Polyester top production optimization method and system based on multivariate data analysis

Abstract

The present invention relates to the technical field of polyester top production, and relates to an optimization method for polyester top production based on multivariate data analysis. The quality control of polyester tops is optimized through feedback control to improve the strength and elasticity indexes of polyester tops. It includes obtaining the fluidity index of raw materials according to the raw material information in the spinning stage of polyester tops; obtaining the compatibility index of additives according to the additive material information in the spinning stage of polyester tops; collecting the temperature data set of the entire process in the spinning stage of polyester tops; obtaining the raw material influence strength prediction value according to the temperature data set and the corresponding fluidity index of raw materials; obtaining the additive influence strength prediction value according to the temperature data set and the compatibility index corresponding to the additive material; fusing the raw material influence strength prediction value and the additive influence strength prediction value to obtain the spinning influence strength prediction index; correcting the draft ratio in the drafting stage of polyester tops according to the spinning influence strength prediction index, and producing polyester tops by implementing the corrected draft ratio.

Description

Technical Field

[0001] The present invention relates to the technical field of polyester top, and in particular to a production optimization method and system for polyester top based on multi - variable data analysis. Background Art

[0002] As a basic material widely used in the textile industry, polyester top is usually blended with wool top to produce fabric materials such as woolen cloth, and is mostly used in high - quality clothing products such as suits and military uniforms. The production process of polyester top involves multiple key links, including melt spinning, fiber drawing, cutting and shaping, and carding and top - making. The process parameters of these links and their mutual cooperation play a decisive role in the quality of the final product, especially having a significant impact on the strength and elasticity of polyester top.

[0003] After a long - term production practice, there is a common problem in the industry: during the entire production process of polyester top, there is a lack of linkage control for each link, resulting in insufficient coordination between production links. For example, the entire production process is executed according to the pre - established process flow and process parameters; this situation makes it difficult to optimize the quality of the finished product in real - time during actual production, increasing production costs and the risk of defective products. In addition, the traditional process flow and parameter setting rely on experience, lacking a scientific analysis and adjustment mechanism, further restricting the flexibility and adaptability of production. Summary of the Invention

[0004] To this end, the technical problem to be solved by the present invention is to overcome the difficulty in real - time optimizing the quality of polyester top finished products in the prior art, and provide a production optimization method and system for polyester top based on multi - variable data analysis, which optimize the quality control of polyester top through feedback control and improve the strength and elasticity indexes of polyester top.

[0005] In the first aspect, to solve the above - mentioned technical problem, the present invention provides a production optimization method for polyester top based on multi - variable data analysis. The production of polyester top includes a spinning stage and a drawing stage. The production optimization method for polyester top includes,

[0006] Obtain the raw material information in the spinning stage of polyester top, and perform a fluidity analysis on the raw material information to obtain the fluidity index of the raw material; wherein, the fluidity index is used to describe the relationship between the viscosity of the raw material and the temperature change;

[0007] Obtain the additive material information in the spinning stage of the polyester top, and perform a compatibility analysis on the additive material information to obtain the compatibility index of the additive; wherein, the compatibility index is used to describe the relationship between the transparency of the additive and the temperature change;

[0008] Collect the temperature data of the entire process in the spinning stage of the polyester top to obtain a temperature data set;

[0009] Perform the strength prediction of the polyester top according to the temperature data set and the corresponding fluidity index of the raw materials, and obtain the raw material influence strength prediction value;

[0010] Perform the strength prediction of the polyester top according to the temperature data set and the corresponding compatibility index of the auxiliary materials, and obtain the auxiliary influence strength prediction value;

[0011] Fuse the raw material influence strength prediction value and the auxiliary influence strength prediction value to obtain the spinning influence strength prediction index;

[0012] Modify the draft ratio in the drafting stage of the polyester top according to the spinning influence strength prediction index, and produce the polyester top with the modified draft ratio.

[0013] In an embodiment of the present invention, it further includes modifying the drafting speed in the drafting stage of the polyester top according to the spinning influence strength prediction index, and producing the polyester top according to the modified drafting speed.

[0014] In an embodiment of the present invention, modifying the draft ratio in the drafting stage of the polyester top according to the spinning influence strength prediction index includes,

[0015] The spinning influence strength prediction index includes an increasing strength index and a decreasing strength index;

[0016] When the spinning influence strength prediction index is the increasing strength index, reduce the draft ratio of the polyester top;

[0017] When the spinning influence strength prediction index is the decreasing strength index, increase the draft ratio of the polyester top.

