Dough sheet rolling process parameter intelligent optimization system

Abstract

The invention relates to the technical field of wheaten food processing, and discloses a dough sheet rolling process parameter intelligent optimization system, which comprises a parameter monitoring end, an intelligent analysis end, a control execution end, a data management end and an early warning notification end, the calendering pressure, the calendering speed, the calendering temperature and the patch thickness are collected in real time through the parameter monitoring end and compared with the dynamically updated standard parameter range, parameter deviation can be recognized in time, the intelligent analysis end deeply analyzes a deviation source through a machine learning algorithm, and a self-adaptive adjustment strategy is generated; the system can deal with the problem of abnormal matching of calendaring pressure and speed, the uniformity of the thickness of the dough sheet is guaranteed, the quality stability and the control precision of the whole calendaring process are improved, production fluctuation caused by parameter mismatching is avoided, and the production efficiency is improved. And the trend prediction function of the intelligent analysis end and the partition fuzzy control method of the execution control end are combined, so that accurate and self-adaptive adjustment of the temperature field is realized.

Description

Technical Field

[0001] This invention relates to the field of pasta processing technology, specifically to an intelligent optimization system for dough sheet rolling process parameters. Background Technology

[0002] People frequently use noodles and other convenient foods in their daily lives. Traditional methods of hand-pulling or processing noodles are labor-intensive and inefficient. The rolling process is widely used in the noodle product processing industry. In noodle production, a dough sheet rolling machine is a machine that continuously rolls cooked dough into sheets of a certain thickness.

[0003] Currently, due to the multi-parameter coupling effect in the sheet calendering process, the parameter monitoring system that collects sheet production data during real-time control of calendering quality cannot identify abnormalities in the matching of calendering pressure and speed in real time. When the parameter coordination deviation persists, it will exacerbate the problem of uneven sheet thickness and make it impossible to guarantee the stability of calendering quality.

[0004] Therefore, an intelligent optimization system for sheet calendering process parameters is proposed to solve the above problems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent optimization system for sheet calendering process parameters, which solves the problem of aggravated sheet thickness unevenness and inability to guarantee the stability of calendering quality mentioned in the background.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent optimization system for sheet calendering process parameters, the system comprising: The parameter monitoring terminal is used to set the standard parameter range for the sheet calendering process and collect the actual process parameters in real time during the calendering process. By comparing the actual parameters with the standard parameters, it can determine whether there is a deviation in the parameters. The standard parameter range is dynamically updated based on historical optimization data. The intelligent analysis terminal is used to analyze the root cause of parameter deviations based on real-time data provided by the parameter monitoring terminal, predict the trend of deviation development, and generate optimized parameter adjustment strategies. The adjustment strategies include pressure adjustment, speed adjustment, and temperature adjustment, and the adjustment strategies are adaptively corrected according to the real-time rolling state. The control execution end is used to receive the optimization parameter adjustment strategy generated by the intelligent analysis end, and adjust the execution instructions of the calendering equipment in real time through the control unit to perform the correction operation of process parameter deviation; The data management terminal is used to store historical process data, optimization strategy records, and system operation logs, and supports data query, backtracking analysis, and model training. The early warning notification terminal is used to remind operators to intervene when abnormal parameters or system malfunctions are detected. The notification mechanism includes status prompts and anomaly level classification.

[0007] Preferably, the parameter monitoring terminal includes a parameter acquisition module, a deviation judgment module, and a standard update module; The parameter acquisition module is used to collect actual process parameters during the sheet rolling process in real time, including rolling pressure, rolling speed, rolling temperature and sheet thickness, and to process the parameter acquisition data using multi-source data fusion technology. The deviation judgment module is used to calculate the parameter deviation value by comparing the difference between the actual collected parameters and the standard parameter range. When the deviation value exceeds the preset threshold, the parameter is judged to be abnormal. The standard update module is used to dynamically adjust the range of standard parameters and update the correspondence between standard parameters and current production conditions based on historical optimization results and real-time process data.

