An online custom four-way system for gypsum board

Abstract

The present application relates to paper gypsum board technical field, the present application provides a kind of paper gypsum board online customization four division system, to solve the problem of material waste caused by unable to analyze sawing process in prior art, system includes data acquisition, sawing prediction, data analysis, sawing verification and sawing optimization etc.Modulus.Through pre-processing historical data training random forest and deep learning model, realize the accurate prediction and control of sawing equipment.After sawing, the system verifies sawing error in real time, and improves sawing accuracy by optimizing control model through genetic algorithm.The present application effectively improves sawing efficiency, reduces material waste, meets the diversified construction demand, and realizes intelligent production.

Description

Technical Field

[0001] This invention relates to the field of paper-faced gypsum board technology, and more particularly to an online custom four-part system for paper-faced gypsum board. Background Technology

[0002] Online custom-cut gypsum board panels to 1 / 4 inch size are widely used in interior decoration and construction projects. When performing this online custom-cutting, mechanical equipment such as fully automatic four-sided cutting saws or high-precision CNC edge-cutting machines is typically used. Fully automatic four-sided cutting saws improve cutting accuracy and efficiency while reducing dust generation, protecting the construction environment and worker health. Before sawing, the lines to be cut are marked on the gypsum board, and then the infrared positioning or CNC adjustment function of the mechanical equipment is used for precise cutting. Custom-cutting methods are not only suitable for cutting large areas of gypsum board but can also be flexibly adjusted according to specific needs to meet different construction requirements.

[0003] Existing online custom gypsum board sawing processes cut gypsum boards according to preset sawing requirements, without conducting pre-analysis before sawing. When cutting large slabs in batches, if the sawing process is not pre-analyzed and calibrated, errors in the sawing process will lead to waste of gypsum board materials. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an online custom four-part gypsum board system. This system solves the problem that existing online custom four-part gypsum board sawing processes cut gypsum boards according to preset sawing requirements, without prior analysis before sawing. When cutting large batches of large boards, without prior analysis and calibration of the sawing process, errors in gypsum board sawing will lead to waste of gypsum board material.

[0005] To solve the above-mentioned technical problems, the specific technical solution of the present invention is as follows:

[0006] This invention provides an online custom four-part system for paper-faced gypsum board, comprising:

[0007] The data acquisition module acquires historical data on raw materials, production processes, equipment status, product specifications, customer needs, and quality control.

[0008] The sawing prediction module preprocesses historical data of raw materials, production processes, equipment status, product specifications, customer demand, and quality control to obtain preprocessed data. The preprocessed data is then used to train a random forest model to obtain a sawing equipment prediction model. The module also receives real-time customer demand data and real-time sawing quality standards, and substitutes the real-time customer demand data into the sawing equipment prediction model to obtain the sawing equipment prediction data.

[0009] The data analysis module acquires historical data of the online customization four-part system for paper-faced gypsum board and historical data of the sawing equipment. It trains a deep learning model with historical data of customer demand, quality control, and the operation of the sawing equipment to obtain a control model for the online customization four-part system for paper-faced gypsum board. It then substitutes real-time customer demand data into the control model to output real-time control parameters for the online customization four-part system for paper-faced gypsum board.

[0010] The sawing verification module transmits the real-time control parameters of the online custom gypsum board four-part system to the sawing equipment in the online custom gypsum board four-part system. After the sawing equipment executes the real-time control parameters of the online custom gypsum board four-part system, it generates the real-time operating data of the first sawing equipment. The real-time operating data of the first sawing equipment is compared with the predicted data of the sawing equipment to obtain the sawing error data.

[0011] The sawing optimization module optimizes the control model of the online customization four-part system for paper-faced gypsum board, resulting in an optimized control model. Real-time customer demand data is then input into the optimized control model to obtain optimized real-time control parameters for the online customization four-part system. Control parameters for the sawing equipment at the position to be corrected are generated, and the sawing equipment executes these control parameters to complete the cutting of the gypsum board.

[0012] Furthermore, in the paper-faced gypsum board online customization four-part system of the present invention, the data acquisition module includes:

[0013] Feature extraction is performed on the four-part online customization system of paper-faced gypsum board to obtain feature data of the four-part online customization system of paper-faced gypsum board. Based on the feature data of the four-part online customization system of paper-faced gypsum board, the data type to be acquired is determined.

[0014] Data is acquired based on the type of data to be acquired, and the acquired data is checked for continuity. If data is missing, a data missing warning message is generated.

[0015] The preset data missing warning information is substituted into the data simulation generation model of the online customization four-part system for paper-faced gypsum board to generate simulated data corresponding to the data missing warning information. The simulated data corresponding to the data missing warning information is then used to fill in the missing data.

[0016] Furthermore, in the paper-faced gypsum board online customization four-part system of the present invention, the sawing prediction module includes:

[0017] Receive real-time customer demand data through data interface or user input interface. The real-time customer demand data includes the size, shape and quantity of gypsum board.

[0018] Receive real-time sawing quality standards, which include sawing accuracy, allowable error range, and surface quality;

[0019] The pre-processed real-time customer demand data is substituted into the pre-trained sawing equipment prediction model. The sawing equipment prediction model is trained based on historical data of raw materials, historical data of production processes, and historical data of equipment status. The sawing equipment prediction model predicts the operating parameters and output results of the sawing equipment based on the input customer demand data.

[0020] The prediction data of the sawing equipment is obtained by calculating the prediction model of the sawing equipment. The prediction data of the sawing equipment includes the prediction parameters of the sawing position, the prediction parameters of the speed, and the prediction parameters of the cutting depth.

[0021] Furthermore, in the paper-faced gypsum board online customization four-part system of the present invention, the data analysis module includes:

[0022] The historical data of the online customization four-part system for paper-faced gypsum board and the historical data of sawing equipment operation are divided into training set and validation set.

[0023] The selected deep learning model is trained using the preprocessed training dataset. During training, the model weights are adjusted using the backpropagation algorithm. The model is validated using the validation set and evaluated using the test set. The model's performance metrics are calculated, and the model is optimized based on the evaluation results to obtain the control model for the online customization of paper-faced gypsum board in four parts.

[0024] Furthermore, in the paper-faced gypsum board online customization four-part system of the present invention, the sawing verification module includes:

[0025] The real-time operating data of the first sawing equipment is obtained from the sawing equipment. The real-time operating data of the first sawing equipment includes the actual position parameters, actual speed parameters, and actual depth parameters of the sawing.

[0026] The real-time operating data of the first sawing equipment is aligned with the predicted data of the sawing equipment in both time and space;

[0027] The real-time operating data of the first sawing equipment is compared with the predicted data of the sawing equipment item by item, and the error of each data item is calculated, that is, the difference between the actual value and the predicted value.

