Abnormity recognition system for preparing flame-retardant polyester fibers

Abstract

The invention discloses an anomaly recognition system for flame-retardant polyester fiber preparation, and relates to the technical field of polyester fiber manufacturing control, and the system comprises a superior-subordinate association establishment module which is used for extracting a key process for flame-retardant polyester fiber preparation and establishing a superior-subordinate association relationship; the stage exception identification node establishing module is used for establishing a plurality of stage exception identification nodes; the stage abnormity identification module is used for carrying out abnormity identification on the key process; the abnormal process backstepping module is used for carrying out abnormal process backstepping; and the product performance compensation module is used for performing performance compensation of the current product and production optimization of subsequent products. The technical problems that in the prior art, in the flame-retardant polyester fiber preparation process, due to the fact that procedure association is complex, rapid positioning is difficult to achieve, performance influences caused by abnormity are irreversible, and the product quality is influenced are solved, and the technical effect that the product quality is improved through accurate tracing of the abnormal procedure and dynamic compensation of the product performance is achieved.

Description

Technical Field

[0001] The invention relates to the technical field of polyester fiber manufacturing control, and in particular to an abnormality identification system for preparing flame-retardant polyester fibers. Background Art

[0002] With the continuous improvement of the level of automation in industrial production, the quality requirements for textiles are becoming increasingly stringent. Flame-retardant polyester fiber, as a high-performance textile material, is widely used in clothing, home textiles and industrial fields. Its preparation process is complex and involves multiple key processes. The process parameters of each process have an important impact on the performance, strength and appearance quality of the final product. However, in the actual production process, due to equipment fluctuations, raw material differences and changes in environmental conditions, some processes are prone to abnormalities, which affects the fiber performance and even leads to batch product failures.

[0003] At present, most abnormality monitoring in fiber preparation processes relies on manual or partial automation means, and it is impossible to monitor each process in the production process in real time, resulting in the inability to quickly locate and respond to abnormalities after they occur, and it is difficult to accurately identify the root cause of the abnormality. Especially in production processes with strong correlations among multiple processes, it is easy to misjudge the source of the abnormality. Even if the abnormality is found, there is a lack of systematic compensation measures, and it is impossible to adjust the parameters of subsequent processes in time to compensate for performance losses, which ultimately leads to product performance failing to meet the standards. Summary of the invention

[0004] The present application provides an abnormality identification system for the preparation of flame-retardant polyester fibers, which is used to solve the technical problems in the prior art that abnormalities in the preparation of flame-retardant polyester fibers are difficult to quickly locate due to complex process associations, and the performance impact caused by the abnormalities is irreversible, affecting product quality.

[0005] The present application provides an abnormality identification system for the preparation of flame-retardant polyester fiber, the system comprising: an upper-lower level association establishment module, used to extract key processes for the preparation of flame-retardant polyester fiber, and establish upper-lower level association relationships for each key process according to the influence relationship between the preceding process and the subsequent process; a stage abnormality identification node establishment module, used to establish multiple stage abnormality identification nodes after multiple key processes based on the upper-lower level association relationship, and the multiple stage abnormality identification nodes are communicatively connected with a central processing unit; a stage abnormality identification module, used to respectively enable the multiple stage abnormality identification nodes, perform abnormality identification of key processes in turn, obtain stage abnormality identification results and send them to the central processing unit; an abnormal process reverse deduction module, used to perform abnormal process reverse deduction by the central processing unit according to the stage abnormality identification results and the upper-lower level association relationship, and obtain the stage abnormal process; a product performance compensation module, used to adjust the subsequent process of the stage abnormal process according to the stage abnormality identification results, perform performance compensation for the current product, and optimize the stage abnormal process to perform production control of subsequent products.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The abnormality identification system for flame-retardant polyester fiber preparation provided in the present application relates to the field of polyester fiber manufacturing control technology. By extracting the key processes of flame-retardant polyester fiber preparation, establishing upper and lower level association relationships and stage abnormality identification nodes, real-time abnormality identification and traceability are realized, and the subsequent process parameters are dynamically adjusted according to the abnormality identification results to compensate for the performance. The abnormal process parameters are optimized according to the traceability results to form a closed-loop control. The technical problems in the prior art that the abnormalities in the flame-retardant polyester fiber preparation process are difficult to locate quickly due to the complex process associations, and the performance impact caused by the abnormalities is irreversible, affecting the product quality, are solved. The system realizes the accurate traceability of abnormal processes and dynamic compensation of product performance through upper and lower level process association analysis, and improves the stability of the production process and the consistency of product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 A schematic diagram of the structure of an abnormality identification system for preparing flame-retardant polyester fibers provided in an embodiment of the present application;

[0010] Figure 2A schematic flow chart of establishing a superior-subordinate relationship between each key process in a superior-subordinate relationship establishment module of an abnormality identification system for preparing flame-retardant polyester fiber provided in an embodiment of the present application.