[0008] Preferably, the parameter acquisition module includes a pressure monitoring unit, a speed monitoring unit, and a temperature monitoring unit; The pressure monitoring unit is used to monitor pressure changes in the rolling process in real time and to process noise data through pressure data normalization. The speed monitoring unit is used to collect the speed parameters of the calendering equipment and calculate the relative speed deviation in combination with the sheet conveying speed; The temperature monitoring unit is used to monitor the temperature distribution in the rolling zone in a non-contact manner and analyze hot spots based on temperature gradients.

[0009] Preferably, the intelligent analysis terminal includes a data preprocessing module, a model analysis module, and a strategy generation module; The data preprocessing module is used to clean, normalize, and extract features from the raw data provided by the parameter monitoring terminal, and to process outliers and redundant information in the data. The model analysis module is used to analyze the correlation between parameter deviation and rolling quality through a neural network model, and output the analysis of the causes of deviation. The strategy generation module is used to generate specific parameter adjustment strategies based on the model analysis results, including adjustment magnitude, adjustment timing, and adjustment priority.

[0010] Preferably, the model analysis module includes a deviation attribution unit and a trend prediction unit; The deviation attribution unit is used to identify the main influencing factors of parameter deviation through association rule mining technology, including equipment wear and raw material changes; The trend prediction unit is used to predict future changes in parameter deviations based on time series analysis and to generate early warning signals in advance.

[0011] Preferably, the control execution terminal includes an instruction parsing module, an execution control module, and a feedback verification module; The instruction parsing module is used to receive the optimization parameter adjustment strategy sent by the intelligent analysis terminal and parse it into executable control instructions; The execution control module is used to adjust the operating parameters of the calendering equipment, including motor speed and heating power, according to the parsed control instructions. The feedback verification module is used to perform verification operations on control commands by monitoring the changes in adjusted parameters in real time. If the adjustment does not meet expectations, it will trigger re-optimization.

[0012] Preferably, the execution control module includes a pressure control unit, a speed control unit, and a temperature control unit; The pressure control unit is used to adjust the rolling pressure according to the adjustment strategy, adopts the fuzzy PID control algorithm to adaptively adjust the pressure parameters, and realizes closed-loop control through the pressure feedback signal. The speed control unit is used to synchronously adjust the rolling speed and the sheet conveying speed. It establishes a speed coupling model based on a multi-motor synchronous control strategy and performs speed synchronization control operations. The temperature control unit is used to adjust the temperature of the rolling zone through a graded temperature control method, and to establish a temperature field equilibrium model using a zoned fuzzy control method to perform temperature field adjustment operations.

[0013] Preferably, the data management terminal includes a data storage module, a query analysis module, and a model update module; The data storage module adopts a distributed storage architecture for data storage, and establishes a storage structure for historical process data, optimization strategy records and system operation logs; The query and analysis module establishes a multi-condition-based process data query mechanism and provides a visual analysis function for process status. The model update module establishes a model retraining mechanism based on newly acquired process data and performs parameter update operations on the machine learning model.

[0014] Preferably, the early warning notification terminal includes an anomaly determination module and an early warning strategy generation module; The anomaly determination module establishes an anomaly level classification mechanism based on the comparison results of parameter deviation values ​​and preset thresholds, dividing the abnormal state into Level 1 warning, Level 2 warning and Level 3 warning; The early warning strategy generation module establishes a corresponding response strategy library based on the anomaly level classification results, and generates an early warning handling plan that includes parameter adjustment suggestions, equipment checklists, and manual intervention prompts.

[0015] Preferably, the early warning notification terminal further includes a response execution module and a feedback calibration module; The response execution module automatically executes the basic operation instructions in the parameter adjustment suggestions according to the early warning strategy generation module output early warning processing plan, and activates the self-test program corresponding to the equipment checklist. The feedback calibration module collects actual process parameter data during the calendering process after the early warning processing scheme is executed in real time, compares and verifies the actual parameters with the threshold conditions of the early warning level, and updates the judgment logic of the anomaly judgment module based on the comparison results.

[0016] Compared with the prior art, the present invention provides an intelligent optimization system for sheet rolling process parameters, which has the following beneficial effects: 1. In this invention, the rolling pressure, rolling speed, rolling temperature, and sheet thickness are collected in real time by the parameter monitoring terminal and compared with the dynamically updated standard parameter range. This allows for timely identification of parameter deviations. The intelligent analysis terminal uses machine learning algorithms to deeply analyze the root causes of deviations and generate adaptive adjustment strategies. This enables the system to cope with abnormal matching of rolling pressure and speed, ensuring not only the uniformity of sheet thickness but also improving the quality stability and control accuracy of the entire rolling process, and avoiding production fluctuations caused by parameter mismatch.

