A polyester cloth production whole process monitoring management method and system

Abstract

The application provides a polyester fabric production whole-process monitoring management method and system. It belongs to the technical field of production process monitoring. The method comprises the following steps: deploying multiple types of sensors at each key node of a polyester fabric production line; collecting various physical parameters and environmental information related to polyester fabric production in real time; simultaneously, integrating an operation state monitoring system of production equipment to obtain equipment data; performing deep fusion processing on multi-source heterogeneous data obtained from a polyester fabric production holographic perception architecture; constructing a data correlation model; based on the model, combining a process flow and quality standards of polyester fabric production, comprehensively evaluating a production situation; through production situation evaluation, generating a production situation evaluation report; a whole-process data traceability system provides traceable basis for quality disputes, raw material problems and the like, enhances the control ability of supply chain risks, simultaneously meets the needs of downstream customers for product traceability, and improves brand trustworthiness.

Description

TECHNICAL FIELD

[0001] The application provides a polyester fabric production whole-process monitoring management method and system, and belongs to the technical field of production process monitoring. BACKGROUND

[0002] As an important synthetic fiber fabric, polyester fabric has been widely used in many fields such as clothing, home textiles and industrial textiles due to its high strength, good wear resistance, strong wrinkle resistance and relatively low price and other advantages. In recent years, with the steady growth of the global economy and the continuous upgrading of the consumer market, the market demand for polyester fabric continues to expand, and the production scale is also growing.

[0003] Traditional production decisions are often based on experience and simple statistical analysis, lacking deep mining and analysis of a large amount of historical data and real-time data. In terms of production planning, it is usually arranged according to past production experience and rough prediction of market demand, which is difficult to accurately predict the dynamic changes of market demand, and is easy to cause the disconnection between production plan and actual demand, resulting in production gap or surplus. In terms of process parameter optimization, due to the lack of scientific optimization method and real-time data feedback, the adjustment of process parameters is often blind and lagging, which cannot realize the optimal control of the production process, affecting the product quality and production efficiency. SUMMARY

[0004] The application provides a polyester fabric production whole-process monitoring management method and system to solve the problems mentioned in the background.

[0005] The application provides a polyester fabric production whole-process monitoring management method, which comprises the following steps:

[0006] S1, deploying multiple types of sensors at each key node of the polyester fabric production line; collecting various physical parameters and environmental information related to polyester fabric production in real time; at the same time, integrating the operation state monitoring system of the production equipment to obtain equipment data; preliminarily integrating the data from different sensors and equipment monitoring systems to build a polyester fabric production holographic perception architecture;

[0007] S2, deeply fusing the multi-source heterogeneous data obtained from the polyester fabric production holographic perception architecture; constructing a data correlation model; based on the model, combining the process flow and quality standards of polyester fabric production, comprehensively evaluating the production situation; generating a production situation evaluation report through production situation evaluation;

[0008] S3, collect the multidimensional data of the historical polyester cloth, train and learn the multidimensional data by using a machine learning algorithm, and build a production demand dynamic prediction model; compare the production demand dynamic prediction result with the current production plan to obtain potential production gaps or excesses;

[0009] S4, according to the production demand dynamic prediction result and the production situation evaluation report, combining the process knowledge of polyester cloth production, using intelligent optimization algorithm to adaptively optimize the production process parameters; in the optimization process, the quality indexes and equipment running states in the production process are monitored in real time, and the optimization strategy is adjusted in time according to the feedback information;

[0010] S5, in the production process of polyester cloth, a unique identification code is given to each batch of products, and all related data in the production process are stored in association with the identification code; a real-time quality traceability system is established, the whole life cycle information of the batch of products is inquired by scanning the identification code; at the same time, the historical quality data are analyzed by using data mining and machine learning technology, and a quality defect early warning model is established;

[0011] S6, comprehensively considering the production demand dynamic prediction result, the production situation evaluation report, the process parameter optimization result and the quality traceability and defect early warning information, using operational research and intelligent scheduling algorithm to intelligently allocate production resources.

[0012] The application provides a polyester cloth production whole-process monitoring management system, which comprises:

[0013] one or more processors;

[0014] a memory for storing one or more programs,

[0015] When the one or more programs are executed by the one or more processors, the one or more processors realize the method in any one of the above.

[0016] The application has the advantages that: the whole-process data is collected in real time through the holographic perception architecture, the problems of information islands in traditional production are solved by combining data fusion and intelligent analysis technology, the production state (such as equipment operation, process parameters, environmental conditions and the like) is visualized and transparentized, production management personnel can intuitively master the overall situation through the situation evaluation report, the decision response speed is greatly improved, the production process is changed from experience-driven to data-driven, and the intelligent and fine degree of the production process is significantly improved.

[0017] The process parameter adaptive optimization function continuously searches for the optimal parameter combination through the intelligent algorithm, dynamically adjusts in combination with real-time quality feedback, effectively reduces the parameter fluctuation of key links such as spinning, weaving and dyeing, and reduces the quality defects (such as color difference, insufficient strength, uneven lines and the like) caused by parameter deviation.

[0018] The quality traceability and defect early warning system realizes "one product one code" full life cycle traceability, can quickly locate the root cause of quality problems, and the defect early warning function can identify abnormalities in advance to avoid batch quality problems, improve product qualification rate, and significantly enhance quality stability.

[0019] The production demand dynamic prediction model accurately predicts market demand, avoids "overproduction" or "short supply", reduces inventory backlog and additional costs of emergency production; combined with intelligent resource allocation, the utilization rate of resources such as raw materials, equipment, and manpower is improved, and problems such as equipment idle rate and manpower waste are reduced.

[0020] The process parameter optimization and equipment state monitoring are coordinated to reduce the equipment failure rate and energy consumption (such as precise control of spinning temperature to reduce energy waste), shorten the production cycle, and significantly reduce the production cost per unit of product.

[0021] Real-time quality defect early warning and equipment operation state monitoring functions can detect potential risks (such as equipment failure and quality abnormalities) in advance, facilitate timely intervention, avoid risk amplification, and reduce production interruption losses.

[0022] The full-process data traceability system provides traceable basis for quality disputes and raw material problems, enhances the control ability of supply chain risks, meets the needs of downstream customers for product traceability, and improves brand trust.

[0023] Through production demand dynamic prediction and plan adjustment mechanism, the demand changes of different specifications and varieties of polyester cloth in the market can be quickly responded, and the production plan can be flexibly adjusted (such as switching varieties and adjusting output). Combined with intelligent resource allocation, efficient coordination of small batch and multi-variety production is realized, the adaptability of enterprises to market fluctuations is enhanced, and the timeliness and accuracy of order delivery are improved. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 The method steps of the present application are shown in the figure;

[0025] Figure 2 The present application is described Figure 1 in detail in the S1 step diagram. DETAILED DESCRIPTION

[0026] The preferred embodiments of the present application are described below in conjunction with the accompanying drawings, and it should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application, and are not used to limit the present application.

[0027] One embodiment of the present application, as shown in Figure 1 the method comprises:

[0028] S1, Deploy multiple types of sensors at each key node of the polyester fabric production line; Real-time collection of various physical parameters and environmental information related to polyester fabric production; At the same time, integrate the operation state monitoring system of the production equipment to obtain equipment data; Preliminary integration of data from different sensors and equipment monitoring systems to build a holographic perception architecture for polyester fabric production;

[0029] S2, Deep fusion processing of multi-source heterogeneous data obtained from the holographic perception architecture for polyester fabric production; Build a data correlation model; Based on the model, combined with the process flow and quality standards of polyester fabric production, conduct a comprehensive assessment of the production situation; Through production situation assessment, generate a production situation assessment report;

[0030] S3, Collect multi-dimensional data of historical polyester fabric; Use machine learning algorithms to train and learn multi-dimensional data to build a production demand dynamic prediction model; Compare and analyze the production demand dynamic prediction results with the current production plan to obtain potential production gaps or excesses;

[0031] S4, According to the production demand dynamic prediction results and the production situation assessment report, combined with the process knowledge of polyester fabric production, use intelligent optimization algorithms to adaptively optimize the production process parameters; In the optimization process, real-time monitoring of quality indicators and equipment operating status in the production process, timely adjustment of optimization strategies according to feedback information;

[0032] S5, In the process of polyester fabric production, give each batch of products a unique identification code, and associate and store all related data in the production process with the identification code; Establish a real-time quality traceability system to query the full life cycle information of the batch of products by scanning the identification code; At the same time, use data mining and machine learning techniques to analyze historical quality data and establish a quality defect early warning model;

[0033] S6, Comprehensive consideration of production demand dynamic prediction results, production situation assessment report, process parameter optimization results, and quality traceability and defect early warning information, use operations research and intelligent scheduling algorithms to intelligently allocate production resources.

