Intelligent scheduling method for high-efficiency wire embedding machine and press integrated production line

Abstract

This invention discloses an intelligent scheduling method for an integrated production line of high-efficiency wire embedding machine and press, belonging to the field of intelligent scheduling technology. It utilizes high-precision sensors to collect real-time operating parameters and raw material properties of the wire embedding machine and press, forming a multi-dimensional dynamic dataset. This dataset is then combined with historical data to construct a predictive model using machine learning. The model calculates the optimal start time for pressing based on factors such as wire embedding speed, pressing force, and wire thickness, and calibrates it in real time. The system triggers dynamic adjustments through deviation analysis to ensure synchronized equipment operation and improve production efficiency. Quality inspection feedback further optimizes model parameters, improves prediction accuracy, and enhances adaptability, thereby achieving continuous improvement in complex production environments and overcoming the limitations of traditional methods in handling dynamic changes. Through real-time data collection and regular updates, it can adapt to fluctuations in machine performance and raw materials, ensuring simultaneous improvement in production quality and efficiency.

Description

Technical Field

[0001] This invention relates to the field of intelligent scheduling technology, specifically to an intelligent scheduling method for an integrated production line of high-efficiency wire embedding machine and press. Background Technology

[0002] Against the backdrop of rapid development in modern manufacturing, wire embedding technology and lamination processes have become indispensable core technologies in many high-precision manufacturing fields. These technologies are widely used in various scenarios such as electronic product assembly, automotive parts production, and new energy. For example, in circuit board manufacturing, wire embedding machines precisely embed wires into predetermined positions, and then a press applies appropriate pressure to ensure a firm bond between the wires and the substrate; in automotive wiring harness assembly, wire embedding and lamination processes are used to achieve efficient wire placement and fixation to meet complex electrical connection requirements; and in new energy battery module production, this technology is used for embedding and compacting electrode connection wires to ensure the high performance and safety of battery components. As the market's requirements for product performance, reliability, and production efficiency continue to increase, the collaborative work of wire embedding machines and presses is increasingly developing towards integration, automation, and intelligence. By building integrated production lines, companies can achieve seamless integration of production processes while reducing manual intervention, thereby improving overall manufacturing efficiency and meeting the needs of large-scale customized production.

[0003] Although the embedding and pressing processes can theoretically achieve efficient synergy, precise synchronization of their operation sequences remains a key technical challenge limiting their performance in actual production environments. Specifically, the embedding machine and pressing machine are affected by various factors during dynamic operation, making it difficult to maintain a stable timing for their coordinated operation. For example, the operating speed of the embedding machine may fluctuate slightly due to mechanical wear, motor response characteristics, or control system delays, while the timing of the pressing machine's actions is limited by the response time of the hydraulic system or the delay in sensor signal transmission. Furthermore, subtle differences in raw material properties (such as wire diameter, hardness, or surface friction coefficient) further complicate timing matching. If the embedding speed increases instantaneously and the pressing machine fails to keep up, the wire may not be adequately secured, leading to poor contact or the risk of detachment; conversely, if the pressing machine starts too early, pressure may be applied before the wire is fully embedded, causing positional displacement or even material damage.

[0004] Traditional scheduling methods, typically based on fixed parameters and preset operation sequences, struggle to adapt to real-time dynamic changes, often leading to reduced production efficiency, decreased product consistency, and even increased scrap rates. Therefore, ensuring precise synchronization between the wire embedding and pressing processes in real-world production scenarios with multiple variables has become a critical technical challenge that urgently needs to be addressed.

[0005] To address this, an intelligent scheduling method for an integrated production line combining high-efficiency wire embedding machines and presses was proposed. Summary of the Invention

[0006] (a) Technical problems to be solved

[0007] To address the shortcomings of existing technologies, this invention provides an intelligent scheduling method for an integrated production line combining a wire embedding machine and a pressing machine. High-precision sensors collect real-time operating parameters and raw material properties from both the embedding and pressing machines, forming a multi-dimensional dynamic dataset. This dataset is then combined with historical data to construct a predictive model using machine learning. The model calculates the optimal pressing start time based on factors such as embedding speed, pressing force, and wire thickness, and calibrates it in real time. The system triggers dynamic adjustments through deviation analysis to ensure synchronized equipment operation and improve production efficiency. Quality inspection feedback further optimizes model parameters, improves prediction accuracy, and enhances adaptability, enabling continuous improvement in complex production environments and overcoming the limitations of traditional methods in handling dynamic changes. Through real-time data collection and regular updates, the system can adapt to fluctuations in machine performance and raw materials, ensuring simultaneous improvement in production quality and efficiency, thus solving the technical problems described in the background art.