[0017] 2. In this invention, the temperature distribution in the rolling zone is monitored and gradient analyzed non-contactly by a temperature monitoring unit. This allows the temperature field state to be sensed. Combined with the trend prediction function of the intelligent analysis terminal and the partitioned fuzzy control method of the execution control terminal, accurate and adaptive adjustment of the temperature field is achieved. This not only allows for real-time assessment and compensation of the adverse effects of temperature fluctuations on the rolling effect, preventing micro-cracks and over-compaction defects on the surface of the sheet, but also improves the uniformity and consistency of the product by balancing the temperature field, ensuring high-quality output.

[0018] 3. In this invention, the systematic storage and management of historical process data, optimization records, and operation logs through the data management terminal provides a solid data foundation for model training and backtracking analysis. The model update module's model retraining mechanism based on new data ensures that the machine learning model can continuously evolve, adapting to changes in raw material ratios and environmental fluctuations. This enables the system to have self-learning capabilities, achieving intelligent optimization and adaptive adjustment of process parameters in multi-batch continuous production, thereby reducing batch-to-batch quality differences and improving production efficiency and yield. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the intelligent optimization system for sheet rolling process parameters according to the present invention. Detailed Implementation

[0020] 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.

[0021] For specific implementation examples, please refer to: Figure 1 A smart optimization system for sheet calendering process parameters, the system comprising: The parameter monitoring terminal is used to set the standard parameter range for the sheet calendering process and collect the actual process parameters in real time during the calendering process. By comparing the actual parameters with the standard parameters, it can determine whether there is a deviation in the parameters. The standard parameter range is dynamically updated based on historical optimization data. The intelligent analysis terminal is used to analyze the root cause of parameter deviations based on real-time data provided by the parameter monitoring terminal, predict the trend of deviation development, and generate optimized parameter adjustment strategies. The adjustment strategies include pressure adjustment, speed adjustment, and temperature adjustment, and the adjustment strategies are adaptively corrected according to the real-time rolling state. The control execution end is used to receive the optimization parameter adjustment strategy generated by the intelligent analysis end, and adjust the execution instructions of the calendering equipment in real time through the control unit to perform the correction operation of process parameter deviation; The data management terminal is used to store historical process data, optimization strategy records, and system operation logs, and supports data query, backtracking analysis, and model training. The early warning notification terminal is used to remind operators to intervene when abnormal parameters or system malfunctions are detected. The notification mechanism includes status prompts and anomaly level classification.

[0022] The parameter monitoring terminal includes a parameter acquisition module, a deviation judgment module, and a standard update module; The parameter acquisition module is used to collect actual process parameters during the sheet rolling process in real time, including rolling pressure, rolling speed, rolling temperature, and sheet thickness. It then uses multi-source data fusion technology to process the acquired parameter data, specifically including: Timestamp alignment is performed for the same type of process parameters from different acquisition sources; Calculate the confidence weight for each data source. The confidence weight is determined based on the historical error rate of that data source. The calculation formula is as follows: ; in, Indicates the first Confidence weights of each data source Indicates the first Historical error rate of each data source Indicates the first Historical error rate of each data source It is a very small positive number. This indicates the total number of data sources participating in the integration; Based on the confidence weights, a weighted average algorithm is used to calculate the fusion value of this type of process parameter. The calculation formula is as follows: ; in, Indicates time The fused process parameter values, Indicates time No. Process parameter values ​​collected from one data source. Indicates time No. Confidence weights of each data source This indicates the total number of data sources participating in the integration; The deviation judgment module is used to calculate the parameter deviation value by comparing the difference between the actual collected parameters and the standard parameter range. When the deviation value exceeds the preset threshold, the parameter is judged to be abnormal. The standard update module is used to dynamically adjust the range of standard parameters and update the correspondence between standard parameters and current production conditions based on historical optimization results and real-time process data.