[0018] In an embodiment of the present invention, it further includes,

[0019] Construct a draft ratio - drafting speed table; wherein, the draft ratio - drafting speed table is used to describe the mapping relationship between the draft ratio and the drafting speed, and define the drafting speed corresponding to the draft ratio in the draft ratio - drafting speed table as the target drafting speed;

[0020] When the spinning influence strength prediction index is the increasing strength index, gradually reduce the draft ratio of the polyester top with a first step length until the drafting speed reaches the target drafting speed.

[0021] In an embodiment of the present invention, when the spinning influence strength prediction index is the decreasing strength index, gradually increase the draft ratio of the polyester top with a second step length until the drafting speed reaches the target drafting speed.

[0022] In one embodiment of the present invention, polyester top strength prediction is performed according to the temperature data set and the corresponding fluidity index of the raw material, and a raw material influence strength prediction value is obtained, including,

[0023] Construct a spinning Bayesian network model;

[0024] Based on the spinning Bayesian network model, analyze the temperature data set and the fluidity index to obtain the viscosity change vector of the raw material at each process node in the spinning stage;

[0025] According to the viscosity of the raw material at the first node and the viscosity change vector, obtain the viscosity of the raw material at each process node in the spinning stage;

[0026] According to the viscosity of the raw material at each node, predict the strength of the polyester spinning to obtain a raw material influence strength prediction value.

[0027] In one embodiment of the present invention, polyester top strength prediction is performed according to the temperature data set and the corresponding compatibility index of the additive material, and an additive influence strength prediction value is obtained, including,

[0028] Based on the spinning Bayesian network model, analyze the temperature data set and the compatibility index to obtain the transparency change vector of the additive at each process node in the spinning stage;

[0029] According to the transparency at the first node and the transparency change vector, obtain the transparency of the additive at each process node in the spinning stage;

[0030] According to the transparency of the additive at each process node in the spinning stage, predict the strength of the polyester spinning to obtain an additive influence strength prediction value.

[0031] In one embodiment of the present invention, the spinning Bayesian network model is trained according to the hierarchical freezing signal;

[0032] Wherein, the spinning Bayesian network model includes a first network layer and a second network layer, the first network layer is connected to the second network layer, the first network layer is used to analyze the temperature data set and the fluidity index to obtain the viscosity change vector of the raw material at each process node in the spinning stage; the second network layer is used to analyze the temperature data set and the compatibility index to obtain the transparency change vector of the additive at each process node in the spinning stage;

[0033] The spinning Bayesian network model selectively outputs one of the first network layer and the second network layer according to the hierarchical freezing signal.

[0034] In one embodiment of the present invention, the predicted spinning influence strength index is obtained by fusing the predicted raw material influence strength value and the predicted additive influence strength value based on weighted average.

[0035] Second, to solve the above technical problems, the present invention also provides a polyester top production optimization system based on multivariate data analysis. The polyester top production includes a spinning stage and a drawing stage. The polyester top production optimization system includes:

[0036] A data acquisition module, which is used to obtain the raw material information and additive material information in the spinning stage of polyester top; and collect the temperature data of the entire process in the spinning stage of polyester top to obtain a temperature data set;

[0037] An analysis module, which is used to perform fluidity analysis on the raw material information to obtain the fluidity index of the raw material; and perform compatibility analysis on the additive material information to obtain the compatibility index of the additive; wherein, the fluidity index is used to describe the relationship between the viscosity of the raw material and the temperature change; the compatibility index is used to describe the relationship between the transparency of the additive and the temperature change;

[0038] A prediction module, which is used to perform polyester top strength prediction according to the temperature data set and the corresponding fluidity index of the raw material to obtain the predicted raw material influence strength value; and perform polyester top strength prediction according to the temperature data set and the corresponding compatibility index of the additive material to obtain the predicted additive influence strength value;

[0039] A fusion module, which is used to fuse the predicted raw material influence strength value and the predicted additive influence strength value to obtain the predicted spinning influence strength index;

[0040] An optimization module, which is used to correct the draw ratio in the drawing stage of the polyester top according to the predicted spinning influence strength index, and produce the polyester top with the corrected draw ratio.

[0041] The above technical solutions of the present invention have the following beneficial effects compared with the prior art:

[0042] The polyester top production optimization method and system based on multivariate data analysis according to the present invention realize the feedback control of the polyester top production process through data driving. By combining the analysis of raw materials, additives and temperature data, the influence of the spinning stage on the strength of polyester top is accurately predicted to dynamically adjust the draw ratio, realizing the production optimization based on real-time feedback data, significantly improving the strength and elasticity of polyester top, so that the final product can better meet the market demand; in addition, through continuous process optimization by real-time feedback, the generation of unqualified products is reduced, thereby improving production efficiency, reducing material waste and production costs.