[0008] (II) Technical Solution

[0009] To achieve the above objectives, the present invention provides the following technical solution: an intelligent scheduling method for an integrated production line of high-efficiency wire embedding machine and press, comprising: when real-time data acquisition is initiated, sensors perform high-frequency measurement and integration operations on the operating parameters of the wire embedding machine and press and the properties of the raw materials to form a real-time dataset. ;

[0010] When the synchronous model is activated, the machine learning algorithm is based on historical real-time datasets. A synchronization model was constructed using regression analysis. Generate the optimal start time for pressing. ;

[0011] When a real-time operational adjustment is triggered, receive the real-time dataset. Combined with synchronization model Calculate the predicted pressing start time To address the discrepancy between prediction and reality Implement dynamic adjustments to optimize the speed of the wire embedding machine and the pressing machine;

[0012] When quality feedback and model optimization are initiated, output quality data is collected and fed back to optimize the synchronous model. By comparing threshold values ​​to identify defects and updating model parameters, the optimized model can be obtained. And redeploy, transforming the data format from detection results into optimized input;

[0013] Preferably, the following operating parameters are monitored and recorded in real time using a high-precision sensor network installed on the wire embedding machine and the press: the wire embedding speed of the wire embedding machine. The pressing force of the press The operation times of the wire embedding machine and the pressing machine are respectively and ; Forming a time series dataset ,in Represents a time set; utilizes material property sensors to measure and record key properties of raw materials in real time: wire diameter. Material hardness ; Form an attribute dataset ;

[0014] Prioritize integrating the collected runtime parameters and attribute data into a real-time dataset:

[0015] Perform outlier detection and data smoothing to generate high-quality standardized datasets. ;

[0016] Preferably, from real-time datasets Key features are extracted, and the order hysteresis of the buried wire speed is introduced. The feature set is defined as The target variable is the optimal start-up time for compression. ,

[0017] Preferably, a synchronization model is constructed using the support vector regression algorithm. Support vector regression algorithm can effectively process feature sets. With target variable The nonlinear relationship between them, the model input is the feature set The output is the optimal start-up time for pressing. ;

[0018] The optimization objective of the support vector regression algorithm is defined as follows:

[0019]

[0020] Constraints:

[0021]

[0022] In the formula: This is the weight vector; For bias; For penalty parameters; Tolerance for error; , These are slack variables; For feature vectors With weight The inner product;

[0023] Preferably, the root mean square error (RMSE) is used to evaluate the model performance. If the RMSE exceeds a preset threshold, the model is optimized by adjusting the parameters of the feature selection or support vector regression algorithm.

[0024] The trained synchronous model Deployed to an intelligent scheduling system, with a periodic update mechanism: utilizing newly collected real-time datasets at fixed intervals. Retrain the model and calculate the RMSE of the new model. If the RMSE of the new model is lower than that of the current model, update the deployment.

[0025] Preferably, the intelligent scheduling system receives real-time datasets. Obtain the synchronization model and with real-time datasets As input, predict the pressing start time. As output; calculate the predicted pressing start time. Compared with actual pressing operation time Timing deviation The formula is:

[0026]

[0027] In the formula: This refers to the actual pressing operation time;

[0028] like If the pressing operation is delayed, the press speed needs to be increased or the press started earlier; if If the pressing operation is performed too early, the speed of the wire embedding machine needs to be slowed down or the start-up delayed; if The operation is consistent with the forecast and no adjustment is needed.

