CN115671874B

CN115671874B - Production method, system and device of multifunctional filter material and storage medium

Info

Publication number: CN115671874B
Application number: CN202211332241.7A
Authority: CN
Inventors: 朱志慧
Original assignee: Suzhou Meeko Environmental Technology Co ltd
Current assignee: Suzhou Meeko Environmental Technology Co ltd
Priority date: 2022-10-28
Filing date: 2022-10-28
Publication date: 2023-10-13
Anticipated expiration: 2042-10-28
Also published as: CN115671874A

Abstract

Embodiments of the present disclosure provide a method, system, apparatus, and storage medium for producing a multifunctional filter material. The method comprises the steps of obtaining stock solution parameters, spinning parameters and solid parameters of the filter material; determining a predicted fiber quality of the filter material according to the stock solution parameter, the spinning parameter and the solid-state parameter; determining a pass rate of the filter material based on the predicted fiber quality; and generating early warning information based on the qualification rate so as to remind a user to carry out parameter adjustment.

Description

Production method, system and device of multifunctional filter material and storage medium

Technical Field

The present disclosure relates to the field of fiber production, and in particular, to a method, a system, a device, and a storage medium for producing a multifunctional filter material.

Background

With the continuous development of the chemical industry, the synthetic fibers realize rapid breakthrough in the aspect of filter materials. Synthetic fibers have been widely used in the field of filter materials because of their excellent properties of corrosion resistance, high temperature resistance, friction resistance, etc. In particular, polytetrafluoroethylene fibers are synthetic fibers prepared by spinning or preparing a film from PTFE and then cutting or fibrillating, and have important roles in high-temperature flue gas filtration.

At present, synthetic fibers still have certain process difficulty in production, and especially in wet spinning, the determination of processing parameters mainly depends on the experience of workers, so that the production process of the filter material is seriously influenced.

Disclosure of Invention

One aspect of embodiments of the present specification provides a method of producing a multifunctional filter material, the method comprising: acquiring stock solution parameters, spinning parameters and solid parameters of the filter material; determining a predicted fiber quality of the filter material according to the stock solution parameter, the spinning parameter and the solid-state parameter; determining a pass rate of the filter material based on the predicted fiber quality; and generating early warning information based on the qualification rate so as to remind a user to carry out parameter adjustment.

Another aspect of embodiments of the present specification provides a production system of a multifunctional filter material, the system comprising: the parameter acquisition module is used for acquiring stock solution parameters, spinning parameters and solid parameters of the filter material; the fiber quality determining module is used for determining the predicted fiber quality of the filter material according to the stock solution parameter, the spinning parameter and the solid-state parameter; a pass rate determination module for determining a pass rate of the filter material based on the predicted fiber quality; and the early warning module is used for generating early warning information based on the qualification rate so as to remind a user to carry out parameter adjustment.

Another aspect of embodiments of the present specification provides an apparatus for producing a multifunctional filter material. The apparatus includes at least one storage medium for storing computer instructions and at least one processor; the at least one processor is configured to execute the computer instructions to implement a method of producing a multifunctional filter material.

Another aspect of embodiments of the present description provides a computer-readable storage medium. The storage medium stores computer instructions that, when executed by a computer, implement a method of producing a multifunctional filter material.

Drawings

The present specification will be further elucidated by way of example embodiments, which will be described in detail by means of the accompanying drawings. The embodiments are not limiting, in which like numerals represent like structures, wherein:

FIG. 1 is a schematic illustration of an application scenario of a production system for a multifunctional filter material according to some embodiments of the present disclosure;

FIG. 2 is an exemplary block diagram of a production system for a multi-functional filter material according to some embodiments of the present disclosure;

FIG. 3 is an exemplary flow chart of a method of producing a multifunctional filter material according to some embodiments of the present disclosure;

FIG. 4 is a schematic illustration of a predictive model shown in accordance with some embodiments of the present disclosure;

FIG. 5 is an exemplary flow chart of a method of determining spinneret parameters and die set parameters according to some embodiments of the present description.

Detailed Description

In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings that are required to be used in the description of the embodiments will be briefly described below. It is apparent that the drawings in the following description are only some examples or embodiments of the present specification, and it is possible for those of ordinary skill in the art to apply the present specification to other similar situations according to the drawings without inventive effort. Unless otherwise apparent from the context of the language or otherwise specified, like reference numerals in the figures refer to like structures or operations.

It will be appreciated that "system," "apparatus," "unit" and/or "module" as used herein is one method for distinguishing between different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

As used in this specification and the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.

A flowchart is used in this specification to describe the operations performed by the system according to embodiments of the present specification. It should be appreciated that the preceding or following operations are not necessarily performed in order precisely. Rather, the steps may be processed in reverse order or simultaneously. Also, other operations may be added to or removed from these processes.

Fig. 1 is a schematic view of an application scenario of a production system of a multifunctional filter material according to some embodiments of the present description.

The application scenario 100 of the production system of the multifunctional filter material may include a processor 110, a processor 120, a storage device 130, a user terminal 140, and a production device 150.

The production system of the multifunctional filter material can be used for producing the multifunctional filter material. In some embodiments, the system may determine the predicted fiber quality of the filter material by obtaining stock parameters, spinning parameters, and solids parameters of the filter material, and then determining the predicted fiber quality of the filter material based on the parameters; then determining the qualification rate of the filter material based on the predicted fiber quality; and finally, generating early warning information based on the qualification rate to remind a user of carrying out parameter adjustment so as to improve the fiber production efficiency.