[0043] In summary, the polyester top production optimization method and system based on multivariate data analysis described herein optimize the quality control of polyester tops through comprehensive data analysis and feedback control, not only improving production efficiency but also significantly improving the strength and elasticity indexes of the products. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to make the content of the present invention easier to be clearly understood, the following further describes the present invention in detail according to specific embodiments of the present invention and in conjunction with the drawings, wherein

[0045] Figure 1 is a flowchart of the polyester top production optimization method based on multivariate data analysis in a preferred embodiment of the present invention;

[0046] Figure 2 is a structural block diagram of the polyester top production optimization system based on multivariate data analysis in a preferred embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] The following further describes the present invention in conjunction with the drawings and specific embodiments, so that those skilled in the art can better understand the present invention and be able to implement it, but the embodiments cited do not limit the present invention.

[0048] Embodiment 1

[0049] The whole process of polyester top production includes melt spinning (hereinafter referred to as "spinning stage"), fiber drawing (hereinafter referred to as "drawing stage"), cutting and shaping, and carding and roving. Among them, melt spinning and fiber drawing are the most critical links affecting the formation of polyester tops. The following describes the specific processes of the spinning stage and the drawing stage of polyester top production:

[0050] I. Spinning stage:

[0051] Usually, polyester (such as polyethylene terephthalate, PET) particles are used as raw materials. After drying treatment to remove moisture and ensure the fluidity of the material during melting; then heating and melting, the polyester particles are heated to a certain temperature (usually between 240 - 290 °C) in a high-temperature melting device to change from a solid state to a flowing molten state; then entering the spinning process, the molten polyester is extruded from the nozzle to form silk threads; the silk threads are quickly solidified in cooling air or coolant to form a preliminary fiber structure. Each process node in the spinning stage is related to temperature. For example, drying, melting, spinning, and cooling are all related to the temperature acting on the raw materials; the inventors of the present application found through continuous production experiments that temperature affects the viscosity of the raw materials by affecting their fluidity, and the viscosity differences of the raw materials at each node in the spinning stage ultimately affect the strength of the polyester tops.

[0052] In addition, in order to improve the relevant properties of the polyester top, it is necessary to add additives; for example, antistatic agents, plasticizers, colorants and other additives are added as needed to improve the properties of the polyester top. As the spinning process progresses, the solid additives melt into a fluid state and react with the raw materials to be compatible, thereby participating in the improvement of the properties of the polyester top. Through continuous production tests, the inventors of this application found that temperature affects the transparency of the additives by influencing their compatibility, and the transparency of the additives at each node of the spinning stage affects their role in participating in the property improvement, and ultimately affects the strength of the polyester top.

[0053] II. Drafting stage

[0054] The fibers wound up in the spinning stage are loosened to ensure that they do not knot during the drafting process; then the initially formed fibers are stretched through a drafting device, and the strength and elasticity of the fibers are affected by setting the draft ratio and the drafting speed. Subsequently, the drafted fibers are shaped by the device to maintain their cross-section and strength.

[0055] Refer to Figure 1 As shown, an embodiment of the present invention discloses an optimization method for polyester top production based on multi-source data analysis. Among them, this optimization method for polyester top production involves the spinning stage and the drafting stage of polyester top production. This optimization method for polyester top production includes,

[0056] Obtain the raw material information of the spinning stage of the polyester top, and perform a fluidity analysis on the raw material information to obtain the fluidity index of the raw materials; wherein, the fluidity index is used to describe the relationship between the viscosity of the raw materials and the temperature change.

[0057] In a specific application scenario, the fluidity indexes of various raw materials are obtained through pre-tests; when there are multiple raw materials, the fluidity indexes of each raw material are obtained through tests, and a raw material fluidity index set is formed; this provides basic data for subsequent production optimization. The fluidity index is also used to optimize the temperature in the spinning stage to ensure that the raw materials have appropriate fluidity during spinning.

[0058] Obtain the additive material information of the spinning stage of the polyester top, and perform a compatibility analysis on the additive material information to obtain the compatibility index of the additives; wherein, the compatibility index is used to describe the relationship between the transparency of the additives and the temperature change;

[0059] In a specific application scenario, the compatibility indexes of various additives are obtained through pre-tests; when there are multiple additives, the compatibility indexes of each additive are obtained through tests, and an additive fluidity index set is formed; this provides basic data for subsequent production optimization. The compatibility index is also used to understand the performance of the additives at different temperatures, so as to optimize the additive addition process and ensure good interaction between the additives and the raw materials.

