Multi-target real-time monitoring system and method for rearview mirror lens manufacturing and processing

Abstract

This invention discloses a multi-target real-time monitoring system and method for rearview mirror lens manufacturing, relating to the field of automotive parts manufacturing technology. It aims to solve the technical problems of existing monitoring systems being unable to effectively distinguish between false anomalies caused by dynamic environmental interference and real defects in the rearview mirror lens pressing process, and lacking a precise process adjustment mechanism based on quantitative data. The system includes a data acquisition module for real-time acquisition of first, second, and third monitoring data for the pressing process. The first monitoring data is visual image data of the lens-plate pressing state; the second monitoring data is dynamic environmental data of the pressing station; and the third monitoring data is pressing process operation data. This invention has the advantages of accurately identifying environmental interference and coupling types in the rearview mirror lens pressing process, distinguishing between false anomalies and real defects, and reducing monitoring misjudgment rate and pressing failure rate.

Description

Technical Field

[0001] This invention relates to the field of automotive parts manufacturing technology, and more specifically, to a multi-target real-time monitoring system and method for the production and processing of rearview mirror lenses. Background Technology

[0002] In the manufacturing process of automotive rearview mirror lenses, the crimping process is a core step that determines the assembly accuracy and performance of the lenses. This process must ensure that the lens and the support plate achieve precise bonding and fixation under the action of automated equipment. However, the production environment of the crimping station is easily affected by a variety of dynamic factors, such as real-time fluctuations in temperature and humidity, changes in dust concentration in the air, instability in light intensity, and abnormal airflow speed. Moreover, these environmental parameters often exhibit coupled interference from multiple dimensions, posing significant challenges to crimping quality monitoring and process control.

[0003] Existing monitoring technologies for the rearview mirror lens pressing process generally suffer from problems such as limited monitoring dimensions and crude judgment logic. On the one hand, they can only independently collect pressing visual images or single process parameters, without establishing a correlation analysis mechanism between environmental parameters, process parameters, and visual data. This makes it difficult to distinguish between "false anomalies caused by environmental interference" (such as image distortion caused by dust adhering to the lens, or temporary fluctuations in process parameters caused by temperature and humidity fluctuations) and "real defects in the pressing process itself" (such as misalignment of the lens and the support plate, loose bonding, or lens damage), frequently resulting in misjudgments or missed judgments. On the other hand, when anomalies are detected, process parameter adjustments rely on experience-based operations, lacking quantitative adjustment basis based on historical data and interference intensity. This makes it impossible to generate precise adjustment schemes for single environmental interference or multi-environment coupled interference, leading to adjustment lag and insufficient accuracy. This not only affects the stability of pressing quality but may also reduce production efficiency due to unnecessary downtime for adjustments. In view of this, we propose a multi-target real-time monitoring system and method for rearview mirror lens production and processing. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-target real-time monitoring system and method for the production and processing of rearview mirror lenses, so as to solve the technical problems that existing monitoring systems cannot effectively distinguish between false anomalies caused by dynamic environmental interference in the rearview mirror lens pressing process and real defects in the pressing process, and lack a precise process adjustment mechanism based on quantitative data.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a multi-target real-time monitoring system for rearview mirror lens production and processing, comprising: The data acquisition module is used to collect the first monitoring data, the second monitoring data and the third monitoring data of the crimping process in real time. The first monitoring data is the visual image data of the crimping status of the lens and the tray, the second monitoring data is the dynamic environmental data of the crimping station, and the third monitoring data is the crimping process operation data. The storage module stores historical production data and the initial adaptation model. The historical production data includes historical environmental interference characteristics, historical crimping defect characteristics, historical environment-process correlation data, and historical interference coupling characteristics. The initial adaptation model is an environment-process basic correlation model built based on knowledge of similar lens crimping processes. The processing module calls the data and model from the storage module to perform fusion judgment on the first, second, and third monitoring data, distinguishing between false anomalies caused by single dynamic environmental interference, coupling of multiple environmental interferences, and real defects in the crimping process, and assisting in judgment based on the initial adaptation model when historical data is insufficient. The control module generates a set of candidate adjustment parameters based on the judgment results of the processing module; The decision-making module prioritizes the candidate adjustment parameter set, selects the optimal adjustment scheme and executes it, and simultaneously receives the post-adjustment monitoring data fed back by the data acquisition module to form a closed-loop control of the pressing quality.

[0006] Preferably, the data acquisition module includes a visual acquisition unit, an environmental acquisition unit, a process acquisition unit, and an interference coupling sensing unit; The visual acquisition unit is set in the monitoring area of ​​the crimping station and is used to capture the crimping process of the lens and the tray, and output the first monitoring data, including the alignment image of the lens and the tray, the crimping gap image and the surface condition image of the lens. The environmental acquisition unit is deployed around the crimping station to capture the dynamic changes of the crimping environment in real time and output the second monitoring data, including the temperature and humidity data, dust concentration data, light intensity data and airflow speed data of the crimping station. The process acquisition unit is connected to the control terminal of the pressing equipment to acquire the operating parameters of the pressing equipment and output the third monitoring data, including the temperature data of each section of the heating element, the pressure data of the pressing cylinder, the pressing time data, and the conveyor belt speed data. The interference coupling sensing unit is used to identify the coordinated change characteristics of multiple environmental parameters in the second monitoring data, and output the coupling type and coupling strength data of multiple environmental interferences. The coupling type includes temperature and humidity-dust coupling and light-airflow coupling.

[0007] Preferably, the storage module includes a historical interference library, a historical defect library, an association database, a coupling feature library, and a model storage library; The historical interference database stores historical environmental interference characteristics, including the environmental data range, visual image noise characteristics, and interference duration corresponding to a single environmental interference. The historical defect database stores historical crimping defect features, including visual image features, process parameter deviation ranges, and defect impact characteristics corresponding to lens and tray misalignment, loose crimping, and lens breakage. The associated database stores historical environment-process correlation data, including records of process parameter adjustments and quality results under single environmental disturbances; The coupling feature library stores historical interference coupling features, including the parameter coordination change law corresponding to multiple environmental interference coupling types, visual image distortion features under coupling state, and adapted process parameter combinations. The model repository stores the initial adaptation model, including a table of environmental parameter thresholds and process parameter baseline values ​​constructed based on the process manual.