[0011] Explanation of reference numerals: upper-lower level association establishing module 10 , stage abnormality identification node establishing module 20 , stage abnormality identification module 30 , abnormal process reverse inference module 40 , product performance compensation module 50 . DETAILED DESCRIPTION

[0012] The present application provides an abnormality identification system for the preparation of flame-retardant polyester fibers, which is used to solve the technical problems in the prior art that abnormalities in the preparation of flame-retardant polyester fibers are difficult to quickly locate due to complex process associations, and the performance impact caused by the abnormalities is irreversible, affecting product quality.

[0013] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0014] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices.

[0015] Embodiment 1, as Figure 1 As shown, the present application provides an abnormality identification system for flame retardant polyester fiber preparation, the system comprising:

[0016] The upper-lower level association establishing module 10 is used to extract the key processes of preparing the flame-retardant polyester fiber, and establish the upper-lower level association relationship between each key process according to the influence of the preceding process on the succeeding process.

[0017] Further, such as Figure 2 As shown, the upper and lower level association establishing module 10 is also used to perform the following steps:

[0018] P11: Extract the key processes for the preparation of flame-retardant polyester fibers, and establish a process influence matrix of each preceding process on the subsequent process in the key process through qualitative analysis and quantitative modeling; P12: Based on the process influence matrix, establish a superior-subordinate relationship between each key process, and the superior-subordinate relationship includes a single association relationship and a multi-association relationship, wherein the multi-association relationship includes multiple association branches, and the association branches correspond to two associated processes; P13: Sort and screen the multiple association branches, and assign contribution coefficients to the multiple association branches according to the sorting results.

[0019] It should be understood that the upper and lower level association establishment module 10 of the present application is responsible for extracting key processes and establishing upper and lower level associations between processes. The core task of this module is to identify the relationship between each process and form an influence matrix by analyzing the impact of the previous process on the subsequent process.

[0020] First, it is necessary to extract the key processes in the preparation of flame-retardant polyester fibers. Key processes refer to process steps that are critical to the performance of the final product, and the execution status of these processes will directly affect product quality and production efficiency. By conducting qualitative analysis (for example, expert experience or historical data analysis) and quantitative modeling (using statistical methods, data mining and other technologies) of each key process, the mutual influence between each process can be systematically identified. At this time, based on the analysis of the cause-effect relationship between each process, a process impact matrix is ​​generated. The process impact matrix is ​​a multidimensional array, and each matrix element represents the degree of influence of the preceding process (i.e., the input process) on the subsequent process (i.e., the output process). This matrix provides basic data support for the subsequent construction of superior-subordinate association relationships.

[0021] After extracting and analyzing the process impact matrix, the next step is to establish the hierarchical relationship between each key process based on the data in the matrix. The hierarchical relationship indicates how a process affects the results of the previous process and further affects the execution of the subsequent process. These relationships can be divided into two categories: single relationship and multi-level relationship.

[0022] Among them, a single correlation refers to the direct impact between the preceding process and the subsequent process. For example, the processing temperature of a polyester fiber directly determines the strength of the fiber, forming a direct single correlation. A multi-correlation relationship involves the combined impact of multiple preceding processes on the subsequent process. For example, the stretching process of polyester fiber is not only affected by temperature, but also by the combined effect of multiple process parameters such as speed and pressure. In this case, multiple processes are connected to each other through multiple correlation branches, thus forming a multi-correlation relationship.

[0023] After establishing the superior-subordinate association relationship, multiple associated branches need to be sorted and screened. Since there may be multiple influencing factors in the association between processes, some associations may have a greater impact on subsequent processes, while others may have a smaller impact. Therefore, in this step, sort each associated branch according to the degree of influence on the subsequent process, and screen out the most influential associated branches. Then, assign a contribution coefficient to each associated branch based on the sorting result. The contribution coefficient is an important indicator to measure the impact of each process or associated branch on the subsequent process, and is usually determined through quantitative analysis (such as regression analysis, variance analysis, etc.). These coefficients play an important role in the subsequent abnormality identification and reverse deduction process, helping the system identify key processes that may cause abnormalities and provide a basis for abnormal compensation.

[0024] Through the above steps, the upper and lower level association establishment module can effectively provide the entire system with a deep understanding of the relationship between processes, thereby providing a scientific basis for subsequent abnormality identification, reverse analysis, performance compensation, etc. This module not only relies on traditional engineering experience and knowledge, but also makes full use of modern data analysis technology, thereby improving the accuracy and robustness of the system.