[0028] All calculated errors are summarized, and sawing error data is generated based on the summarized error results.

[0029] Furthermore, in the paper-faced gypsum board online customization four-part system of the present invention, the sawing optimization module includes:

[0030] The real-time operating data of the first sawing equipment is compared with the real-time sawing quality standard. When the sawing error data exceeds the real-time sawing quality standard, the operating data of the data acquisition terminal is obtained. The data acquisition terminal is then inspected based on this data. If the inspection is successful, a genetic algorithm is used to optimize the control model of the online custom gypsum board four-part system based on the sawing error data, historical data of the online custom gypsum board four-part system, and historical operating data of the sawing equipment. This results in an optimized control model. Real-time customer demand data is then substituted into the optimized control model to obtain the optimized real-time control parameters for the online custom gypsum board four-part system. The control parameters of the real-time online customization four-part system for gypsum board are substituted into the prediction model of the sawing equipment to generate real-time operating data of the second sawing equipment. The real-time operating data of the second sawing equipment is compared with the real-time sawing quality standard. If the real-time operating data of the second sawing equipment meets the real-time sawing quality standard range, the image data of the gypsum board to be cut is collected. The image data of the gypsum board to be cut is compared with the real-time customer demand data to obtain the position to be corrected in the image data of the gypsum board to be cut. The position to be corrected in the image data of the gypsum board to be cut is transmitted to the optimized control model of the online customization four-part system for gypsum board to generate the control parameters of the sawing equipment for the position to be corrected. The sawing equipment executes the control parameters of the sawing equipment for the position to be corrected to complete the cutting of the gypsum board to be cut.

[0031] Furthermore, in the paper-faced gypsum board online customization four-part system of the present invention, the sawing optimization module includes:

[0032] Based on sawing error data and real-time sawing quality standards, determine the optimization objectives;

[0033] Acquire sawing error data, historical data from the online custom gypsum board four-part system, and historical operating data of the sawing equipment.

[0034] The data was used as input for the genetic algorithm optimization to train and optimize the control model of the online customization four-part system for paper-faced gypsum board.

[0035] Receive and set the parameters of the genetic algorithm, which include population size, selection, crossover, mutation probability, and number of iterations;

[0036] Initialize the population by generating an initial population using a random or specific method, where each individual represents a possible combination of control model parameters.

[0037] Each individual in the population is evaluated, and its fitness value is calculated;

[0038] Individuals with high fitness values ​​are selected as parents to generate the next generation. The selection operation can be carried out using roulette wheel selection or tournament selection methods.

[0039] Perform a crossover operation on the selected parent generation to generate new offspring individuals;

[0040] Mutation operations are performed on offspring individuals to introduce new genes, and the newly generated offspring individuals replace the old population to form a new generation of population.

[0041] Repeat the selection, crossover, and mutation processes until the predetermined number of iterations is reached or the optimization objective is met.

[0042] The individual with the highest fitness value was selected from the final population and used as the parameter for the optimized four-part system control model of online customization of paper-faced gypsum board;

[0043] The optimized control model for the online customization of gypsum board in four-part systems is obtained by updating the parameters of the optimized control model.

[0044] The beneficial effects of this invention are:

[0045] This invention, by introducing a data acquisition module, a sawing prediction module, and a data analysis module, enables precise prediction and analysis before the sawing operation, thereby improving sawing accuracy and efficiency. Real-time calibration and optimization mechanisms ensure that the sawing operation maintains a high level of precision at all times, reducing material waste caused by sawing errors.

[0046] By performing pre-analysis and prediction before sawing, this invention can effectively avoid waste of gypsum board material caused by errors in the large-scale cutting process. The real-time sawing verification and optimization mechanism further ensures the accuracy of each sawing operation and reduces the scrap rate.

[0047] This invention enables customized sawing operations based on real-time customer needs, meeting diverse requirements. Optimized sawing parameters and precise cutting results improve product quality, thereby increasing customer satisfaction. By integrating various advanced data processing and analysis technologies (such as random forest models, deep learning models, and genetic algorithms), this invention significantly enhances the system's intelligence and automation levels. It reduces the need for manual intervention, lowers the possibility of human error, and improves production efficiency and consistency.

[0048] This invention establishes a closed-loop optimization mechanism that continuously optimizes and improves the system control model by constantly collecting and analyzing data during the sawing process. This continuous improvement capability enables the system to adapt to ever-changing customer needs and production environments, maintaining its competitiveness and flexibility.

[0049] This invention enables rapid response to real-time changes in customer demand, adjusting sawing parameters and production plans to improve production flexibility and responsiveness. It helps businesses maintain a competitive edge in a highly competitive market environment and quickly meet market demands.

[0050] In summary, by introducing a variety of advanced data processing and analysis technologies, this invention establishes an efficient, intelligent, and flexible online customization four-part system for paper-faced gypsum board, which significantly improves the precision and efficiency of sawing, reduces material waste, enhances customer satisfaction, and increases the system's intelligence and automation level. At the same time, through continuous optimization and improvement mechanisms, this invention also promotes increased production flexibility and response speed. Attached Figure Description

[0051] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on the drawings without creative effort.

[0052] Figure 1 A schematic diagram of the functional modules of the online customization four-part system for paper-faced gypsum board provided in an embodiment of the present invention. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention. The technical solutions provided by various embodiments of this invention will be described in detail below with reference to the accompanying drawings. To better understand the objectives of this invention, it will be described in further detail below.

[0054] This invention provides an online custom four-part system for paper-faced gypsum board, comprising:

[0055] The data acquisition module acquires historical data on raw materials, production processes, equipment status, product specifications, customer needs, and quality control.

[0056] In the online four-part customization system for paper-faced gypsum board, the system is responsible for collecting and integrating various historical data to support subsequent data analysis and sawing prediction. Specifically, the data acquisition module needs to acquire the following types of historical data:

[0057] Historical raw material data records relevant information about the paper-faced gypsum board raw materials used in the past, such as the source, batch, and quality parameters (e.g., density, strength, moisture content). Analyzing this historical data allows us to understand the stability and quality fluctuations of the raw materials, thus providing a reference for setting sawing parameters.

[0058] Historical production process data: This data reflects the process parameters and conditions at each stage of gypsum board production, such as temperature, humidity, pressure, and curing time. This data is crucial for understanding how variables in the production process affect the final quality and sawing performance of the gypsum board.

[0059] Historical Equipment Status Data: This data records the status information of the sawing equipment during its past operation, such as wear and tear, fault records, and maintenance records. Analyzing this data allows for prediction of the equipment's current performance, timely detection of potential problems, and prevention of material waste caused by equipment malfunctions during the sawing process.

[0060] Historical Product Specification Data: This data describes the dimensions, shape, thickness, and other specifications of different batches or types of gypsum board. This data is crucial for understanding customer needs and customizing sawing solutions.