[0023] The parameter acquisition module includes a pressure monitoring unit, a speed monitoring unit, and a temperature monitoring unit; The pressure monitoring unit is used to monitor pressure changes in real time during the rolling process and to process noise data through pressure data normalization. The speed monitoring unit is used to collect the speed parameters of the calendering equipment and calculate the relative speed deviation in conjunction with the sheet conveying speed, specifically including: The linear speed of the calendering rolls in the calendering equipment and the linear speed of the conveyor belt in the sheet conveying equipment are collected simultaneously. The collected calender roll linear speed and conveyor belt linear speed are timestamped and aligned. Based on the aligned velocity data, the relative velocity deviation is calculated using the following formula: ; in, This indicates the linear velocity of the calendering roll. This indicates the linear speed of the conveyor belt. Indicates relative velocity deviation; The temperature monitoring unit is used to monitor the temperature distribution in the rolling zone in a non-contact manner and analyze hot spots based on the temperature gradient.

[0024] The intelligent analysis module includes a data preprocessing module, a model analysis module, and a strategy generation module; The data preprocessing module is used to clean, normalize, and extract features from the raw data provided by the parameter monitoring terminal, and to handle outliers and redundant information in the data. Specifically, it includes: Clean the raw data, identify and process missing values ​​and format errors in the data, and detect, replace and remove abnormal data values ​​that exceed the physical reasonable range based on preset rules; The cleaned data is normalized to linearly transform process parameter values ​​from different monitoring points and with different dimensions into a unified numerical range. Feature extraction is performed on the normalized data. By calculating statistical feature quantities and analyzing the correlation between parameters, a feature dataset for model analysis is constructed, and highly correlated redundant feature information is removed. The model analysis module is used to analyze the correlation between parameter deviations and rolling quality through a neural network model, and outputs an analysis of the causes of deviations, specifically including: The parameter deviation values ​​are used as input features of the input layer, and the input features include pressure deviation. Speed ​​deviation Temperature deviation and thickness deviation ; The input features are nonlinearly transformed using at least one hidden layer. The hidden layer employs the ReLU activation function, the mathematical expression of which is: ; in, Indicates the first Layer The output value of each neuron Indicates the connection of the first Layer The first neuron and the second Layer The weights of each neuron, Indicates the first Layer The activation value of each neuron. Indicates the first Layer Bias of each neuron For indexes of neural network layers; The analysis results of the causes of output deviations are presented in the output layer, and the results are expressed in probability form to represent the likelihood of each potential influencing factor. The strategy generation module is used to generate specific parameter adjustment strategies based on the model analysis results, including adjustment magnitude, adjustment timing, and adjustment priority.

[0025] The model analysis module includes a bias attribution unit and a trend prediction unit; The deviation attribution unit is used to identify the main influencing factors of parameter deviations through association rule mining techniques, including equipment wear and raw material variations, specifically including: Building a transactional database Each record contains a set of concurrently occurring parameter deviation events and production condition attributes; The Apriori algorithm is used to find frequent itemsets, which are those that satisfy the minimum support threshold. Project collection, itemset support The calculation formula is: ; in, Representation Itemset Support Represents a transactional database Includes itemsets The number of transactions, Represents a transactional database The total number of transactions in the process; Association rules are generated based on frequent itemsets. And calculate its confidence level. The calculation formula is: ; in, Representing association rules confidence level Representation Itemset and The support of the union; Confidence level higher than preset threshold The production condition attributes corresponding to the rules are the main influencing factors; The trend prediction unit is used to predict future changes in parameter deviations based on time series analysis and to generate early warning signals, specifically including: Historical parameter deviation data series Perform a stationarity test; if the sequence is not stationary, then... The sequence after difference processing is: ; Identifying the structural parameters of ARIMA models based on autocorrelation and partial autocorrelation plots. ; Using the identified ARIMA The model makes rolling predictions of future parameter deviations. The general expression of the model is: ; in, Indicates time The parameter deviation value, For the shift operator, In the autoregressive part, the first... The lag operator of order, Indicating the moving average part of the first... The lag operator of order, These are the autoregressive coefficients. The moving average coefficient is... Indicates time The white noise error term, Let the order be the autoregressive order. Let be the difference order. The moving average order; The control execution end includes an instruction parsing module, an execution control module, and a feedback verification module; The instruction parsing module is used to receive the optimization parameter adjustment strategy sent by the intelligent analysis terminal and parse it into executable control instructions; The execution control module is used to adjust the operating parameters of the calendering equipment, including motor speed and heating power, according to the parsed control instructions; The feedback verification module is used to verify the control commands by monitoring the changes in the adjusted parameters in real time. If the adjustment does not meet expectations, it will trigger a re-optimization.