In the application scenario 100 of the production system of the multifunctional filter material, the processor 110 may communicate with the storage device 130, the user terminal 140, and the production device 150 through the network 120 to provide various functions of the production of the multifunctional filter material, and the storage device 130 may store all information of the production process. In some embodiments, the user terminal 140 may send production instructions to the production device 150 and receive the pre-alarm information from the processor 110. The processor 110 may obtain the production parameters of the filter material, process the same, and send alert parameter adjustment information to the user terminal 140. The information transfer relationship between the above devices is merely an example, and the present application is not limited thereto.

In some embodiments, the processor 110, the user terminal 140, and possibly other system components may include a storage device 130.

The processor 110 may process data and/or information obtained from other devices or system components. The processor may execute program instructions to perform one or more of the functions described in this disclosure based on such data, information, and/or processing results. In some embodiments, the processor 110 may contain one or more sub-processing devices (e.g., single-core processing devices or multi-core processing devices).

The network 120 may connect components of the system and/or connect the system with external resource components. Network 120 enables communication between components and other parts of the system to facilitate the exchange of data and/or information. In some embodiments, network 120 may be any one or more of a wired network or a wireless network. For example, network 120 may include a cable network, a fiber optic network, a telecommunications network, the internet, a Local Area Network (LAN), a Wide Area Network (WAN), or any combination thereof. The network connection between the parts can be in one of the above-mentioned ways or in a plurality of ways. In some embodiments, the network may be a point-to-point, shared, centralized, etc. variety of topologies or a combination of topologies.

Storage device 130 may be used to store data and/or instructions. Storage device 130 may include one or more storage components, each of which may be a separate device or may be part of another device. In some embodiments, the storage device 130 may include Random Access Memory (RAM), read-only memory (ROM), mass storage, removable storage, and the like. In some embodiments, the storage device 130 may be implemented on a cloud platform.

User terminal 140 refers to one or more terminal devices or software used by a user. In some embodiments, one or more users of the user terminal 140 may be used, including users who directly use the service, as well as other related users. In some embodiments, the user terminal 140 may be one or any combination of mobile device 140-1, tablet computer 140-2, laptop computer 140-3, desktop computer 140-4, and the like, as well as other input and/or output enabled devices.

In some embodiments, tablet device 140-2 may be an industrial tablet computer or the like. In some embodiments, desktop computer 140-4 may be an industrial computer or the like. In some embodiments, a user may refer to a manufacturer or other production participant. The above examples are only intended to illustrate the broad scope of the user terminal 140 devices and not to limit the scope thereof.

The production facility 150 may be used for the production of multifunctional filter materials. The production equipment 150 is used for raw material preparation, spinning, solidification, antistatic treatment, opening, needling, molding and the like. In some embodiments, the production apparatus 150 may include preparing a stock solution, i.e., preparing a raw material that forms the fibers; the fiber mother solution is sprayed out through a spinneret orifice under a certain pressure; the sprayed stock solution silk is soaked in solid solution, etc., to form PTFE fiber. In some embodiments, the production equipment 150 then performs antistatic treatment, opening, needling, forming steps on the fibers to form a finished product, which may be a PTFE needled filter cloth (felt), or the like.

FIG. 2 is an exemplary block diagram of a production system for a multi-functional filter material according to some embodiments of the present disclosure.

As shown in fig. 2, the production system 200 of the multifunctional filter material may include a parameter acquisition module 210, a fiber quality determination module 220, a yield determination module 230, and an early warning module 240.

The parameter acquisition module 210 may be configured to acquire a dope parameter, a spinning parameter, and a solids parameter of the filter material. For more on the dope parameters, the spinning parameters and the die set parameters, see fig. 3 for a description thereof.

In some embodiments, the parameter acquisition module 210 may be further configured to: acquiring a stock solution parameter, a preset spinning parameter and a preset solid type parameter of a filter material; predicting the predicted solidification duration of the filter material in the solidification process according to the preset spinning parameters and the preset solidification parameters; judging whether the predicted fixation time length meets a preset fixation condition or not; if yes, taking the preset spinning parameters and the preset shaping parameters which meet the preset shaping conditions as spinning parameters and shaping parameters; if not, determining a preset spinning parameter and/or a parameter adjustment value of the preset solid type parameter according to the predicted solid type duration and the preset solid type condition, and adjusting the preset spinning parameter and/or the preset solid type parameter according to the parameter adjustment value so as to determine the spinning parameter and the solid type parameter.

Wherein, determining the preset spinning parameter and/or the parameter adjustment value of the preset shaping parameter according to the predicted shaping time length and the preset shaping condition may further include: determining a fixed time length difference value according to a preset fixed time length range and a predicted fixed time length in preset fixed conditions; determining a preset spinning parameter and/or a predicted adjustment value of the preset fixing parameter according to the fixing time difference value, and updating the preset spinning parameter and the preset fixing parameter; updating the predicted die fixing time based on the updated preset spinning parameters and the preset die fixing parameters, and judging whether the updated predicted die fixing time meets the preset die fixing condition; repeating the steps until the updated predicted shaping time meets the preset shaping condition, and determining the parameter adjustment value according to the predicted adjustment value meeting the preset shaping condition. For more on determining the spinning parameters and the solids parameters, see fig. 5 and its associated description.

The fiber quality determination module 220 may be configured to determine a predicted fiber quality of the filter material based on the dope parameters, the spin parameters, and the solids parameters. For more on predicting fiber quality see fig. 3 and its associated description.

In some embodiments, the fiber quality determination module 220 may be further configured to: and processing the stock solution parameters, the spinning parameters and the solid type parameters based on a prediction model, and determining the predicted fiber quality of the filter material, wherein the prediction model is a machine learning model. For more on the predictive model see fig. 4 and its associated description.