[0025] Furthermore, a process influence matrix of each preceding process on the subsequent process in the key process is established. Step P11 of the embodiment of the present application further includes:

[0026] P11-1: Obtain the key production parameters and output key parameters of each key process, as well as the sequence of each key process; P11-2: According to the sequence, traverse and extract the output key parameters of the preceding process and the key production parameters of the succeeding process, conduct qualitative analysis, and establish the impact relationship between the output key parameters of each process and its subsequent processes; P11-3: Through correlation analysis, quantify the impact relationship between the output key parameters of each process and its subsequent processes, and establish the process impact matrix.

[0027] Optionally, by analyzing the impact of each key process on subsequent processes, a detailed process impact matrix is ​​constructed to provide a scientific basis for the anomaly identification system.

[0028] First, it is necessary to extract all the key processes from the production process of flame-retardant polyester fiber. Each process has its key production parameters and output key parameters. The key production parameters refer to the variables controlled by each process in the production process, such as temperature, pressure, speed, etc. These parameters directly affect the execution of the process and the performance of the product. The output key parameters refer to the key indicators of the product or process status obtained after the process is completed, such as fiber strength, fiber uniformity, coating thickness, etc. These parameters reflect the effect of process execution.

[0029] The execution order between different processes is crucial to the entire production process. This order determines how the results of the previous process affect the input of the subsequent process, which in turn affects the quality and performance of the final product. At this stage, by sorting out the relevant parameters of each process and determining the execution order of the processes, the foundation for subsequent analysis can be laid.

[0030] Next, according to the execution order of each process, extract the output key parameters of the preceding process and the production key parameters of the subsequent process one by one. Through qualitative analysis, that is, based on expert experience, process rules and historical data, determine whether the output parameters of the preceding process will directly or indirectly affect the production parameters of the subsequent process. The goal of this stage is to establish the influence relationship between the preceding and following processes. For example, the output of a preceding process (such as temperature, concentration) may directly affect the execution effect of the subsequent process (such as stretching process, coating process). In this process, the key is to reasonably match the output of each process with the input of the subsequent process and to preliminarily establish the process influence relationship.

[0031] Finally, correlation analysis is performed, which is the core of quantitative analysis. Correlation analysis uses mathematical and statistical methods (such as Pearson correlation coefficient, regression analysis, variance analysis, etc.) to quantitatively evaluate the degree of correlation between the output key parameters of each process and the key production parameters of its subsequent processes. Through this analysis, it can be determined which output parameters have a strong impact on the subsequent processes and which parameters have a weaker impact. Correlation analysis is a mathematical method used to reveal linear or nonlinear relationships between different variables. In this application, correlation analysis helps identify process links that may have a significant impact on product performance by measuring the trend of changes in parameters between processes.

[0032] Based on the results of correlation analysis, a process impact matrix is ​​established. Each element in the matrix represents the strength of association between the output parameter of one process and the input parameter of another process. The process impact matrix can not only quantify the impact of each process on the subsequent processes, but also provide reliable data support for the establishment of subsequent superior-subordinate association relationships.

[0033] The stage abnormality identification node establishment module 20 is used to establish multiple stage abnormality identification nodes after multiple key processes based on the upper-lower level association relationship, and the multiple stage abnormality identification nodes are communicatively connected with the central processing unit.

[0034] Optionally, the stage abnormality identification node establishment module 20 of the present application has the main function of establishing multiple stage abnormality identification nodes after multiple key processes based on the aforementioned superior-subordinate association relationship, and communicating with the central processing unit. The design purpose of this module is to set abnormality identification points at each stage of the key process in the production process, to detect potential abnormalities in a timely manner, so as to take measures as soon as possible to prevent the abnormalities from further affecting product quality or production efficiency.

[0035] Specifically, first, according to the upper-lower level association relationship constructed above, a corresponding stage abnormality identification node is set for each key process. The function of these nodes is to identify whether there are abnormal situations by real-time monitoring and analyzing the operation data of the process. Each stage abnormality identification node is designed according to the dependency relationship between processes to ensure that potential problems can be discovered in time at each stage of the key process.

[0036] The superior-subordinate relationship plays a key role in this stage. Through the established process impact matrix, the system can understand how the production process of each key process affects each other and which parameters have a greater impact on the subsequent processes. Based on this analysis, corresponding abnormality identification nodes are set in the subsequent stages of each key process. These nodes will judge whether the process is running normally according to the fluctuations of the actual process data and provide timely feedback. For example, in the preparation process of polyester fiber, the temperature control of a certain process has a greater impact on the subsequent stretching process. Therefore, an abnormality identification node will be established after this process to monitor in real time whether the temperature exceeds the predetermined range, so as to ensure that the stretching process can proceed as expected.

[0037] After the abnormal identification nodes at each stage are established, they are connected to the central processing unit for communication. The central processing unit is the brain of the entire abnormal identification system, responsible for collecting information from nodes at each stage and performing centralized processing, analysis and decision-making. Each abnormal identification node can feed back the monitoring results of each stage (such as abnormal parameters, exceeding standard information, etc.) to the central processing unit through real-time data transmission with the central processing unit for rapid response.