[0060] Collect the temperature data of the entire process in the polyester top-spinning stage to obtain a temperature data set;

[0061] In a specific application scenario, based on the temperature sensing module, collect the temperature data of the entire process in the polyester top-spinning stage in real time to provide time-series data for subsequent strength prediction.

[0062] Perform polyester top strength prediction according to the temperature data set and the corresponding fluidity index of the raw material to obtain the raw material influence strength prediction value;

[0063] In a specific application scenario, according to the temperature data set collected in real time and combined with the fluidity index corresponding to the current raw material, predict the influence of the viscosity difference of the raw material on the polyester spinning strength in the spinning stage, providing a decision-making basis for subsequent dynamic adjustment of the draft ratio. In addition, the selection and use of raw materials can be optimized according to the predicted influence on the polyester spinning strength.

[0064] Perform polyester top strength prediction according to the temperature data set and the corresponding compatibility index of the auxiliary material to obtain the auxiliary influence strength prediction value;

[0065] In a specific application scenario, according to the temperature data set collected in real time and combined with the compatibility index corresponding to the current auxiliary, predict the influence of the transparency difference of the auxiliary on the polyester spinning strength in the spinning stage, providing a decision-making basis for subsequent dynamic adjustment of the draft ratio.

[0066] Fuse the raw material influence strength prediction value and the auxiliary influence strength prediction value to obtain the spinning influence strength prediction index;

[0067] In a specific application scenario, by fusing the raw material influence strength prediction value and the auxiliary influence strength prediction value, a comprehensive spinning strength prediction index is formed, so as to make a more scientific adjustment decision.

[0068] According to the spinning influence strength prediction index, correct the draft ratio in the drafting stage of the polyester top, and produce the polyester top with the corrected draft ratio. Correcting the draft ratio realizes production optimization based on real-time feedback data, and can flexibly adjust production parameters to adapt to the actual situation, thereby improving the strength and elasticity of the polyester top.

[0069] The polyester top production optimization method and system based on multivariate data analysis according to the present invention realizes the feedback control of the polyester top production process through data driving. By combining the analysis of raw materials, additives, and temperature data, it accurately predicts the influence of the spinning stage on the strength of the polyester top and dynamically adjusts the drafting ratio to achieve production optimization based on real-time feedback data, significantly improving the strength and elasticity of the polyester top, so that the final product can better meet the market demand. In addition, through continuous process optimization based on real-time feedback, the generation of unqualified products is reduced, thereby improving production efficiency and reducing material waste and production costs.

[0070] In other embodiments of the present invention, it further includes correcting the drafting speed in the drafting stage of the polyester top according to the strength prediction index affecting spinning, and producing the polyester top according to the corrected drafting speed.

[0071] In specific application scenarios, the drafting ratio and drafting speed are corrected simultaneously according to the strength prediction index affecting spinning to further optimize the production process and improve the strength and elasticity of the polyester top.

[0072] Furthermore, correcting the drafting ratio in the drafting stage of the polyester top according to the strength prediction index affecting spinning includes,

[0073] The strength prediction index affecting spinning includes an index for increasing strength and an index for decreasing strength;

[0074] In specific application scenarios, the process flow of the polyester top and the process parameters of each process flow are preset, and the strength of the polyester top produced based on the set process parameters is detected to obtain a strength reference value. When other stages except the spinning stage and the drafting stage remain unchanged, the polyester top is produced based on the above method, and a strength prediction value is obtained according to the strength prediction index affecting spinning. The strength prediction value is compared with the strength reference value. If the strength prediction value is less than the strength reference value, the strength prediction index affecting spinning is an index for decreasing strength; if the strength prediction value is greater than the strength reference value, the strength prediction index affecting spinning is an index for increasing strength. If the strength prediction value is equal to the strength reference value, the current drafting ratio and drafting speed are maintained to execute the production of the polyester top.

[0075] It should be noted that: The drafting ratio is the ratio of the fiber length before and after drafting, and its characterization is as follows: Among them, Dr is the drafting ratio; L 牵伸前 is the length before drafting; L 牵伸后 is the length after drafting. The drafting speed is the speed at which the fiber passes through the drafting equipment.

[0076] The drafting ratio can characterize the compactness of the internal structure of the fiber, thereby affecting the strength and modulus of the fiber.