[0061] Historical Customer Demand Data: This data records customers' specific requirements for gypsum board products, such as size, shape, quantity, and delivery date. Analyzing this historical data allows us to understand market demand trends and changes, providing a basis for developing more precise sawing plans.

[0062] Historical Quality Control Data: Quality control data records quality information for gypsum board products during production and final inspection, such as dimensional deviations, surface quality, and strength test results. This data is crucial for evaluating product pass rates and quality control levels, and also provides feedback for optimizing sawing parameters.

[0063] The data acquisition module collects and integrates this historical data, providing a foundation for subsequent sawing prediction, data analysis, and system control.

[0064] The sawing prediction module preprocesses historical data of raw materials, production processes, equipment status, product specifications, customer demand, and quality control to obtain preprocessed data. The preprocessed data is then used to train a random forest model to obtain a sawing equipment prediction model. The module also receives real-time customer demand data and real-time sawing quality standards, and substitutes the real-time customer demand data into the sawing equipment prediction model to obtain the sawing equipment prediction data.

[0065] The sawing prediction module, a key function of the online custom gypsum board system, is to predict the sawing process using historical data, thereby providing guidance and optimization suggestions for actual sawing operations. The following are the detailed steps of how this module works:

[0066] The sawing prediction module preprocesses various historical data obtained from the data acquisition module. Preprocessing steps may include data cleaning (removing outliers and handling missing values), data transformation (such as normalization and standardization), and feature selection (selecting key factors that influence sawing prediction). Preprocessed data will be more accurate, consistent, and suitable for subsequent model training.

[0067] The random forest model is trained using preprocessed data. Random forest is an ensemble learning method that improves the accuracy and stability of predictions by constructing multiple decision trees and combining their predictions. During training, the model learns the relationship between various factors (such as raw material quality, production process parameters, equipment status, etc.) and sawing results from historical data, thereby establishing a prediction model for sawing equipment.

[0068] When the system receives real-time customer demand data (such as the size, shape, and quantity of gypsum board) and real-time sawing quality standards (such as sawing accuracy, allowable error range, and surface quality), the sawing prediction module will input this real-time data into the pre-trained sawing equipment prediction model.

[0069] The sawing prediction module, based on input data and patterns learned from historical data, predicts the expected operating parameters and output results of the sawing equipment under current conditions; this is known as the sawing equipment prediction data. This prediction data may include key parameters such as sawing position, speed, and cutting depth. Based on the prediction data output by the sawing equipment prediction model, the system can adjust the sawing equipment parameters, such as adjusting the saw blade position, speed, and cutting depth, to ensure that the sawing operation is performed according to the expected accuracy and quality standards.

[0070] Through the sawing prediction module, the system can accurately predict and rehearse the sawing process before the actual sawing operation, thereby avoiding deviations and errors during the actual sawing process, reducing material waste, and improving production efficiency and product quality. At the same time, the sawing prediction module can continuously update and optimize the prediction model based on historical and real-time data to adapt to different batches, specifications, and requirements of gypsum board sawing tasks.

[0071] The data analysis module acquires historical data of the online customization four-part system for paper-faced gypsum board and historical data of the sawing equipment. It trains a deep learning model with historical data of customer demand, quality control, and the operation of the sawing equipment to obtain a control model for the online customization four-part system for paper-faced gypsum board. It then substitutes real-time customer demand data into the control model to output real-time control parameters for the online customization four-part system for paper-faced gypsum board.

[0072] The data analysis module in the online custom gypsum board four-part system is responsible for processing and analyzing various historical data to build an efficient system control model, thereby guiding the actual production process. The following is a detailed explanation of the workflow and specific functions of the data analysis module:

[0073] The data analysis module first collects necessary data from both internal and external sources. This data includes historical operational data of the online custom gypsum board four-part system, historical operational data of the sawing equipment, historical customer demand data, and historical quality control data.

[0074] The collected raw data often needs to be preprocessed to ensure data quality and consistency. This includes steps such as data cleaning (removing noise and handling missing values), data transformation (such as normalization and standardization), and feature selection.

[0075] The deep learning model is trained using preprocessed data. Deep learning is a powerful machine learning tool capable of learning hidden patterns and rules from complex data. In this process, historical customer demand data, historical quality control data, historical system data, and historical sawing equipment operation data are used to train the model to establish the relationship between inputs (customer demand) and outputs (system control parameters).

[0076] By using optimization techniques such as backpropagation, the parameters of the deep learning model are continuously adjusted to minimize prediction errors and improve the model's prediction accuracy.

[0077] When new customer requests are input, the data analysis module feeds this real-time customer demand data into the pre-trained online custom gypsum board four-part system control model. Based on the input data, the model quickly calculates the optimal system control parameters. These parameters directly guide the operation of the sawing equipment to ensure that the produced gypsum board products meet customer requirements.

[0078] The data analysis module will also continuously receive new operational and quality control data, which will be used for the model's continuous learning and improvement. Over time, the model will be able to automatically adapt to changes in the production process, maintaining the accuracy and effectiveness of its predictive and control capabilities.

[0079] Through the data analysis module, the online customization system for paper-faced gypsum board can achieve intelligent and automated production control, improve production efficiency, reduce production costs, and enhance the stability and consistency of product quality.

[0080] The sawing verification module transmits the real-time control parameters of the online custom gypsum board four-part system to the sawing equipment in the online custom gypsum board four-part system. After the sawing equipment executes the real-time control parameters of the online custom gypsum board four-part system, it generates the real-time operating data of the first sawing equipment. The real-time operating data of the first sawing equipment is compared with the predicted data of the sawing equipment to obtain the sawing error data.

[0081] The sawing verification module in the online custom four-part system for paper-faced gypsum board is responsible for verifying and optimizing the sawing process to ensure consistency between the actual sawing operation and the expected target. The following is the specific workflow of the sawing verification module:

[0082] The sawing verification module first receives the real-time control parameters of the online custom gypsum board four-part system output from the data analysis module. These control parameters are then transmitted to the sawing equipment within the online custom gypsum board four-part system.

[0083] The sawing equipment executes sawing operations based on the received control parameters, generating real-time operating data for the first sawing equipment. This data includes key information such as the actual sawing position parameters, actual speed parameters, and actual depth parameters.

[0084] The sawing verification module compares the real-time operating data of the first sawing device with the predicted data of the sawing device output by the previous sawing prediction module. The comparison is strictly aligned in time and space to ensure the accuracy and comparability of the data.

[0085] By comparing the control parameters of the real-time online customization four-part system for gypsum board, the differences between the actual operating data and the predicted data are calculated item by item to obtain the sawing error data. These error data reflect the deviation between the actual sawing operation and the expected target.