[0034] The working principle of the above technical solution is: at each key node of the polyester fabric production line, such as the raw material storage area, the spinning workshop, the weaving section, the printing and dyeing area, and the finished product packaging line, etc., a variety of types of sensors are deployed; including but not limited to temperature sensors, humidity sensors, pressure sensors, tension sensors, color sensors and weight sensors, etc., to collect real-time various physical parameters and environmental information related to polyester fabric production; at the same time, the running state monitoring system of the production equipment is integrated to obtain the data such as the speed, power, running time of the equipment; these data from different sensors and equipment monitoring systems are preliminarily integrated to build a holographic perception architecture for polyester fabric production; the architecture takes data flow as the core, connects each data source into an organic whole, and provides comprehensive and accurate basic data for subsequent data analysis and processing;

[0035] The multi-source heterogeneous data obtained from the holographic perception architecture for polyester fabric production is deeply fused and processed; the data cleaning technology is used to remove noise data and outliers, and the data standardization method is used to unify different dimensional data to the same scale. Then, the data correlation analysis algorithm is used to mine the potential relationship between the data, and a data correlation model is constructed; based on the model, combined with the process flow and quality standards of polyester fabric production, the production situation is comprehensively evaluated; the evaluation content includes whether the production progress is normal, whether the equipment operation is stable, whether the product quality meets the requirements, etc.; through the production situation evaluation, a production situation evaluation report is generated, which displays the overall situation of the current polyester fabric production with intuitive charts and detailed data analysis, providing strong support for production decision-making.

[0036] Collect historical polyester fabric production data, market sales data, industry dynamic information, and seasonal factors, etc. multidimensional data; use machine learning algorithms to train and learn the multidimensional data, and build a production demand dynamic prediction model; the model can automatically adjust the prediction parameters according to the real-time input data, and realize accurate prediction of the production demand of polyester fabric in the future period. The prediction results include the yield demand, delivery period requirements, etc. information of different specifications and different varieties of polyester fabric; compare and analyze the production demand dynamic prediction results with the current production plan, and find out the potential production gap or excess situation in time, and provide the basis for the adjustment of production plan;

[0037] According to the production demand dynamic prediction result and the production situation evaluation report, combined with the process knowledge and expert experience of polyester fabric production, the intelligent optimization algorithm is used to adaptively optimize the production process parameters; for the temperature, pressure, speed and other parameters in the spinning process, the warp and weft density, tension control and other parameters in the weaving process, and the dye concentration, temperature, time and other parameters in the printing and dyeing process, through simulation and actual production test, the optimal parameter combination is found; in the optimization process, the quality indicators and equipment running state in the production process are monitored in real time, and the optimization strategy is adjusted in time according to the feedback information; the optimized process parameters are transmitted to the production equipment control system to realize automatic adjustment and precise control of process parameters, improve the production quality and production efficiency of polyester fabric;

[0038] In the production process of polyester fabric, a unique identification code is given to each batch of products, and all related data in the production process, such as raw material information, process parameters, equipment running data, quality detection data, etc., are stored in association with the identification code; a real-time quality traceability system is established, and the whole life cycle information of the product can be quickly queried by scanning the identification code; at the same time, by using data mining and machine learning technology, the historical quality data is analyzed, and a quality defect early warning model is established; this model can monitor the quality indicators in the production process in real time, and when the quality indicators deviate from the normal range, it can send out early warning signals in time, and point out the possible types and causes of quality defects. Production managers can take timely measures to adjust and improve according to the early warning information to avoid the expansion of quality problems;

[0039] Considering the production demand dynamic prediction result, production situation evaluation report, process parameter optimization result and quality traceability and defect early warning information, the production resources are intelligently allocated by using operational research and intelligent scheduling algorithm. Production resources include raw materials, production equipment, human resources, etc. According to the priority and urgency of production tasks, the supply plan of raw materials is reasonably arranged, the scheduling scheme of production equipment is optimized, and the post distribution of human resources is allocated. Through intelligent allocation of production resources, efficient cooperation of production process is realized, utilization rate of production resources is improved, and production cost is reduced. At the same time, production collaborative optimization mechanism is established to strengthen information communication and coordination among various production links to ensure the smooth progress of the whole polyester fabric production process.

[0040] The effect of the above technical scheme is that by deploying various sensors and equipment monitoring systems, real-time collection of various data in the production process is realized, and accurate and real-time basic data for subsequent data analysis and processing is provided. This greatly reduces the human intervention and delay in the production process, thereby improving the production efficiency.

[0041] By using deep fusion processing and data correlation analysis algorithm, the production situation can be comprehensively evaluated, and potential problems such as production progress lag and equipment operation instability can be found in time, so as to effectively avoid the risks that may occur in the production process. In addition, the quality defect early warning system can also help to find quality problems in time and avoid rework or waste in the later production.

[0042] By establishing a production demand dynamic prediction model and an intelligent optimization algorithm through machine learning algorithm, the future production demand can be accurately predicted, and the production process parameters can be automatically adjusted. This makes the production plan and adjustment more accurate, reduces unnecessary loss and overproduction in the production process. In the production process, the quality indicators are monitored in real time and optimized and adjusted according to the feedback, effectively improving the quality of polyester fabric. In addition, the establishment of the quality traceability system makes each batch of products have complete quality information, which can quickly locate and handle any quality problems, further ensuring the consistency and stability of the products.

[0043] Through intelligent scheduling algorithm and optimization of production resources, the supply of raw materials, the use of production equipment and the allocation of human resources are reasonably arranged, which reduces resource waste and reduces production cost. At the same time, the efficient cooperation of the production process further improves the resource utilization rate and avoids repeated waste of resources. Combined with historical data, real-time data and external factors, the system can quickly respond to changes in market demand and adjustments in production environment, making production more flexible and efficient, and being able to quickly respond to emergencies and meet market demand.

[0044] In one embodiment of the present application, the S1 comprises:

[0045] S11, survey and analysis at each key node of the polyester fabric production line; and according to the production characteristics and monitoring requirements of each node, determine the type of sensor to be deployed;

[0046] S12, according to the selected sensor type and deployment position, carry out sensor installation and debugging work; at the same time, configure data acquisition equipment to convert the analog signals collected by the sensor into digital signals, and carry out preliminary storage and processing;

[0047] S13, integrate the running state monitoring system of the production equipment, obtain the key data of the equipment through equipment monitoring technology; use industrial Ethernet to seamlessly connect the equipment monitoring system with the data acquisition network, and synchronously transmit the equipment running data and sensor collected data;

[0048] S14, preliminarily integrate the data collected from different sensors and equipment monitoring systems, use data fusion technology to associate and match the data from multiple data sources, eliminate data redundancy and conflicts; build a polyester fabric production holographic perception architecture, and connect each data source into an organic whole.

[0049] The working principle of the above technical solution is: detailed investigation and analysis are carried out at each key node of the polyester fabric production line, such as raw material storage area, spinning workshop, weaving section, printing and dyeing area, and finished product packaging line, etc.; and according to the production characteristics and monitoring requirements of each node, the type of sensor required is determined; for example, in the raw material storage area, a temperature and humidity sensor is deployed to monitor the temperature and humidity of the raw material storage environment to prevent the raw material from being damp or deteriorated; in the spinning workshop, temperature sensors, pressure sensors and tension sensors are installed to monitor the temperature, pressure and fiber tension during the spinning process, respectively, to ensure stable spinning quality; in the weaving section, warp and weft density sensors and tension sensors are set to real-time master the warp and weft density and tension of the fabric; in the printing and dyeing area, color sensors and temperature sensors are equipped to accurately control the color of the dye and the printing and dyeing temperature; in the finished product packaging line, a weight sensor is used to ensure the accuracy of the finished product packaging weight;

[0050] According to the selected sensor type and deployment position, the installation and debugging of the sensor are carried out; ensure that the sensor is installed firmly and accurately in position, and can stably and reliably collect data; at the same time, configure data acquisition equipment to convert the analog signals collected by the sensor into digital signals and carry out preliminary storage and processing; establish a data acquisition network to ensure that the data of each sensor can be transmitted to the data processing center in real time and accurately;

[0051] The integrated production equipment operation state monitoring system acquires key data such as the speed, power and running time of the equipment through equipment monitoring technology; the industrial Ethernet is used to seamlessly connect the equipment monitoring system with the data acquisition network for synchronous transmission of the equipment operation data and sensor collected data;

[0052] The data collected from different sensors and equipment monitoring systems are preliminarily integrated, the data fusion technology is adopted to associate and match the data from multiple data sources, and the data redundancy and conflict are eliminated; a polyester fabric production holographic perception architecture is constructed, which takes data flow as the core and connects each data source into an organic whole.

[0053] The effect of the above technical solution is: through real-time monitoring of each link in the production process, it is ensured that each step is carried out in the best state. Greatly reduces the delay in the production process, improves the overall production efficiency. Through the integration of sensors and equipment monitoring systems, potential equipment failures or production progress problems can be identified in real time, and timely intervention can be made to reduce the risks that may occur in production.

[0054] With the help of machine learning and data analysis technology, production demand is predicted and process parameters are automatically adjusted. The accuracy of production decision-making is improved, making the production process more intelligent and adaptive. By monitoring the quality indicators in real time during the production process and adjusting according to feedback, the product quality is ensured to meet the standards at every stage, reducing the defect rate.