The pass rate determination module 230 may be configured to determine a pass rate of the filter material based on the predicted fiber quality. For more details on yield, see FIG. 3 and its associated description.

The early warning module 240 may be configured to generate early warning information based on the qualification rate to alert a user to perform parameter adjustment. For more information about the pre-warning information see fig. 3 and its associated description.

Some embodiments of the present disclosure also provide an apparatus for producing a multi-functional filter material, the apparatus comprising a processor and a memory; the memory is configured to store instructions that, when executed by the processor, cause the apparatus to perform operations corresponding to the method of producing a multifunctional filter material according to any of the embodiments of the present specification.

Some embodiments of the present description also provide a computer-readable storage medium storing computer instructions that, when read by a computer in the storage medium, the computer performs a method of producing a multifunctional filter material.

It should be noted that the above description of the production system of the multifunctional filter material and the modules thereof is for convenience of description only and is not intended to limit the present description to the scope of the illustrated embodiments. It will be appreciated by those skilled in the art that, given the principles of the system, various modules may be combined arbitrarily or a subsystem may be constructed in connection with other modules without departing from such principles. In some embodiments, the parameter acquisition module 210, the fiber quality determination module 220, the qualification rate determination module 230, and the pre-warning module 240 disclosed in fig. 2 may be different modules in a system, or may be one module to implement the functions of two or more modules. For example, each module may share one memory module, or each module may have a respective memory module. Such variations are within the scope of the present description.

Fig. 3 is an exemplary flow chart of a method of producing a multifunctional filter material according to some embodiments of the present description. In some embodiments, the process 300 may be performed by the production system 200 of the multifunctional filter material.

As shown in fig. 3, the process 300 may include the steps of:

step 310, obtaining the stock solution parameters, spinning parameters and solid parameters of the filter material. In some embodiments, step 310 may be performed by parameter acquisition module 210.

In some embodiments, the filter material may refer to a material made of synthetic fibers for processing filtration. For example, the filter material may include filter cloth, mask core, and the like. The filter materials referred to in this specification may include synthetic fibers produced in a wet spinning process. For example, the filter material may be a filter material made of Polytetrafluoroethylene (PTFE) fibers.

The dope parameters may refer to parameters of the filter material related to the dope in the wet spinning process. The spinning solution is a solution with certain composition, certain viscosity and good spinnability, wherein the spinning solution is prepared by dissolving a fiber-forming polymer (such as PTFE) of synthetic fibers in a proper solvent (namely an organic solvent used for dispersing the fiber-forming polymer). The parameters of the stock solution can comprise parameters such as the temperature of the stock solution, the concentration of the stock solution, the proportion of the solvent and the solute, and the like.

In some embodiments, the staff member may preset the stock solution information. For example, a worker may input stock solution information to the storage device 130 through the user terminal 140 (e.g., an operation panel). When the parameter acquisition module 210 needs the parameter, it may be invoked directly from the storage device 130.

The spinning parameters may refer to parameters related to the spinning dope in the spinning process. That is, the spinning dope can be fed into the spinneret under a certain pressure and ejected under the metering of the metering pump, thereby forming an unshaped fiber. The spinning parameters can include spinning hole diameter, spinning speed, swelling ratio and the like.

In some embodiments, the worker may preset the spinning parameters. For example, a worker may input preset spinning parameters to the storage device 130 via the user terminal 140 (e.g., an operation panel). When the parameter acquisition module 210 needs the parameter, the spinning parameter may be preset directly from the storage device 130.

In some embodiments, the spin parameters may also be adjusted based on the solids condition. For example, when the fixing time period is different from the preset fixing condition, the spinning parameters may be adjusted differently. For more description of determining the spin parameters, see FIG. 5 and its associated disclosure.

The solidification parameters can refer to the relevant parameters in the process of resolidifying the spinning dope after spinning. That is, after the spinning dope is sprayed out through the spinneret, the dope can be coagulated in a solid-type liquid, and in the coagulation process, a solvent in the dope trickles is diffused to the solid-type liquid, and a coagulant in the solid-type liquid is permeated to the trickles, so that the dope trickles reach a critical concentration, and the dope trickles are precipitated in a coagulation bath to form fibers. Wherein, the solid type parameters can comprise relevant parameters such as solid type liquid amount, concentration, temperature and the like. The solvent in the solid-liquid is the same as the solvent in the spinning dope.

In some embodiments, the staff may preset the fixation parameters. For example, a worker may input preset fixed parameters to the storage device 130 through the user terminal 140 (e.g., an operation panel). When the parameter acquisition module 210 needs the parameter, the fixed type parameter may be preset directly from the storage device 130.

In some embodiments, the fixation parameters may also be adjusted based on the fixation conditions. For example, when the fixation time period is different from the preset condition, the fixation parameters may be adjusted differently. For more description of determining the fixation parameters, see fig. 5 and its related content.

Step 320, determining predicted fiber quality of the filter material based on the dope parameters, the spinning parameters, and the solids parameters. In some embodiments, step 320 may be performed by the fiber quality determination module 220.

The fiber quality may be an evaluation of the fiber after coagulation. For example, fiber quality may be characterized by linear density, uniformity, length, and like related parameters. Predicting fiber quality may refer to predicting fiber quality after the coagulation process based on dope parameters, spinning parameters, and die set parameters.

In some embodiments, the predicted fiber quality may be determined based on historical production data. For example, the historical dope parameters, the historical spinning parameters and the historical die set parameters during the historical production can be recorded, the actual fiber pattern can be obtained through the image sensor, and the fiber quality can be detected based on the image recognition system. Thereby forming the corresponding relation between each historical fiber quality and the historical stock solution parameter, the historical spinning parameter and the historical solid-state parameter. When the prediction is performed, the historical fiber quality with the closest calling condition can be used as the predicted fiber quality based on the corresponding relation.

