time monitoring and prediction of product defect rate, and adjustment of standardized process parameters based on the predicted product defect rate; S6. Correct the standardized process parameters. When the standardized process parameters are judged to be abnormal, an alarm is sent to the process person in charge. In step S1, the initial process parameters in the process production line are collected by the sensor group, and the initial process parameters are denoised, filtered and normalized to obtain standardized process parameters. The standardized process parameters are {A1, A2, ..., A i ,…,A I }, where I represents the number of standardized process parameters, A i represents the i-th standardized process parameter; In step S2, based on the process flow and historical data, the process production line is divided into P process stages, and the p-th process stage is analyzed, where p=1,2,…,P, and the standard stage parameters of the product in the p-th process stage are {B1,B2,…,B q …,B Q }, where Q represents the number of standard stage parameters of the product produced in the pth process stage, B q represents the qth standard stage parameter of the product produced in the pth process stage, and the actual stage parameter of the product in the pth process stage is {b1, b2, ..., b q …,b Q }, b q It represents the qth actual stage parameter of the product produced in the pth process stage, and the stage product deviation degree W of the pth process stage is obtained. p : ； Calculate the standardized process parameter A i Correlation coefficient w for the degree of product deviation at the pth process stage ip : ； Where N represents the total N production lines in the historical data, n represents the nth production line in the historical data, and A i,n represents the i-th standardized process parameter in the n-th process production line, a i W represents the average value of the i-th standardized process parameter in N production lines, p,n It represents the product deviation degree of the pth process stage in the production of the nth process line, w p It represents the average value of the product deviation degree of the pth process stage in the N-times process production line production, and then obtains the standardized process parameter A i The relevant weight W for the degree of product deviation at the pth process stage ip : ； When W ip When <W0, determine the standardized process parameter A i Is not relevant to the pth process stage, otherwise, the standardized process parameter A i is the standardized process parameter related to the process stage, where W0 is the established correlation coefficient threshold; In step S3, a product defect rate impact analysis model is established: ； Among them, v p Indicates the stage product deviation degree W of the pth process stage p The coefficient of influence on product defect rate, W p,n It represents the product deviation degree of the pth process stage in the production of the nth process line, w p Y represents the average value of the product deviation degree of the pth process stage in the N-time process production line production, n represents the product defect rate in the nth process production line, y represents the average product defect rate in the Nth process production line, and then the stage product deviation degree W of the pth process stage is obtained p The relevant weight V for product defect rate p : 。 2. The multi-source process parameter mapping and supervision method based on a big data model according to claim 1 is characterized in that: In step S4, the product deviation degree F of the pth process stage in production is predicted based on the standardized process parameters. p The product deviation degree of the pth process stage is the sum of the standardized process parameters multiplied by the corresponding weights of the product deviation degree, where the standardized process parameters A p The weight W associated with the degree of product deviation ip Correspondingly, the product defect rate in production is predicted based on the product deviation degree in the production stage, and the product defect rate in production is the sum of the product deviation degree in production and the corresponding relevant weight of the product defect rate, where the product deviation degree in production F p and product defect rate V p correspond.

3. The multi-source process parameter mapping and supervision method based on a big data model according to claim 2 is characterized in that: In step S5, during the production process of the process line, the product defect rate G is monitored and predicted in real time. When G < G0, the system determines that the product defect rate meets the requirements, and G0 is the product defect rate threshold set by the system; when G ≥ G0, the system determines that the product defect rate does not meet the requirements, and monitors the impact value H of the pth process stage on the defect rate. p , H p =v p *W p , one by one into p = 1, 2, ..., P, and obtain the impact of P process stages on the defect rate {H1, H2, ..., H u ,…,H P }, where the maximum process stage’s impact on defect rate is H u , so the process stage corresponding to the maximum impact value is the u-th process stage, and the standardized process parameter A is obtained by analyzing the u-th process stage. i Impact value J on the degree of product deviation at the u-th process stage iu =w ip *A i , one by one into i = 1, 2, ..., I, and then get the impact value of I standardized process parameters on the product deviation degree of the u-th process stage {J 1u ,J 2u ,…,J xu ,…,J Iu }, where the maximum standardized process parameter’s influence on the product deviation degree of the u-th process stage is J xu , that is, the xth standardized process parameter has the greatest impact on the product deviation degree of the uth process stage, and the xth standardized process parameter is corrected to the optimal xth standardized process parameter.

4. The multi-source process parameter mapping and supervision method based on a big data model according to claim 3 is characterized by: In step S6, after the xth standardized process parameter is corrected, the predicted product defect rate is monitored again. The predicted product defect rate at this time is G1. If G1 < G0, wait for a set monitoring interval and then monitor the predicted product defect rate again. If G1 ≥ G0, correct the standardized process parameter again until the predicted product defect rate is less than the set product defect rate threshold.

5. The multi-source process parameter mapping and supervision method based on a big data model according to claim 4 is characterized in that: When the number of times the standardized process parameters are continuously adjusted is less than or equal to the preset number threshold, the counting continues until the predicted product defect rate is less than the preset product defect rate threshold or the number of times the standardized process parameters are continuously adjusted is greater than the preset number threshold; When the number of times the standardized process parameters are continuously adjusted exceeds a preset threshold, the initially set standardized process parameters are judged to be abnormal, and a parameter abnormality alarm is issued to the person in charge of the process.

6. A multi-source process parameter mapping supervision system based on a big data model, the system being applied to a multi-source process parameter mapping supervision method based on a big data model according to any one of claims 1 to 5, characterized in that: The system includes: a parameter acquisition and standardization module, a process stage division and parameter correlation analysis module, a defect rate impact analysis module, a parameter deviation analysis and defect rate prediction module, a real-time monitoring and parameter correction module, and a parameter anomaly processing module; The parameter acquisition and standardization module is used to acquire the initial process parameters in the process production line through the sensor group, and process the initial process parameters to obtain standardized process parameters; The process stage division and parameter correlation analysis module is used to divide the process production line into process stages based on the process flow and historical data, analyze the degree of product deviation in the process stages, and calculate the correlation between the standardized process parameters and the degree of product deviation in the process stages; The defect rate impact analysis module is used to analyze the correlation between the stage product deviation degree and the product defect rate in the process stage; The parameter deviation analysis and defect rate prediction module is used to predict the degree of product deviation in the process stage of production based on the standardized process parameters, and to predict the product defect rate in production based on the degree of product deviation in the process stage; The real-time monitoring and parameter correction module is used to monitor the predicted product defect rate G in real time and adjust the standardized process parameters according to the predicted product defect rate; The parameter abnormality processing module is used to correct the standardized process parameters and alert the process responsible person when it is determined that the standardized process parameters are abnormal.

7. The multi-source process parameter mapping and monitoring system based on a big data model according to claim 6 is characterized in that: The sensor group includes a temperature sensor, a pressure sensor, a flow sensor and a rotation speed sensor.

Images