[0086] The sawing verification module evaluates the calculated sawing error data. If the error is within an acceptable range, it indicates that the sawing operation is basically as expected and no further optimization is needed. If the error exceeds a predetermined threshold, the sawing verification module will trigger an optimization process, feeding the error data back to the sawing optimization module so that the system control model can be adjusted and improved.

[0087] The sawing verification module continuously monitors the equipment's operating status and sawing results throughout the entire sawing process. When an anomaly or deviation is detected, the module immediately issues an alarm and takes intervention measures to prevent further escalation of the error. Simultaneously, the module records monitoring data and feedback information, providing valuable information for subsequent data analysis and model optimization.

[0088] Through the operation of the sawing verification module, the online customization four-part system for paper-faced gypsum board can achieve precise control and continuous optimization of the sawing process, thereby producing high-quality gypsum board products that meet customer requirements.

[0089] The sawing optimization module optimizes the control model of the online customization four-part system for paper-faced gypsum board, resulting in an optimized control model. Real-time customer demand data is then input into the optimized control model to obtain optimized real-time control parameters for the online customization four-part system. Control parameters for the sawing equipment at the position to be corrected are generated, and the sawing equipment executes these control parameters to complete the cutting of the gypsum board.

[0090] The sawing optimization module, part of the online customization system for paper-faced gypsum board, is tasked with continuously optimizing the system control model to improve the accuracy of sawing operations. The following is a detailed workflow of the sawing optimization module:

[0091] The sawing optimization module first identifies deficiencies or deviations in the current four-part control model of the online customization system for paper-faced gypsum board based on previously collected sawing error data, actual operating data, and customer feedback. Utilizing advanced optimization algorithms such as gradient descent, genetic algorithms, or particle swarm optimization, the module iteratively updates the original system control model, striving to find model parameters that more accurately reflect the relationship between customer needs and the actual operation of the sawing equipment. During the optimization process, the module fully considers multiple factors, including raw material characteristics, differences in production processes, changes in equipment status, and fluctuations in customer demand, to ensure the model's comprehensiveness and adaptability.

[0092] After model optimization, the sawing optimization module uses historical or simulated data to validate the new model and evaluate whether its prediction accuracy and stability meet the expected standards. Only when the new model demonstrates significantly better performance than the original model will it be adopted as the official control model for the online customization of paper-faced gypsum board in a four-part system.

[0093] When new customer requirements are input, the sawing optimization module incorporates this data into the already optimized system control model. The optimized system control model quickly calculates the optimized real-time online customization control parameters for the four-part gypsum board system based on the customer's needs. These parameters are closer to actual production requirements and can effectively reduce sawing errors.

[0094] Based on the optimized system control parameters, the sawing optimization module further generates sawing equipment control parameters for the position to be corrected. These parameters specifically guide the sawing equipment in various operational details when performing the cutting task, such as sawing position, speed, depth, and cutting path.

[0095] The sawing equipment executes the cutting operation based on the received control parameters for the position to be corrected. During the cutting process, the sawing optimization module continuously monitors the equipment's operating status and cutting effect, collecting real-time data for subsequent analysis and further optimization.

[0096] The sawing optimization module periodically or based on actual needs triggers new optimization cycles, continuously fine-tuning and improving the system control model. Through this continuous optimization process, the online custom four-part system for paper-faced gypsum board can gradually approach the ideal sawing effect, improving product quality and production efficiency.

[0097] In summary, the sawing optimization module provides precise and efficient sawing parameter guidance for the online customization four-part system of paper-faced gypsum board by continuously optimizing the system control model, thereby realizing refined control and management of the sawing process.

[0098] Specifically, the data acquisition module of the online custom four-part paper-faced gypsum board system of the present invention includes:

[0099] Feature extraction is performed on the four-part online customization system of paper-faced gypsum board to obtain feature data of the four-part online customization system of paper-faced gypsum board. Based on the feature data of the four-part online customization system of paper-faced gypsum board, the data type to be acquired is determined.

[0100] Data is acquired based on the type of data to be acquired, and the acquired data is checked for continuity. If data is missing, a data missing warning message is generated.

[0101] The preset data missing warning information is substituted into the data simulation generation model of the online customization four-part system for paper-faced gypsum board to generate simulated data corresponding to the data missing warning information. The simulated data corresponding to the data missing warning information is then used to fill in the missing data.

[0102] A comprehensive feature extraction process is performed on the four-part online customization system for paper-faced gypsum board. This process may involve in-depth analysis of the attributes, behaviors, and interrelationships of each component of the system (such as raw material processing, production technology, equipment status, product specifications, customer needs, and quality control). Using feature extraction techniques (such as principal component analysis, linear discriminant analysis, and independent component analysis), a new feature set reflecting the system characteristics is constructed from the original data; this is the feature data of the four-part online customization system for paper-faced gypsum board.

[0103] Based on the extracted system feature data and the actual needs of system operation, determine the specific data types that need to be acquired. These data types may include, but are not limited to, raw material quality parameters, production process parameters, equipment status parameters, customer order information, and quality control indicators.

[0104] Based on the determined data type to be acquired, data is retrieved from the corresponding data source (such as sensors, databases, file systems, etc.) in real time or at scheduled intervals. During data acquisition, the accuracy and completeness of the data must be ensured, while also considering its timeliness and security.

[0105] The acquired data undergoes continuous analysis to determine if any data is missing or anomaly-prone. This typically involves analyzing the data's timestamps, numerical ranges, and trends. If missing or anomaly-prone data is detected, timely data loss warnings or anomaly alarms should be generated for subsequent processing.

[0106] When missing data is detected, the system generates a data missing warning message based on preset rules and thresholds. This information may include the type of missing data, the time range, and the degree of missing data.

[0107] The generated data missing warning information is then incorporated into the data simulation model of the online custom gypsum board four-part system. This model can simulate and generate reasonable values ​​corresponding to the missing data based on the system's historical data and current operating status.

[0108] Simulated data is used to fill in missing data to ensure the normal operation of the system and the continuity of data analysis. During the data filling process, the accuracy and reliability of the simulated data must be considered to avoid misleading system decisions.

[0109] The data acquisition module of the online customization four-part system for paper-faced gypsum board achieves comprehensive, accurate, and timely acquisition of the data required by the system through steps such as feature extraction, data type determination, data acquisition, continuity detection, data missing warning, and simulated data filling.

[0110] Specifically, the paper-faced gypsum board online customization four-part system of the present invention includes a sawing prediction module comprising:

[0111] Receive real-time customer demand data through data interface or user input interface. The real-time customer demand data includes the size, shape and quantity of gypsum board.