[0029] Preferably, to eliminate timing deviations The speed of the wire embedding machine or press is dynamically adjusted according to the magnitude of the deviation. The adjustment strategy is as follows:

[0030] Press speed adjustment: If Increase press speed The adjustment range is calculated using the following formula:

[0031]

[0032] The adjusted press speed is:

[0033]

[0034] Wire embedding machine speed adjustment: If Slow down the speed of the burying machine The adjustment range is:

[0035]

[0036] The adjusted speed of the burying machine is:

[0037]

[0038] In the formula: The press speed before adjustment; The speed of the burying machine before adjustment. For the adjustment range of the press speed, The speed adjustment range of the wire laying machine. This is the press speed adjustment coefficient, determined by the press performance and process requirements; This is the speed adjustment coefficient for the wire burying machine, determined by the performance of the machine. For the adjusted press speed, The adjusted speed of the burying machine;

[0039] Preferably, the adjusted operating data is recorded, including the adjusted wire embedding speed. Press speed Actual pressing start time and time-series cumulative error The calculation formula is:

[0040]

[0041] In the formula: The sliding window size represents the time range for evaluation. Forgetting factor, ; For quality deviations (such as the failure rate of wire positioning accuracy or bonding strength); This is the quality sensitivity coefficient, with a value greater than 0, which controls the amplification effect of quality on error. The index of the sliding window;

[0042] Build and adjust dataset Defined as ,in: The adjusted actual pressing start time is used to measure the deviation between the adjusted operation and the prediction.

[0043] Preferably, output quality data is collected in real time, including wire position accuracy. and compressive strength Preset quality standard threshold: wire position accuracy standard and compressive strength standard ;

[0044] The quality qualification conditions are and If the conditions are not met, the invalid data will be recorded, including the real-time dataset at the corresponding time. Adjusting the dataset and predicted pressing start time ;

[0045] Preferably, for non-conforming outputs, a quality deviation index is calculated. The quality parameters of the product are represented by vectors. This indicates that the standard mass vector is The quality deviation vector is defined as:

[0046]

[0047] Quality Deviation Indicators The definition is as follows:

[0048]

[0049] In the formula: The meaning is: through the quality influence matrix For the deviation vector Perform the transformation and then calculate its Euclidean norm. , used to measure the static deviation of quality parameters;

[0050] matrix Characterizing the interaction between different quality parameters; parameters These are the weighting coefficients for dynamic terms;

[0051] Preferably, a quality feedback threshold is set. ,like This triggers synchronous model optimization and collects feedback datasets. It includes real-time datasets corresponding to defective outputs. Adjusting the dataset Predicted pressing start time and quality data Sample weights are based on Direct proportional distribution;

[0052] Based on feedback dataset Incremental support vector regression is used to optimize the synchronization model. The updated formula is as follows:

[0053]

[0054] In the formula: For the model weight vector, This is the optimized weight vector; To adjust the pressing start time; For learning rate, For the sample Lagrange multipliers and quality deviation index Proportional;

[0055] This refers to the actual pressing start time. To predict the pressing start time, For kernel functions; To provide the number of feedback samples; the optimized model Redeploy if and If the proportion increases, the optimized synchronization model will be retained. Otherwise, adjust the learning rate. or quality feedback threshold Re-execute the synchronization model optimization;

[0056] (III) Beneficial Effects

[0057] This invention provides an intelligent scheduling method for an integrated production line combining a high-efficiency wire embedding machine and a press, which has the following beneficial effects:

[0058] By acquiring high-frequency real-time data, the system comprehensively perceives the dynamic changes in operating parameters such as wire embedding speed and pressing force, as well as the properties of raw materials, providing a reliable data foundation for the synchronization model. The synchronization model employs a support vector regression algorithm, combined with lag characteristics, to accurately predict the optimal start time for pressing, overcoming the limitations of traditional fixed-rule scheduling. Real-time operation adjustments utilize a dynamic mechanism driven by timing deviations to flexibly adjust machine speed, ensuring precise synchronization between the wire embedding and pressing processes, significantly improving production cycle time and consistency.

[0059] The quality feedback and model optimization process utilizes data such as wire position accuracy and bonding strength, and employs incremental support vector regression (ISVR) with quality deviation weighted sampling to optimize the model online, continuously improving prediction accuracy and quality control effectiveness. A closed-loop verification mechanism ensures the effectiveness of the optimization, significantly reducing the scrap rate and resolving the quality instability issues caused by multivariate interference in the background technology.