[0008] Preferably, the fusion determination process of the processing module includes interference type identification, false anomaly elimination, real defect determination, and data insufficiency adaptation; In the interference type identification process, the processing module compares the second monitoring data and the coupling data output by the interference coupling sensing unit with the historical interference database and the coupling feature database to identify whether the current interference is single-environment interference or multi-environment interference coupling, and determines the interference type and intensity. In the process of eliminating false anomalies, the processing module calls the corresponding historical interference features or coupling features according to the identified interference type, and removes the image distortion area and parameter fluctuation noise caused by environmental interference from the first monitoring data. In the determination of real defects, the processing module compares the first monitoring data after excluding false anomalies with the visual features of the historical defect database, and at the same time compares the third monitoring data with the process features of the historical defect database. When both match, it is determined to be a real defect. In the data-sufficient adaptation process, if the current interference type or coupling type is not recorded in the historical data, the processing module calls the initial adaptation model and uses the basic association rules of similar lenses to help determine the false anomaly and the real defect. To quantify the results of interference type identification and actual defect determination, the processing module implements the determination logic through the following algorithm formula: Interference intensity calculation: For interference in a single environment, the interference intensity... The formula is calculated by weighting the deviation rate between the current environmental parameters and the standard parameters and the duration of the disturbance: ; when When the interference is significant, it is considered a single, significant disturbance. in, For single environmental interference intensity, As the parameter deviation rate weight, These are the current environmental parameter values. These are standard values ​​for environmental parameters. Weighted by the duration of the disturbance. For the duration of the interference, For a single interference intensity threshold; Coupling interference determination: Coupling strength is determined for interference from multiple environments. The formula for calculating the synergy coefficient of the deviations of various environmental parameters is as follows: ; when This was determined to be significant coupling interference; in, For multi-environment interference coupling strength, For the first The weights of each environmental parameter, For the first Current values ​​of each environmental parameter For the first Standard values ​​for each environmental parameter For the first , The synergy coefficient of each environmental parameter This is the threshold for coupling interference strength; True Defect Determination: Defect Matching Degree The product of visual feature matching degree and process feature matching degree is calculated using the following formula: ; when When this occurs, it is determined to be a genuine defect; in, For defect matching degree, For visual feature matching degree, For process feature matching degree, This is the defect matching threshold.

[0009] Preferably, the processing module further includes a dynamic association update submodule. The dynamic association update submodule is used to associate and store the current interference type, coupling type, process parameter adjustment record and pressing quality result to the corresponding database after the data shortage adaptation step is executed. At the same time, iteratively optimize the initial adaptation model based on the new data. When the number of records of a certain interference type or coupling type reaches a preset threshold, it automatically migrates it from the management scope of the initial adaptation model to the historical database.

[0010] Preferably, the process by which the control module generates a set of candidate adjustment parameters includes single disturbance adjustment and coupled disturbance adjustment; In the single interference adjustment, if the processing module identifies it as a single environmental interference without a real defect, the control module calls the process parameter adaptation range corresponding to the interference type in the associated database to generate at least 3 sets of differentiated adjustment parameter combinations; if a real defect exists, the control module generates a targeted adjustment parameter combination based on the repair scheme of the corresponding defect in the historical defect database. In the coupling interference adjustment, if the processing module identifies multi-environment interference coupling, the control module calls the process parameter combination record corresponding to the coupling type in the coupling feature library, and generates a multi-dimensional collaborative adjustment parameter combination by combining the influence weight of each environmental parameter. The collaborative adjustment parameter combination includes the linkage adjustment scheme of heating temperature, pressing pressure and pressing time. To accurately generate adjustment parameters, the control module calculates the adjustment amount using the following algorithm formula: Adjustment amount of process parameters for a single disturbance: Adjustment amount of process parameters for a single environmental disturbance. Based on interference strength The historical correlation coefficient is determined by the following formula: ; ; When generating differentiated combinations, by adjusting The value of , Get at least 3 sets ; in, This refers to the adjustment amount of process parameters under a single disturbance. The historical environment-process correlation coefficient. For single environmental interference intensity, To adjust the process parameter values, These are the current process parameter values. This represents the deviation of the correlation coefficient. Coupling interference process parameter adjustment amount: Multi-dimensional coordinated adjustment amount to address coupling interference. Based on coupling strength The weights of each parameter are determined by the following formula: ; ; in, For coupling interference, the first Adjustment amount of each process parameter For the first The correlation coefficient between each process parameter and coupling interference. For multi-environment interference coupling strength, For the first The influence weight of each process parameter For coupling interference, the first The values ​​after adjusting each process parameter For the first Current values ​​of each process parameter.

[0011] Preferably, the priority ranking process of the decision-making module includes the construction of evaluation indicators, weight allocation, and scheme selection; In the construction of the evaluation indicators, the decision-making module constructs an evaluation indicator system including the degree of improvement in crimping quality, the range of parameter adjustment, the rate of change in energy consumption, and the speed of adjustment response. In the weight allocation, the decision module dynamically allocates indicator weights according to the current production needs. When pursuing quality stability, the weight of the crimping quality improvement is increased; when pursuing production efficiency, the weight of the adjustment response speed is increased. In the preferred scheme, the decision module quantifies and scores the candidate adjustment parameter set based on evaluation indicators and weights, selects the adjustment scheme with the highest comprehensive score as the optimal scheme, and sends it to the crimping device for execution. To quantitatively select the optimal solution, the decision-making module calculates a comprehensive score using the following algorithm formula: Standardized score for each single indicator: Each evaluation indicator is standardized to eliminate the influence of dimensions. The formula is: Positive indicators: ; Negative indicators: ; in, For the first Standardized scores of each evaluation indicator For the first The actual value of each evaluation indicator For the first The historical maximum value of each evaluation indicator For the first The historical minimum value of each evaluation indicator; Overall Score: A weighted sum of standardized scores based on indicator weights, using the following formula: ; choose The largest candidate solution is considered the optimal solution. If multiple solutions exist... If they are the same, prioritize adjusting the response speed. The largest solution in, The overall score of the candidate adjustment schemes. For the first The weight of each evaluation indicator.

[0012] Preferably, the system further includes an emergency intervention module, which is connected to the decision-making module and the processing module; If the optimal adjustment scheme output by the decision module is executed, and the processing module determines that the crimping quality still does not meet the standard and the degree of defect exceeds the preset severe threshold, the emergency intervention module will automatically trigger graded emergency measures: the first-level measure is to suspend the crimping of the current batch of lenses; the second-level measure is to link the upstream material supply system to suspend material conveying; at the same time, an emergency report is generated, which includes the interference evolution curve, the adjustment parameters of each time, and the defect spread risk assessment.