[0077] When the spinning influence strength prediction index is to increase the strength index, reduce the draft ratio of the polyester top; when the spinning influence strength prediction index is to decrease the strength index, increase the draft ratio of the polyester top.

[0078] In a specific application scenario, when the spinning influence strength prediction index is to increase the strength index, it indicates that the fiber structure obtained by spinning already has a relatively high initial strength. By reducing the draft ratio to balance the high strength and avoid unnecessary fracture risks; when the spinning influence strength prediction index is to decrease the strength index, it indicates that the initial strength of the fiber structure obtained by spinning is low. By increasing the draft ratio to improve the strength of the fiber to ensure that the final product meets the required performance standards; in this way, it is ensured that the produced polyester top will neither break due to excessive drafting nor have insufficient strength due to insufficient drafting.

[0079] Furthermore, it also includes constructing a draft ratio-drafting speed table; wherein, the draft ratio-drafting speed table is used to describe the mapping relationship between the draft ratio and the drafting speed, and define the drafting speed corresponding to the draft ratio in the draft ratio-drafting speed table as the target drafting speed;

[0080] In a specific application scenario, there is a certain internal relationship between the draft ratio and the drafting speed. The control logic of the draft ratio is to achieve the expected fiber properties, such as strength, modulus, fineness, etc. The control logic of the drafting speed is to ensure that the fibers can be evenly stressed during the drafting process, avoiding fiber breakage caused by too fast speed or low efficiency caused by too slow speed. Within a certain range, the draft ratio and the drafting speed have a positive correlation, that is, when increasing the draft ratio, the drafting speed is correspondingly increased to improve production efficiency; when reducing the draft ratio, the drafting speed is reduced to avoid fiber breakage. According to historical production experience, establish a draft ratio-drafting speed table to provide data support for subsequent decision-making.

[0081] When the spinning influence strength prediction index is to increase the strength index, gradually reduce the draft ratio of the polyester top in the first step length until the drafting speed reaches the target drafting speed.

[0082] In a specific application scenario, when the spinning influence strength prediction index is to increase the strength index, perform production with the goal of reducing the draft ratio. Determine the total target reduction amount of the draft ratio according to the actual increased strength index, and gradually reduce the draft ratio to the total target reduction amount in a certain step length, so as to slowly change the draft ratio and avoid the impact of sudden changes in the draft ratio on other properties. As the draft ratio decreases, the drafting speed is correspondingly reduced. Due to the certain lag in data transmission and processing, in order to overcome the adverse impact of this lag on production control, when the changed draft ratio reaches the target drafting speed corresponding to it in the draft ratio-drafting speed table, stop adjusting the draft ratio to avoid over-adjustment.

[0083] In one embodiment of the present invention, when the spinning influence strength prediction index is a reduced strength index, the draft ratio of the polyester top is gradually increased in a second step length until the draft speed reaches the target draft speed.

[0084] In a specific application scenario, when the spinning influence strength prediction index is a reduced strength index, production is carried out with the goal of increasing the draft ratio. The target total increase in the draft ratio is determined according to the actual reduced strength index, and the draft ratio is gradually increased to the target total increase in a certain step length, thereby slowly changing the draft ratio to avoid the impact of sudden changes in the draft ratio on other properties. As the draft ratio increases, the draft speed is correspondingly increased. Due to the certain lag in data transmission and processing, in order to overcome the adverse impact of this lag on production control, when the changed draft ratio reaches the corresponding target draft speed in the draft ratio-draft speed table, the adjustment of the draft ratio is stopped to avoid over-adjustment.

[0085] In a further embodiment of the present invention, polyester top strength prediction is performed according to the temperature data set and the corresponding fluidity index of the raw material, and a raw material influence strength prediction value is obtained, including:

[0086] Construct a spinning Bayesian network model;

[0087] In a specific application scenario, relevant variables are determined, including temperature data, the fluidity index of the raw material composition, the compatibility index of the additive material, the strength of the polyester top, etc.; a spinning Bayesian network model is established according to domain knowledge. Each node represents a variable, and the edge represents the conditional dependence relationship between variables. The relationship between the fluidity index and temperature, raw material composition, and polyester top strength, as well as the relationship between the compatibility index and temperature, additive composition, and polyester top strength, are determined; the conditional probability is estimated from the historical data set, and the parameters in the spinning Bayesian network are filled.

[0088] Based on the spinning Bayesian network model, the temperature data set and the fluidity index are analyzed to obtain the viscosity change vector of the raw material at each process node in the spinning stage;

[0089] In a specific application scenario, the spinning Bayesian network is used to comprehensively analyze the temperature data set and the raw material fluidity index to obtain the viscosity change vector of the raw material at each process node; the viscosity change vector describes the viscosity change amount between the next process node and the previous process node.