[0055] By optimizing resource allocation and scheduling, unnecessary resource waste is reduced, and the collaborative efficiency of the production process is improved, significantly reducing production costs. The system can quickly respond to changes in market demand and production environment adjustments, enhancing the flexibility of production, enabling it to quickly respond to emergencies and meet changing market demands.

[0056] One embodiment of the present application, the S14, comprises:

[0057] S141, the collected various types of raw data are preprocessed, and data mining and correlation analysis techniques are used to deeply analyze the internal relationships between different data sources; according to the logical relationship of the production process and the cooperative working principle between devices, a data correlation model is established;

[0058] S142, using data fusion algorithm, the data after correlation matching is deeply analyzed; at the same time, for the conflict data existing in different data sources, according to the preset rules and priority, the conflict data is processed;

[0059] S143, based on the above processed data, a data fusion model for polyester fabric production is constructed; the data is fused and optimized;

[0060] S144, taking the fused data stream as the core, a holographic perception architecture for polyester fabric production is constructed; through data interface and network protocol connection, an organic whole is formed, and real-time data sharing and interaction are carried out.

[0061] The working principle of the above technical solution is: the collected various types of raw data are preprocessed, and data mining and correlation analysis techniques are used to deeply analyze the internal relationships between different data sources; according to the logical relationship of the production process and the cooperative working principle between devices, a data correlation model is established; through the model, the data from different sensors and device monitoring systems are precisely matched to determine their corresponding relationship in time and space. For example, the temperature, pressure and tension data of the spinning workshop are associated with the corresponding time point device running parameters to comprehensively analyze the influence of various factors in the production process on product quality;

[0062] A data fusion algorithm is used to perform in-depth analysis on the associated and matched data; redundant information in the data is identified and eliminated to avoid repeated calculations and storage, thereby improving data processing efficiency; at the same time, conflicting data that may exist from different data sources are processed according to preset rules and priorities; for example, when multiple sensors measure the same physical quantity, if there are differences in the measurement results, the final data value to be used is determined based on factors such as the accuracy and reliability of the sensors.

[0063] Based on the processed data, a data fusion model for polyester fabric production is constructed. This model comprehensively considers various variables and factors in the production process, employing advanced algorithms such as machine learning and deep learning to fuse and optimize the data. By continuously training and adjusting the model parameters, the model's ability to simulate and predict the production process is improved, providing more accurate and comprehensive data support for production decisions. For example, the data fusion model can be used to predict the quality indicators of polyester fabric, identify potential quality problems in advance, and adjust production parameters in a timely manner.

[0064] A holographic sensing architecture for polyester fabric production is constructed, centered on the fused data stream. This architecture connects various data sources, including sensors and equipment monitoring systems, into an organic whole through data interfaces and network protocols, enabling real-time data sharing and interaction. This allows production managers to grasp the status and information of each stage of the production process from a global perspective. Simultaneously, this architecture is scalable and flexible, easily integrating new data sources and equipment to adapt to the continuous development and innovation of polyester fabric production technology. For example, when introducing new production equipment or monitoring technologies, only corresponding configuration and integration within the holographic sensing architecture are needed to collect and analyze the new data.

[0065] The above technical solution achieves the following results: by using data fusion and preprocessing techniques, it reduces the generation of redundant data, optimizes data storage and processing flows, and improves the overall efficiency of data processing. By utilizing preset rules and priorities to handle conflicting data, it effectively eliminates errors caused by data inconsistency, ensuring the accuracy and reliability of the data.

[0066] By constructing a data fusion model and combining it with machine learning and deep learning technologies, quality indicators in the production process can be simulated and predicted more accurately, potential problems can be identified in advance, and the occurrence of quality defects can be reduced. Based on the fused data stream, production managers can understand the production status in real time from a global perspective, providing more accurate and comprehensive data support for production decisions and improving the scientific nature and efficiency of decision-making.

[0067] Through automated data processing and intelligent production monitoring, the need for human intervention is reduced, and the production risk caused by human error is reduced. The holographic perception architecture has good scalability and flexibility, can quickly adapt to technical updates and equipment introduction, and maintain efficient operation of the system, thereby improving the flexibility of the production process and the ability to respond to emergencies. With the advancement of technology and the continuous optimization of data fusion models, production systems will become increasingly intelligent, capable of automatically adjusting and optimizing production processes, and thus achieving a more efficient and environmentally friendly production model.

[0068] In one embodiment of the present application, the S142 comprises:

[0069] For the data matched by association, calculate various statistical indicators, identify highly correlated data by analyzing the statistical correlation between different data sequences, and analyze the data at the semantic level in combination with the professional knowledge base in the production field; determine whether there is semantically repeated data;

[0070] For time series data, analyze its trend and periodic characteristics; if multiple data sequences show similar trend and periodic patterns within the same time interval, and the similarity exceeds the reasonable error range, it is determined as redundant data; for statistically correlated or semantically repeated redundant data, use data aggregation method for processing; according to the type and characteristics of the data, select the aggregation function, and for redundant data on time series, use data sampling method to eliminate; determine the sampling interval according to the change frequency and importance of the data;

[0071] Use the established data association model and prediction algorithm to predict and replace redundant data; by analyzing the change law of historical data and current data, establish a prediction model, and use the predicted value to replace part of the redundant data;

[0072] For the data matched by association, check its consistency between different data sources; by comparing the attributes of the data, obtain the conflicting data; formulate conflict judgment rules in combination with the business rules and process requirements of polyester fabric production; establish normal data mode and abnormal data mode using historical data; by comparing the current data with the historical data mode, find the data that does not conform to the normal mode, and determine whether it is conflict data;

[0073] Evaluate the data source of the conflict data to determine the time effectiveness priority of the data; determine the business importance priority of the data according to its importance in the polyester fabric production business; use machine learning algorithm to fuse the conflict data; collect a large amount of historical conflict data and corresponding correct processing results to construct a training data set;

[0074] The model learns the processing mode of the conflict data by training the training data set through the neural network; when new conflict data appears, the data is input into the trained model to obtain the fused result.

[0075] The working principle of the above technical solution is that, for the data matched by association, statistical indicators such as mean, variance, and covariance are calculated; by analyzing the statistical correlation between different data sequences, highly correlated data is identified; for example, in polyester fabric production, if the data collected by two temperature sensors at different positions show almost the same mean value and very small variance difference over a long period of time, and the covariance is close to the variance value, it can be preliminarily determined that the two sets of data are redundant; combined with the professional knowledge base in the production field, the data is analyzed at the semantic level; considering the actual meaning of the data and the logical relationship in the production process, it is determined whether there is semantically repeated data; for example, when describing the running state of the equipment, "normal operation" and "equipment fault-free operation" have high semantic similarity, if they appear at the same time in the data and are used to express the state of the same equipment, they can be determined as redundant information.

[0076] For time series data, analyze its trend and periodic characteristics; if multiple data sequences show similar trend and periodic patterns in the same time interval, and the similarity exceeds the reasonable error range, it is determined as redundant data; for example, when monitoring the tension change in the polyester fabric production process, if the tension data of two adjacent stations shows almost synchronous rising and falling trend in multiple production cycles, it is considered as redundant; for redundant data with statistical correlation or semantic repetition, data aggregation method is used for processing; according to the type and characteristics of the data, select aggregation functions such as sum, average, maximum, and minimum. For example, for redundant temperature data collected by multiple temperature sensors, the average value can be calculated as the representative temperature of the region, thereby reducing the data volume and retaining the key information; for redundant data on time series, data sampling method is used to eliminate; according to the change frequency and importance of the data, determine the sampling interval; for slowly changing data with less impact on production, the sampling interval can be appropriately increased; for frequently changing and critical data, keep a small sampling interval. For example, when monitoring the vibration data of the polyester fabric production equipment, if the equipment vibration changes smoothly in normal operation state, a larger sampling interval can be used; when the equipment has abnormal vibration, the sampling interval is immediately reduced to obtain more detailed data.