In some embodiments, the fiber quality determination module 220 may also train a machine learning model based on the historical production data and determine a predicted fiber quality based on the machine learning model. The predicted fiber quality of the filter material can be determined based on the stock solution parameters, the spinning parameters and the solid type parameters, wherein the prediction model is a machine learning model. For more on the predictive model see fig. 4 and its associated description.

Step 330, determining the pass rate of the filter material based on the predicted fiber quality. In some embodiments, step 330 may be performed by the yield determination module 230.

The qualification rate may refer to the qualification rate of the filter material after the filter material has been processed. I.e. the qualification rate of the filter material obtained by the subsequent processing treatment based on the predicted fiber quality. For example, if the subsequent processing may require quality of the fibers, then the solidified fibers may have unacceptable portions of the filter material during the subsequent processing due to predicted fluctuations in the quality of the fibers that may result in fluctuations in the quality of the filter material.

In some embodiments, the material score of the filter material may be utilized to determine the pass rate of the filter material. I.e. the material score of the filter material can be determined from the predicted fiber quality; and determining the qualification rate of the filter material according to the material score, wherein the material score is positively correlated with the qualification rate. Wherein, the predicted fiber quality can be positively correlated with the material score of the filter material, and when the predicted fiber quality meets the filter fiber processing requirement (e.g., approaches the fiber quality preset by the fiber processing requirement), the higher the material score of the filter material, the higher the corresponding qualification rate.

In some embodiments, the material score of the filter material can be characterized by a first score, a second score, and a third score. Wherein the material score of the filter material may be a weighted sum of the first score, the second score, and the third score.

The first score is directly related to the predicted fiber quality itself, from which the first score may be determined. For example, the first score is positively correlated with the predicted fiber quality. The better the predicted fiber quality, the higher the first score.

In some embodiments, the first score may characterize the effect of a difference in predicted fiber quality from a preset fiber quality on the filter material. The preset fiber quality may refer to a fiber quality required for producing a filter material satisfying a user's demand or a fiber quality preset by a user.

In some embodiments, the first score may be determined based on a first similarity of the predicted fiber quality to a preset fiber quality. Wherein the higher the first similarity, the higher the first score. In some embodiments, the first similarity may be characterized according to a similarity (e.g., 1- | predicted linear density-predicted linear density|/predicted linear density) of each parameter of the predicted fiber quality (e.g., predicted linear density, predicted uniformity, predicted length) to each parameter of the predicted fiber quality (e.g., predicted linear density, predicted uniformity, predetermined length).

The second score, the third score may reflect the effect of the subsequent process parameters on the material quality of the filter material. The subsequent process parameters may refer to the process parameters of the processing process of the fibers of the filter material after the solidification process. For example, the fibers of the filter material may be subjected to an antistatic treatment, an opening process, and a needling process after the coagulation process to obtain a finished product (e.g., PTFE needled filter cloth). The corresponding subsequent process parameters may include at least one of antistatic treatment parameters (such as antistatic treatment time, antistatic treatment mode, etc.), opening parameters (such as opening time), and needling parameters (such as needling times).

In some embodiments, standard subsequent process parameters may be recorded in a second database. The standard subsequent process parameters may refer to the subsequent process parameters that should be used under various production conditions in order to produce acceptable filter materials. In some embodiments, the production conditions may include a combination of various dope parameters, spin parameters, and die parameters. Namely, the second database can record the subsequent process parameters (such as opening parameters, needling parameters and the like) which are needed to be adopted for producing qualified filter materials under the production conditions of different stock solution parameters, spinning parameters and solid parameters.

In some embodiments, when the second database is constructed, different raw liquid parameters, spinning parameters and solid type parameters can be combined to be used as a main key of the second database, and subsequent process parameters in actual production are used as standard subsequent process parameters of the main key. Thereby constructing the corresponding relation between the production conditions (different stock solution parameters, spinning parameters and solid type parameters) and the standard follow-up process parameters.

In some embodiments, the production conditions may also include actual fiber quality produced with different combinations of dope parameters, spin parameters, and die parameters. When the second database is constructed, the actual fiber quality produced under the combination of different stock solution parameters, spinning parameters and solid parameters is recorded as data matched with the main key. Thereby recording the corresponding relation between the actual fiber quality and the standard follow-up technological parameters.

The second score may reflect the predicted impact of fiber quality on subsequent processing. In some embodiments, the second score may characterize the effect of a difference in the first standard subsequent process parameter from the preset subsequent process parameter on the filter material. Where the first standard subsequent process may refer to the subsequent process parameters that should be taken to produce a qualified filter material if the predicted fiber quality is taken as production conditions. The preset subsequent process parameters may refer to subsequent process parameters required for producing a filter material satisfying the user's requirements or subsequent process parameters preset by the user. In some embodiments, the first standard follow-up process parameter may be queried from the second database based on the predicted fiber quality and by a correspondence of the actual fiber quality to the standard follow-up process parameter.

In some embodiments, the second score may be determined based on a second similarity of the first standard subsequent process parameter to the predetermined subsequent process parameter. Wherein the higher the second similarity, the higher the second score. In some embodiments, the second similarity may be characterized according to a similarity (e.g., 1- | first standard opening parameter-preset opening parameter|/preset opening parameter) of each of the first standard subsequent process parameters (e.g., first standard opening parameter, first standard needling parameter, etc.) to each of the preset subsequent process parameters (e.g., preset opening parameter, etc.).