[0112] Receive real-time sawing quality standards, which include sawing accuracy, allowable error range, and surface quality;

[0113] The pre-processed real-time customer demand data is substituted into the pre-trained sawing equipment prediction model. The sawing equipment prediction model is trained based on historical data of raw materials, historical data of production processes, and historical data of equipment status. The sawing equipment prediction model predicts the operating parameters and output results of the sawing equipment based on the input customer demand data.

[0114] The prediction data of the sawing equipment is obtained by calculating the prediction model of the sawing equipment. The prediction data of the sawing equipment includes the prediction parameters of the sawing position, the prediction parameters of the speed, and the prediction parameters of the cutting depth.

[0115] The sawing prediction module receives gypsum board customization requests from customers in real time through data interfaces (such as APIs, database connections, etc.) or user input interfaces (such as web forms, mobile application interfaces, etc.). Customer request data specifically includes the dimensions (length, width, thickness), shape (such as rectangles, squares, special shapes, etc.), and required quantity of the gypsum boards.

[0116] Meanwhile, the sawing prediction module also receives quality standards related to the sawing operation, which are usually provided directly by the quality control department or the customer. The sawing quality standards include the sawing accuracy requirements (such as the straightness and perpendicularity of the cutting line), the allowable error range (such as dimensional deviation and angular deviation), and the surface quality requirements (such as smoothness and no cracks).

[0117] The received customer demand data and sawing quality standards are preprocessed, including data cleaning (removing invalid or erroneous data), data transformation (such as standardizing size units), and data normalization (scaling data to a specific range) to ensure data consistency and processability.

[0118] Load the pre-trained prediction model for the sawing equipment. This model is trained based on a large amount of historical data (including historical data on raw materials, production processes, and equipment status), and can reflect the operating patterns and output results of the sawing equipment under different conditions.

[0119] The preprocessed customer demand data is used as input into the sawing equipment prediction model. This input data provides the model with the necessary information to make predictions.

[0120] The sawing equipment prediction model predicts the operating parameters and output results of the sawing equipment based on the input customer demand data and its internal learning algorithm and parameters.

[0121] Operating parameters may include sawing position parameters (such as start point, end point, cutting path, etc.), speed parameters (such as cutting speed, acceleration, etc.), and cutting depth parameters (such as cutting depth, cutting layers, etc.).

[0122] The model calculates predictive data for the sawing equipment. This data is crucial for subsequent sawing operations, guiding the equipment to cut according to predetermined parameters and paths.

[0123] Predictive data from the sawing equipment is transmitted to the sawing execution module or operators to guide actual sawing operations. During the sawing process, actual operating data of the sawing equipment can be collected in real time or periodically and compared with the predicted data to evaluate the accuracy of the predictions and the performance of the model. Based on the comparison results and feedback, the sawing equipment prediction model can be continuously optimized and updated to improve its prediction accuracy and adaptability.

[0124] Through the above process, the sawing prediction module can accurately predict the sawing operation in the online customization four-part system of paper-faced gypsum board, providing strong support for the automated and intelligent operation of the system.

[0125] Specifically, the data analysis module of the online custom four-part paper-faced gypsum board system of the present invention includes:

[0126] The historical data of the online customization four-part system for paper-faced gypsum board and the historical data of sawing equipment operation are divided into training set and validation set.

[0127] The selected deep learning model is trained using the preprocessed training dataset. During training, the model weights are adjusted using the backpropagation algorithm. The model is validated using the validation set and evaluated using the test set. The model's performance metrics are calculated, and the model is optimized based on the evaluation results to obtain the control model for the online customization of paper-faced gypsum board in four parts.

[0128] This system compiles historical data from four parts of the online custom gypsum board system, covering various parameter records during the production process, such as raw material information, production process parameters, and equipment status. It also collects historical data on sawing equipment operation, including sawing parameters (position, speed, depth, etc.), equipment status (temperature, vibration, etc.), and maintenance records. Furthermore, it organizes historical customer demand data, recording customer customization requirements for gypsum board dimensions, shapes, quantities, etc., and summarizes historical quality control data, including quality inspection results, analysis of the causes of non-conforming products, and improvement measures.

[0129] The collected data is divided into training, validation, and test sets. The training set is used for model learning, the validation set is used for model training to adjust hyperparameters, and the test set is used for final performance evaluation. Data partitioning should follow the principles of randomness, representativeness, and balance to ensure the reliability and generalization ability of model training.

[0130] Choose a deep learning model based on the characteristics and requirements of the online customization four-part system for paper-faced gypsum board. This could be a convolutional neural network (CNN) for image processing (such as identifying surface defects in gypsum board), a recurrent neural network (RNN) for time series prediction (such as predicting equipment failure), or other types of deep learning models.

[0131] Data preprocessing involves preprocessing the training, validation, and test sets, including data cleaning (removing outliers and filling in missing values), data transformation (such as normalization and standardization), and feature selection (extracting features useful for model prediction).

[0132] Model training: The selected deep learning model is trained using the preprocessed training dataset. During training, the model weights are adjusted using the backpropagation algorithm to minimize the error between the model's predictions and the actual data.

[0133] Set reasonable hyperparameters such as the number of training epochs, learning rate, and batch size, and adjust these hyperparameters based on the performance of the validation set.

[0134] During training, the model is periodically validated using a validation set to assess whether its performance meets expectations. If the validation set performance declines, it may indicate overfitting, requiring adjustments to the model structure or hyperparameters.

[0135] The trained model is comprehensively evaluated using a test set, and performance metrics such as accuracy, recall, F1 score, and mean squared error are calculated. These performance metrics reflect the model's ability to generalize on unknown data.

[0136] Based on the evaluation results, the model is optimized. This may include adjusting the model structure (such as increasing or decreasing the number of network layers, changing the activation function, etc.), adding data augmentation techniques (such as image processing techniques such as rotation, scaling, and flipping), or adopting ensemble learning methods (such as random forests, gradient boosting, etc.) to improve the model's performance.

[0137] The optimized model was selected as the control model for the four-part online customization system of gypsum board and deployed in the actual production environment. The model can receive customer demand data and quality control data in real time, and output optimized sawing parameters and production process parameters to guide the online customization production of gypsum board. Simultaneously, the data analysis module continuously monitors the model's performance in actual production and continuously optimizes and updates the model based on feedback data.

[0138] Specifically, the paper-faced gypsum board online customization four-part system of the present invention includes a sawing and verification module comprising:

[0139] The real-time operating data of the first sawing equipment is obtained from the sawing equipment. The real-time operating data of the first sawing equipment includes the actual position parameters, actual speed parameters, and actual depth parameters of the sawing.

[0140] The real-time operating data of the first sawing equipment is aligned with the predicted data of the sawing equipment in both time and space;

[0141] The real-time operating data of the first sawing equipment is compared with the predicted data of the sawing equipment item by item, and the error of each data item is calculated, that is, the difference between the actual value and the predicted value.