[0077] The established data correlation model and prediction algorithm are used to predict and replace redundant data; by analyzing the change law of historical data and current data, a prediction model is established, and the prediction value is used to replace part of the redundant data, thereby reducing the storage and calculation amount of actual data; for example, when predicting a certain process parameter in the production process of polyester fabric, according to the data at the previous time points and the known production conditions, the prediction value at the current time point is calculated by using the prediction model, and if the error between the prediction value and the actually collected data is within the allowable range, the prediction value can replace the actual data;

[0078] The consistency of the data after correlation matching is checked between different data sources; by comparing the value range, unit, precision and other attributes of the data, the conflict data is obtained; for example, in the production of polyester fabric, the measurement units of the same production parameter (such as spinning speed) by different equipment monitoring systems may be different, which will lead to data conflict if not converted uniformly; conflict judgment rules are formulated in combination with the business rules and process requirements of polyester fabric production; for example, according to the production process regulations, the temperature in the spinning process should fluctuate within a certain range, if the temperature data collected by a certain sensor exceeds this range and conflicts with other related data (such as heating power, cooling water flow, etc.), it can be judged as conflict data; the normal data mode and abnormal data mode are established by using historical data; by comparing the current data with the historical data mode, the data that does not conform to the normal mode is found, and whether it is conflict data is judged; for example, the equipment running parameters and product quality data in the past period of polyester fabric production are analyzed to establish the data distribution range and correlation model under normal production mode. When the current data deviates from the model, further analysis is made to determine whether it is conflict data;

[0079] The data sources of the conflicting data are evaluated, and the accuracy priority of the data is determined according to factors such as the accuracy, reliability, and calibration of the data acquisition equipment; the data of the data source with high accuracy is preferentially adopted, and the data of the data source with low accuracy is discarded or modified. For example, in the production of polyester fabric, if the data collected by a high-precision sensor that has been calibrated periodically conflicts with the data collected by a low-precision sensor that has not been calibrated, the data of the high-precision sensor is preferentially adopted; the time effectiveness priority of the data is determined by considering the collection time and update frequency of the data; for production processes with high real-time requirements, the latest collected data is preferentially adopted. For example, when monitoring the real-time running state of the polyester fabric production equipment, if the data provided by two data sources conflicts, the most recently collected data is preferentially adopted to ensure the timeliness and accuracy of production decisions; the business importance priority of the data is determined according to the importance of the data in the polyester fabric production business; for data that affects product quality, production safety, and key production indicators, a higher priority is given. For example, when controlling the fiber fineness of polyester fabric, process parameter data related to fiber fineness (such as spinneret hole diameter, melt temperature, etc.) have higher business importance, which should be preferentially considered when processing conflicting data; for data that conflicts but is of equal importance, a weighted average fusion method is used for processing; according to factors such as the accuracy, timeliness, and business importance of the data, each data source is assigned a corresponding weight, and the weighted average value is calculated as the fused data. For example, when fusing the thickness data of polyester fabric collected by two different sensors, if the accuracy, timeliness, and business importance of the two sensors are similar, the same weight can be assigned to them according to the actual situation, and the weighted average value is calculated as the final thickness data; for complex conflicting data, the knowledge and reasoning ability of an expert system are used for processing; an expert system is established, and expert knowledge and experience in polyester fabric production are encoded into rules. When conflicting data occurs, the expert system makes inferences and judgments according to the rules, selects the most appropriate data, or gives fusion suggestions. For example, when processing conflicts between multiple process parameters in the production of polyester fabric, the expert system can analyze the mutual influence relationship between the parameters according to production process rules and historical experience to determine the optimal parameter combination; machine learning algorithms are used to fuse conflicting data; a large amount of historical conflicting data and corresponding correct processing results are collected to construct a training data set;

[0080] The training data set is trained through a neural network, so that the model learns the processing mode of conflicting data; when new conflicting data occurs, the data is input into the trained model to obtain the fused result. For example, a neural network model is used to fuse device fault diagnosis data in the production of polyester fabric to improve the accuracy and reliability of fault diagnosis.

[0081] The effect of the above technical solution is that: through statistical analysis, aggregation, sampling and prediction of the data matched by association, the storage and calculation amount of redundant data can be effectively reduced, the data processing process is optimized, and the efficiency of data analysis in the production process is improved. By analyzing the correlation and semantic repetition of data, redundant data can be identified and eliminated to ensure the accuracy and simplicity of the data. In addition, by evaluating the accuracy, timeliness and business importance of the data source, the negative impact of conflicting data on the decision-making process can be reduced.

[0082] In processing redundant and conflicting data, high-precision, high-timeliness and high-business importance data sources are preferred to ensure that the final data is more accurate and reliable. In addition, by combining the processing capabilities of machine learning and expert systems, conflicting data can be intelligently integrated to improve the accuracy and reliability of the final results.

[0083] By establishing a data correlation model and a prediction algorithm, redundant data can be predicted and replaced to provide the required information in real time during actual production, support real-time decision-making and optimize operations, and improve production efficiency and product quality. Using statistical analysis of data and machine learning models, abnormal data in production can be better identified, reducing uncertainty caused by data conflicts, redundancies or errors, and improving production stability.

[0084] Through expert systems and machine learning models, data can be intelligently analyzed and processed to improve the intelligent level of the production process, achieve automated management, reduce human intervention, and reduce human errors. Combined with semantic analysis and business rules, data can be accurately interpreted from a statistical perspective and effectively interpreted from a business and production perspective to provide valuable information support for production decisions. Through trend analysis and periodic feature identification of time series data, the optimization direction of equipment and processes can be identified to reduce equipment failure and process instability, further improving production efficiency and quality.

[0085] One embodiment of the present application, S2, comprises:

[0086] S21, deeply clean the multi-source heterogeneous data obtained from the production of polyester fabric in the holographic perception architecture; through a machine learning algorithm, identify and remove noise data and outliers;

[0087] S22, use a data standardization method to unify data of different dimensions to the same scale; through data standardization processing, eliminate the dimensional differences between different data;

[0088] S23, use a data correlation analysis algorithm to mine the potential relationship between data, use association rule mining to find the internal relationship between data in different production links; build a data correlation model;

[0089] S24, based on the constructed data correlation model, combining the process flow and quality standards of polyester fabric production, the production situation is evaluated; through the preset evaluation index and threshold, the real-time monitoring and analysis of various data in the production process are carried out;

[0090] S25, when the data exceeds the normal range, a warning signal is sent; according to the production situation evaluation result, a production situation evaluation report is generated.

[0091] The working principle of the above technical scheme is: the multi-source heterogeneous data obtained from the polyester fabric production holographic perception architecture is deeply cleaned; through machine learning algorithm, noise data and outliers are identified and removed; for example, the continuous data collected by the sensor is smoothed by using mean filtering, median filtering and other methods, and the abnormal data points caused by sensor failure or external interference are removed; at the same time, the data is processed for missing values, and interpolation method, mean substitution method and other methods are used to fill the missing data, to ensure the integrity and accuracy of the data;

[0092] The data standardization method is used to unify the data of different dimensions to the same scale; through data standardization processing, the dimensional difference between different data is eliminated;

[0093] The data correlation analysis algorithm is used to mine the potential relationship between data, and the association rule mining is used to find the internal relationship between data in different production links; for example, by analyzing the relationship between temperature, pressure parameters in the spinning process and the warp density, weft density, tension parameters in the weaving section, a data correlation model is constructed;

[0094] Based on the constructed data correlation model, combining the process flow and quality standards of polyester fabric production, the production situation is evaluated; the evaluation content includes whether the production progress is normal, whether the equipment operation is stable, whether the product quality meets the requirements, etc.; through the preset evaluation index and threshold, the real-time monitoring and analysis of various data in the production process are carried out;

[0095] When the data exceeds the normal range, a warning signal is sent; according to the production situation evaluation result, a production situation evaluation report is generated, which displays the overall situation of the current polyester fabric production with intuitive charts and detailed data analysis, providing strong support for production decision-making.

[0096] The effect of the above technical scheme is: through deep cleaning of multi-source heterogeneous data, using mean filtering, median filtering and other methods to remove noise data and outliers, ensuring the accuracy of the data, thereby providing high-quality data support for subsequent analysis and decision-making. Through smoothing the continuous data collected by the sensor, the influence of abnormal data points caused by equipment failure or external interference is reduced, thereby reducing the interference of these abnormal data on the analysis result.

[0097] By filling in missing values and uniformly standardizing the data, the consistency of the data at different levels and dimensions is ensured, the analysis errors caused by data missing or dimension difference are avoided, and the comparability and stability of the data are enhanced. The data correlation analysis algorithm is used to construct the data correlation model, which can deeply mine the internal relationship between different production links, make each link of the production process more transparent, and help the production manager understand the interaction and influence between each link.

[0098] By real-time monitoring and evaluation of production situation, including production progress, equipment running state and product quality, potential problems can be found in time and effective measures can be taken to reduce potential risks in the production process. Based on the constructed correlation model and evaluation report, the management personnel can more clearly understand the production situation, rely on intuitive charts and data analysis to make more scientific and accurate production decisions.

[0099] Through real-time data monitoring and early warning mechanism, abnormalities can be found in time and corresponding remedial measures can be taken to ensure the smooth progress of the production process, reduce downtime, optimize resource allocation and improve overall production efficiency. Through automatic data processing and evaluation, the need for manual intervention is reduced, the incidence of human error is reduced, and the automation level of production management is improved.

[0100] Through real-time monitoring and data analysis of the production process, product quality can be effectively detected to ensure that the produced polyester fabric meets the quality standards, and the accuracy and timeliness of quality control are enhanced. Through fine processing and analysis of data, quick response to various changes in the production process is provided, providing higher flexibility and adaptability to help the production line adjust and optimize according to different needs.

[0101] An embodiment of the present application, the S23, comprises:

[0102] Preliminary exploratory analysis is performed on the cleaned and standardized data; the linear correlation degree between the data is quantified from the numerical point of view;

[0103] According to the characteristics and analysis requirements of the polyester fabric production data, the association rule mining algorithm is selected; the optimized association rule mining algorithm is used to mine frequent item sets in the polyester fabric production data;

[0104] Based on the mined frequent item sets, association rules are generated; and the generated association rule model is verified, the data set is divided into training set and test set by cross-validation method, the association rules are generated on the training set, and then the effectiveness of these rules is verified on the test set.