[0060] By ensuring data reliability through high-precision sensors and data preprocessing technology, and combining this with strategies for regular model updates and dynamic adjustments, the system can flexibly respond to changes in machine performance and raw material properties. This adaptive capability enables the system to maintain high-efficiency operation even in complex production environments, overcoming the limitations of traditional methods in dealing with dynamic changes. Attached Figure Description

[0061] Figure 1 This is a schematic diagram of the intelligent scheduling process of the integrated production line of the present invention. Detailed Implementation

[0062] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] Please see Figure 1 This invention provides an intelligent scheduling method for an integrated production line combining a high-efficiency wire embedding machine and a press, comprising:

[0064] Step 1: When real-time data acquisition is initiated, the sensors perform high-frequency measurement and integration operations on the operating parameters of the wire embedding machine and the press, along with the properties of the raw materials, to form a real-time dataset. ;

[0065] Step one includes the following:

[0066] Step 101: Acquisition of Operating Parameters

[0067] The following operating parameters are monitored and recorded in real time using a high-precision sensor network installed on the wire embedding machine and the pressing machine:

[0068] The burying speed of the burying machine The pressing force of the press The operation times of the wire embedding machine and the pressing machine are respectively and ;

[0069] The data acquisition frequency is set higher than the production cycle time to capture instantaneous fluctuations in parameters. The operating parameters are aligned using timestamps to form a time-series dataset. ,in Represents a time set;

[0070] Using material property sensors to measure and record key properties of raw materials in real time: wire diameter Material hardness These attributes may change due to batch or environmental factors, so continuous monitoring is required; the attribute data is also aligned with timestamps to form an attribute dataset. ;

[0071] By collecting the operating parameters of the wire embedding machine and the press in real time using high-precision sensors, it is possible to capture minute fluctuations in machine performance. This not only provides dynamic and comprehensive basic data for subsequent synchronization models, but also improves the timeliness of data through high-frequency acquisition, ensuring the system's real-time perception of production status and thus improving the accuracy of predictions.

[0072] By using material property sensors to monitor key properties of raw materials in real time and aligning operational parameter data with timestamps, it is possible to detect batch changes in raw materials or the impact of environmental factors, thereby enhancing robustness and adaptability to changes in the production environment and further improving prediction accuracy and production stability.

[0073] Step 102: Data Integration and Preprocessing

[0074] The collected operating parameters and attribute data are integrated into a real-time dataset. To ensure the quality of the collected data, outlier detection and data smoothing are performed; for real-time datasets... Standardization is performed using a min-max standardization method to scale each parameter to... The interval, the formula is:

[0075]

[0076] in These are the original parameter values. For standardized values, and These are the minimum and maximum values ​​of the parameter within the sampling period, respectively; the standardized dataset is denoted as... ;

[0077] Collect operating parameters of the wire embedding machine and the press. and raw material properties And through integration, preprocessing, and standardization, a high-quality standardized dataset is generated. Scaling the data to a minimum-maximum scale using min-max normalization. The interval eliminates differences in parameter dimensions, improving the convergence speed and stability of model training.

[0078] Step 2: When the synchronous model is activated, the machine learning algorithm is based on the historical real-time dataset. A synchronization model was constructed using regression analysis. Generate the optimal start time for pressing. ;

[0079] Step two includes the following:

[0080] Step 201: Feature Engineering and Data Preparation

[0081] From real-time datasets Extract key features, including the embedding speed. Pressing force , burying machine operation time Press operation time wire diameter Material hardness ;

[0082] To capture time-series dependencies, hysteresis features are introduced, such as the order hysteresis of the burial speed. Reflecting the impact of historical states on current synchronization, the final feature set is defined as follows: The target variable is the optimal start-up time for compression. It is calculated by the difference between the time of completion of the buried wire and the time of start of the pressing in historical data;

[0083] For example, if the burying speed A sudden increase in pressure necessitates initiating the pressing process earlier to match the schedule, resulting in a lag characteristic. This can help the model identify this trend of speed change; set the window size to 5 time units (e.g., 5 seconds), at each time point The system extracts data from the previous time point from the window, calculates and updates it in real time. ;

[0084] Example: If the current time seconds, then use The embedding speed data is displayed in seconds, and the window continuously scrolls to keep the data up-to-date.