[0142] All calculated errors are summarized, and sawing error data is generated based on the summarized error results.

[0143] To acquire real-time operating data of the first sawing device, the sawing verification module first obtains the operating data of the first sawing device in real time through a data interface or directly from the sawing device. This data includes the actual position parameters of the sawing (such as the cutting start point, end point, path, etc.), the actual speed parameters (such as cutting speed, acceleration, etc.), and the actual depth parameters (such as cutting depth, layers, etc.).

[0144] To ensure the accuracy of the comparison, the sawing verification module aligns the real-time operating data of the first sawing device with the predicted data of the sawing device output by the previous sawing prediction module in both time and space. This means that it is necessary to ensure that the two sets of data correspond to sawing operations within the same point in time or time period, and that the data items can be matched one-to-one.

[0145] The sawing verification module compares the real-time operating data of the first sawing equipment with the predicted data of the sawing equipment item by item. For each data item (such as position, speed, and depth), its actual value is compared with the predicted value.

[0146] During the comparison process, the sawing verification module calculates the error for each data point, that is, the difference between the actual value and the predicted value. These errors may be expressed in the form of absolute value, relative value, or percentage, depending on the nature of the data and the system requirements.

[0147] The sawing verification module summarizes all calculated errors. This may include statistical indicators such as the average, standard deviation, maximum, and minimum values ​​of the calculated errors, in order to comprehensively evaluate the operating performance of the sawing equipment and the accuracy of the prediction model.

[0148] Based on the summarized error results, the sawing verification module generates sawing error data. This data not only reflects the deviation between the actual operation of the sawing equipment and the prediction, but can also be used for subsequent analysis, optimization, and decision-making processes. For example, if the error exceeds a set threshold, the system can trigger an alarm or adjust the sawing parameters to reduce the error.

[0149] Sawing error data can be transmitted to other modules of the system (such as the data analysis module and control module) for further processing and analysis. Based on the sawing error data, the system can optimize and adjust the operating parameters of the sawing equipment, the parameters of the prediction model, or the entire production process to improve production efficiency and product quality. Simultaneously, the sawing verification module can provide a real-time feedback mechanism to help operators understand the operating status of the sawing equipment and the accuracy of the prediction model in a timely manner, thereby enabling them to make corresponding adjustments and decisions.

[0150] Specifically, the paper-faced gypsum board online customization four-part system of the present invention includes a sawing optimization module comprising:

[0151] The real-time operating data of the first sawing equipment is compared with the real-time sawing quality standard. When the sawing error data exceeds the real-time sawing quality standard, the operating data of the data acquisition terminal is obtained. The data acquisition terminal is then inspected based on this data. If the inspection is successful, a genetic algorithm is used to optimize the control model of the online custom gypsum board four-part system based on the sawing error data, historical data of the online custom gypsum board four-part system, and historical operating data of the sawing equipment. This results in an optimized control model. Real-time customer demand data is then substituted into the optimized control model to obtain the optimized real-time control parameters for the online custom gypsum board four-part system. The control parameters of the real-time online customization four-part system for gypsum board are substituted into the prediction model of the sawing equipment to generate real-time operating data of the second sawing equipment. The real-time operating data of the second sawing equipment is compared with the real-time sawing quality standard. If the real-time operating data of the second sawing equipment meets the real-time sawing quality standard range, the image data of the gypsum board to be cut is collected. The image data of the gypsum board to be cut is compared with the real-time customer demand data to obtain the position to be corrected in the image data of the gypsum board to be cut. The position to be corrected in the image data of the gypsum board to be cut is transmitted to the optimized control model of the online customization four-part system for gypsum board to generate the control parameters of the sawing equipment for the position to be corrected. The sawing equipment executes the control parameters of the sawing equipment for the position to be corrected to complete the cutting of the gypsum board to be cut.

[0152] The sawing error detection and optimization module first receives sawing error data from the sawing verification module and compares this data with the real-time sawing quality standard. If the sawing error data exceeds the threshold set by the real-time sawing quality standard, the subsequent optimization process is triggered.

[0153] After confirming that the sawing error exceeds the standard, the module acquires and verifies the operational data from the data acquisition terminal. The purpose of the verification is to ensure that the data acquisition terminal itself is free from faults or errors, thereby eliminating sawing errors caused by data acquisition problems. If the data acquisition terminal passes the verification, subsequent optimization steps continue; if it fails, the data acquisition terminal needs to be repaired or replaced.

[0154] Data on sawing errors, historical data from the online customization system for paper-faced gypsum board, and historical operating data of sawing equipment are collected and used as inputs for the optimization algorithm.

[0155] A genetic algorithm was applied to optimize the control model of a four-part online customization system for paper-faced gypsum board. The genetic algorithm is a search algorithm that simulates natural selection and genetic mechanisms, suitable for solving complex optimization problems.

[0156] Through the iterative process of the genetic algorithm, the parameters of the control model are continuously adjusted to minimize the sawing error and meet the real-time sawing quality standards.

[0157] After optimization using a genetic algorithm, an optimized control model for the online customization of gypsum board in four-part systems was obtained. Real-time customer demand data was then input into the optimized control model to generate optimized real-time control parameters for the online customization of gypsum board in four-part systems.

[0158] These control parameters are then input into the sawing equipment prediction model to generate real-time operating data for the second sawing equipment. This includes sawing position parameters, speed parameters, and cutting depth parameters, which are used to guide subsequent sawing operations.

[0159] The real-time operating data of the second sawing equipment is compared with the real-time sawing quality standard to verify whether the sawing result meets the standard. If the sawing result meets the standard, image data of the cut gypsum board is collected. This image data is compared with real-time customer demand data to identify the areas in the image data of the cut gypsum board that need correction (such as dimensional deviations, shape discrepancies, etc.).

[0160] The position to be corrected is transmitted to the optimized control model, generating control parameters for the sawing equipment at that position. The sawing equipment then performs correction operations based on these parameters to achieve precise cutting of the plasterboard.

[0161] The entire sawing optimization module is a cyclical optimization process. By continuously detecting sawing errors, optimizing the control model, generating control parameters, verifying sawing results, and correcting deviations, the system achieves continuous improvement and optimization of the online customization four-part system for paper-faced gypsum board. This optimization mechanism not only improves production efficiency and product quality but also enhances the system's adaptability and robustness, enabling it to better cope with changes in customer needs and uncertainties in the production process.

[0162] Specifically, the paper-faced gypsum board online customization four-part system of the present invention includes a sawing optimization module comprising:

[0163] Based on sawing error data and real-time sawing quality standards, determine the optimization objectives;

[0164] Acquire sawing error data, historical data from the online custom gypsum board four-part system, and historical operating data of the sawing equipment.

[0165] The data was used as input for the genetic algorithm optimization to train and optimize the control model of the online customization four-part system for paper-faced gypsum board.