[0105] A calculation verification index is used to evaluate the performance of the model on unknown data; if the model performance does not meet the preset requirements, the reasons are analyzed and the algorithm parameters are adjusted or the algorithm is reselected for optimization.

[0106] The working principle of the above technical solution is that before formally using a complex algorithm to mine data correlations, a preliminary exploratory analysis is performed on the cleaned and standardized data; by drawing a data scatter plot matrix, the relationship between the data of different production links is intuitively presented, for example, the distribution trend of the data points of the spinning temperature and the weaving tension in the graph is observed to determine whether there are linear, nonlinear or no obvious correlation; at the same time, the simple correlation coefficients between the data are calculated, such as the Pearson correlation coefficient and the Spearman rank correlation coefficient, to quantify the degree of linear correlation between the data from the numerical point of view and quickly locate the data pairs that may have correlations to provide direction guidance for subsequent in-depth mining; for example, if it is found that the absolute value of the correlation coefficient of the spinning pressure and the printing color deviation is large, it can be listed as the key object of association analysis;

[0107] According to the characteristics and analysis requirements of the polyester fabric production data, the association rule mining algorithm is selected; for example, the classic Apriori algorithm is suitable for small data sets and simple rules, and can efficiently find frequent item sets and association rules; and the FP-Growth algorithm has higher efficiency in processing large-scale data sets, and reduces the number of database scans by constructing an FP tree. Considering that the polyester fabric production data may have multi-dimensional, massive and complex characteristics, the selected algorithm can be optimized. For example, the Apriori algorithm is improved by using parallel computing technology to divide the data set into multiple subsets and mine frequent item sets in parallel on different computing nodes, and finally combine the results to greatly improve the running efficiency of the algorithm; the optimized association rule mining algorithm is used to mine frequent item sets in the polyester fabric production data; a minimum support threshold is set, which represents the minimum proportion of an item set appearing in all data; only the item set that meets the minimum support requirement is considered as a frequent item set. For example, when analyzing the data of the spinning workshop and the weaving section, the minimum support is set to 0.1, and the algorithm will find production parameter combinations that appear in at least 10% of the data records, such as (spinning temperature higher than the threshold value, weaving warp and weft density in a certain range), these frequent item sets reflect the situation that different production link data often appear at the same time, and are the basis for building association rules;

[0108] Based on the mined frequent item sets, association rules are generated. The form of the association rules is "if A, then B", where A and B are different production parameters or events. The support, confidence and lift of each association rule are calculated. The support represents the frequency of the simultaneous occurrence of A and B; the confidence represents the probability of the occurrence of B under the condition of the occurrence of A; and the lift measures the effectiveness of the rule. If the lift is greater than 1, it indicates that there is a positive correlation between A and B. According to the preset confidence and lift thresholds, association rules with practical significance are selected. For example, if the confidence is set to 0.7 and the lift is set to 1.2, the association rule "if the spinning pressure exceeds the standard value, then the probability of the increase in the number of fiber breaks during the weaving process is greater than 70%, and the rule can effectively reflect the relationship between the two" is selected. The generated association rule model is verified to ensure its accuracy and reliability. The cross-validation method is used to divide the data set into a training set and a test set. The association rules are generated on the training set, and then the effectiveness of the rules is verified on the test set.

[0109] The verification indicators such as the accuracy and recall rate of the rules are calculated to evaluate the performance of the model on unknown data. If the performance of the model does not meet the preset requirements, the reasons are analyzed and the algorithm parameters are adjusted or the algorithm is reselected for optimization. For example, if it is found that the accuracy of the association rules on the test set is low, the minimum support threshold can be appropriately increased to reduce the number of frequent item sets and improve the quality of the rules. Alternatively, other association rule mining algorithms such as the Eclat algorithm can be tried to rebuild the association model.

[0110] The effect of the above technical solution is that through exploratory data analysis before formal algorithm application, the relationship between production link data can be intuitively observed, and potential association patterns can be found, providing higher accuracy and direction for subsequent in-depth analysis. Through preliminary exploratory analysis of the data, such as calculating the correlation coefficient and drawing a scatter plot, potential associated data pairs can be quickly located, thereby reducing the amount of complex data that needs to be further mined.

[0111] Through optimization of the Apriori algorithm, parallel computing technology and efficient use of the FP-Growth algorithm, the data processing and mining efficiency can be greatly improved, especially when facing large-scale and complex data sets. The optimized algorithm can quickly find frequent item sets. By accurately setting the minimum support, confidence and lift indicators, association rules with practical significance can be selected from a large amount of data, thereby improving the accuracy of data association analysis and ensuring that the mined rules have practical application value.

[0112] By calculating the verification indexes such as support, confidence and lift in the generated association rules, unreliable or invalid rules can be screened out, and the probability of adopting false rules is reduced. Using the effective association rules mined, management personnel can optimize the production process according to these rules, thereby improving the scientificity and reliability of decision-making and ensuring that the optimization scheme of each production link has effectiveness.

[0113] By dividing the data set into training set and test set by using cross-validation method, the accuracy of the model is verified on different data sets, which can ensure that the generated association rules have higher robustness and avoid the risk of overfitting. During the verification process, if the performance of the model does not meet the expected performance, the model can be quickly optimized by adjusting the algorithm parameters or selecting a more suitable algorithm such as Eclat algorithm, which reduces the complexity of rebuilding the model.

[0114] Through the rules revealed by data association analysis, production managers can flexibly adjust production parameters such as spinning temperature and weaving tension, optimize the production process and improve the flexibility and adaptability of the production line. By mining and applying actual effective association rules, production parameters can be accurately adjusted to reduce the generation of unqualified products, thereby improving the quality control level of products and reducing quality fluctuations in the production process.

[0115] An embodiment of the present application, the S3, comprises:

[0116] S31, acquire the multidimensional data of the historical polyester fabric, and train and learn the multidimensional data by using a machine learning algorithm;

[0117] S32, divide the historical data into a training set and a test set, continuously adjust the parameters and structure of the model, build a production demand dynamic prediction model, automatically adjust the prediction parameters according to the real-time input data, and predict the production demand of the polyester fabric in the future period of time;

[0118] S33, compare and analyze the production demand dynamic prediction result with the current production plan; by establishing a comparative analysis model, calculate the difference between the predicted demand and the production plan, and adjust the production plan according to the comparative analysis result.

[0119] The working principle of the above technical solution is as follows: collect multi-dimensional data of historical polyester fabric, including time (such as polyester fabric demand in different seasons and months), market factors (including market demand trends, competitor dynamics, changes in consumer preferences, etc.), price information (raw material price fluctuations, polyester fabric finished product price trends), production-related data (such as past production volume, production efficiency, equipment operating conditions, etc.), and other factors that may affect polyester fabric production demand. Use machine learning algorithms to train and learn the collected multi-dimensional historical data. Machine learning algorithms can automatically mine potential patterns, rules and relationships from data, and continuously adjust the parameters inside the algorithm to make the model as accurate as possible to fit the characteristics reflected by the historical data, so as to have the ability to understand and analyze the polyester fabric production demand related data, and lay the foundation for subsequent prediction work.

[0120] The collected historical data is divided into training set and test set according to a certain proportion (such as common 70%-30% or 80%-20% etc.). The training set is used to train the machine learning model, so that the model learns the patterns and rules in the data; the test set is used to evaluate the performance of the model on unseen data, and to test the generalization ability of the model. In the training process, the parameters and structure of the model are constantly adjusted. Parameter adjustment aims to optimize the numerical settings inside the model to better fit the data; structure adjustment may involve changing the model's hierarchy, number of nodes, etc. to find the most suitable model architecture for polyester fabric production demand prediction. Through repeated iteration of training and adjustment, a production demand dynamic prediction model is constructed. The model has the ability to automatically adjust the prediction parameters according to real-time input data. When new real-time data (such as the latest market dynamics, production data, etc.) is input, the model can quickly respond and use these new data to update and optimize the prediction parameters, thereby achieving accurate prediction of polyester fabric production demand in the future period. This dynamic adjustment mechanism enables the model to adapt to changing market environments and production conditions, improving the timeliness and accuracy of the prediction.

[0121] The predicted results from the production demand dynamic prediction model are compared and analyzed with the current production plan. By establishing a special comparative analysis model and using specific algorithms and rules, the differences between the predicted demand and the production plan are calculated. These differences may be reflected in the production quantity, production time arrangement, product types, etc., and can directly reflect the degree of mismatch between the current production plan and the market demand prediction. According to the results of the comparative analysis, the production plan is adjusted. If the predicted demand is higher than the current production plan, the production quantity may need to be increased, the production time may need to be advanced, or the product types may need to be adjusted to meet market demand; conversely, if the predicted demand is lower than the current production plan, the production quantity may need to be reduced, the production resource allocation may need to be optimized, or the production rhythm may need to be adjusted to avoid inventory accumulation and resource waste. Through this dynamic adjustment mechanism based on data prediction, the production plan can better meet market demand, improving the production efficiency and economic benefits of enterprises.