[0038] During the communication connection process, the system will use a real-time data transmission protocol to ensure that the data transmitted by each node can be delivered to the central processing unit in a timely manner and processed in the shortest possible time. Through this connection method, the central processing unit can accurately grasp the operating status of each key process and issue an alarm signal or adjust the production control strategy in a timely manner based on the data fed back by the node. Ensure that the entire production process can identify and correct abnormalities at the earliest stage, minimizing production losses caused by abnormalities.

[0039] The stage abnormality identification module 30 is used to respectively enable the multiple stage abnormality identification nodes, perform abnormality identification of key processes in turn, obtain stage abnormality identification results and send them to the central processing unit.

[0040] Furthermore, the stage abnormality identification module 30 is also used to perform the following steps:

[0041] P31: Based on the multiple stage abnormality identification nodes, locate the first-order stage abnormality identification node, perform abnormality identification of the first key process, and obtain the abnormality identification result of the first process; P32: Determine whether there is an abnormality in the abnormality identification result of the first process; P33: If so, send the abnormality identification result of the first process to the central processing unit for abnormality processing; P34: If not, postpone to the second stage abnormality identification node to perform abnormality identification and abnormality judgment of the second key process, and so on, traverse and enable the multiple stage abnormality identification nodes, and perform abnormality identification of key processes in turn.

[0042] It should be understood that the core task of the stage abnormality identification module 30 of the present application is to enable and perform abnormality identification of key processes in sequence through multiple stage abnormality identification nodes, and send the abnormality identification results to the central processing unit for further processing and response. The purpose of this module is to ensure that abnormalities can be discovered in time and countermeasures can be taken by gradually checking the operating status of each key process, thereby ensuring the smoothness of the production process and the stability of product quality.

[0043] First, the first-order stage abnormality identification node will be located according to the pre-set rules. This node is usually the first key process in the entire production process and is responsible for starting the first step of abnormality identification. The node performs abnormality detection by receiving real-time data, such as production parameters, sensor monitoring data, etc. For example, in the preparation process of flame-retardant polyester fiber, the first stage may be the spinning process, and the key parameters of this process such as temperature, pressure and speed need to be monitored in real time. The function of the first-order stage abnormality identification node is to analyze whether these parameters are within the standard range. Once an abnormality is found, it can trigger subsequent abnormality identification and processing.

[0044] When the abnormal identification node in the first sequence stage completes the abnormal identification of the first process, it analyzes and judges the identification results of the process to check whether the abnormal identification results of the first process meet the preset abnormal judgment criteria. The criteria for abnormal judgment can be based on multiple factors, such as: deviation threshold, judging whether the process parameters exceed the specified deviation range. Historical data comparison, compare the current process parameters with historical normal data to see if there are significant differences. For example, if the temperature exceeds the set range or the pressure fluctuates too much in the spinning process, these may be judged as abnormalities and trigger subsequent processing.

[0045] If it is determined that the abnormal identification result of the first process is abnormal, the abnormal processing of the first process can be directly performed, and the abnormal identification result can be sent to the central processing unit, and the central processing unit will perform the abnormal processing. For example, if the temperature is found to be too high in the spinning process, the system will automatically adjust the power of the heater, or notify the operator through the alarm system to check the equipment condition.

[0046] If the abnormality identification result of the first process is judged to be normal, it will be postponed to the second stage abnormality identification node to perform abnormality identification and judgment of the second key process. This process will be carried out in sequence, traversing and enabling all stage abnormality identification nodes to perform abnormality identification of each key process. Each identification will determine whether there is an abnormality in the current process. If there is no abnormality, it will be postponed to the next process; if there is an abnormality, the abnormality identification result will be immediately passed to the central processing unit for processing. This step-by-step identification method can effectively avoid the lag problem in the production process, thereby improving production efficiency and ensuring product quality.

[0047] The abnormal process reverse deduction module 40 is used for the central processing unit to reverse the abnormal process according to the stage abnormality identification result and the upper and lower level association relationship to obtain the stage abnormal process.

[0048] Furthermore, the abnormal process reverse inference module 40 is further configured to perform the following steps:

[0049] P41: Match the corresponding stage superior-subordinate relationship according to the key process code corresponding to the stage abnormality identification result; P42: Extract the stage abnormality index according to the stage abnormality identification result; P43: Based on the superior-subordinate relationship, trace the stage abnormality index to determine multiple related branches; P44: Perform an abnormality score according to the contribution coefficients of the multiple related branches, and select the previous process with high contribution as the stage abnormal process according to the abnormality score result. The stage abnormal process can be one or more.