[0013] Preferably, the data acquisition module further includes a time series analysis subunit, which is used to perform time series modeling on the second monitoring data and the third monitoring data to predict the environmental parameter change trend and process parameter drift trend within a preset time period in the future. The processing module receives the prediction results from the timing analysis subunit, identifies potential interference risks in advance, and provides a basis for the control module to generate a set of preventive adjustment parameters. To achieve parameter trend prediction, the time series analysis subunit constructs a prediction model using the following algorithm: Environmental parameter trend prediction: Predicting future trends using a linear regression model. Environmental parameter values ​​within the time period The formula is: ; when Exceeding At that time, it was determined to be a potential interference risk; in, For the future Predicted values ​​of environmental parameters within the time period These are the current environmental parameter values. For regression coefficients, To predict duration, This refers to the allowable fluctuation range of environmental parameters; Process parameter drift prediction: based on environmental parameter predictions Predicting process parameter drift using historical correlation models The formula is: ; Control module based on Generate preventive adjustment parameters To offset the effects of parameter drift in advance; in, For the future The amount of process parameter drift within a time period. This is the process parameter drift coefficient. For preventative adjustment parameters, This represents the current process parameter value.

[0014] A multi-target real-time monitoring method for rearview mirror lens manufacturing includes the following steps: S1. Data acquisition steps: The first monitoring data, the second monitoring data, the third monitoring data, and the interference coupling data are collected in real time through the data acquisition module. The time series analysis subunit predicts the changing trends of environmental and process parameters and then outputs the results. S2. Historical and Model Calling Steps: The processing module calls the historical production data and initial adaptation model from the storage module; S3, Fusion Judgment Steps: The processing module identifies the type of interference, eliminates false anomalies based on historical data or the initial adaptation model, determines the real defects, and at the same time dynamically associates and updates the sub-modules to record new data and iterate the model. S4. Candidate parameter generation steps: Based on the judgment results, the control module generates differentiated adjustment parameter combinations for single disturbances and multi-dimensional collaborative adjustment parameter combinations for coupled disturbances, forming a candidate adjustment parameter set. S5. Optimal Solution Decision-Making Steps: The decision-making module constructs an evaluation index system and dynamically allocates weights. After scoring the candidate adjustment parameter set, it selects the optimal adjustment solution. S6. Closed-loop control steps: After executing the optimal solution, collect the monitoring data after adjustment and re-evaluate. If the standard is met, update the historical data. If the standard is not met and the defect is serious, trigger emergency intervention. S7. Continuous optimization steps: Repeat steps S1-S7, combine the timing prediction results to achieve preventive adjustments, and continuously optimize the accuracy of crimping quality control.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention constructs an integrated data acquisition module that combines "visual acquisition, environmental acquisition, process acquisition, and interference coupling perception." Combined with the processing module's fusion judgment logic based on historical data and the initial adaptation model, it can accurately identify the type and intensity of environmental interference, effectively eliminate false anomalies, and accurately determine real defects. Simultaneously, the control module generates a quantified set of candidate adjustment parameters based on the interference type, and the decision module selects the optimal solution through multi-index weighted scoring, achieving precise adjustment of process parameters, significantly reducing the monitoring misjudgment rate and crimping failure rate, and improving production efficiency.

[0016] 2. The dynamic association update submodule of the processing module of this invention can associate and store new interference types, coupling characteristics, adjustment records and quality results in the database after each adaptation judgment and adjustment, and iteratively optimize the initial adaptation model. When the records of a certain type of interference / coupling reach the threshold, they are automatically transferred to the historical database management, which solves the problem of "insufficient adaptability of the initial adaptation model to new interference scenarios", realizes the dynamic upgrading of the monitoring and adjustment mechanism, and improves the system's adaptability to diverse production scenarios.

[0017] 3. The time series analysis subunit included in the data acquisition module of this invention can predict the trend of environmental parameter changes and process parameter drift within a preset time period through time series modeling. The processing module identifies potential interference risks in advance based on the prediction results, and the control module generates preventive adjustment parameters accordingly. This solves the problem that "post-event adjustment cannot avoid short-term quality fluctuations", realizes the transformation from "passively responding to interference" to "actively preventing risks", and further improves the stability of crimping quality and the continuity of the production process. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the system framework of the present invention.

[0019] Figure 2 This is a schematic diagram of the system framework hierarchy and core process of the present invention.

[0020] Figure 3 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0021] To facilitate understanding of the technical solution of the present invention by those skilled in the art, the technical solution of the present invention will now be further described in conjunction with the accompanying drawings.

[0022] Example 1, such as Figure 1 and Figure 2 As shown, the present invention provides a multi-target real-time monitoring system for rearview mirror lens manufacturing and processing, comprising: The data acquisition module is used to collect the first monitoring data, the second monitoring data and the third monitoring data of the crimping process in real time. The first monitoring data is the visual image data of the crimping status of the lens and the tray, the second monitoring data is the dynamic environmental data of the crimping station, and the third monitoring data is the crimping process operation data. The storage module stores historical production data and the initial adaptation model. The historical production data includes historical environmental interference characteristics, historical crimping defect characteristics, historical environment-process correlation data, and historical interference coupling characteristics. The initial adaptation model is an environment-process basic correlation model built based on knowledge of similar lens crimping processes. The processing module calls the data and model from the storage module to perform fusion judgment on the first, second, and third monitoring data, distinguishing between false anomalies caused by single dynamic environmental interference, coupling of multiple environmental interferences, and real defects in the crimping process, and assisting in judgment based on the initial adaptation model when historical data is insufficient. The control module generates a set of candidate adjustment parameters based on the judgment results of the processing module; The decision-making module prioritizes the candidate adjustment parameter set, selects the optimal adjustment scheme and executes it, and simultaneously receives the post-adjustment monitoring data fed back by the data acquisition module to form a closed-loop control of the pressing quality.

[0023] In an embodiment of the present invention, the data acquisition module includes a visual acquisition unit, an environmental acquisition unit, a process acquisition unit, and an interference coupling sensing unit; The visual acquisition unit is set in the monitoring area of ​​the crimping station and is used to capture the crimping process of the lens and the tray, and output the first monitoring data, including the alignment image of the lens and the tray, the crimping gap image and the surface condition image of the lens. The environmental acquisition unit is deployed around the crimping station to capture the dynamic changes of the crimping environment in real time and output the second monitoring data, including the temperature and humidity data, dust concentration data, light intensity data and airflow speed data of the crimping station. The process acquisition unit is connected to the control terminal of the pressing equipment to acquire the operating parameters of the pressing equipment and output the third monitoring data, including the temperature data of each section of the heating element, the pressure data of the pressing cylinder, the pressing time data, and the conveyor belt speed data. The interference coupling sensing unit is used to identify the coordinated change characteristics of multiple environmental parameters in the second monitoring data, and output the coupling type and coupling strength data of multiple environmental interferences. The coupling type includes temperature and humidity-dust coupling and light-airflow coupling.