[0090] According to the raw material viscosity at the first node and the viscosity change vector, the raw material viscosity at each process node in the spinning stage is obtained;

[0091] In a specific application scenario, the raw material viscosity of the second node is obtained based on the raw material viscosity of the first node (i.e., the first node) and the viscosity change amount between the first node and the second node. The raw material viscosity of the third node is obtained based on the raw material viscosity of the second node and the viscosity change amount between the second node and the third node, and so on, to obtain the raw material viscosities of all process nodes in the spinning stage.

[0092] Based on the raw material viscosities of the respective nodes, the strength of the polyester spinning is predicted to obtain a raw material influence strength prediction value.

[0093] In a specific application scenario, a spinning Bayesian network model is trained according to historical data to obtain a strength prediction model. The raw material viscosities of all process nodes in the current raw material spinning stage are learned through the strength prediction model, and the predicted strength of the polyester spinning produced by the current raw material spinning is obtained. The predicted strength is compared with the actual strength of the historical data to obtain a raw material influence strength prediction value.

[0094] In an embodiment of the present invention, polyester top strength prediction is performed according to the temperature data set and the compatibility index corresponding to the auxiliary material to obtain an auxiliary material influence strength prediction value, including

[0095] Analyzing the temperature data set and the compatibility index based on the spinning Bayesian network model to obtain a transparency change vector of the auxiliary material at each process node in the spinning stage;

[0096] In a specific application scenario, the spinning Bayesian network is used to comprehensively analyze the temperature data set and the auxiliary material compatibility index to obtain a transparency change vector of the auxiliary material at each process node; the transparency change vector describes the transparency change amount between the next process node and the previous process node.

[0097] Based on the transparency of the first node and the transparency change vector, the transparency of the auxiliary material at each process node in the spinning stage is obtained;

[0098] In a specific application scenario, the transparency of the auxiliary material of the second node is obtained based on the transparency of the auxiliary material of the first node (i.e., the first node) and the transparency change amount between the first node and the second node. The transparency of the auxiliary material of the third node is obtained based on the transparency of the auxiliary material of the second node and the transparency change amount between the second node and the third node, and so on, to obtain the transparency of the auxiliary material of all process nodes in the spinning stage.

[0099] Based on the transparency of the auxiliary material at each process node in the spinning stage, the strength of the polyester spinning is predicted to obtain an auxiliary material influence strength prediction value.

[0100] In a specific application scenario, the original transparency of all process nodes in the current auxiliary agent spinning stage is learned through the strength prediction model, the predicted strength of the polyester fiber spun in the current auxiliary agent spinning production is obtained, the predicted strength is compared with the actual strength of the historical data, and the predicted value of the strength affected by the auxiliary agent is obtained.

[0101] Further, the spinning Bayesian network model is trained according to the hierarchical freezing signal;

[0102] Wherein, the spinning Bayesian network model includes a first network layer and a second network layer, the first network layer is connected to the second network layer, the first network layer is used to analyze the temperature data set and the fluidity index, and obtain the viscosity change vector of the raw material at each process node in the spinning stage; the second network layer is used to analyze the temperature data set and the compatibility index, and obtain the transparency change vector of the auxiliary agent at each process node in the spinning stage;

[0103] The spinning Bayesian network model outputs one of the first network layer and the second network layer according to the hierarchical freezing signal.

[0104] In a specific application scenario, through the hierarchical freezing signal, the features of different network layers are selectively output, allowing the selection of the features of different network layers according to specific requirements during the training process, so as to train in stages and gradually adjust the strength prediction model, making it better adapt to different complex data types, and being able to obtain better feature representations when processing various data, thereby further improving the accuracy of strength prediction.

[0105] In an embodiment of the present invention, the spinning influence strength prediction index is obtained based on weighted average fusion of the raw material influence strength prediction value and the auxiliary agent influence strength prediction value.

[0106] In a specific application scenario, a specific weight is respectively assigned to the raw material influence strength prediction value and the auxiliary agent influence strength prediction value according to the contribution degree to the final result, and they are multiplied by their corresponding weights and then added; obtaining the spinning influence strength prediction index through weighted average fusion can improve the prediction accuracy and enhance the robustness.