[0026] The execution control module includes a pressure control unit, a speed control unit, and a temperature control unit; The pressure control unit is used to adjust the rolling pressure according to the adjustment strategy. It adopts a fuzzy PID control algorithm to adaptively adjust the pressure parameters and realizes closed-loop control through pressure feedback signals. Specifically, it includes: Real-time acquisition of pressure setpoint Compared with the actual pressure value Calculate pressure deviation and its rate of change And these two are used as inputs to the fuzzy controller, and the calculation formula is as follows: ; ; in, In time Pressure deviation, Set the pressure value. In time The actual pressure value, In time The rate of change of pressure deviation; Based on the fuzzy rule table, the adjustment amount of the PID control parameters is calculated through fuzzy inference; The adjusted PID parameters are applied to the pressure control loop to generate control commands. The calculation formula is as follows: ; in, Indicates time The control output, Indicates time Pressure deviation, , , These are the initial proportional, integral, and derivative coefficients of the PID controller. , , This refers to the adjustment amount of the PID parameters. For integration variables; The speed control unit is used to synchronously adjust the rolling speed and the sheet conveying speed. Based on a multi-motor synchronous control strategy, a speed coupling model is established to execute speed synchronization control operations, specifically including: Establish speed command for main motor Feedback from motor speed The coupling relationship equation between them: ; in, This indicates the speed command for the main motor. Indicates the first The speed feedback value from the motor, Indicates the first Speed ​​tracking error of the motor For motor indexing; A cross-coupled control structure is adopted to integrate the speed error signals of each motor. Coupling compensation is performed, and the synchronization error after compensation is... The calculation formula is: ; in, This indicates the total number of motors under synchronous control. Indicates the first Synchronization error after compensation by the motor Indicates the first Speed ​​tracking error of the motor Indicates all The average value of the speed tracking error of the motor; Based on the compensated synchronization error Combined with the original tracking error Calculate the control output of each motor separately. To achieve speed synchronization, the calculation formula is as follows: ; in, Indicates the first The control output of the trolley motor, Indicates tracking error The control gain coefficient, Indicates synchronization error The control gain coefficient; The temperature control unit is used to adjust the temperature of the rolling zone through a graded temperature control method. It establishes a temperature field equilibrium model using a zoned fuzzy control method and executes temperature field adjustment operations, specifically including: The rolling zone is divided into Each independent control sub-region; Set temperature setpoints for each sub-zone and the actual temperature of each region Deviation from set value As input to the partitioned fuzzy controller, that is: ; in, Indicates the first Temperature control deviation in each area For the first Temperature setpoints for each area For the first The actual temperature value of each region For the index of the temperature control area, ; Each zone's fuzzy controller independently calculates the adjustment amount of heating power. To achieve a balanced temperature field in a coordinated manner, the first The final control output of each region for: ; in, For the first Heating power adjustment for each area For the first The reference heating power for each region For the first The final control output for each region.

[0027] The data management module includes a data storage module, a query and analysis module, and a model update module; The data storage module adopts a distributed storage architecture for data storage, establishing a storage structure for historical process data, optimization strategy records, and system operation logs, specifically including: Historical process data, optimization strategy records, and system operation logs are segmented according to time series. Different data shards are stored on different physical nodes, and a data shard index table is created; Data shards are located and accessed using a consistent hashing algorithm; The query and analysis module establishes a multi-condition-based process data query mechanism and provides a visual analysis function for process status; The model update module establishes a model retraining mechanism based on newly acquired process data, performing parameter update operations on the machine learning model, specifically including: Set a model performance degradation threshold, specifically by obtaining the baseline accuracy of the original machine learning model on a historical validation dataset. Based on performance tolerance requirements, a relative attenuation coefficient is set. The model performance degradation threshold is calculated according to the following formula. : ; in, This represents the baseline accuracy obtained by evaluating the model on the historical validation dataset. This represents the allowable relative performance degradation coefficient. This is the performance degradation threshold; The calculated threshold As a trigger condition, when the model's accuracy on the new validation dataset... satisfy When this happens, model retraining is triggered; The newly added process data is combined with some historical data to form a new training dataset; Incremental training of the existing machine learning model is performed using a new training dataset to update its model parameters.