[0085] By extracting key features and introducing lag features to capture temporal dependencies, the model's ability to perceive production dynamics is significantly enhanced. The high-dimensional design enriches the data representation capabilities, enabling the prediction of the optimal start-up time for pressing. It provides robust support and improves the model's prediction resolution;

[0086] Step 202: Model Selection and Training

[0087] A synchronization model is constructed using the support vector regression algorithm. Support vector regression algorithm can effectively process feature sets. With target variable The nonlinear relationship between them, the model input is the feature set The output is the optimal start-up time for pressing. The optimization objective of the support vector regression algorithm is defined as follows:

[0088]

[0089] Constraints:

[0090]

[0091] In the formula: The weight vector determines the contribution of features to the prediction; As a bias, adjust the baseline of the predicted values; The penalty parameter controls the model's tolerance to error, and its value range is [not specified]. ; To tolerate error, an acceptable range of prediction error is defined, with a range of values. ; , The slack variable measures the excess error and has a range of values. ; For feature vectors With weight The inner product;

[0092] A synchronization model is constructed using the support vector regression algorithm. By optimizing the objective function, the feature set can be effectively fitted. Optimal start-up time for pressing The nonlinear relationship. By minimizing the weight norm. Including error penalty terms, we seek the optimal start-up time for features and pressing. Nonlinear mapping between parameters ensures a balance between prediction accuracy and model complexity, and penalizes the parameters. And determined through grid search and cross-validation to optimize the model's generalization ability;

[0093] If the burying speed Optimal start-up time for pressing The relationship is non-linear (e.g., the compression time decreases as the speed increases to a certain extent). Support vector regression algorithms can capture this trend through kernel functions (e.g., radial basis functions).

[0094] It should be noted that the Support Vector Regression algorithm model uses a Radial Basis Function (RBF) kernel, and the kernel function has the following form:

[0095]

[0096] Parameter settings: The result was obtained through optimization using grid search combined with 5-fold cross-validation.

[0097] Step 203: Model Evaluation and Optimization

[0098] The root mean square error (RMSE) is used to evaluate model performance, and the calculation formula is as follows:

[0099]

[0100] In the formula: To predict the pressing start time for the model, This refers to the actual pressing start time. Sample size;

[0101] RMSE quantifies the deviation between predicted and actual values, guiding model optimization. If RMSE exceeds a preset threshold, adjustments can be made to feature selection or support vector regression algorithm parameters (e.g., by increasing penalty parameters). (Or reduce) optimize the model until the accuracy requirements are met;

[0102] Model performance is quantified using root mean square error (RMSE) and by dynamically adjusting parameters (such as penalty parameters). (And tolerance error) to optimize model accuracy. The RMSE evaluation results provide clear directions for optimization (such as reducing the error from 0.05 seconds to 0.03 seconds), ensuring that the prediction error is controlled within an acceptable range for production, and improving the practicality and reliability of the model.

[0103] Step 204, Model Deployment and Update Mechanism

[0104] The trained synchronous model Deployed to the intelligent scheduling system to adapt to changes in the production environment (such as the hardness of raw materials) (Fluctuations) Set up a periodic update mechanism: Use newly collected real-time datasets at fixed intervals. Retrain the model and calculate the RMSE of the new model. If the RMSE of the new model is lower than that of the current model, update the deployment.

[0105] The trained synchronous model The model is deployed to the intelligent scheduling system and kept adaptable to changes in the production environment through a regular update mechanism. After the update, if the RMSE improves, the model is automatically deployed to ensure the long-term stability and prediction accuracy of the system.

[0106] During use, a synchronous model for wire embedding and pressing was constructed and optimized. Extracting multidimensional feature sets using feature engineering The optimal start time for pressing is predicted using the support vector regression algorithm. Accuracy and adaptability are ensured through RMSE evaluation and regular update mechanisms.

[0107] Step 3: When a real-time operation adjustment is triggered, receive the real-time dataset. Combined with synchronization model Calculate the predicted pressing start time To address the discrepancy between prediction and reality Implement dynamic adjustments to optimize the speed of the wire embedding machine and the pressing machine;

[0108] Step three includes the following:

[0109] Step 301: Receive real-time data and model predictions

[0110] The intelligent scheduling system receives real-time datasets. Including the speed of wire burial Pressing force Wire embedding machine operation time Press operation time wire diameter Material hardness ; Obtain the synchronization model and with real-time datasets As input, predict the pressing start time. The calculation formula is:

[0111]

[0112] In the formula: To predict the pressing start time, this indicates the ideal point in time when the press should begin operation;

[0113] When in use, the intelligent scheduling system receives real-time datasets. Using a synchronization model Predicting the optimal start-up time for pressing The prediction process has a fast response speed and seamlessly connects with the production cycle, ensuring that the system can respond to dynamic changes in a timely manner and provide an accurate time reference for subsequent adjustments.