[0122] The effect of the above technical solution is that by training historical polyester fabric data using machine learning algorithms, a more accurate production demand dynamic prediction model can be constructed, improving the prediction accuracy of future production demand and better meeting market demand. By dividing historical data into training and test sets, the parameters and structure of the model can be continuously adjusted to reduce the difference between production demand prediction and actual production plan, reducing the risk of plan errors and resource waste.

[0123] By adjusting the parameters of the prediction model in real time, the production demand prediction can be dynamically adjusted according to the latest input data, enhancing the flexibility and adaptability of the production plan and ensuring that the production plan can respond to changes in market demand at any time. By establishing a comparative analysis model, the difference between predicted demand and production plan can be monitored in real time, potential problems in production scheduling can be discovered in a timely manner, and rapid adjustments can be made, improving the efficiency and response speed of production scheduling.

[0124] Through accurate production demand prediction and production plan adjustment, resource allocation can be optimized, unnecessary waste in the production process can be reduced, and the consumption of raw materials, labor and time resources can be reduced. Accurate production demand prediction and dynamic production plan adjustment enable enterprises to better meet customer demand in a changing market environment, thereby enhancing the market competitiveness and customer satisfaction of enterprises.

[0125] By comparing and analyzing the differences between predicted demand and production plan, data-driven decision support can be provided for decision-makers, ensuring that production plan adjustments are based on reliable data analysis and improving the scientificity and accuracy of decisions. By optimizing production plans and resource allocation, additional costs caused by overproduction or underproduction can be reduced, thereby reducing overall production costs.

[0126] By continuously monitoring the differences between predicted and actual demand, better control over production progress and quality can be achieved, enhancing the controllability and stability of the production process. By adjusting production demand forecasts and production plans in real time, businesses can respond more quickly to fluctuations in market demand, ensuring timely adjustments to production lines, thereby improving the speed of the enterprise's response to external changes.

[0127] One embodiment of the present application, the S4, comprises:

[0128] S41, through the network crawler, the process knowledge of polyester fabric production is obtained, and the experience and skills in the production process are obtained through the expert knowledge base;Integrate the process knowledge and experience and skills to form a complete production process parameter optimization knowledge base;

[0129] S42, according to the characteristics of polyester fabric production and the target of process parameter optimization, intelligent optimization algorithm selection is carried out, and the production process parameters are self-adaptively optimized;

[0130] S43, in the optimization process, different parameter combinations are simulated and tested by using simulation technology;By establishing a mathematical model of the production process, the production effect and quality index under different parameter combinations are predicted by using computer simulation software for simulation;

[0131] S44, according to the simulation results, the optimal parameter combination is selected for actual production test;In the actual production process, the quality index and equipment running state in the production process are monitored in real time, and the related data are recorded;

[0132] S45, according to the feedback information of actual production test, the optimization strategy is adjusted in time;If there is a difference between the actual production effect and the simulation result, the reason is analyzed and the parameter combination is further optimized;

[0133] S46, the optimized process parameters are transmitted to the production equipment control system for automatic adjustment and control of the process parameters;And through the communication interface with the production equipment, the optimized parameters are sent to the equipment controller in real time.

[0134] The working principle of the above technical solution is: using web crawler technology, according to the preset rules and keywords, automatically searching, filtering and extracting the process knowledge related to polyester fabric production from the vast amount of information on the Internet. These knowledge covers the theoretical knowledge, technical specifications and operation points of each link from raw material selection, spinning process, weaving process to finishing process. Through the expert knowledge base, the rich experience and practical skills accumulated by senior experts in the industry in the production process of polyester fabric are gathered. The expert knowledge base can integrate successful cases and problem solving solutions from different enterprises and different production scenarios to provide practical guidance for process parameter optimization. The process knowledge obtained through the web crawler and the experience and skills collected from the expert knowledge base are systematically integrated. Remove redundant and invalid information, classify, summarize and organize the knowledge, and form a complete production process parameter optimization knowledge base. This knowledge base provides a comprehensive theoretical and practical basis for subsequent process parameter optimization.

[0135] In-depth analysis of the characteristics of polyester fabric production, such as the continuity of the production process, the complexity of the process parameters, the multidimensionality of the quality indicators, etc., and clearly define the goals of process parameter optimization, such as improving production efficiency, reducing production cost, improving product quality, etc. According to these characteristics and goals, select the most suitable algorithm from numerous intelligent optimization algorithms, such as genetic algorithm, particle swarm optimization algorithm, neural network algorithm, etc. Use the selected intelligent optimization algorithm to adaptively optimize the production process parameters. The algorithm will automatically adjust the value range and optimization direction of the parameters according to the real-time data in the production process and the optimization goal. Through continuous iteration calculation and parameter adjustment, the optimal process parameter combination is found to realize the optimization of the production process.

[0136] Based on the physical and chemical principles of polyester fabric production, a mathematical model of the production process is established. This model can accurately describe the quantitative relationship between each process parameter and the production effect, quality indicators in the production process, providing a theoretical basis for simulation. Use computer simulation software to input different process parameter combinations into the established mathematical model for simulation test. Through the high-speed calculation and visualization function of the simulation software, predict the production effect under different parameter combinations, such as yield, production cycle, etc., and quality indicators, such as the strength, elongation, color fastness of polyester fabric, etc. Simulation technology can quickly evaluate the feasibility of different parameter combinations without actual production, providing a reference for actual production.

[0137] According to the simulation results, the parameter combination with the best prediction effect is selected for actual production test. In the actual production environment, the production equipment is set according to the selected parameter combination, and the production of polyester fabric is carried out. In the actual production process, various sensors and monitoring equipment are used to monitor the quality indicators and equipment running status in real time, such as temperature, pressure, speed, tension, etc. At the same time, the monitored data is recorded in time to provide data support for subsequent analysis and optimization.

[0138] The results of actual production test are compared with the simulation results. If there is a difference between the actual production effect and the simulation results, the reasons for the difference are analyzed in depth, which may be caused by factors such as inaccurate mathematical model, difference between actual production environment and simulation environment, unstable equipment performance, etc. According to the results of difference analysis, the optimization strategy is adjusted in time. The mathematical model is corrected and improved to improve its accuracy; considering the influence factors of actual production environment, the parameter combination is further optimized to ensure that the actual production effect reaches the expected target.

[0139] The process parameters verified and optimized by actual production are transmitted to the production equipment control system in real time through the communication interface with the production equipment. The communication interface can adopt wired or wireless mode to ensure the stability and timeliness of parameter transmission. After receiving the optimized parameters, the production equipment control system automatically adjusts the running parameters of the equipment to realize accurate control of the production process parameters. Through automatic adjustment and control, it ensures that the production process is always in the optimal state, improves the stability of production and the consistency of product quality.

[0140] The effect of the above technical solution is: by integrating process knowledge and experience, a complete production process parameter optimization knowledge base is constructed, which makes the adjustment of production process parameters more scientific and accurate, thereby improving the stability of production process and product quality. Through intelligent optimization algorithm, the production process parameters are adaptively optimized, which can automatically adjust the parameters according to the characteristics of polyester fabric production, reduce manual intervention and errors, and thus reduce the uncertainty and volatility in the production process.

[0141] By using simulation technology to simulate different parameter combinations, the best parameter combination can be predicted before actual production, reducing the number and time of production tests, thereby significantly improving production efficiency. In the optimization process, the production effect and quality indicators under different parameter combinations are predicted through simulation test, which can accurately adjust the key parameters in the production process and improve the consistency and quality stability of the product.

[0142] Through the optimization and simulation of the production process, the waste and defective products caused by improper parameter setting are reduced, the waste of raw materials and the energy consumption in the production process are reduced, thereby effectively reducing the production cost. Through real-time monitoring of the quality indicators and equipment running state in the production process, and automatically transmitting the optimized process parameters to the production equipment control system, real-time adjustment can be made in the production process, making the production process more flexible and more adaptable.

[0143] Through the feedback information of simulation and actual production test, data support is provided for the management layer, ensuring that the decision basis is more scientific and reliable, and reducing the risk brought by experience judgment. Through pre-test by simulation, the effect of different parameter combinations can be evaluated in a risk-free environment, avoiding a large number of trial and error in actual production, and reducing the cost brought by trial and error.

[0144] According to the actual production feedback, the optimization strategy can be adjusted in time, which can ensure that the process optimization process is continuously iterated and improved, and the sustainability and long-term effect of process optimization are enhanced. Through the communication interface with the production equipment, the optimized process parameters are transmitted in real time, realizing the automatic adjustment and control of the production process, improving the automation degree of the production equipment, and reducing manual intervention and operation errors.

[0145] An embodiment of the present application, the S5, comprises:

[0146] S51, in the production process of polyester fabric, a unique identification code is given to each batch of products, all related data in the production process are stored in association with the identification code, a product information database is established, and the associated data is stored in the database;

[0147] S52, a real-time quality traceability system is established, the whole life cycle information of the batch of products is queried by scanning the identification code, and historical quality data is analyzed by using data mining and machine learning technology;

[0148] S53, a quality defect early warning model is established by using a classification algorithm, the quality indicators in the production process are monitored in real time, when the quality indicators deviate from the normal range, an early warning signal is sent, and the possible quality defect type and cause are pointed out;

[0149] S54, the production management personnel take measures to adjust and improve according to the early warning information.