[0028] The early warning notification module includes an anomaly detection module and an early warning strategy generation module; The anomaly detection unit establishes an anomaly level classification mechanism based on the comparison results of parameter deviation values ​​and preset thresholds, dividing anomaly states into Level 1 warning, Level 2 warning, and Level 3 warning, specifically including: Set parameter deviation threshold ranges corresponding to different warning levels, with the threshold ranges divided based on the absolute value of the deviation; The calculated parameter deviation value is matched with its corresponding threshold interval; Based on the matching results, the current abnormal status is determined to be the corresponding Level 1, Level 2, and Level 3 warnings; The early warning strategy generation module establishes a corresponding response strategy library based on the anomaly level classification results, and generates early warning handling solutions including parameter adjustment suggestions, equipment inspection checklists, and manual intervention prompts.

[0029] The early warning notification terminal also includes a response execution module and a feedback calibration module; The response execution module generates an early warning processing plan based on the early warning strategy, automatically executes the basic operation instructions in the parameter adjustment suggestions, and activates the self-test program corresponding to the equipment checklist. The feedback calibration module collects the actual process parameter data during the calendering process after the early warning processing plan is executed in real time, compares and verifies the actual parameters with the threshold conditions of the early warning level, and updates the judgment logic of the anomaly judgment module based on the comparison results.

[0030] The operation steps of this intelligent optimization system for sheet calendering process parameters are as follows: Step 1: Real-time parameter monitoring and data acquisition: The system first sets the standard parameter range for the sheet calendering process through the parameter monitoring terminal, and then collects the actual process parameters in real time, including calendering pressure, calendering speed, calendering temperature, and sheet thickness. The parameter acquisition module uses multi-source data fusion technology to align the timestamps and calculate confidence weights for data from different sources, and obtains the fused process parameter values ​​through a weighted average algorithm. The deviation judgment module compares the actual collected parameters with the standard parameter range, calculates the parameter deviation value, and determines the parameter to be abnormal when the deviation value exceeds a preset threshold, providing an accurate data foundation for subsequent intelligent analysis.

[0031] Step 2: Data Preprocessing and Feature Extraction The data preprocessing module cleans, normalizes, and extracts features from the raw data provided by the parameter monitoring station. Specifically, this includes identifying and handling missing values ​​and format errors in the data, removing outlier values ​​that exceed physically reasonable ranges; linearly transforming process parameter values ​​from different monitoring points and with different dimensions to a unified numerical range; and constructing a feature dataset for model analysis by calculating statistical features and analyzing the correlation between parameters, while removing highly correlated redundant features to provide high-quality data input for the model analysis module.

[0032] Step 3: Intelligent Analysis and Strategy Generation The intelligent analysis unit, based on preprocessed data, analyzes the correlation between parameter deviations and rolling quality using a neural network model, outputting a deviation cause analysis. The deviation attribution unit uses association rule mining technology to identify the main influencing factors of parameter deviations, while the trend prediction unit predicts future changes in parameter deviations based on time series analysis. The strategy generation module generates specific parameter adjustment strategies based on the model analysis results, including pressure adjustment, speed adjustment, and temperature adjustment. These adjustment strategies are adaptively corrected based on real-time rolling conditions, forming targeted optimization schemes.

[0033] Step 4: Execution of Control Commands and Real-time Adjustment: The control execution unit receives the optimized parameter adjustment strategy generated by the intelligent analysis unit, and the instruction parsing module parses it into executable control instructions. The execution control module adjusts the operating parameters of the calendering equipment according to the parsed instructions: the pressure control unit uses a fuzzy PID control algorithm to adaptively adjust the pressure parameters; the speed control unit establishes a speed coupling model based on a multi-motor synchronous control strategy to achieve synchronization between calendering speed and conveying speed; the temperature control unit establishes a temperature field equilibrium model through a partitioned fuzzy control method. The feedback verification module monitors the changes in the adjusted parameters in real time to verify the execution effect of the control instructions.