The third score may reflect the effect of the production conditions (stock solution parameters, spinning parameters, and solids parameters) on the subsequent processing process. In some embodiments, the third score may characterize the effect of a difference in the second standard subsequent process parameter from the preset subsequent process parameter on the filter material. The second standard subsequent process may refer to a subsequent process parameter that should be adopted for producing qualified filter materials under the condition that the raw liquid parameter, the spinning parameter and the solid type parameter are taken as production conditions. In some embodiments, the second standard subsequent process parameters may be queried from the second database based on the stock solution parameters, the spinning parameters, and the solids type parameters, and by correspondence of the production conditions (combination of each stock solution parameter, spinning parameter, and solids type parameter) to the standard subsequent process parameters.

In some embodiments, the third score may be determined based on a second similarity of the second standard subsequent process parameter to the predetermined subsequent process parameter. Wherein the higher the third similarity, the higher the third score. In some embodiments, the third similarity may be characterized according to the similarity (e.g., 1- | second standard opening parameter-preset opening parameter|/preset opening parameter) of each of the second standard subsequent process parameters (e.g., second standard opening parameter, second standard needling parameter, etc.) to each of the preset subsequent process parameters (e.g., preset opening parameter, etc.).

In some embodiments, when determining the material score of the filter material based on the first score, the second score, and the third score, the material score may be determined based on the weight of the first score (e.g., 0.4), the weight of the second score (e.g., 0.3), and the weight of the third score (e.g., 0.3). For example, the material score of the filter material = 0.4 x first score +0.3 x second score +0.3 x third score. In some embodiments, the weight of the score may be determined from the actual impact of the impact factors characterized by the respective scores on the filter material. For example, the first score characterizes a factor that affects fiber quality severely, which affects filter material quality, so the first score may be weighted higher (e.g., 0.4). In some embodiments, the weights of the individual scores may be preset by the staff based on historical production experience. In some embodiments, the yield may be converted by a material score of the filter material. For example, the material score of the filter material may be converted to a percent form (e.g., 100% of the material score of the filter material/upper material score limit) and used as a pass rate for the filter material.

In some embodiments, the yield may also be related to actual production. For example, the processing engineering of each process link can be detected in real time during production, and the qualification rate is adjusted according to the difference between the processing result of each link and the preset/predicted parameters. For example, the actual fiber quality may be detected in real time and compared to the predicted fiber quality, and the yield may be adjusted or recalculated based on the comparison of the predicted fiber quality to the actual fiber quality. For example, when the actual fiber quality actually detected differs greatly from the predicted fiber quality, the yield may be lowered based on the difference value. For another example, when the actual detected fiber quality and the predicted fiber quality differ greatly, the actual fiber quality may be used as the predicted fiber quality to recalculate the second score, and the qualification rate may be updated based on the updated second score.

And 340, generating early warning information based on the qualification rate to remind a user to adjust parameters.

In some embodiments, when the qualification rate meets the pre-warning condition (e.g., is below a qualification rate threshold), it may be indicated that the filter material produced based on the current process parameters fails to meet the user's expectations, and the process parameters should be adjusted.

The alert information may be alert information triggered in response to the yield meeting the alert condition. In some embodiments, the pre-warning information may include specific content of the current processing recipe, qualification rate, etc. related information reflecting current processing process parameters. In some embodiments, the pre-warning information may also include advice on parameter adjustments. The parameter adjustment can comprise adjustment of parameters such as a stock solution parameter, a spinning parameter, a solid type parameter and the like and adjustment of subsequent process parameters.

Based on the production method of the multifunctional filter material provided by the specification, the fiber quality can be predicted by analyzing the stock solution parameter, the spinning parameter and the solid parameter, and the qualification rate of the filter material can be estimated. Therefore, whether each process parameter is proper or not can be determined before the production of the filter material, and the trial-and-error cost of the production of the filter material is reduced. In addition, early warning can be performed in time in the production process of the filter material, and the process parameters are adjusted, so that the error correction capability and the adaptability in the production process of the filter material are improved.

FIG. 4 is a schematic illustration of a predictive model shown in accordance with some embodiments of the present description.

As shown in fig. 4, the prediction model 400 may include a first embedded layer 410, a second embedded layer 420, a third embedded layer 430, and a first prediction layer 440. The predictive model 400 may process, among other things, the dope parameters, the spin parameters, and the solids parameters to determine a predicted fiber quality for the filter material.

As shown in fig. 4, dope parameters, spinning parameters, and solid parameters may be input into a predictive model 400 to obtain a predicted fiber quality. The dope parameters are input into the first embedding layer 410, and the dope characteristics are obtained through the treatment of the first embedding layer 410. The spinning parameters and the characteristics of the stock solution are input into the second embedded layer 420, and the spinning characteristics are obtained through the treatment of the second embedded layer 420. The solid type parameters, the spinning characteristics and the stock solution characteristics are input into the third embedded layer 430, and the solid type characteristics are determined through the treatment of the third embedded layer 430. The solid type characteristic, the spinning characteristic and the dope characteristic are input into the first prediction layer 440, and the predicted fiber quality is obtained through the treatment of the first prediction layer 440.

The first embedding layer 410 may be used to process the dope parameters and extract feature data in the dope parameters to generate dope features. The input of the first embedding layer may be a raw liquid parameter, and the output may be a raw liquid characteristic. The stock solution features may be normalized feature vectors. For example, a dope feature may include a plurality of elements, and the element value of each element may reflect the specifics of the dope parameters. Illustratively, the characteristics of the stock solution may include four elements of solute type, solvent type, temperature and concentration, and the actual condition of the corresponding parameters may be characterized by the corresponding element values. For example, the solute type may be characterized (e.g., 1) according to the number of the actual solute (e.g., PTFE) in the stock solution in a preset list of solutes.