[0107] Embodiment 2

[0108] As Figure 2 shown, an embodiment of the present invention discloses a polyester top production optimization system based on multivariate data analysis. The polyester top production includes a spinning stage and a drawing stage. The polyester top production optimization system includes,

[0109] A data acquisition module, which is used to obtain the raw material information and auxiliary agent material information in the spinning stage of the polyester top; and collect the temperature data of the whole process in the spinning stage of the polyester top to obtain a temperature data set;

[0110] An analysis module for performing a fluidity analysis on the raw material information to obtain a fluidity index of the raw material; and performing a compatibility analysis on the auxiliary material information to obtain a compatibility index of the auxiliary; wherein the fluidity index is used to describe the relationship between the viscosity of the raw material and the temperature change; the compatibility index is used to describe the relationship between the transparency of the auxiliary and the temperature change;

[0111] A prediction module for performing a prediction of the strength of the polyester top according to the temperature data set and the corresponding fluidity index of the raw material to obtain a predicted value of the strength affected by the raw material; and performing a prediction of the strength of the polyester top according to the temperature data set and the corresponding compatibility index of the auxiliary material to obtain a predicted value of the strength affected by the auxiliary;

[0112] A fusion module for fusing the predicted value of the strength affected by the raw material and the predicted value of the strength affected by the auxiliary to obtain a predicted index of the strength affected by spinning;

[0113] An optimization module for correcting the draft ratio in the drafting stage of the polyester top according to the predicted index of the strength affected by spinning and producing the polyester top with the corrected draft ratio.

[0114] The polyester top production optimization system based on multi - variable data analysis disclosed in the embodiments of the present invention can effectively realize the production optimization of the polyester top, and the technical effects can be as described in the above embodiments, which will not be elaborated here.

[0115] In summary, for the polyester top production optimization method and system based on multi - variable data analysis of the present invention, through data - driven feedback control of the polyester top production process, combined with the analysis of raw materials, auxiliaries and temperature data, it accurately predicts the impact of the spinning stage on the strength of the polyester top to dynamically adjust the draft ratio, realizes production optimization based on real - time feedback data, significantly improves the strength and elasticity of the polyester top, so that the final product can better meet the market demand; in addition, through continuous process optimization with real - time feedback, it reduces the generation of unqualified products, thereby improving production efficiency, reducing material waste and production costs.

[0116] The polyester top production optimization method and system based on multi - variable data analysis of the present invention optimize the quality control of the polyester top through comprehensive data analysis and feedback control, not only improving production efficiency, but also significantly improving the strength and elasticity indexes of the product.

[0117] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0118] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0119] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0120] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0121] Obviously, the above embodiments are only examples given for clear illustration and are not limitations on the implementation manners. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation manners here. And the obvious changes or modifications derived therefrom are still within the protection scope of the present invention.

Claims

1. An optimization method for polyester top production based on multivariate data analysis. Polyester top production includes a spinning stage and a drawing stage, and is characterized in that: The optimization method for producing polyester top includes obtaining the raw material information in the spinning stage of polyester top and performing a fluidity analysis on the raw material information to obtain the fluidity index of the raw materials; wherein, the fluidity index is used to describe the relationship between the viscosity of the raw materials and the temperature change; obtaining the auxiliary material information in the spinning stage of polyester top and performing a compatibility analysis on the auxiliary material information to obtain the compatibility index of the auxiliaries; wherein, the compatibility index is used to describe the relationship between the transparency of the auxiliaries and the temperature change; collecting the temperature data of the entire process in the spinning stage of polyester top to obtain a temperature data set; performing a strength prediction of polyester top according to the temperature data set and the corresponding fluidity index of the raw materials to obtain a raw material influence strength prediction value; performing a strength prediction of polyester top according to the temperature data set and the corresponding compatibility index of the auxiliary materials to obtain an auxiliary influence strength prediction value; fusing the raw material influence strength prediction value and the auxiliary influence strength prediction value to obtain a spinning influence strength prediction index; correcting the draft ratio in the drafting stage of polyester top according to the spinning influence strength prediction index and producing polyester top with the corrected draft ratio.

2. The polyester top production optimization method based on multivariate data analysis according to claim 1, wherein: It also includes correcting the drafting speed in the drafting stage of polyester top according to the spinning influence strength prediction index and producing polyester top with the corrected drafting speed.

3. The polyester top production optimization method based on multivariate data analysis according to claim 1 or 2, characterized in that: Correcting the draft ratio in the drafting stage of polyester top according to the spinning influence strength prediction index includes the spinning influence strength prediction index includes an increasing strength index and a decreasing strength index; when the spinning influence strength prediction index is the increasing strength index, reducing the draft ratio of polyester top; when the spinning influence strength prediction index is the decreasing strength index, increasing the draft ratio of polyester top.