[0034] Step 5: Early Warning Notification and Feedback Calibration: When the early warning notification module detects abnormal parameters or system malfunctions, the anomaly determination unit classifies the abnormal state into different early warning levels, and the strategy generation unit generates corresponding early warning handling plans based on the anomaly level. The response execution unit automatically executes the basic operation instructions in the parameter adjustment suggestions and activates the self-test program corresponding to the equipment checklist. The feedback calibration unit collects the actual process parameter data after the early warning handling plan is executed in real time, compares and verifies the actual parameters with the threshold conditions of the early warning level, and updates the judgment logic of the anomaly determination unit based on the comparison results, thereby achieving system self-optimization.

[0035] Step Six: Data Management and Model Updates The data management module employs a distributed storage architecture to store historical process data, optimization strategy records, and system operation logs, supporting data querying and backtracking analysis. The model update module establishes a model retraining mechanism based on newly acquired process data. When the model's accuracy on a newly added validation dataset falls below a performance degradation threshold, the model retraining process is triggered. By combining new data with historical data to create a new training dataset, the original machine learning model is incrementally trained, continuously optimizing model parameters and ensuring the system's continuous learning capability and adaptability.

[0036] Step 7: Closed-loop optimization and continuous improvement: The system forms a complete closed-loop optimization process through the collaborative work of modules such as parameter monitoring, intelligent analysis, control execution, early warning notification, and data management. The standard update module dynamically adjusts the standard parameter range based on historical optimization results and real-time process data, enabling the system to adapt to changes in production conditions. The entire system achieves continuous optimization and control of the wafer calendering process parameters through a cyclical mechanism of real-time data acquisition, intelligent algorithm analysis, and automatic adjustment, ensuring the stability of the production process and the consistency of product quality.

[0037] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0038] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A smart optimization system for sheet calendering process parameters, characterized in that: The system includes: The parameter monitoring terminal is used to set the standard parameter range for the sheet calendering process and collect the actual process parameters in real time during the calendering process. By comparing the actual parameters with the standard parameters, it can determine whether there is a deviation in the parameters. The standard parameter range is dynamically updated based on historical optimization data. The intelligent analysis terminal is used to analyze the root cause of parameter deviations based on real-time data provided by the parameter monitoring terminal, predict the trend of deviation development, and generate optimized parameter adjustment strategies. The adjustment strategies include pressure adjustment, speed adjustment, and temperature adjustment, and the adjustment strategies are adaptively corrected according to the real-time rolling state. The control execution end is used to receive the optimization parameter adjustment strategy generated by the intelligent analysis end, and adjust the execution instructions of the calendering equipment in real time through the control unit to perform the correction operation of process parameter deviation; The data management terminal is used to store historical process data, optimization strategy records, and system operation logs, and supports data query, backtracking analysis, and model training. The early warning notification terminal is used to remind operators to intervene when abnormal parameters or system malfunctions are detected. The notification mechanism includes status prompts and anomaly level classification.

2. The intelligent optimization system for sheet calendering process parameters according to claim 1, characterized in that: The parameter monitoring terminal includes a parameter acquisition module, a deviation judgment module, and a standard update module; The parameter acquisition module is used to collect actual process parameters during the sheet rolling process in real time, including rolling pressure, rolling speed, rolling temperature and sheet thickness, and to process the parameter acquisition data using multi-source data fusion technology. The deviation judgment module is used to calculate the parameter deviation value by comparing the difference between the actual collected parameters and the standard parameter range. When the deviation value exceeds the preset threshold, the parameter is judged to be abnormal. The standard update module is used to dynamically adjust the range of standard parameters and update the correspondence between standard parameters and current production conditions based on historical optimization results and real-time process data.

3. The intelligent optimization system for sheet calendering process parameters according to claim 2, characterized in that: The parameter acquisition module includes a pressure monitoring unit, a speed monitoring unit, and a temperature monitoring unit; The pressure monitoring unit is used to monitor pressure changes in the rolling process in real time and to process noise data through pressure data normalization. The speed monitoring unit is used to collect the speed parameters of the calendering equipment and calculate the relative speed deviation in combination with the sheet conveying speed; The temperature monitoring unit is used to monitor the temperature distribution in the rolling zone in a non-contact manner and analyze hot spots based on temperature gradients.