[0050] Optionally, the abnormal process reverse analysis module 40 of the present application is responsible for realizing the reverse analysis of the abnormal process by combining the stage abnormality identification results and the upper and lower level association relationship to determine which processes may cause the current abnormal phenomenon. Through detailed tracing analysis, this module can not only find out the source of the abnormality, but also select the key process that is most likely to cause the abnormality according to the contribution coefficient of each associated branch, providing a basis for subsequent troubleshooting and repair.

[0051] First, obtain the key process code in the stage anomaly identification result. Each process has a unique code in the production process to facilitate system tracking and management. Based on the code, the system will search for the corresponding associated data in the pre-built superior-subordinate relationship. Specifically, the system will find all the preceding and subsequent processes related to the current process, as well as the mutual influence relationship between these processes. For example, if the current anomaly is detected in the third process, the system will find the superior-subordinate relationship of the third process based on the code, and further determine whether the second and fourth processes may affect the anomaly of the third process.

[0052] Furthermore, according to the stage abnormality identification results, the stage abnormality indicators, i.e., the key performance indicators (KPIs) related to the current process abnormality, are extracted. These indicators are usually obtained by collecting and analyzing the real-time data of the process. For example, abnormal values ​​of parameters such as temperature, pressure, flow, and speed may be involved. By extracting these abnormal indicators, the system can more accurately understand the nature and extent of the abnormality, and provide key data support for subsequent reverse analysis. For example, if the pressure of the third process is too high, the pressure indicator will be extracted as an abnormal indicator.

[0053] Next, the abnormal indicators of the stage are traced and analyzed by using the upper and lower level association relationship. Specifically, the system will analyze the relationship between the abnormal indicators and the previous processes, and find out all the previous processes that may affect the current process. Each previous process may affect the operation of the subsequent process in some way, forming multiple related branches. For example, in the production process, the abnormality of the third process may be caused by improper temperature control in the previous second process, or it may be affected by the quality problems of raw materials in the first process. Based on these factors, the system will determine multiple possible related branches, that is, the chain of previous processes that affect the current abnormality.

[0054] Finally, the anomaly score is performed based on the contribution coefficients of the multiple associated branches. The calculation of the contribution coefficient can be based on multiple factors, such as the degree of direct influence between processes: if the parameters of process 1 directly affect the anomaly of process 3, the contribution coefficient of process 1 is higher. Parameter sensitivity: Some production parameters (such as temperature or pressure) may have a more significant impact on anomalies than other parameters. Historical data comparison: The system will determine which processes are most often associated with the current anomaly in past anomaly cases based on the analysis of historical data.

[0055] Through comprehensive analysis of these factors, the system assigns a contribution coefficient to each associated branch and scores each process based on these coefficients. Ultimately, the system will select those preceding processes with higher contributions as stage abnormality processes, so the stage abnormality processes can be one or more. These processes may be the root cause of the current abnormality. For example, if the abnormality contributions of process 1 and process 2 to process 3 are 0.8 and 0.6 respectively, and process 1 contributes more to the abnormality, the system will mark process 1 as a stage abnormality process as a possible root cause.

[0056] By scoring the anomalies of multiple related branches, the system can select the most likely preceding process as the source of the anomaly, providing an important basis for troubleshooting during the production process. This process not only improves the efficiency of production fault diagnosis, but also provides data support for subsequent optimization, ensuring the stability of the production process and the controllability of product quality.

[0057] Furthermore, anomaly scoring is performed according to the contribution coefficients of the multiple associated branches. Step P44 of the embodiment of the present application further includes:

[0058] P44-1: Score the abnormal processes corresponding to the multiple associated branches to determine the previous processes with high contribution. The scoring formula is as follows: S i =α·C i +β·D i ; Among them, S i is the abnormality score of process i, C i is the contribution coefficient of process i to subsequent abnormalities, D i is the coefficient of the degree to which the current parameters of process i deviate from the normal value, and α and β are weight factors.

[0059] In a possible embodiment of the present application, a formula is used to quantitatively score the abnormal processes corresponding to multiple associated branches to determine the preceding process with the highest contribution to the subsequent abnormality. This scoring process not only relies on the correlation between processes, but also combines the deviation degree and weight factor of the current process parameters to ensure the scientificity and accuracy of the scoring.

[0060] The scoring formula is as follows: S i =α·C i +β·D i ; Among them, S i is the abnormality score of process i, which is an indicator for comprehensively evaluating the impact of process i on subsequent abnormalities. The higher the score, the more significant the impact of process i on the current abnormality. iIt is the contribution coefficient of process i to subsequent abnormalities, and is a quantitative indicator to measure the impact of process i on subsequent abnormalities. It is usually obtained through the process impact matrix in the upper and lower level association relationship. Processes with higher contribution coefficients are usually the focus of abnormality tracing. i It is the coefficient of the degree of deviation of the current parameter of process i from the normal value, reflecting whether the process parameter deviates from the standard range. The deviation degree between the process parameter and the normal value is calculated through the parameter data monitored in real time. The greater the deviation, the higher the possibility of abnormality in the process. α and β are weight factors used to adjust the relative importance of the contribution coefficient and the degree of parameter deviation in the score. Their values ​​can be adjusted according to the specific production process or actual needs. For example, for processes that are highly dependent on process influence relationships, α may take a higher value; while for processes that are sensitive to real-time parameter fluctuations, β may have a greater weight.