[0166] Receive and set the parameters of the genetic algorithm, which include population size, selection, crossover, mutation probability, and number of iterations;

[0167] Initialize the population by generating an initial population using a random or specific method, where each individual represents a possible combination of control model parameters.

[0168] Each individual in the population is evaluated, and its fitness value is calculated;

[0169] Individuals with high fitness values ​​are selected as parents to generate the next generation. The selection operation can be carried out using roulette wheel selection or tournament selection methods.

[0170] Perform a crossover operation on the selected parent generation to generate new offspring individuals;

[0171] Mutation operations are performed on offspring individuals to introduce new genes, and the newly generated offspring individuals replace the old population to form a new generation of population.

[0172] Repeat the selection, crossover, and mutation processes until the predetermined number of iterations is reached or the optimization objective is met.

[0173] The individual with the highest fitness value was selected from the final population and used as the parameter for the optimized four-part system control model of online customization of paper-faced gypsum board;

[0174] The optimized control model for the online customization of gypsum board in four-part systems is obtained by updating the parameters of the optimized control model.

[0175] Define the optimization objectives based on sawing error data and real-time sawing quality standards. These objectives typically include minimizing sawing errors, improving cutting accuracy, and meeting specific size and shape requirements.

[0176] Acquire sawing error data, which reflects the deviation between the current sawing operation and the ideal state. Collect historical data from the online custom gypsum board four-part system, including past sawing operation records and system configuration parameters. Acquire historical operating data of the sawing equipment, including equipment performance parameters and maintenance records.

[0177] The above data is used as input data for the genetic algorithm optimization, which is used to train and optimize the control model of the online customization four-part system for paper-faced gypsum board.

[0178] Genetic Algorithm Parameter Setting: This section receives and sets the parameters of the genetic algorithm, which have a significant impact on the algorithm's performance and convergence speed.

[0179] Population size: Determines the number of individuals in the initial population.

[0180] Selection probability: determines the probability that an excellent individual will be selected in a selection operation.

[0181] Crossover probability: controls the frequency of crossover operations.

[0182] Mutation probability: determines the probability of mutation operations occurring, and introduces new genes to increase the diversity of the population.

[0183] Number of iterations: Sets the maximum number of iterations the algorithm can run, which can be used as one of the stopping conditions.

[0184] Initialize the population using a random or specific method, where each individual represents a possible combination of control model parameters. These parameter combinations will serve as the starting point for the genetic algorithm's optimization.

[0185] Each individual in the population is evaluated, and its fitness value is calculated. The fitness value is usually determined based on the sawing error data and the objective function of the control model, reflecting the individual's performance in the optimization problem.

[0186] Selection involves choosing individuals with superior fitness values ​​as parents to generate the next generation. Selection can employ methods such as roulette wheel selection or tournament selection to ensure the transmission of superior genes.

[0187] Crossover is a process of performing a crossover operation on selected parents to generate new offspring. Crossover occurs by exchanging partial gene segments between parents, resulting in offspring with new characteristics.

[0188] Mutation involves performing mutation operations on offspring individuals to introduce new genes. Mutation operations can randomly change certain gene values ​​of individuals to increase population diversity and explore new solution spaces.

[0189] Update the population by replacing the old population with newly generated offspring individuals to form a new generation. Repeat the selection, crossover, and mutation process until the predetermined number of iterations is reached or the optimization objective is met.

[0190] The individual with the highest fitness value is selected from the final population and used as the parameter for the optimized control model of the online custom gypsum board four-part system. These parameters are used to update the control model, resulting in the optimized control model for the online custom gypsum board four-part system. This model will guide subsequent sawing operations to improve cutting accuracy and meet customer requirements.

[0191] The sawing optimization module is a continuous optimization process. In practical applications, the parameters and optimization objectives of the genetic algorithm can be continuously adjusted based on feedback from the sawing operation and changes in customer needs to achieve better optimization results.

[0192] Through the above process, the sawing optimization module uses a genetic algorithm to optimize the control model of the online customization four-part system for paper-faced gypsum board, improving the accuracy and stability of the sawing operation and providing a strong guarantee for the efficient operation of the system and customer satisfaction.

[0193] This invention solves the problems existing in the current online custom-made gypsum board sawing process through the following technical solution:

[0194] Historical data on raw materials, production processes, equipment status, product specifications, customer demand, and quality control were acquired. This data provides the foundation for subsequent analysis and prediction. The acquired data was preprocessed, and a random forest model was trained using the preprocessed data to obtain a prediction model for the sawing equipment.

[0195] The system receives real-time customer demand data and real-time sawing quality standards, and then inputs the real-time customer demand data into the sawing equipment prediction model to obtain the sawing equipment prediction data. This prediction data includes parameters such as sawing position, speed, and cutting depth, providing preliminary analysis for subsequent sawing operations.

[0196] Historical data from the online custom gypsum board four-part system and the sawing equipment's operation history are acquired. This data (including historical customer demand data and quality control data) is used to train a deep learning model, resulting in a control model for the online custom gypsum board four-part system. Real-time customer demand data is then fed into this control model to output real-time control parameters for the online custom gypsum board four-part system. These parameters provide precise guidance for the actual operation of the sawing equipment.

[0197] The real-time online customization system control parameters for gypsum board are transmitted to the sawing equipment. After execution, the sawing equipment generates real-time operating data. This real-time operating data is compared with the predicted data of the sawing equipment to obtain the sawing error data. This step achieves real-time calibration of the sawing process, ensuring the accuracy of the sawing operation.

[0198] When the sawing error data exceeds the real-time sawing quality standard, a genetic algorithm is used to optimize the control model of the online customization four-part system for paper-faced gypsum board, based on the sawing error data, system historical data, and equipment operation historical data. The optimized model yields more precise control parameters and generates control parameters for the sawing equipment at the location to be corrected. The sawing equipment then executes these parameters to precisely correct the gypsum board being cut.

[0199] Through the above technical solution, this invention achieves pre-analysis before gypsum board sawing and real-time calibration during the sawing process, thereby effectively avoiding the waste of gypsum board materials caused by sawing errors. Simultaneously, by continuously optimizing the system control model, the accuracy and efficiency of sawing are improved, meeting diverse construction requirements.