[0114] Step 302: Calculate timing deviation

[0115] Calculate the predicted pressing start time Compared with actual pressing operation time Timing deviation The formula is:

[0116]

[0117] In the formula: Timing deviation represents the difference between the actual operation time and the predicted time. This refers to the actual pressing operation time;

[0118] like If the pressing operation is delayed, the press speed needs to be increased or the press started earlier; if If the pressing operation is performed too early, the speed of the wire embedding machine needs to be slowed down or the start-up delayed; if The operation is consistent with the forecast and requires no adjustment; for example, if Second, seconds, then The second indicates a 0.2-second delay in pressing, which needs to be accelerated to make up for the time difference.

[0119] When using it, calculate the prediction time. Compared with actual operation time Timing deviation It can provide a clear direction and range of adjustment, and the quantitative analysis of deviations enhances the scientific nature and pertinence of the adjustment, ensuring the efficiency and controllability of operation optimization.

[0120] Step 303: Dynamically adjust the machine speed

[0121] To eliminate timing deviations The speed of the wire embedding machine or press is dynamically adjusted according to the magnitude of the deviation. The adjustment strategy is as follows:

[0122] Press speed adjustment: If Increase press speed The adjustment range is calculated using the following formula:

[0123]

[0124] The adjusted press speed is:

[0125]

[0126] Wire embedding machine speed adjustment: If Slow down the speed of the burying machine The adjustment range is:

[0127]

[0128] The adjusted speed of the burying machine is:

[0129]

[0130] In the formula: The press speed before adjustment; The speed of the burying machine before adjustment. For the adjustment range of the press speed, The speed adjustment range of the wire laying machine. This is the press speed adjustment coefficient, determined by the press performance and process requirements; This is the speed adjustment coefficient for the wire burying machine, determined by the performance of the machine. For the adjusted press speed, The adjusted speed of the burying machine;

[0131] in: and The calculation method is as follows:

[0132]

[0133] In the formula: and : These represent the maximum speed variation of the press and the wire embedding machine, respectively (unit: m / s). and : These are the maximum response times of the press and the wire embedding machine, respectively (unit: seconds).

[0134] Example: If the maximum speed of the press changes meters per second, response time seconds, then Meters per second.

[0135] Based on time series deviation Dynamically adjust the speed of the wire embedding machine or press, such as adjusting the press speed. Upgraded to To compensate for the lag, and in conjunction with equipment performance parameters, the wire embedding and pressing processes are precisely synchronized, the adjustment strategy is flexible, and the system's adaptability and response speed are improved.

[0136] Step 304: Adjusting, recording, and transmitting the effects.

[0137] Record the adjusted operating data, including the adjusted wire embedding speed. Press speed Actual pressing start time and time-series cumulative error The calculation formula is:

[0138]

[0139] In the formula: The sliding window size represents the time range for evaluation. Forgetting factor, ; For quality deviations (such as the failure rate of wire positioning accuracy or bonding strength); This is the quality sensitivity coefficient, with a value greater than 0, which controls the amplification effect of quality on error. The index of the sliding window;

[0140] Integrate this data into an adjusted dataset Defined as ,in: The adjusted actual pressing start time is used to measure the deviation between the adjusted operation and the prediction.

[0141] When using it, record the adjusted running data (such as...). and time-series cumulative error It provides complete data support for quality feedback and model optimization. The comprehensiveness and systematic nature of the records ensure the effectiveness of the feedback mechanism and promote continuous improvement of the production process.

[0142] Step 4: When quality feedback and model optimization are initiated, collect output quality data and feed it back to optimize the synchronous model. By comparing threshold values ​​to identify defects and updating model parameters, the optimized model can be obtained. And redeploy, transforming the data format from detection results into optimized input;

[0143] Step 401: Quality Data Collection and Evaluation

[0144] Output quality data, including wire position accuracy, is collected in real time through quality inspection equipment. and compressive strength Preset quality standard threshold: wire position accuracy standard and compressive strength standard The quality qualification conditions are: and If the conditions are not met, the invalid data will be recorded, including the real-time dataset at the corresponding time. Adjusting the dataset and predicted pressing start time ;

[0145] Output quality data, such as wire position accuracy, is collected in real time through quality inspection equipment. and compressive strength Compared with standard thresholds, it accurately identifies non-conforming products, and real-time monitoring improves the timeliness of quality control, providing a reliable basis for deviation analysis and model optimization.