[0024] In an embodiment of the present invention, the storage module includes a historical interference library, a historical defect library, an association database, a coupling feature library, and a model storage library; The historical interference database stores historical environmental interference characteristics, including the environmental data range, visual image noise characteristics, and interference duration corresponding to a single environmental interference. The historical defect database stores historical crimping defect features, including visual image features, process parameter deviation ranges, and defect impact characteristics corresponding to lens and tray misalignment, loose crimping, and lens breakage. The associated database stores historical environment-process correlation data, including records of process parameter adjustments and quality results under single environmental disturbances; The coupling feature library stores historical interference coupling features, including the parameter coordination change law corresponding to multiple environmental interference coupling types, visual image distortion features under coupling state, and adapted process parameter combinations. The model repository stores the initial adaptation model, including a table of environmental parameter thresholds and process parameter baseline values ​​constructed based on the process manual. In an embodiment of the present invention, the fusion determination process of the processing module includes interference type identification, false anomaly elimination, real defect determination, and data insufficiency adaptation. In the interference type identification process, the processing module compares the second monitoring data and the coupling data output by the interference coupling sensing unit with the historical interference database and the coupling feature database to identify whether the current interference is single-environment interference or multi-environment interference coupling, and determines the interference type and intensity. In the process of eliminating false anomalies, the processing module calls the corresponding historical interference features or coupling features according to the identified interference type, and removes the image distortion area and parameter fluctuation noise caused by environmental interference from the first monitoring data. In the determination of real defects, the processing module compares the first monitoring data after excluding false anomalies with the visual features of the historical defect database, and at the same time compares the third monitoring data with the process features of the historical defect database. When both match, it is determined to be a real defect. In the data-sufficient adaptation process, if the current interference type or coupling type is not recorded in the historical data, the processing module calls the initial adaptation model and uses the basic association rules of similar lenses to help determine the false anomaly and the real defect. To quantify the results of interference type identification and actual defect determination, the processing module implements the determination logic through the following algorithm formula: Interference intensity calculation: For interference in a single environment, the interference intensity... The formula is calculated by weighting the deviation rate between the current environmental parameters and the standard parameters and the duration of the disturbance: ; when When the interference is significant, it is considered a single, significant disturbance. in, The intensity of a single environmental interference (quantifying the degree of impact of a single environmental interference on crimping). Weighting the parameter deviation rate (adjusting the importance of "degree of deviation of environmental parameters" in the calculation of interference intensity); This represents the current environmental parameter values ​​(real-time collected environmental data from the crimping station, such as temperature, humidity, and dust concentration). These are the standard values ​​for environmental parameters (the baseline values ​​for environmental parameters required for the crimping process). Weighting for interference duration (adjusting the importance of "interference duration" in interference intensity calculation); Duration of disturbance (the time from when environmental parameters were detected to deviate from the standard value to the current moment, in minutes); The threshold for the intensity of a single interference (the critical value for determining whether a single interference needs to be addressed as a priority). By weighting the "parameter deviation rate" and "interference duration" in two dimensions, the impact of a single environmental interference on the crimping process is quantified. First, the proportion of the current environmental parameter deviating from the standard value is calculated, and then the duration of the interference is combined with the weight allocation to highlight the contribution of the two to the interference intensity. Finally, by comparing with the preset threshold, it is determined whether the interference needs to be addressed. Coupling interference determination: Coupling strength is determined for interference from multiple environments. The formula for calculating the synergy coefficient of the deviations of various environmental parameters is as follows: ; when This was determined to be significant coupling interference; in, The coupling strength of multiple environmental interferences (quantifying the degree of superimposed influence of the coordinated changes of multiple environmental parameters on the crimping). For the first Weights of environmental parameters (adjusting the contribution ratio of different environmental parameters in coupling interference); For the first Current value of an environmental parameter (real-time collected data of a specific environmental parameter, such as temperature or airflow speed); For the first A standard value for an environmental parameter (a baseline value for a specific environmental parameter); For the first , The synergy coefficient of an environmental parameter (reflects the degree of mutual influence when two environmental parameters change in tandem, with a value of 0-1, determined by historical data). This is the coupling interference strength threshold (the critical value for determining whether coupling interference requires special attention). First, calculate the deviation rate of each environmental parameter and assign it a corresponding weight. Then, multiply it by the coordination coefficient between parameters (to reflect the superimposed effect of parameter linkage on the crimping). The overall strength of the coupling interference is obtained by summing, so as to distinguish between "multiple independent interferences" and "coordinated coupling interference" and avoid misjudgment caused by ignoring parameter linkage. True Defect Determination: Defect Matching Degree The product of visual feature matching degree and process feature matching degree is calculated using the following formula: ; when When this occurs, it is determined to be a genuine defect; in, Defect matching degree (quantifying the degree of matching between the current crimping state and historical defect characteristics). Visual feature matching degree (the similarity between the current crimped visual image and the visual features of historical defects, with a value of 0-1, where 1 indicates a perfect match). This represents the process feature matching degree (the degree of similarity between the current process parameters and the historical defect process features, with a value of 0-1, where 1 indicates a perfect match). This is the defect matching threshold (the critical value for determining whether a defect is real). By multiplying the "visual feature matching degree" and the "process feature matching degree", "dual feature cross-verification" is achieved. Only when the visually observed pressing state (such as bonding gap and alignment) and process parameters (such as heating temperature and pressing pressure) both match the historical defect characteristics are they judged as real defects, thus eliminating the monitoring error caused by "single feature mismatch". The above algorithm formula can accurately quantify the intensity of environmental interference and the degree of defect matching, which not only avoids missed or misjudged interference caused by "qualitative judgment", but also ensures the accuracy of defect judgment through dual feature cross-validation. It effectively solves the problem of misjudgment in monitoring under dynamic environmental interference, provides a reliable judgment basis for subsequent process parameter adjustment, and ensures the stability of crimping quality.