[0061] In actual application, the system first obtains the contribution coefficient C of each preceding process through the upper and lower level association relationship established in the early stage. i For example, in the preparation of flame-retardant polyester fiber, if the temperature fluctuation of a process has a greater impact on the coating uniformity of the subsequent process, the contribution coefficient of this process is higher. At the same time, the operating parameters of the process are monitored in real time, and the degree of deviation from the normal value is calculated to obtain the degree coefficient D i For example, if the normal range of the process temperature setting is 180-200°C, and the current temperature is 210°C, then the deviation coefficient D i The severity of the out-of-range condition will be reflected. Finally, the contribution coefficient and parameter deviation coefficient are multiplied by their corresponding weight factors α and β respectively, and the weighted sum is calculated to obtain the abnormal score S of each process. i The processes with high scores are the key targets for abnormality investigation. This scoring process provides a scientific basis for abnormality tracing and can accurately locate abnormal processes with high contribution, thereby effectively improving the efficiency of troubleshooting and ensuring the stability of the production process.

[0062] The product performance compensation module 50 is used to adjust the subsequent processes of the stage abnormal process according to the stage abnormality identification result to compensate for the performance of the current product, and at the same time, optimize the stage abnormal process to control the production of subsequent products.

[0063] Specifically, the product performance compensation module 50 of the present application mainly adjusts the key parameters of the subsequent process for the abnormal process in the stage abnormality identification result to achieve dynamic compensation for the current product performance. At the same time, it also improves the subsequent production process by optimizing the abnormal process, reduces the recurrence of the same abnormality, and thus improves the overall quality and efficiency of production.

[0064] Furthermore, the product performance compensation module 50 is further configured to perform the following steps:

[0065] P51: According to the abnormality identification result of the stage, the abnormal parameter type and the corresponding abnormal parameter deviation value of the abnormal process of the stage are extracted; P52: Based on the abnormal parameter type and the corresponding abnormal parameter deviation value, product performance impact analysis is performed to obtain a performance impact index; P53: According to the performance impact index, performance compensation parameter association is performed, and relevant parameters of the subsequent process are extracted as performance compensation associated parameters; P54: A performance compensation model is established, and performance compensation of the current product is performed according to the performance compensation model and the performance compensation associated parameters.

[0066] Wherein, the performance compensation model includes a performance balance formula:

[0067] Among them, P c is the performance compensation value of the current product, P a is the direct effect of abnormal process on performance, Δp i The compensation value for performance adjustment of the subsequent process parameters, w i is the compensation weight of the subsequent process for this performance.

[0068] It should be understood that by analyzing the results of stage abnormality identification and applying the performance compensation model, the key parameters of the subsequent process can be dynamically adjusted to compensate for the product performance deviation caused by the abnormal process and ensure that the product performance meets the target requirements.

[0069] Next, based on the extracted abnormal parameter types and deviation values, the specific impact of the abnormality on product performance is calculated to obtain the performance impact index. The performance impact index is used to quantify the negative impact of abnormal processes on target performance indicators (such as strength, heat resistance, etc.).

[0070] First, based on the results of stage abnormality identification, the abnormal parameter types (such as temperature, pressure, speed, etc.) of the abnormal process and their corresponding parameter deviation values ​​are extracted. The abnormal parameter type indicates the specific process parameter that causes the product performance deviation, while the parameter deviation value quantifies the degree of deviation of the current parameter from the normal value. For example, if the temperature deviation value of a process is +15°C, it means that the current temperature is significantly beyond the set range, which may have a significant impact on subsequent performance.

[0071] According to the extracted abnormal parameter type and its deviation value, the module uses the performance impact analysis model to evaluate the specific impact of the abnormality on the target performance index and calculates the performance impact index. The performance impact index is a quantitative indicator used to characterize the degree of negative impact of abnormal parameters on the current product performance (such as strength, heat resistance, etc.). For example, temperature deviation may cause the tensile strength of the fiber to decrease by 10%. At this time, the performance impact index will reflect the proportion or absolute value of this strength decrease.

[0072] After clarifying the performance impact index, the module determines the performance compensation associated parameters in the subsequent process based on the relationship between the upper and lower processes. These parameters are key variables that have a significant regulatory effect on performance in the subsequent process, such as coating thickness, processing speed, etc. Through parameter adjustment, the subsequent process can compensate for the negative impact of abnormal processes on product performance. For example, if the polyester fiber is insufficient in strength due to abnormal stretching, the strength of the final product can be improved by adjusting the thickness of the coating process.