The second inlay 420 may be used to process the dope characteristics as well as the spin parameters to determine the spin characteristics. Wherein the input of the second embedding layer 420 may be the dope characteristics and the spinning parameters, and the output may be the spinning characteristics. The spin feature may be a normalized feature vector. For example, the spinning characteristics may include a plurality of elements, and the element value of each element may reflect a particular processing instance of the spinning process. Illustratively, the spinning characteristic may include two elements, such as a spinning aperture and a spinning speed, and the actual condition of the corresponding parameter may be characterized by the corresponding element value. For example, the spinneret aperture may be characterized by the diameter size of the exit aperture in the spray head.

The third embedded layer 430 may be used to process dope characteristics, spin characteristics, and solids parameters to determine solids characteristics. The third embedding layer 430 may have a raw liquid characteristic, a spinning characteristic, and a solid-type parameter as input, and a solid-type characteristic as output. The fixed features may be normalized feature vectors. For example, the solid-type feature may include a plurality of elements, and the element value of each element may reflect a particular processing instance of the solid-type process. For example, the solid-type characteristic may include three elements of solid-type agent type, solid-type liquid temperature, and solid-type liquid concentration, and the actual condition of the corresponding parameter may be characterized by the corresponding element value. For example, the solid-liquid concentration may be characterized by the ratio of solid-forming agent to solution in the solid-liquid.

In some embodiments, the first, second, and third embedded layers 410, 420, 430 may be feature extraction models. For example, the first, second, and third embedded layers 410, 420, 430 may be deep neural networks (Deep Neural Networks, DNN) that may process corresponding process parameters to generate corresponding feature vectors.

The first predictive layer 440 may be used to process the dope characteristics, the spin characteristics, and the solid characteristics to determine a predicted fiber quality. The input of the first prediction layer 440 may be a dope characteristic, a spinning characteristic, and a solid characteristic, and the output may be a predicted fiber quality. For more details on predicting fiber quality, see the relevant description of step 320.

In some embodiments, first prediction layer 440 may be a feature selection model (e.g., a probabilistic predictive model, a classifier, a feature selector, etc.). The predicted fiber quality can be determined by feature selection according to the solid features, the spinning features and the dope features.

In some embodiments, considering that the first, second, and third embedded layers 410, 420, and 430 do not have direct characterizations capabilities themselves, the first, second, and third embedded layers 410, 420, and 430 are trained in conjunction with the first prediction layer 440 such that the feature vectors output thereby meet the computational requirements of the first prediction layer 440.

In some embodiments, the sample data of the joint training includes historical dope parameters, historical spinning parameters, and historical die set parameters in the historical production data, with the label being the historical fiber quality at the sample parameters. During training, the historical dope parameters are input into the first embedding layer 410 to obtain the historical dope characteristics and serve as training sample data, the historical spinning parameters are input into the second embedding layer 420 to obtain the historical spinning characteristics and serve as training sample data, the historical dope characteristics and the historical solid parameters are input into the third embedding layer 430 to obtain the historical solid characteristics and serve as training sample data, and the historical dope characteristics and the historical solid characteristics are input into the first prediction layer 440 to output predicted fiber quality. The parameters of the first, second, third and first prediction layers 410, 420, 430 are synchronously updated based on the historical fiber quality at the sample parameters and the price of the predicted fiber quality output by the first prediction layer 440 to construct a loss function. Through the parameter updating, a trained first embedded layer 410, second embedded layer 420, third embedded layer 430, and first prediction layer 440 are obtained.

As shown in fig. 4, the predictive model 400 also includes a second predictive layer 450. The second prediction model 450 may be connected to the output layers of the second embedded layer 420 and the third embedded layer 430, obtain the spinning characteristics output by the second embedded layer 420 and the solid characteristics output by the third embedded layer 430, and determine the predicted solid duration based on the solid characteristics and the spinning characteristics. The predicted duration of the die bonding process may refer to the predicted duration of execution of the die bonding process (e.g., 10 minutes) in the current environment (i.e., spinning parameters and die bonding parameters).

The second prediction layer 450 may be used to process the spinning characteristics as well as the solids characteristics to determine a predicted solids duration. The input of the second prediction layer 450 is the spinning characteristic and the solid type characteristic, and the output is the predicted solid type duration.

In some embodiments, the second prediction layer 450 may be a feature selection model (e.g., a probabilistic predictive model, a classifier, a feature selector, etc.). The predicted solid type duration can be determined by feature selection according to solid type features and spinning features.

In some embodiments, the second prediction layer 450 may be trained in conjunction with the first, second, and third embedded layers 410, 420, 430. The sample label can be a historical fixed time length under the sample parameter. The joint training process of the second prediction layer 450 is similar to that of the first prediction layer 440, and will not be described again.

In some embodiments, first prediction layer 440 and second prediction layer 450 may be co-trained. Then, correspondingly, a plurality of loss terms are included in the loss function involved in the training of the predictive model 400, e.g., the loss function includes 3 loss terms. I.e. the loss function may comprise a first loss term, a second loss term and a third loss term. The first loss term corresponds to the error of the first prediction layer, the second loss term corresponds to the error of the second prediction layer, and the third loss term is determined based on the first loss term and the second loss term and used for representing the error of each embedded layer. In actual training, the parameters of the first prediction layer 440, the parameters of the second prediction layer 450, and the parameters of the first, second, and third embedding layers 410, 420, and 430 may be updated based on the first loss term, the second loss term, and the third loss term.

In some embodiments, the third penalty term may be a sum of weights of the first penalty term and the second penalty term, i.e., the third penalty term is a sum of products of the first penalty term and the second penalty term and their reference weights. The reference weight refers to the relative importance of the parameter error characterized by the corresponding loss term in the overall loss function.