[0146] Step 402, Quality Feedback and Deviation Analysis

[0147] For defective products, calculate the quality deviation index. This is used to quantify the degree of quality defects and introduces matrix operations and dynamic change analysis. It is assumed that the quality parameters of the product can be represented by a vector, for example... (representing positional accuracy and compressive strength respectively), the standard mass vector is The quality deviation vector is defined as:

[0148]

[0149] Based on this, the quality deviation index The definition is as follows:

[0150]

[0151] In the formula: The meaning is: through the quality influence matrix For the deviation vector Perform the transformation and then calculate its Euclidean norm. , used to measure the static deviation of quality parameters;

[0152] matrix For one The matrix represents the interaction between different quality parameters, such as positional accuracy. Possibly affecting the compressive strength Indirect effects can be generated, and matrix elements can be determined through process analysis or data fitting. The meaning is: the rate of change of the integral quality deviation vector over time (i.e., The Euclidean norm of the parameter captures the dynamic trend of quality defects. This is the weighting coefficient for dynamic items, with a value greater than 0. It is used to balance the importance of static bias and dynamic changes and can be adjusted according to actual needs; time interval. arrive This indicates the time frame for the evaluation, such as a specific stage in the production process.

[0153] Through quality deviation index By quantifying the severity of quality defects, nonlinear design improves the accuracy and relevance of feedback, highlights significant quality issues, and provides scientific guidance for model optimization.

[0154] Step 403: Model Optimization Triggering and Data Preparation

[0155] Set quality feedback threshold ,like This triggers synchronous model optimization and collects feedback datasets. It includes real-time datasets corresponding to defective outputs. Adjusting the dataset Predicted pressing start time and quality data To enhance the model's adaptability to quality issues, sample weights are adjusted according to... Direct proportional distribution;

[0156] Based on feedback dataset Incremental support vector regression is used to optimize the synchronization model. The updated formula is as follows:

[0157]

[0158] In the formula: For the model weight vector, This is the optimized weight vector; To adjust the pressing start time; For learning rate, For the sample Lagrange multipliers and quality deviation index Proportional;

[0159] This refers to the actual pressing start time. To predict the pressing start time, Kernel function, used to measure samples With current input Similarity; This refers to the number of feedback samples;

[0160] The optimized model Redeploy and verify the quality improvement effects in subsequent production; if and If the proportion increases, the optimized synchronization model will be retained. Otherwise, adjust the learning rate. or quality feedback threshold Re-execute the synchronous model optimization in step 404;

[0161] Set quality feedback threshold (e.g., 0.1), only when Optimization is triggered in real-time to avoid resource waste. Based on quality deviation metrics. The weighted sampling mechanism makes the model pay more attention to serious quality problems, improving optimization efficiency and targeting; incremental support vector regression is used to optimize the synchronous model online. By dynamically updating the weights Adapting to production changes, the deviation-oriented optimization mechanism ensures continuous improvement in model prediction accuracy and enhances system stability. The optimized model... Deploy to step 3 and verify the quality improvement effect in production (e.g.) The deviation was reduced to 0.05 mm. The closed-loop verification mechanism ensures the effectiveness of optimization, improving the consistency of production quality and system reliability.