Claims

1. A four-part online customization system for paper-faced gypsum board, characterized in that, include: The data acquisition module acquires historical data on raw materials, production processes, equipment status, product specifications, customer needs, and quality control. The sawing prediction module preprocesses historical data of raw materials, production processes, equipment status, product specifications, customer demand, and quality control to obtain preprocessed data. The preprocessed data is then used to train a random forest model to obtain a sawing equipment prediction model. The module also receives real-time customer demand data and real-time sawing quality standards, and substitutes the real-time customer demand data into the sawing equipment prediction model to obtain the sawing equipment prediction data. The data analysis module acquires historical data of the online customization four-part system for paper-faced gypsum board and historical data of the sawing equipment. It trains a deep learning model with historical data of customer demand, quality control, and the operation of the sawing equipment to obtain a control model for the online customization four-part system for paper-faced gypsum board. It then substitutes real-time customer demand data into the control model to output real-time control parameters for the online customization four-part system for paper-faced gypsum board. The sawing verification module transmits the real-time control parameters of the online custom gypsum board four-part system to the sawing equipment in the online custom gypsum board four-part system. After the sawing equipment executes the real-time control parameters of the online custom gypsum board four-part system, it generates the real-time operating data of the first sawing equipment. The real-time operating data of the first sawing equipment is compared with the predicted data of the sawing equipment to obtain the sawing error data. The sawing optimization module optimizes the control model of the online customization four-part system for paper-faced gypsum board, resulting in an optimized control model. Real-time customer demand data is then input into the optimized control model to obtain optimized real-time control parameters for the online customization four-part system. Control parameters for the sawing equipment at the position to be corrected are generated, and the sawing equipment executes these control parameters to complete the cutting of the gypsum board.

2. The online customization four-part system for paper-faced gypsum board as described in claim 1, characterized in that, The data acquisition module includes: Feature extraction is performed on the four-part online customization system of paper-faced gypsum board to obtain feature data of the four-part online customization system of paper-faced gypsum board. Based on the feature data of the four-part online customization system of paper-faced gypsum board, the data type to be acquired is determined. Data is acquired based on the type of data to be acquired, and the acquired data is checked for continuity. If data is missing, a data missing warning message is generated. The preset data missing warning information is substituted into the data simulation generation model of the online customization four-part system for paper-faced gypsum board to generate simulated data corresponding to the data missing warning information. The simulated data corresponding to the data missing warning information is then used to fill in the missing data.

3. The online customization four-part system for paper-faced gypsum board as described in claim 1, characterized in that, The sawing prediction module includes: Receive real-time customer demand data through data interface or user input interface. The real-time customer demand data includes the size, shape and quantity of gypsum board. Receive real-time sawing quality standards, which include sawing accuracy, allowable error range, and surface quality; The pre-processed real-time customer demand data is substituted into the pre-trained sawing equipment prediction model. The sawing equipment prediction model is trained based on historical data of raw materials, historical data of production processes, and historical data of equipment status. The sawing equipment prediction model predicts the operating parameters and output results of the sawing equipment based on the input customer demand data. The prediction data of the sawing equipment is obtained by calculating the prediction model of the sawing equipment. The prediction data of the sawing equipment includes the prediction parameters of the sawing position, the prediction parameters of the speed, and the prediction parameters of the cutting depth.

4. The online customization four-part system for paper-faced gypsum board as described in claim 1, characterized in that, The data analysis module includes: The historical data of the online customization four-part system for paper-faced gypsum board and the historical data of sawing equipment operation are divided into training set and validation set. The selected deep learning model is trained using the preprocessed training dataset. During training, the model weights are adjusted using the backpropagation algorithm. The model is validated using the validation set and evaluated using the test set. The model's performance metrics are calculated, and the model is optimized based on the evaluation results to obtain the control model for the online customization of paper-faced gypsum board in four parts.

5. The online customization four-part system for paper-faced gypsum board as described in claim 1, characterized in that, The sawing verification module includes: The real-time operating data of the first sawing equipment is obtained from the sawing equipment. The real-time operating data of the first sawing equipment includes the actual position parameters, actual speed parameters, and actual depth parameters of the sawing. The real-time operating data of the first sawing equipment is aligned with the predicted data of the sawing equipment in both time and space; The real-time operating data of the first sawing equipment is compared with the predicted data of the sawing equipment item by item, and the error of each data item is calculated, that is, the difference between the actual value and the predicted value. All calculated errors are summarized, and sawing error data is generated based on the summarized error results.

6. The online customization four-part system for paper-faced gypsum board as described in claim 1, characterized in that, The sawing optimization module includes: The real-time operating data of the first sawing equipment is compared with the real-time sawing quality standard. When the sawing error data exceeds the real-time sawing quality standard, the operating data of the data acquisition terminal is obtained. The data acquisition terminal is then inspected based on this data. If the inspection is successful, a genetic algorithm is used to optimize the control model of the online custom gypsum board four-part system based on the sawing error data, historical data of the online custom gypsum board four-part system, and historical operating data of the sawing equipment. This results in an optimized control model. Real-time customer demand data is then substituted into the optimized control model to obtain the optimized real-time control parameters for the online custom gypsum board four-part system. The control parameters of the real-time online customization four-part system for gypsum board are substituted into the prediction model of the sawing equipment to generate real-time operating data of the second sawing equipment. The real-time operating data of the second sawing equipment is compared with the real-time sawing quality standard. If the real-time operating data of the second sawing equipment meets the real-time sawing quality standard range, the image data of the gypsum board to be cut is collected. The image data of the gypsum board to be cut is compared with the real-time customer demand data to obtain the position to be corrected in the image data of the gypsum board to be cut. The position to be corrected in the image data of the gypsum board to be cut is transmitted to the optimized control model of the online customization four-part system for gypsum board to generate the control parameters of the sawing equipment for the position to be corrected. The sawing equipment executes the control parameters of the sawing equipment for the position to be corrected to complete the cutting of the gypsum board to be cut.

7. The online customization four-part system for paper-faced gypsum board as described in claim 6, characterized in that, The sawing optimization module includes: Based on sawing error data and real-time sawing quality standards, the optimization objectives are determined; Acquire sawing error data, historical data from the online custom gypsum board four-part system, and historical operating data of the sawing equipment. The data was used as input for the genetic algorithm optimization to train and optimize the control model of the online customization four-part system for paper-faced gypsum board. Receive and set the parameters of the genetic algorithm, which include population size, selection, crossover, mutation probability, and number of iterations; Initialize the population by generating an initial population using a random or specific method, where each individual represents a possible combination of control model parameters. Each individual in the population is evaluated, and its fitness value is calculated; Individuals with high fitness values ​​are selected as parents to generate the next generation. The selection operation can be carried out using roulette wheel selection or tournament selection methods. Perform a crossover operation on the selected parent generation to generate new offspring individuals; Mutation operations are performed on offspring individuals to introduce new genes, and the newly generated offspring individuals replace the old population to form a new generation of population. Repeat the selection, crossover, and mutation processes until the predetermined number of iterations is reached or the optimization objective is met. The individual with the highest fitness value was selected from the final population and used as the parameter for the optimized four-part system control model of online customization of paper-faced gypsum board; The optimized control model for the online customization of gypsum board in four-part systems is obtained by updating the parameters of the optimized control model.

Images