[0150] The working principle of the above technical solution is: in the production process of polyester fabric, a unique identification code is assigned to each batch of products; the identification code can take the form of a bar code, a two-dimensional code, or an RFID tag; all relevant data during production, such as raw material information, process parameters, equipment operation data, quality detection data, etc., are associated with the identification code and stored; a product information database is established, and the associated data is stored in the database;

[0151] A real-time quality traceability system is established, and by scanning the identification code, the full life cycle information of the batch of products can be quickly queried; data mining and machine learning techniques are used to analyze historical quality data;

[0152] A quality defect early warning model is established using classification algorithms to monitor quality indicators in real time, such as fiber strength, fabric density, dyeing rate, etc.; when the quality indicators deviate from the normal range, an early warning signal is sent out in time, and the possible types and causes of quality defects are pointed out; for example, if the strength of the fiber in the spinning process is detected to be lower than the set threshold, the early warning model will issue an alarm and suggest possible causes such as excessive temperature, insufficient pressure, etc.

[0153] Production managers can quickly take measures to adjust and improve based on the early warning information. For potential quality defects, production process parameters are adjusted in time, equipment maintenance and maintenance are strengthened, and raw materials are re-inspected. At the same time, quality defect events are recorded and analyzed, lessons are learned, and the quality management system and production process are continuously improved to avoid the expansion and repeated occurrence of quality problems.

[0154] The effect of the above technical solution is: by assigning a unique identification code to each batch of products and associating all production-related data with the identification code, an efficient product information database is established, and the traceability of product life cycle information is improved, making each batch of products more transparent in quality traceability, management, and maintenance. The real-time quality traceability system allows production managers to quickly query historical data of products, identify potential quality problems in time and take corrective measures, reducing the risk of production interruptions and customer complaints due to quality problems.

[0155] By combining data mining and machine learning techniques to analyze historical quality data, potential quality problems can be identified in advance, and a quality defect early warning model is established through classification algorithms to monitor quality indicators in real time, improving the accuracy of quality management. The real-time early warning system can issue an alarm in time and accurately point out the type and possible cause of the quality defect, helping production managers quickly take measures to adjust and improve, improving the response speed and flexibility of the production process.

[0156] Through the early warning model, timely warnings and possible causes can be provided to intervene and adjust before quality problems occur, thereby reducing the frequency of quality defects and reducing waste and defective product rates. By recording and analyzing quality defect events, not only can lessons be learned, but the quality management system and production process can also be continuously improved, thereby enhancing the continuous improvement capability of the production process and gradually improving the process level.

[0157] By timely adjusting production process parameters, strengthening equipment maintenance, re-inspecting raw materials, and other measures, it ensures that various parameters in the production process are maintained within the ideal range, enhances the controllability of the production process, and reduces the impact of external factors on product quality. Through the early warning model, problems can be discovered and corrected in a timely manner, effectively reducing the number of rework and repair, thereby saving production costs and resource waste caused by quality defects.

[0158] By establishing a transparent and efficient quality traceability system, customers can more confidently choose and use products, enhancing customer trust and satisfaction with the production enterprise, and helping to improve the market competitiveness of the brand. Real-time monitoring of key quality indicators and early warning during the production process help ensure product quality consistency and stability, reducing quality fluctuations caused by human error and equipment problems.

[0159] One embodiment of the present application, the S6, comprises:

[0160] S61, comprehensively consider the production demand dynamic prediction result, production situation evaluation report, process parameter optimization result and quality traceability and defect early warning information, integrate and analyze these multi-source information;

[0161] S62, use operational research and intelligent scheduling algorithm to intelligently allocate production resources; adopt linear programming to optimize the supply plan of raw materials, and use genetic algorithm to optimize the scheduling scheme of production equipment;

[0162] S63, establish a production collaborative optimization mechanism to strengthen information communication and coordination between various production links; through the establishment of a production collaborative management platform, real-time sharing and transmission between various production links are realized;

[0163] S64, evaluate the effect of intelligent allocation of production resources, establish an evaluation index system, collect and analyze relevant data regularly to evaluate the actual effect of intelligent allocation of production resources; according to the evaluation results, continuously improve and optimize the production resource allocation strategy and collaborative optimization mechanism.

[0164] The working principle of the above technical solution is: comprehensively considering the production demand dynamic prediction result, the production situation evaluation report, the process parameter optimization result and the quality traceability and defect early warning information, integrating and analyzing these multi-source information; determining the decision factors of production resource allocation, such as the priority and urgency of production tasks, the supply situation of raw materials, the running state of production equipment, the skill level of human resources, etc.; through the comprehensive evaluation of these factors, providing scientific basis for intelligent allocation of production resources;

[0165] Intelligent allocation of production resources is achieved by using operational research and intelligent scheduling algorithm; linear programming is used to optimize the supply plan of raw materials, ensuring timely supply and reasonable inventory of raw materials; genetic algorithm is used to optimize the scheduling scheme of production equipment, improving the utilization rate and production efficiency of equipment; at the same time, according to the characteristics of production tasks and the skill level of human resources, the post distribution of human resources is reasonably allocated to realize the optimal allocation of human resources;

[0166] Establish a production collaborative optimization mechanism to strengthen the information communication and coordination among various production links; through the establishment of a production collaborative management platform, real-time sharing and transmission of production plan, process parameter, quality detection and other information among various production links are realized;

[0167] The effect of intelligent allocation of production resources is evaluated, and an evaluation index system is established, such as production efficiency improvement rate, resource utilization rate improvement rate, production cost reduction rate, etc.; by regularly collecting and analyzing relevant data, the actual effect of intelligent allocation of production resources is evaluated; according to the evaluation result, the production resource allocation strategy and collaborative optimization mechanism are continuously improved and optimized.

[0168] The effect of the above technical solution is: through comprehensive analysis of production demand dynamic prediction, production situation evaluation report, process parameter optimization result and quality traceability and defect early warning information, the decision basis of production resource allocation is ensured to be more comprehensive, accurate and scientific, so as to improve the rationality of resource allocation and the effectiveness of production plan. Through operational research and intelligent scheduling algorithm, especially using linear programming to optimize the supply plan of raw materials, timely supply and reasonable inventory of raw materials are ensured, and the risk of production interruption and waste caused by untimely supply or inventory accumulation of raw materials is reduced.

[0169] The scheduling scheme of production equipment is optimized by using genetic algorithm, which improves the utilization rate and production efficiency of equipment, reduces the idle time of equipment and the loss of production process, and ensures the smoothness and efficiency of production process. Through reasonable allocation of post distribution of human resources, combined with the characteristics of production tasks and the skill level of employees, the optimal allocation of human resources is realized, ensuring the matching of skill requirements of each post and production tasks, improving the production efficiency and the work satisfaction of employees.

[0170] By establishing a production collaborative optimization mechanism, the information communication and coordination between various production links are strengthened, avoiding information silos and repeated labor, and reducing errors and costs caused by poor communication in the production process. Through the establishment of a production collaborative management platform, the real-time sharing and transmission of production plan, process parameters, quality detection and other information are realized, so that each link of production can quickly respond to changing demands, and the flexibility and adaptability of production planning are enhanced.

[0171] Through the optimization of intelligent scheduling algorithm, the actual situation of various resources is considered comprehensively, so that the resource utilization in the production process is more efficient, the production efficiency is improved, and the resource waste phenomenon is greatly reduced. Through intelligent allocation and optimization management of production resources, the use of production equipment, raw materials and human resources is more refined, reducing energy waste, equipment wear and tear and unnecessary labor costs in the production process, thereby reducing production costs.

[0172] Through the establishment of an evaluation mechanism for the effect of intelligent allocation of production resources, the key indicators of production efficiency, resource utilization rate and production cost are evaluated regularly to ensure continuous optimization and adjustment of production scheduling strategies, and the continuous improvement capability of the production process is enhanced. Through real-time sharing and transmission of production information, production management becomes more transparent, and relevant decisions and operations become more controllable, which helps to improve the control ability of management personnel on the production process and the ability to identify potential problems in advance.

[0173] One embodiment of the present application is a polyester fabric production whole process monitoring and management system, comprising:

[0174] One or more processors;

[0175] Memory for storing one or more programs,

[0176] Wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of the above.