[0162] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0163] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0164] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0165] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0166] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An intelligent scheduling method for a high-efficiency wire embedding machine and press integrated production line, characterized by: Comprising, When real-time data acquisition is initiated, the sensors perform high-frequency measurements and integration operations on the operating parameters of the laying head and the press and on the properties of the raw material, forming a real-time data set ; When the synchronization model is activated, the machine learning algorithm is based on the historical real-time data set Synchronization model is built by regression analysis , generating the best pressing start time ; wherein the collected operating parameters and attribute data are integrated into a real-time data set , extracting key features from the real-time data set , introducing the order lag of the buried line speed Define the feature set , the target variable is the best pressing start time ; Adopt support vector regression algorithm to construct synchronization model , support vector regression algorithm can effectively handle the nonlinear relationship between the feature set and the target variable , the model input is the feature set , and the output is the best pressing start time ; Receiving real-time data set when real-time operation adjustment trigger Combining synchronization model Calculating predicted press start time , for predicting and actual deviation Implementing dynamic adjustment, optimizing speed of burying line machine and press Computing a predicted press start time from an actual press operation time with a timing offset , as: ， In the formula: is the timing deviation, representing the difference between the actual operation and the predicted time, is the actual pressing operation time; When quality feedback and model optimization are initiated, collect output quality data and feedback optimization synchronization model , identify defects and update model parameters, and obtain an optimized model And redeploy, so that the data form is transformed from the detection result to the optimization input. 2.The intelligent scheduling method according to claim 1, characterized in that: Through the high-precision sensor network installed on the wire embedding machine and the pressing machine, the operating parameters are monitored and recorded in real time: the stitch speed of the stitcher , the pressing force of the press , the operating time of the stitcher and the press, respectively and ; forming a time series dataset wherein denotes a set of times; Real-time measurement and recording of key attributes of the raw material using material attribute sensors: wire diameter , material hardness ; forming an attribute dataset . 3.The intelligent scheduling method according to claim 2, characterized in that: The root mean square error (RMSE) is used to evaluate the model performance, and if the RMSE exceeds the preset threshold, the model is optimized by adjusting the feature selection or the support vector regression algorithm parameters; The trained synchronization model Deployed to an intelligent scheduling system, with a periodic update mechanism: utilizing newly collected real-time datasets at fixed intervals. Retrain the model and calculate the RMSE of the new model. If the RMSE of the new model is lower than that of the current model, update the deployment. 4.The intelligent scheduling method according to claim 3, characterized in that: Calculating predicted press start time Deviation from actual press operation time Timing deviation If Press operation lags, speed up press or start earlier; if Press operation leads, slow down stitcher or start later; if Operation is on schedule, no adjustment needed. 5.The intelligent scheduling method according to claim 4, characterized in that: According to the size of the deviation, the speed of the wire embedding machine or the pressing machine is dynamically adjusted, and the adjustment strategy is as follows: Press speed adjustment: if , increase press speed , adjustment range is: ; adjusted press speed is: ; wire embedding machine speed adjustment: if , slow down wire embedding machine speed , adjustment range is: ; adjusted wire embedding machine speed is: ; wherein: is the press speed before adjustment; is the stitcher speed before adjustment, is the press speed adjustment amplitude, is the stitcher speed adjustment amplitude, is the press speed adjustment coefficient; is the stitcher speed adjustment coefficient, determined by stitcher performance; is the press speed after adjustment, is the stitcher speed after adjustment. 6.The intelligent scheduling method according to claim 5, characterized in that: Recording adjusted run data, including adjusted wire embedding speed , press speed , actual press start time , and timing cumulative error , building adjustment data set , defined as ; Timing accumulation error The formula for calculating the timing accumulation error is: ， wherein: is a sliding window size, is a forgetting factor, is a quality bias, is a quality sensitivity coefficient, is an index of the sliding window. 7.The intelligent scheduling method according to claim 6, characterized in that: Real-time acquisition of output quality data, including wire position accuracy and press-in strength , preset quality standard threshold: wire position accuracy standard and press-in strength standard ; Quality acceptance conditions are and If not, record non-acceptance data, including real-time data set at corresponding time , adjustment data set and predicted press start time . 8.The intelligent scheduling method according to claim 7, characterized in that: For non-conforming output, a quality deviation index is calculated The quality parameters of the product are represented by a vector The standard quality vector is The quality deviation vector is defined as: ; mass deviation indicator is defined as follows: ， wherein: the meaning of is: by the mass influence matrix transforming the deviation vector and then calculating its Euclidean norm , the static deviation of the quality parameter for measuring; the matrix characterizes the mutual influence between different quality parameters; the parameter is the dynamic term weight coefficient. 9.The intelligent scheduling method according to claim 8, characterized in that: Setting quality feedback threshold , if , triggering synchronization model optimization; Collecting feedback datasets , containing real-time datasets corresponding to subpar outputs , adjusting datasets , predicting press start times and quality data , sample weights are assigned in direct proportion to ​ Feedback dataset , incremental support vector regression is used to optimize the synchronization model ; the optimized model is redeployed, if and the proportion of the promotion, the optimized synchronization model is retained, otherwise the learning rate or the quality feedback threshold is adjusted, and the synchronization model optimization is re-executed; Based on feedback dataset , employing incremental support vector regression to optimize the synchronization model .

Images