[0025] In an embodiment of the present invention, the processing module further includes a dynamic association update submodule. The dynamic association update submodule is used to associate and store the current interference type, coupling type, process parameter adjustment record and pressing quality result to the corresponding database after the data shortage adaptation step is executed. At the same time, iteratively optimizes the initial adaptation model based on the new data. When the number of records of a certain interference type or coupling type reaches a preset threshold, it automatically migrates it from the management scope of the initial adaptation model to the historical database to improve the accuracy of subsequent judgments.

[0026] In embodiments of the present invention, the process by which the control module generates a set of candidate adjustment parameters includes single disturbance adjustment and coupled disturbance adjustment; In the single interference adjustment, if the processing module identifies it as a single environmental interference without a real defect, the control module calls the process parameter adaptation range corresponding to the interference type in the associated database to generate at least 3 sets of differentiated adjustment parameter combinations; if a real defect exists, the control module generates a targeted adjustment parameter combination based on the repair scheme of the corresponding defect in the historical defect database. In the coupling interference adjustment, if the processing module identifies multi-environment interference coupling, the control module calls the process parameter combination record corresponding to the coupling type in the coupling feature library, and generates a multi-dimensional collaborative adjustment parameter combination by combining the influence weight of each environmental parameter. The collaborative adjustment parameter combination includes the linkage adjustment scheme of heating temperature, pressing pressure and pressing time. To accurately generate adjustment parameters, the control module calculates the adjustment amount using the following algorithm formula: Adjustment amount of process parameters for a single disturbance: Adjustment amount of process parameters for a single environmental disturbance. Based on interference strength The historical correlation coefficient is determined by the following formula: ; ; When generating differentiated combinations, by adjusting The value of , Get at least 3 sets ; in, This refers to the adjustment amount of process parameters under a single disturbance (the incremental adjustment of process parameters required for a single environmental disturbance). Historical environment-process correlation coefficient (extracted from historical data, representing the adjustment range of process parameters corresponding to a single unit of interference intensity). For single environmental interference intensity; The adjusted process parameter values ​​(actual values ​​of the adjusted process parameters are executed). This is the current process parameter value (the real-time value of the process parameters before adjustment). This is the correlation coefficient deviation (used to fine-tune the correlation coefficient for generating differentiated adjustment schemes). Based on a single disturbance intensity, combined with the historical environment-process correlation coefficient (which reflects the range of process parameters that need to be adjusted per unit disturbance intensity), the increment of the process parameters that need to be adjusted is directly calculated. Then, multiple sets of differentiated adjustment schemes are generated by fine-tuning the correlation coefficient, providing sufficient candidates for the subsequent selection of the optimal scheme. Coupling interference process parameter adjustment amount: Multi-dimensional coordinated adjustment amount to address coupling interference. Based on coupling strength The weights of each parameter are determined by the following formula: ; ; in, For coupling interference, the first The adjustment amount of the first process parameter (for coupling interference, the first...) (The incremental adjustment required for each process parameter); For the first The correlation coefficient between the process parameter and coupling interference (extracted from historical data, representing the effect of unit coupling interference intensity on the first process parameter) (the influence range of each process parameter). Coupling strength for multiple environmental interferences; For the first The influence weight of the first process parameter (adjusted for the first) The importance of each process parameter in coupling interference modulation, with a weight sum of 1); For coupling interference, the first The value of the adjusted process parameter (the first value after performing coordinated adjustment) (actual values ​​of each process parameter); For the first Current value of the process parameter (before adjustment) (Real-time values ​​of each process parameter). Based on the coupling interference strength, the basic adjustment range is first determined by the correlation coefficient between process parameters and coupling interference. Then, it is multiplied by the parameter influence weight (reflecting the importance of different process parameters to the crimping quality) to obtain the coordinated adjustment amount of each process parameter. This ensures that the multi-parameter linkage adjustment can accurately respond to coupling interference and avoid parameter adjustment conflicts. The above algorithm formula is based on historical data and previous interference judgment results to ensure that the adjustment amount of process parameters is accurately matched with the interference intensity. This avoids quality problems caused by "insufficient adjustment" and prevents resource waste caused by "over-adjustment". At the same time, by generating multiple sets of differentiated or coordinated adjustment schemes, it provides support for the subsequent decision-making module to select the optimal scheme, effectively solving the problems of low accuracy and easy conflict in process parameter adjustment under dynamic environmental interference, and further ensuring the pressing quality.

[0027] In an embodiment of the present invention, the priority ranking process of the decision-making module includes evaluation index construction, weight allocation, and scheme optimization; In the construction of the evaluation indicators, the decision-making module constructs an evaluation indicator system including the degree of improvement in crimping quality, the range of parameter adjustment, the rate of change in energy consumption, and the speed of adjustment response. In the weight allocation, the decision module dynamically allocates indicator weights according to the current production needs. When pursuing quality stability, the weight of the crimping quality improvement is increased; when pursuing production efficiency, the weight of the adjustment response speed is increased. In the preferred scheme, the decision module quantifies and scores the candidate adjustment parameter set based on evaluation indicators and weights, selects the adjustment scheme with the highest comprehensive score as the optimal scheme, and sends it to the crimping device for execution. To quantitatively select the optimal solution, the decision-making module calculates a comprehensive score using the following algorithm formula: Standardized score for each single indicator: Each evaluation indicator is standardized to eliminate the influence of dimensions. The formula is: Positive indicators (such as the degree of quality improvement): ; Negative indicators (such as adjustment range, energy consumption change rate): ; in, For the first Standardized scores of each evaluation indicator (indicator values ​​after dimensionless measurement, ranging from 0 to 1). For the first The actual value of each evaluation indicator (candidate solution in the th) Specific performance data for each indicator, such as quality improvement rate and energy consumption change rate). For the first The historical maximum value of each evaluation indicator (the largest data value that the indicator has ever appeared in historical production). For the first The historical minimum value of an evaluation indicator (the lowest data value that the indicator has ever recorded in historical production).

[0028] Overall Score: A weighted sum of standardized scores based on indicator weights, using the following formula: ; choose The largest candidate solution is considered the optimal solution. If multiple solutions exist... If they are the same, prioritize adjusting the response speed. The largest solution in, The overall score of the candidate adjustment scheme (quantifying the overall quality of the candidate scheme, with a value of 0-1); For the first The weight of the evaluation indicator (adjusted) The importance of each indicator in the comprehensive evaluation, with a weight sum of 1, is determined by production demand. Based on current production needs (such as prioritizing quality or efficiency), weights are assigned to each standardized indicator. The comprehensive score of each candidate adjustment scheme is obtained by weighted summation. The scheme with the highest score is the optimal scheme. If the scores are the same, the adjustment response speed is used as a secondary criterion to ensure that the scheme selection meets both production goals and efficiency. By standardizing indicators, we can eliminate differences in dimensions and avoid evaluation bias caused by different ranges of indicator values. By dynamically allocating weights, we can adapt to different production needs and ensure that the selected solutions are consistent with actual production goals. Finally, by quantifying the merits of solutions through comprehensive scoring, we can achieve scientific screening of candidate adjustment solutions, solve the problem of "subjectivity in solution selection during multi-objective adjustment", improve the rationality and pertinence of process parameter adjustment, and further ensure the balance between crimping quality and production efficiency.