[0177] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A method for monitoring and managing the entire production process of polyester fabric, characterized in that, The method includes: S1. Deploy various types of sensors at key nodes of the polyester fabric production line; collect various physical parameters and environmental information related to polyester fabric production in real time; at the same time, integrate the operation status monitoring system of the production equipment to obtain equipment data; and initially integrate the data from different sensors and equipment monitoring systems to build a holographic perception architecture for polyester fabric production. S2. Deeply fuse the multi-source heterogeneous data obtained from the holographic perception architecture of polyester fabric production; construct a data association model; based on this model, and in conjunction with the process flow and quality standards of polyester fabric production, conduct a comprehensive assessment of the production status; generate a production status assessment report through the production status assessment. S3. Collect multi-dimensional data on historical polyester fabrics; use machine learning algorithms to train and learn from the multi-dimensional data to build a dynamic production demand prediction model; compare and analyze the dynamic production demand prediction results with the current production plan to obtain potential production gaps or surpluses. S4. Based on the dynamic forecast results of production demand and the production status assessment report, combined with the process knowledge of polyester fabric production, intelligent optimization algorithms are used to adaptively optimize the production process parameters. During the optimization process, the quality indicators and equipment operating status of the production process are monitored in real time, and the optimization strategy is adjusted in a timely manner based on the feedback information. S5. During the production of polyester fabric, each batch of products is assigned a unique identification code, and all relevant data in the production process is linked and stored with the identification code; a real-time quality traceability system is established, which allows querying the entire life cycle information of the batch of products by scanning the identification code; at the same time, data mining and machine learning technologies are used to analyze historical quality data and establish a quality defect early warning model. S6. Taking into account the dynamic forecast results of production demand, production status assessment reports, process parameter optimization results, and quality traceability and defect early warning information, the production resources are intelligently allocated using operations research and intelligent scheduling algorithms.

2. The method for monitoring and managing the entire production process of polyester fabric according to claim 1, characterized in that, S1 includes: S11. Conduct surveys and analyses at each key node of the polyester fabric production line; and determine the types of sensors to be deployed based on the production characteristics and monitoring requirements of each node. S12. Install and debug the sensors according to the selected sensor type and deployment location; at the same time, configure the data acquisition equipment to convert the analog signals collected by the sensors into digital signals, and perform preliminary storage and processing. S13. An integrated production equipment operation status monitoring system acquires key equipment data through equipment monitoring technology; and uses industrial Ethernet to seamlessly connect the equipment monitoring system with the data acquisition network to synchronously transmit equipment operation data and sensor-collected data. S14. The data collected from different sensor and equipment monitoring systems will be initially integrated. Data fusion technology will be used to associate and match data from multiple data sources to eliminate data redundancy and conflicts. A holographic perception architecture for polyester fabric production will be constructed to connect the various data sources into an organic whole.

3. The method for monitoring and managing the entire production process of polyester fabric according to claim 2, characterized in that, S14 includes: S141. Preprocess the collected raw data, and use data mining and correlation analysis techniques to deeply analyze the intrinsic relationships between different data sources; establish a data correlation model based on the logical relationship of the production process and the collaborative working principle between equipment. S142. A data fusion algorithm is used to perform in-depth analysis on the matched data; at the same time, conflicting data from different data sources are processed according to preset rules and priorities. S143. Based on the data processed above, construct a data fusion model for polyester fabric production; fuse and optimize the data. S144. Using the fused data stream as the core, construct a holographic perception architecture for polyester fabric production; connect it into an organic whole through data interfaces and network protocols to achieve real-time data sharing and interaction.

4. The method for monitoring and managing the entire production process of polyester fabric according to claim 3, characterized in that, S142 includes: After matching the data, various statistical indicators are calculated, and highly correlated data is identified by analyzing the statistical correlation between different data sequences. Combined with a professional knowledge base in the production field, semantic analysis is performed on the data to determine whether there is semantically repetitive data. For time series data, analyze its changing trends and periodic characteristics; if multiple data series exhibit similar changing trends and periodic patterns within the same time interval, and this similarity exceeds a reasonable error range, they are identified as redundant data; for statistically correlated or semantically repetitive redundant data, data aggregation methods are used for processing; based on the data type and characteristics, an aggregation function is selected, and data sampling methods are used to eliminate redundant data in time series; the sampling interval is determined based on the data's frequency of change and importance. By using established data association models and prediction algorithms, redundant data is predicted and replaced; by analyzing the changing patterns of historical and current data, a prediction model is established, and the predicted values ​​are used to replace some of the redundant data. After matching the data, check its consistency across different data sources; identify conflicting data by comparing data attributes; formulate conflict judgment rules based on business rules and process requirements for polyester fabric production; establish normal and abnormal data patterns using historical data; and identify data that does not conform to the normal pattern by comparing the current data with the historical data patterns to determine whether it is conflicting data. The data sources of conflicting data are evaluated to determine the timeliness priority of the data; the business importance priority of the data is determined according to its importance in the polyester fabric production business; machine learning algorithms are used to fuse the conflicting data; a large amount of historical conflicting data and corresponding correct processing results are collected to construct a training dataset. The training dataset is trained using a neural network, enabling the model to learn how to handle conflict data. When new conflict data appears, the data is input into the trained model to obtain the fused result.

5. The method for monitoring and managing the entire production process of polyester fabric according to claim 1, characterized in that, The S2 includes: S21. Perform deep cleaning on the multi-source heterogeneous data obtained from the holographic perception architecture of polyester fabric production; identify and remove noisy data and outliers through machine learning algorithms; S22. Use data standardization methods to unify data of different dimensions to the same scale; eliminate the differences in dimensions between different data through data standardization processing; S23. Use data association analysis algorithms to mine potential relationships between data, and use association rule mining to find the inherent connections between data in different production stages; construct a data association model. S24. Based on the constructed data association model, combined with the process flow and quality standards of polyester fabric production, the production status is evaluated; through preset evaluation indicators and thresholds, various data in the production process are monitored and analyzed in real time. S25. When the data exceeds the normal range, issue an early warning signal; generate a production status assessment report based on the production status assessment results.

6. The method for monitoring and managing the entire production process of polyester fabric according to claim 1, characterized in that, The S3 includes: S31. Obtain multi-dimensional data of historical polyester fabrics, and use machine learning algorithms to train and learn from the multi-dimensional data; S32. Divide historical data into training and testing sets, continuously adjust the parameters and structure of the model, construct a dynamic prediction model for production demand, automatically adjust prediction parameters based on real-time input data, and predict the production demand of polyester fabric in the future. S33. Compare and analyze the dynamic forecast results of production demand with the current production plan; calculate the difference between the forecast demand and the production plan by establishing a comparative analysis model, and adjust the production plan according to the comparative analysis results.

7. The method for monitoring and managing the entire production process of polyester fabric according to claim 1, characterized in that, The S4 includes: S41. Obtain process knowledge of polyester fabric production through web crawlers, and acquire experience and skills in the production process through expert knowledge bases; integrate process knowledge, experience and skills to form a complete knowledge base for optimizing production process parameters; S42. Based on the characteristics of polyester fabric production and the goal of optimizing process parameters, select an intelligent optimization algorithm and adaptively optimize the production process parameters. S43. During the optimization process, simulation technology is used to conduct simulation experiments on different parameter combinations; by establishing a mathematical model of the production process, computer simulation software is used to conduct simulations to predict the production effect and quality indicators under different parameter combinations. S44. Based on the simulation results, select the optimal parameter combination for actual production testing; during the actual production process, monitor the quality indicators and equipment operating status in real time and record relevant data. S45. Adjust and optimize strategies in a timely manner based on feedback from actual production tests; if there are differences between actual production results and simulation results, analyze the reasons and further optimize the parameter combinations. S46. Transmit the optimized process parameters to the production equipment control system for automatic adjustment and control of the process parameters; and send the optimized parameters to the equipment controller in real time through the communication interface with the production equipment.

8. The method for monitoring and managing the entire production process of polyester fabric according to claim 1, characterized in that, The S5 includes: S51. In the production process of polyester fabric, assign a unique identification code to each batch of products; associate and store all relevant data in the production process with the identification code; establish a product information database and store the associated data in the database; S52. Establish a real-time quality traceability system to query the entire lifecycle information of a batch of products by scanning the identification code; and use data mining and machine learning techniques to analyze historical quality data. S53. A quality defect early warning model is established using a classification algorithm to monitor quality indicators in the production process in real time. When a quality indicator is detected to deviate from the normal range, an early warning signal is issued, and the possible types of quality defects and their causes are indicated. S54. Production management personnel shall take measures to adjust and improve based on the early warning information.

9. The method for monitoring and managing the entire production process of polyester fabric according to claim 1, characterized in that, The S6 includes: S61. Taking into account the dynamic forecast results of production demand, the production status assessment report, the optimization results of process parameters, and the quality traceability and defect early warning information, integrate and analyze these multi-source information. S62. Apply operations research and intelligent scheduling algorithms to intelligently allocate production resources; use linear programming to optimize the raw material supply plan and use genetic algorithms to optimize the scheduling scheme of production equipment. S63. Establish a production collaboration optimization mechanism to strengthen information communication and coordination among various production links; and establish a production collaboration management platform for real-time sharing and transmission among various production links. S64. Evaluate the effectiveness of intelligent allocation of production resources, establish an evaluation index system, and evaluate the actual effectiveness of intelligent allocation of production resources by regularly collecting and analyzing relevant data; based on the evaluation results, continuously improve and optimize the production resource allocation strategy and collaborative optimization mechanism.

10. A monitoring and management system for the entire production process of polyester fabric, comprising: One or more processors; Memory, used to store one or more programs. Wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1 to 9.

Images