4. The polyester top production optimization method based on multivariate data analysis according to claim 3, characterized in that: It also includes constructing a draft ratio - drafting speed table; wherein, the draft ratio - drafting speed table is used to describe the mapping relationship between the draft ratio and the drafting speed, and defining the drafting speed corresponding to the draft ratio in the draft ratio - drafting speed table as the target drafting speed; when the spinning influence strength prediction index is the increasing strength index, gradually reducing the draft ratio of polyester top with a first step length until the drafting speed reaches the target drafting speed.

5. The polyester top production optimization method based on multivariate data analysis according to claim 4, characterized in that: when the spinning influence strength prediction index is the decreasing strength index, gradually increasing the draft ratio of polyester top with a second step length until the drafting speed reaches the target drafting speed.

6. The polyester top production optimization method based on multivariate data analysis according to claim 1, characterized in that: Performing a strength prediction of polyester top according to the temperature data set and the corresponding fluidity index of the raw materials to obtain a raw material influence strength prediction value includes constructing a spinning Bayesian network model; analyzing the temperature data set and the fluidity index based on the spinning Bayesian network model to obtain the viscosity change vector of the raw materials at each process node in the spinning stage; obtaining the raw material viscosity at each process node in the spinning stage according to the raw material viscosity at the first node and the viscosity change vector; predicting the strength of the polyester spinning according to the raw material viscosity at each node to obtain a raw material influence strength prediction value.

7. The polyester top production optimization method based on multivariate data analysis according to claim 6, characterized in that: Execute the strength prediction of polyester top according to the temperature data set and the compatibility index corresponding to the auxiliary material, and obtain the predicted value of the strength affected by the auxiliary material, including: Analyze the temperature data set and the compatibility index based on the spinning Bayesian network model to obtain the transparency change vector of the auxiliary material at each process node in the spinning stage; Obtain the transparency of the auxiliary material at each process node in the spinning stage according to the transparency of the first node and the transparency change vector; Predict the strength of the polyester spinning according to the transparency of the auxiliary material at each process node in the spinning stage, and obtain the predicted value of the strength affected by the auxiliary material.

8. The polyester top production optimization method based on multivariate data analysis according to claim 7, characterized in that: Train the spinning Bayesian network model according to the hierarchical freezing signal; Among them, the spinning Bayesian network model includes a first network layer and a second network layer. The first network layer is connected to the second network layer. The first network layer is used to analyze the temperature data set and the fluidity index to obtain the viscosity change vector of the raw material at each process node in the spinning stage; the second network layer is used to analyze the temperature data set and the compatibility index to obtain the transparency change vector of the auxiliary material at each process node in the spinning stage; The spinning Bayesian network model selects one of the first network layer and the second network layer to output according to the hierarchical freezing signal.

9. The polyester top production optimization method based on multivariate data analysis according to claim 1, characterized in that: Based on weighted average, fuse the predicted value of the strength affected by the raw material and the predicted value of the strength affected by the auxiliary material to obtain the predicted index of the strength affected by spinning.

10. A polyester top production optimization system based on multivariate data analysis. The production of polyester tops includes a spinning stage and a drawing stage, and is characterized in that: The polyester top production optimization system includes: A data acquisition module, which is used to obtain the raw material information and auxiliary material information in the spinning stage of polyester top; and collect the temperature data of the whole process in the spinning stage of polyester top to obtain a temperature data set; An analysis module, which is used to perform fluidity analysis on the raw material information to obtain the fluidity index of the raw material; and perform compatibility analysis on the auxiliary material information to obtain the compatibility index of the auxiliary material; among them, the fluidity index is used to describe the relationship between the viscosity of the raw material and the temperature change; the compatibility index is used to describe the relationship between the transparency of the auxiliary material and the temperature change; A prediction module, which is used to execute the strength prediction of polyester top according to the temperature data set and the fluidity index corresponding to the raw material to obtain the predicted value of the strength affected by the raw material; and execute the strength prediction of polyester top according to the temperature data set and the compatibility index corresponding to the auxiliary material to obtain the predicted value of the strength affected by the auxiliary material; A fusion module, which is used to fuse the predicted value of the strength affected by the raw material and the predicted value of the strength affected by the auxiliary material to obtain the predicted index of the strength affected by spinning; An optimization module, which is used to correct the draft ratio in the drafting stage of the polyester top according to the predicted index of the strength affected by spinning, and produce the polyester top with the corrected draft ratio.

Images