[0073] Furthermore, based on the above data, a performance compensation model is established, and the performance of the current product is dynamically compensated using the model. The performance compensation model calculates the compensation result using the performance balance formula:

[0074] Among them, P c is the performance compensation value of the current product, indicating the performance level of the product after adjustment. a It is the direct impact of abnormal process on performance, that is, the negative deviation value of performance. Δp i Adjust the compensation value of the performance for the subsequent process parameters, such as the effect of increasing the coating thickness on the improvement of tensile strength. i It is the compensation weight of the subsequent process for the performance, reflecting the contribution of different processes to the performance compensation.

[0075] Through model calculation, the system can determine the specific parameter adjustment range of the subsequent process, generate a subsequent parameter adjustment plan, and apply the adjustment plan to the current production process. For example, if the temperature deviation causes a 5% drop in strength, the performance compensation model may recommend increasing the coating thickness by 0.05mm and extending the curing time by 3 seconds to restore the strength to the target range. This intelligent compensation method not only improves the consistency of product quality, but also significantly enhances the stability and flexibility of the production system.

[0076] Further, according to the performance compensation model and in combination with the performance compensation associated parameters, the performance compensation of the current product is performed. Step P54 of the embodiment of the present application further includes:

[0077] P54-1: Determine a performance compensation target, which includes performance constraint values ​​of multiple product performance indicators; P54-2: Use the performance compensation model to perform compensation calculations on the performance compensation-related parameters based on the performance compensation target, generate a subsequent parameter adjustment plan, and perform performance compensation on the current product according to the subsequent parameter adjustment plan.

[0078] Optionally, during the performance compensation process, it is first necessary to determine the performance compensation target, that is, the performance constraint values ​​of multiple product performance indicators set according to production requirements. The performance constraint value is the lower limit or target range that the product performance must meet, which is used to guide subsequent compensation calculations. These values ​​can be extracted from national standards, customer needs, or internal quality specifications. For example, the performance constraint value of tensile strength may be set according to the requirements of the final application scenario. If the product involves multiple performance indicators (such as meeting both strength and flexibility), the system will balance these targets to ensure that other performance is not deviated during the compensation process.

[0079] After determining the performance compensation target, the system will calculate the compensation parameters based on the performance compensation model. The performance compensation model is a mathematical model that comprehensively analyzes the impact of abnormal processes and the adjustment capabilities of subsequent processes. By inputting the abnormal parameter deviation value and performance constraint value, the system uses the compensation model to calculate the required adjustment amount and obtain the subsequent parameter adjustment plan.

[0080] Once the subsequent parameter adjustment plan is generated, the system will send the adjustment instructions to the equipment or operation end of the relevant process, and monitor the effect of the adjustment in real time. For example, in the coating process, the equipment will automatically increase the coating thickness or adjust the spraying speed to achieve the compensation target. After the adjustment plan is executed, the system will monitor the actual effect to detect whether the performance compensation target is achieved. If it is not achieved, the system will further iterate and adjust to ensure that the performance of the current product is restored to the standard range. At the same time, the module has intelligent and closed-loop control capabilities, which can not only compensate for anomalies, but also provide experience support for subsequent production, thereby optimizing the overall production process.

[0081] In summary, the embodiments of the present application have at least the following technical effects:

[0082] This application extracts the key processes of flame-retardant polyester fiber preparation, establishes the upper-lower level association relationship and stage abnormality identification nodes, realizes real-time monitoring and abnormality identification between processes, enables identification nodes in sequence, obtains stage abnormality identification results and transmits them to the central processing unit, locates abnormal processes through abnormal tracing, and adjusts the subsequent process parameters based on the performance compensation model to dynamically compensate product performance. At the same time, the abnormal process parameters are optimized to form a closed-loop control to ensure the stability of subsequent production and the consistency of product quality.

[0083] The technical effect of accurately tracing abnormal processes and dynamically compensating product performance through correlation analysis between upper and lower level processes has been achieved, thereby improving the stability of the production process and the consistency of product quality.

[0084] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0085] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0086] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. An abnormality identification system for flame retardant polyester fiber production, characterized in that: The system comprises: The upper-lower level association establishment module is used to extract the key processes of flame-retardant polyester fiber preparation, and establish the upper-lower level association relationship between each key process according to the influence of the previous process on the subsequent process; A stage abnormality identification node establishment module is used to establish multiple stage abnormality identification nodes after multiple key processes based on the upper-lower level association relationship, and the multiple stage abnormality identification nodes are communicatively connected with the central processing unit; A stage abnormality identification module is used to respectively enable the multiple stage abnormality identification nodes, perform abnormality identification of key processes in turn, obtain stage abnormality identification results and send them to the central processing unit; An abnormal process reverse deduction module is used for the central processing unit to reverse the abnormal process according to the abnormal identification result of the stage and in combination with the upper and lower level association relationship to obtain the abnormal process of the stage; The product performance compensation module is used to adjust the subsequent processes of the stage abnormal process according to the stage abnormality identification result to compensate for the performance of the current product. At the same time, the stage abnormal process is optimized to control the production of subsequent products.