[0029] In an embodiment of the present invention, the system further includes an emergency intervention module, which is connected to the decision-making module and the processing module; If the optimal adjustment scheme output by the decision module is executed, and the processing module determines that the crimping quality still does not meet the standard and the degree of defect exceeds the preset severe threshold, the emergency intervention module will automatically trigger graded emergency measures: the first-level measure is to suspend the crimping of the current batch of lenses and retain the workstation status for manual inspection; the second-level measure is to link the upstream material supply system to suspend material conveying to avoid the accumulation of defects; at the same time, an emergency report is generated, which includes the interference evolution curve, the adjustment parameters of each time, and the defect propagation risk assessment.

[0030] In an embodiment of the present invention, the data acquisition module further includes a time series analysis subunit, which is used to perform time series modeling on the second monitoring data and the third monitoring data to predict the environmental parameter change trend and process parameter drift trend within a future preset time period. The processing module receives the prediction results from the timing analysis subunit, identifies potential interference risks in advance, and provides a basis for the control module to generate a set of preventive adjustment parameters, thereby achieving an upgrade from passive adjustment to active prevention. To achieve parameter trend prediction, the time series analysis subunit constructs a prediction model using the following algorithm: Environmental parameter trend prediction: Predicting future trends using a linear regression model. Environmental parameter values ​​within the time period The formula is: ; (i.e., quadratic regression model) ),when Exceeding At that time, it was determined to be a potential interference risk; in, For the future Predicted values ​​of environmental parameters within a time period (expected environmental parameter values ​​at a future point in time); The current environmental parameter values ​​(real-time environmental parameter data at the time of prediction); The regression coefficients are derived from fitting historical environmental parameter time series data. Corresponding coefficient of the first term, The coefficients of the quadratic term reflect the trend of parameter changes. For the predicted duration (the expected length of future time, in minutes); The allowable fluctuation range of environmental parameters (the acceptable fluctuation range of environmental parameters for the crimping process); A quadratic regression model is used to predict the trend of environmental parameter changes within a certain time period in the future, based on the regression coefficients fitted between the current environmental parameter values ​​and historical time series data. The quadratic term reflects the possible nonlinear changes of environmental parameters (such as accelerated rise or fall of temperature and humidity), avoiding trend misjudgment caused by linear prediction. Furthermore, by comparing with the allowable fluctuation range, potential disturbance risks can be identified in advance. Process parameter drift prediction: based on environmental parameter predictions Predicting process parameter drift using historical correlation models The formula is: ; Control module based on Generate preventive adjustment parameters To offset the effects of parameter drift in advance; in, For the future The drift of process parameters within a time period (the extent to which process parameters deviate from the baseline value due to environmental changes at a certain point in the future). This is the process parameter drift coefficient (extracted from historical data, representing the drift magnitude of process parameters caused by a unit environmental parameter deviation). Preventive adjustment parameters (process parameter values ​​that need to be set in advance to offset future process parameter drift); The current process parameter values ​​(real-time process parameter data during prediction); Based on the predicted environmental parameter values, combined with the process parameter drift coefficient (which reflects the drift range of process parameters caused by a unit deviation of environmental parameters), the potential drift of process parameters in the future is calculated, and then preventive adjustment parameters are generated to offset the impact of environmental changes on process parameters in advance, thus achieving an upgrade from "passive response" to "proactive prevention". By using a quadratic regression model to accurately predict the changing trends of environmental parameters, potential interference risks can be identified in advance, avoiding delays in response to sudden interference. Based on the prediction results, preventive adjustment parameters are generated to offset the impact of environmental changes on process parameters in advance, solving the problem of "quality fluctuations caused by readjustment after interference occurs", realizing proactive prevention and control of the crimping process, and further improving the stability of crimping quality and the continuity of the production process.

[0031] Example 2, as follows Figure 3 As shown, the present invention provides a multi-target real-time monitoring method for rearview mirror lens manufacturing, comprising the following steps: S1. Data Acquisition Steps: The data acquisition module collects the first monitoring data (visual image data of lens and tray pressing status), the second monitoring data (dynamic environmental data of pressing station), the third monitoring data (pressing process operation data), and interference coupling data in real time. After the time series analysis subunit predicts the changing trends of environmental and process parameters, the data is output. S2. Historical and Model Calling Steps: The processing module calls the historical production data (including historical interference, defects, correlation and coupling features) and the initial adaptation model from the storage module; S3. Fusion Judgment Steps: The processing module identifies the type of interference (single / coupled), eliminates false anomalies based on historical data or the initial adaptation model, determines the real defects, and at the same time dynamically associates and updates the sub-modules to record new data and iterates the model. S4. Candidate parameter generation steps: Based on the judgment results, the control module generates differentiated adjustment parameter combinations for single disturbances and multi-dimensional collaborative adjustment parameter combinations for coupled disturbances, forming a candidate adjustment parameter set. S5. Optimal Solution Decision-Making Steps: The decision-making module constructs an evaluation index system and dynamically allocates weights. After scoring the candidate adjustment parameter set, it selects the optimal adjustment solution. S6. Closed-loop control steps: After executing the optimal solution, collect the monitoring data after adjustment and re-evaluate. If the standard is met, update the historical data. If the standard is not met and the defect is serious, trigger emergency intervention. S7. Continuous optimization steps: Repeat steps S1-S7, combine the timing prediction results to achieve preventive adjustments, and continuously optimize the accuracy of crimping quality control.

[0032] The embodiments disclosed in this invention are preferred embodiments, but are not limited thereto. Those skilled in the art can easily understand the spirit of this invention based on the above embodiments and make different extensions and variations, but as long as they do not depart from the spirit of this invention, they are all within the protection scope of this invention.