In some embodiments, the reference weights for the first and second penalty terms may be empirically preset or determined based on other factors. For example, the weight may be set according to the training sample data amount, and the larger the sample data amount, the larger the reference weight.

In some embodiments, the reference weights may also be calculated based on the contribution/importance of the individual outputs in subsequent calculations. For example, when the predicted fixed time length can be used for subsequent adjustment parameters, the predicted fixed time length has higher importance in subsequent calculation, and the reference weight of the second loss term can be adjusted to enable the features determined by the embedded layer to better meet the parameter requirements of the predicted fixed time length, and the accuracy of the predicted fixed time length is improved. I.e. the reference weight of the first penalty term is smaller than the reference weight of the second penalty term.

Based on the prediction model provided by some embodiments of the present disclosure, the predicted fiber quality and the predicted solid-form duration may be determined based on a machine learning algorithm, so that the calculation accuracy of the predicted fiber quality and the predicted solid-form duration is improved. In addition, the model corresponding to the predicted fiber quality and the predicted solid-state duration can be determined based on joint learning, so that the relevance among the models is improved, the full utilization of training data is realized, and the calculation accuracy of the predicted fiber quality and the predicted solid-state duration is further improved.

FIG. 5 is an exemplary flow chart of a method of determining spinneret parameters and die set parameters according to some embodiments of the present description. In some embodiments, the process 500 may be performed by the parameter acquisition module 210.

As shown in fig. 5, the process 500 may include the steps of:

step 510, obtaining a stock solution parameter, a preset spinning parameter and a preset solid type parameter of the filter material. The preset spinning parameters and the preset fixing parameters can be initial values of parameters set by staff according to past experience. For more details of dope parameters, spinning parameters, and solids parameters, see step 310 and the description thereof.

Step 520, predicting the predicted solidification duration of the filter material in the solidification process according to the preset spinning parameters and the preset solidification parameters.

The predicted solid-form duration may refer to the solidification duration of the fiber during the solidification process, i.e., after the solidification process of the predicted solid-form duration, the fiber is formed.

In some embodiments, the preset spinning parameters and the predicted die-fixing time periods corresponding to the preset die-fixing parameters can be retrieved according to the historical data. And when the filter material is qualified, each spinning parameter, the fixed type parameter and the historical fixed type duration construct a first database. The first database includes a correspondence between a reference solid-state duration and a first reference process parameter (such as a stock solution parameter, a spinning parameter, and a solid-state parameter). For example, the first reference process parameter may be used as a primary key of the first database, and the reference solid-state time length when the filter material is qualified may be generated according to the filter material as corresponding data of the primary key. And in determining the predicted solid-form duration, the corresponding solid-form duration can be searched from the first database according to the preset spinning parameters and the preset solid-form parameters to serve as the predicted solid-form duration.

In some embodiments, the retrieval may also be performed in the first database according to the preset spinning parameters and the vector similarity (or vector distance) of the preset die parameters to the respective reference process parameters. For example, a preset parameter vector is constructed according to the preset spinning parameters and the preset fixed parameters, a reference parameter vector is constructed according to the reference process parameters in the first database, and a vector distance (e.g., euclidean distance x) or a vector similarity (e.g., 1/(1+x) between the preset parameter vector and each reference process parameter is calculated during searching ² ) And determining the predicted solid time length according to the reference process parameters with the nearest vector distance or the highest vector similarity.

In some embodiments, the predicted solids time period may also be determined based on a predictive model. For more on the predictive model see fig. 4 and its associated description.

Step 530, determining whether the predicted fixed time length satisfies a preset fixed condition.

The preset fixation condition may be a fixation condition desired by a user. For example, the preset-type condition may include a preset-type time range allowed by the user. Judging whether the predicted solid-type duration meets the preset solid-type condition or not, wherein the judging can be characterized as judging whether the predicted solid-type duration is within the preset solid-type time range or not. If yes, the preset fixation condition is satisfied. If not, the preset fixed condition is not satisfied.

Step 540, if the predicted shaping duration satisfies the preset shaping condition, taking the preset spinning parameters and the preset shaping parameters satisfying the preset shaping condition as the spinning parameters and the shaping parameters.

If the predicted die-fixing time period does not meet the preset die-fixing condition, determining a parameter adjustment value of the preset spinning parameter and/or the preset die-fixing parameter according to the predicted die-fixing time period and the preset die-fixing condition, and adjusting the preset spinning parameter and/or the preset die-fixing parameter according to the parameter adjustment value to determine the spinning parameter and the die-fixing parameter, in step 550.

The parameter adjustment value may reflect the adjustment amplitude for each specific parameter in the preset spinning parameters and/or the preset solids parameters. For example, the parameter adjustment value may include an adjustment value for the solid-state liquid temperature or concentration. In some embodiments, the measured duration determined based on the preset spinning parameter and/or the preset die-fixing parameter adjusted by the parameter adjustment value may satisfy the preset die-fixing condition, and be used as the spinning parameter and the die-fixing parameter.

In some embodiments, the preset spinning parameters and/or the preset solids parameters may be adjusted multiple times, and the parameter adjustment values may be characterized as a summation of the adjustment values over the multiple adjustments. Wherein each adjustment can be considered an iterative process. For example, the parameter adjustment value may include 2 adjustments, wherein in the first adjustment, the solid-liquid temperature increases by 5 ℃ and the spinning speed decreases by 1m/s. In the second adjustment, the solid-liquid temperature was raised by 2℃and the spinning speed was raised by 1m/s. The parameter adjustment value can be characterized as a 7 c (5 c +2 c) rise in solid-liquid temperature with unchanged spinning speed.