2. The abnormality identification system for flame retardant polyester fiber preparation according to claim 1, characterized in that: The upper and lower level association establishment module is also used for: The key processes for the preparation of flame-retardant polyester fibers were extracted, and the process influence matrix of each preceding process on the subsequent process in the key process was established through qualitative analysis and quantitative modeling; Based on the process impact matrix, a superior-subordinate association relationship is established between each key process, wherein the superior-subordinate association relationship includes a single association relationship and a multi-association relationship, wherein the multi-association relationship includes a plurality of association branches, and the association branches correspond to two associated processes; The multiple associated branches are sorted and screened, and contribution coefficients are respectively assigned to the multiple associated branches according to the sorting results.

3. The abnormality identification system for flame retardant polyester fiber preparation according to claim 2, characterized in that: The upper and lower level association establishment module is also used for: Obtain the key production parameters and output key parameters of each key process, as well as the sequence of each key process; According to the sequence, the output key parameters of the preceding process and the production key parameters of the subsequent process are traversed and extracted, and a qualitative analysis is performed to establish the correlation between the output key parameters of each process and the influence of the subsequent process; Through correlation analysis, the output key parameters of each process and their impact on subsequent processes are quantified, and the process impact matrix is ​​established.

4. The abnormality identification system for flame retardant polyester fiber preparation according to claim 1, characterized in that: The stage abnormality identification module is also used for: Based on the multiple stage abnormality identification nodes, locate the first-order stage abnormality identification node, perform abnormality identification of the first key process, and obtain the first process abnormality identification result; Determining whether the abnormality identification result of the first process is abnormal; If so, sending the first process abnormality identification result to the central processing unit for abnormality processing; If not, the process is postponed to the second-stage abnormality identification node to perform abnormality identification and abnormality determination of the second key process. Similarly, the plurality of stage abnormality identification nodes are traversed and enabled, and abnormality identification of the key processes is performed in turn.

5. The abnormality identification system for flame retardant polyester fiber preparation according to claim 2, characterized in that: The abnormal process reverse inference module is also used for: According to the key process codes corresponding to the abnormal identification results of the said stages, the corresponding upper and lower level association relationships of the stages are matched; Extracting stage abnormality indicators according to the stage abnormality identification results; Based on the superior-subordinate association relationship, the abnormal indicators of the stage are traced back and analyzed to determine multiple association branches; Anomaly scoring is performed according to the contribution coefficients of the multiple associated branches, and a preceding process with high contribution is selected according to the anomaly scoring result as a stage abnormal process, and the stage abnormal process may be one or more.

6. The abnormality identification system for flame retardant polyester fiber preparation according to claim 5, characterized in that: The abnormal process reverse inference module is also used for: The abnormal processes corresponding to the multiple associated branches are scored to determine the previous processes with high contribution. The scoring formula is as follows: S i =α·C i +β·D i ; Among them, S i is the abnormality score of process i, C i is the contribution coefficient of process i to subsequent abnormalities, D i is the coefficient of the degree to which the current parameters of process i deviate from the normal value, and α and β are weight factors.

7. The abnormality identification system for flame retardant polyester fiber preparation according to claim 1, characterized in that: The product performance compensation module is also used for: According to the abnormality identification result of the stage, the abnormal parameter type and the corresponding abnormal parameter deviation value of the abnormal process of the stage are extracted; Based on the abnormal parameter type and the corresponding abnormal parameter deviation value, perform product performance impact analysis to obtain a performance impact index; According to the performance impact index, the performance compensation parameter association is performed, and the relevant parameters of the subsequent process are extracted as the performance compensation association parameters; A performance compensation model is established, and performance compensation of the current product is performed according to the performance compensation model and in combination with the performance compensation associated parameters.

8. The abnormality identification system for flame retardant polyester fiber preparation according to claim 7, characterized in that: The product performance compensation module is also used for: Determining a performance compensation target, wherein the performance compensation target includes performance constraint values ​​of multiple product performance indicators; Using the performance compensation model, based on the performance compensation target, compensation calculation is performed on the performance compensation associated parameters, a subsequent parameter adjustment plan is generated, and performance compensation is performed on the current product according to the subsequent parameter adjustment plan.

9. The abnormality identification system for flame retardant polyester fiber preparation according to claim 7, characterized in that: The performance compensation model includes a performance balance formula: Among them, P c is the performance compensation value of the current product, P a is the direct effect of abnormal process on performance, Δp i The compensation value for performance adjustment of the subsequent process parameters, w i is the compensation weight of the subsequent process for this performance.

Images