Claims

1. A multi-target real-time monitoring system for rearview mirror lens manufacturing, characterized in that, include: The data acquisition module is used to collect the first monitoring data, the second monitoring data and the third monitoring data of the crimping process in real time. The first monitoring data is the visual image data of the crimping status of the lens and the tray, the second monitoring data is the dynamic environmental data of the crimping station, and the third monitoring data is the crimping process operation data. The storage module stores historical production data and the initial adaptation model. The historical production data includes historical environmental interference characteristics, historical crimping defect characteristics, historical environment-process correlation data, and historical interference coupling characteristics. The initial adaptation model is an environment-process basic correlation model built based on knowledge of similar lens crimping processes. The processing module calls the data and model from the storage module to perform fusion judgment on the first, second, and third monitoring data, distinguishing between false anomalies caused by single dynamic environmental interference, coupling of multiple environmental interferences, and real defects in the crimping process, and assisting in judgment based on the initial adaptation model when historical data is insufficient. The control module generates a set of candidate adjustment parameters based on the judgment results of the processing module; The decision-making module prioritizes the candidate adjustment parameter set, selects the optimal adjustment scheme and executes it, and simultaneously receives the post-adjustment monitoring data fed back by the data acquisition module to form a closed-loop control of the pressing quality.

2. The multi-target real-time monitoring system for rearview mirror lens production and processing according to claim 1, characterized in that, The data acquisition module includes a visual acquisition unit, an environmental acquisition unit, a process acquisition unit, and an interference coupling sensing unit. The visual acquisition unit is set in the monitoring area of ​​the crimping station and is used to capture the crimping process of the lens and the tray, and output the first monitoring data, including the alignment image of the lens and the tray, the crimping gap image and the surface condition image of the lens. The environmental acquisition unit is deployed around the crimping station to capture the dynamic changes of the crimping environment in real time and output the second monitoring data, including the temperature and humidity data, dust concentration data, light intensity data and airflow speed data of the crimping station. The process acquisition unit is connected to the control terminal of the pressing equipment to acquire the operating parameters of the pressing equipment and output the third monitoring data, including the temperature data of each section of the heating element, the pressure data of the pressing cylinder, the pressing time data, and the conveyor belt speed data. The interference coupling sensing unit is used to identify the coordinated change characteristics of multiple environmental parameters in the second monitoring data, and output the coupling type and coupling strength data of multiple environmental interferences. The coupling type includes temperature and humidity-dust coupling and light-airflow coupling.

3. A multi-target real-time monitoring system for rearview mirror lens production and processing according to claim 2, characterized in that, The storage module includes a historical interference library, a historical defect library, an association database, a coupling feature library, and a model storage library; The historical interference database stores historical environmental interference characteristics, including the environmental data range, visual image noise characteristics, and interference duration corresponding to a single environmental interference. The historical defect database stores historical crimping defect features, including visual image features, process parameter deviation ranges, and defect impact characteristics corresponding to lens and tray misalignment, loose crimping, and lens breakage. The associated database stores historical environment-process correlation data, including records of process parameter adjustments and quality results under single environmental disturbances; The coupling feature library stores historical interference coupling features, including the parameter coordination change law corresponding to multiple environmental interference coupling types, visual image distortion features under coupling state, and adapted process parameter combinations. The model repository stores the initial adaptation model, including a table of environmental parameter thresholds and process parameter baseline values ​​constructed based on the process manual.

4. A multi-target real-time monitoring system for rearview mirror lens manufacturing and processing according to claim 3, characterized in that, The fusion judgment process of the processing module includes interference type identification, false anomaly elimination, real defect judgment, and data insufficiency adaptation. In the interference type identification process, the processing module compares the second monitoring data and the coupling data output by the interference coupling sensing unit with the historical interference database and the coupling feature database to identify whether the current interference is single-environment interference or multi-environment interference coupling, and determines the interference type and intensity. In the process of eliminating false anomalies, the processing module calls the corresponding historical interference features or coupling features according to the identified interference type, and removes the image distortion area and parameter fluctuation noise caused by environmental interference from the first monitoring data. In the determination of real defects, the processing module compares the first monitoring data after excluding false anomalies with the visual features of the historical defect database, and at the same time compares the third monitoring data with the process features of the historical defect database. When both match, it is determined to be a real defect. In the data-sufficient adaptation process, if the current interference type or coupling type is not recorded in the historical data, the processing module calls the initial adaptation model and uses the basic association rules of similar lenses to help determine the false anomaly and the real defect. To quantify the results of interference type identification and actual defect determination, the processing module implements the determination logic through the following algorithm formula: Interference intensity calculation: For interference in a single environment, the interference intensity... The formula is calculated by weighting the deviation rate between the current environmental parameters and the standard parameters and the duration of the disturbance: ； when When the interference is significant, it is considered a single, significant disturbance. in, For single environmental interference intensity, As the parameter deviation rate weight, These are the current environmental parameter values. These are standard values ​​for environmental parameters. For the duration of the disturbance, weight, For the duration of the interference, For a single interference intensity threshold; Coupling interference determination: Coupling strength is determined for interference from multiple environments. The formula for calculating the synergy coefficient of each environmental parameter deviation is as follows: ； when This was determined to be significant coupling interference; in, For multi-environment interference coupling strength, For the first The weights of each environmental parameter, For the first Current values ​​of each environmental parameter For the first Standard values ​​for each environmental parameter For the first , The synergy coefficient of each environmental parameter, This is the threshold for coupling interference strength; True Defect Determination: Defect Matching Degree The product of visual feature matching degree and process feature matching degree is calculated using the following formula: ； when At that time, it was determined to be a genuine defect; in, For defect matching degree, For visual feature matching degree, For process feature matching degree, This is the defect matching threshold.

5. A multi-target real-time monitoring system for rearview mirror lens production and processing according to claim 4, characterized in that, The processing module also includes a dynamic association update submodule. After the data shortage adaptation step is executed, the dynamic association update submodule is used to associate and store the current interference type, coupling type, process parameter adjustment record and crimping quality result to the corresponding database. At the same time, iteratively optimizes the initial adaptation model based on the new data. When the number of records of a certain interference type or coupling type reaches a preset threshold, it automatically migrates it from the management scope of the initial adaptation model to the historical database.