4. The intelligent optimization system for sheet calendering process parameters according to claim 1, characterized in that: The intelligent analysis terminal includes a data preprocessing module, a model analysis module, and a strategy generation module; The data preprocessing module is used to clean, normalize, and extract features from the raw data provided by the parameter monitoring terminal, and to process outliers and redundant information in the data. The model analysis module is used to analyze the correlation between parameter deviation and rolling quality through a neural network model, and output the analysis of the causes of deviation. The strategy generation module is used to generate specific parameter adjustment strategies based on the model analysis results, including adjustment magnitude, adjustment timing, and adjustment priority.

5. The intelligent optimization system for sheet calendering process parameters according to claim 4, characterized in that: The model analysis module includes a deviation attribution unit and a trend prediction unit; The deviation attribution unit is used to identify the main influencing factors of parameter deviation through association rule mining technology, including equipment wear and raw material changes; The trend prediction unit is used to predict future changes in parameter deviations based on time series analysis and to generate early warning signals in advance.

6. The intelligent optimization system for sheet calendering process parameters according to claim 1, characterized in that: The control execution terminal includes an instruction parsing module, an execution control module, and a feedback verification module; The instruction parsing module is used to receive the optimization parameter adjustment strategy sent by the intelligent analysis terminal and parse it into executable control instructions; The execution control module is used to adjust the operating parameters of the calendering equipment, including motor speed and heating power, according to the parsed control instructions. The feedback verification module is used to perform verification operations on control commands by monitoring the changes in adjusted parameters in real time. If the adjustment does not meet expectations, it will trigger re-optimization.

7. The intelligent optimization system for sheet calendering process parameters according to claim 6, characterized in that: The execution control module includes a pressure control unit, a speed control unit, and a temperature control unit; The pressure control unit is used to adjust the rolling pressure according to the adjustment strategy, adopts the fuzzy PID control algorithm to adaptively adjust the pressure parameters, and realizes closed-loop control through the pressure feedback signal. The speed control unit is used to synchronously adjust the rolling speed and the sheet conveying speed. It establishes a speed coupling model based on a multi-motor synchronous control strategy and performs speed synchronization control operations. The temperature control unit is used to adjust the temperature of the rolling zone through a graded temperature control method, and to establish a temperature field equilibrium model using a zoned fuzzy control method to perform temperature field adjustment operations.

8. The intelligent optimization system for sheet calendering process parameters according to claim 1, characterized in that: The data management terminal includes a data storage module, a query analysis module, and a model update module; The data storage module adopts a distributed storage architecture for data storage, and establishes a storage structure for historical process data, optimization strategy records and system operation logs; The query and analysis module establishes a process data query mechanism based on multiple conditions and provides a visual analysis function for process status. The model update module establishes a model retraining mechanism based on newly acquired process data and performs parameter update operations on the machine learning model.

9. The intelligent optimization system for sheet calendering process parameters according to claim 1, characterized in that: The early warning notification terminal includes an anomaly detection module and an early warning strategy generation module; The anomaly determination module establishes an anomaly level classification mechanism based on the comparison results of parameter deviation values ​​and preset thresholds, dividing the abnormal state into Level 1 warning, Level 2 warning and Level 3 warning; The early warning strategy generation module establishes a corresponding response strategy library based on the anomaly level classification results, and generates an early warning handling plan that includes parameter adjustment suggestions, equipment checklists, and manual intervention prompts.

10. The intelligent optimization system for sheet calendering process parameters according to claim 9, characterized in that: The early warning notification terminal also includes a response execution module and a feedback calibration module; The response execution module automatically executes the basic operation instructions in the parameter adjustment suggestions according to the early warning strategy generation module output early warning processing plan, and activates the self-test program corresponding to the equipment checklist. The feedback calibration module collects actual process parameter data during the calendering process after the early warning processing scheme is executed in real time, compares and verifies the actual parameters with the threshold conditions of the early warning level, and updates the judgment logic of the anomaly judgment module based on the comparison results.

Images