6. A multi-target real-time monitoring system for rearview mirror lens production and processing according to claim 4, characterized in that, The process by which the control module generates a candidate set of adjustment parameters includes single disturbance adjustment and coupled disturbance adjustment. In the single interference adjustment, if the processing module identifies it as a single environmental interference without a real defect, the control module calls the process parameter adaptation range corresponding to the interference type in the associated database to generate at least 3 sets of differentiated adjustment parameter combinations; if a real defect exists, the control module generates a targeted adjustment parameter combination based on the repair scheme of the corresponding defect in the historical defect database. In the coupling interference adjustment, if the processing module identifies multi-environment interference coupling, the control module calls the process parameter combination record corresponding to the coupling type in the coupling feature library, and generates a multi-dimensional collaborative adjustment parameter combination by combining the influence weight of each environmental parameter. The collaborative adjustment parameter combination includes the linkage adjustment scheme of heating temperature, pressing pressure and pressing time. To accurately generate adjustment parameters, the control module calculates the adjustment amount using the following algorithm formula: Adjustment amount of process parameters for a single disturbance: Adjustment amount of process parameters for a single environmental disturbance. Based on interference strength The historical correlation coefficient is determined by the following formula: ； ； When generating differentiated combinations, by adjusting The value of , Get at least 3 sets ; in, This refers to the adjustment amount of process parameters under a single disturbance. The historical environment-process correlation coefficient. For single environmental interference intensity, To adjust the process parameter values, These are the current process parameter values. This represents the deviation of the correlation coefficient. Coupling interference process parameter adjustment amount: Multi-dimensional coordinated adjustment amount to address coupling interference. Based on coupling strength The weights of each parameter are determined by the following formula: ； ； in, For coupling interference, the first Adjustment amount of each process parameter For the first The correlation coefficient between each process parameter and coupling interference. For multi-environment interference coupling strength, For the first The influence weight of each process parameter For coupling interference, the first The values ​​after adjusting each process parameter For the first Current values ​​of each process parameter.

7. A multi-target real-time monitoring system for rearview mirror lens manufacturing and processing according to claim 1, characterized in that, The priority ranking process of the decision-making module includes the construction of evaluation indicators, weight allocation, and scheme optimization; In the construction of the evaluation indicators, the decision-making module constructs an evaluation indicator system including the degree of improvement in crimping quality, the range of parameter adjustment, the rate of change in energy consumption, and the speed of adjustment response. In the weight allocation, the decision module dynamically allocates indicator weights according to the current production needs. When pursuing quality stability, the weight of the crimping quality improvement is increased; when pursuing production efficiency, the weight of the adjustment response speed is increased. In the preferred scheme, the decision module quantifies and scores the candidate adjustment parameter set based on evaluation indicators and weights, selects the adjustment scheme with the highest comprehensive score as the optimal scheme, and sends it to the crimping device for execution. To quantitatively select the optimal solution, the decision-making module calculates a comprehensive score using the following algorithm formula: Standardized score for each single indicator: Each evaluation indicator is standardized to eliminate the influence of dimensions. The formula is: Positive indicators: ； Negative indicators: ； in, For the first Standardized scores of each evaluation indicator For the first The actual value of each evaluation indicator For the first The historical maximum value of each evaluation indicator For the first The historical minimum value of each evaluation indicator; Overall Score: A weighted sum of standardized scores based on indicator weights, using the following formula: ； choose The largest candidate solution is considered the optimal solution. If multiple solutions exist... If they are the same, prioritize adjusting the response speed. The largest solution in, The overall score of the candidate adjustment schemes. For the first The weight of each evaluation indicator.

8. A multi-target real-time monitoring system for rearview mirror lens manufacturing and processing according to claim 1, characterized in that, The system also includes an emergency intervention module, which is connected to the decision-making module and the processing module. If the optimal adjustment scheme output by the decision module is executed, and the processing module determines that the crimping quality still does not meet the standard and the degree of defect exceeds the preset severe threshold, the emergency intervention module will automatically trigger graded emergency measures: the first-level measure is to suspend the crimping of the current batch of lenses; the second-level measure is to link the upstream material supply system to suspend material conveying; at the same time, an emergency report is generated, which includes the interference evolution curve, the adjustment parameters of each time, and the defect spread risk assessment.

9. A multi-target real-time monitoring system for rearview mirror lens manufacturing and processing according to claim 1, characterized in that, The data acquisition module also includes a time series analysis subunit, which is used to perform time series modeling on the second and third monitoring data to predict the environmental parameter change trend and process parameter drift trend within a preset time period in the future. The processing module receives the prediction results from the timing analysis subunit, identifies potential interference risks in advance, and provides a basis for the control module to generate a set of preventive adjustment parameters. To achieve parameter trend prediction, the time series analysis subunit constructs a prediction model using the following algorithm: Environmental parameter trend prediction: Predicting future trends using a linear regression model. Environmental parameter values ​​within the time period The formula is: ； when Exceeding At that time, it was determined to be a potential interference risk; in, For the future Predicted values ​​of environmental parameters within the time period These are the current environmental parameter values. For regression coefficients, To predict duration, This refers to the allowable fluctuation range of environmental parameters; Process parameter drift prediction: based on environmental parameter predictions Predicting process parameter drift using historical correlation models The formula is: ； Control module based on Generate preventive adjustment parameters To offset the effects of parameter drift in advance; in, For the future The amount of process parameter drift within a time period. This is the process parameter drift coefficient. For preventative adjustment parameters, This represents the current process parameter value.

10. A method for real-time monitoring of multiple targets in the production and processing of rearview mirror lenses, applied to the real-time monitoring system for multiple targets as described in any one of claims 1-9, characterized in that, Includes the following steps: S1. Data acquisition steps: The first monitoring data, the second monitoring data, the third monitoring data, and the interference coupling data are collected in real time through the data acquisition module. The time series analysis subunit predicts the changing trends of environmental and process parameters and then outputs the results. S2. Historical and Model Calling Steps: The processing module calls the historical production data and initial adaptation model from the storage module; S3, Fusion Judgment Steps: The processing module identifies the type of interference, eliminates false anomalies based on historical data or the initial adaptation model, determines the real defects, and at the same time dynamically associates and updates the sub-modules to record new data and iterate the model. S4. Candidate parameter generation steps: Based on the judgment results, the control module generates differentiated adjustment parameter combinations for single disturbances and multi-dimensional collaborative adjustment parameter combinations for coupled disturbances, forming a candidate adjustment parameter set. S5. Optimal Solution Decision-Making Steps: The decision-making module constructs an evaluation index system and dynamically allocates weights. After scoring the candidate adjustment parameter set, it selects the optimal adjustment solution. S6. Closed-loop control steps: After executing the optimal solution, collect the monitoring data after adjustment and re-evaluate. If the standard is met, update the historical data. If the standard is not met and the defect is serious, trigger emergency intervention. S7. Continuous optimization steps: Repeat steps S1-S7, combine the timing prediction results to achieve preventive adjustments, and continuously optimize the accuracy of crimping quality control.

Images