As shown in fig. 5, for a certain iteration process, the following steps may be included:

and 551, determining a fixed time difference value according to a preset fixed time range and a predicted fixed time in a preset fixed condition. The predicted fixed time length in this step may refer to the predicted fixed time length after the previous iteration.

The difference in the fixed time may reflect a difference between the predicted fixed time length and a certain time length in a preset fixed time length range (e.g., a median value of the preset fixed time length range, an average time length of the filtering material in the history processing, etc.). For example, when the predicted solid time period is 1min20s and the preset solid time period is 50s-1min10s, the solid time difference may be 20s, which is the difference between the predicted solid time period and the intermediate value (1 min) of the preset solid time period.

Step 552, determining a preset spinning parameter and/or a predicted adjustment value of the preset die-fixing parameter according to the die-fixing time difference, and updating the preset spinning parameter and the preset die-fixing parameter.

In some embodiments, according to the historical data (e.g., the first database), the reference process parameters (e.g., the reference spinning parameter and the reference solid-state parameter) that match the solid-state duration difference and the predicted solid-state duration are found from the historical data, the parameter differences between the reference spinning parameter and the reference solid-state parameter and the preset spinning parameter and the preset solid-state parameter are used as the predicted adjustment values, and the reference spinning parameter and the reference solid-state parameter are used as the preset spinning parameter and the preset solid-state parameter after the present iteration. The reference process parameters matched with the predicted solid-state duration may refer to reference process parameters that are the same as or similar to other processes (e.g., the same as the original liquid parameters). For example, when the difference of the die set time is 20s, each process parameter with the die set time length of 1min can be searched from the historical data, then the reference process parameters with the same original liquid parameter and the like are determined, and the difference between each parameter (such as 65 ℃ C. Of the die set temperature) in the reference process parameters and the preset spinning parameter (such as 60 ℃ C. Of the preset die set temperature) is taken as a parameter adjustment value (5 ℃ C. Of the rising die set temperature).

In some embodiments, the actual die-retention time period change caused by the spinning parameter and/or the die-retention parameter can be adjusted according to the historical data, and a preset relation table is constructed. And filling corresponding data of the actual fixed time length change serving as a main key in a database or a data table by taking the adjustment value of the spinning parameter and/or the fixed parameter as the main key of the preset relation table so as to generate the preset relation table. For example, the parameters of the spinning parameters and/or the die fixing parameters (such as the die fixing temperature is reduced by 2 ℃ each time) can be gradually reduced by a preset step length, so as to generate a plurality of groups of adjustments of the spinning parameters and/or the die fixing parameters, which are used as the main keys of the preset relation table, then the actual die fixing time length is detected, and the data of the corresponding main keys (such as the change of the die fixing time length caused by the reduction of 2 ℃ each time) are filled according to the corresponding actual die fixing time length change, so as to form the preset relation table.

In some embodiments, when the adjustment value is determined based on the preset relationship table, the adjustment value of at least one set of preset spinning parameters and/or preset die-fixing parameters corresponding to the actual die-fixing time period change may be determined from the preset relationship table according to the matching relationship between the die-fixing time period difference value and the actual die-fixing time period change in the preset relationship table.

Step 553, updating the predicted die-fixing time based on the updated preset spinning parameters and the preset die-fixing parameters, and judging whether the updated predicted die-fixing time meets the preset die-fixing condition.

In some embodiments, updated preset spinning parameters and preset die-fixing parameters may be input into the prediction model to determine updated predicted die-fixing time periods. For more on determining the predicted solids time based on the pre-set model, see fig. 4 and its associated description.

In some embodiments, step 530 may be re-performed based on the updated predicted fixed time period. And determining a parameter adjustment value according to the predicted adjustment value meeting the preset fixed type condition.

The parameter determination method provided by the application can realize the self-adaptive condition of the parameter based on the important index (such as the preset fixed condition). By analogy, the self-adaptive adjustment of each technological parameter can be realized, and then each controllable parameter in the production process of the filter material can be automatically determined, so that the accuracy and the intelligent degree in the production process are improved.

While the basic concepts have been described above, it will be apparent to those skilled in the art that the foregoing detailed disclosure is by way of example only and is not intended to be limiting. Although not explicitly described herein, various modifications, improvements, and adaptations to the present disclosure may occur to one skilled in the art. Such modifications, improvements, and modifications are intended to be suggested within this specification, and therefore, such modifications, improvements, and modifications are intended to be included within the spirit and scope of the exemplary embodiments of the present application.

Meanwhile, the specification uses specific words to describe the embodiments of the specification. Reference to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic is associated with at least one embodiment of the present description. Thus, it should be emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various positions in this specification are not necessarily referring to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of the present description may be combined as suitable.

Furthermore, the order in which the elements and sequences are processed, the use of numerical letters, or other designations in the description are not intended to limit the order in which the processes and methods of the description are performed unless explicitly recited in the claims. While certain presently useful inventive embodiments have been discussed in the foregoing disclosure, by way of various examples, it is to be understood that such details are merely illustrative and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements included within the spirit and scope of the embodiments of the present disclosure. For example, while the system components described above may be implemented by hardware devices, they may also be implemented solely by software solutions, such as installing the described system on an existing server or mobile device.

Likewise, it should be noted that in order to simplify the presentation disclosed in this specification and thereby aid in understanding one or more inventive embodiments, various features are sometimes grouped together in a single embodiment, figure, or description thereof. This method of disclosure, however, is not intended to imply that more features than are presented in the claims are required for the present description. Indeed, less than all of the features of a single embodiment disclosed above.

In some embodiments, numbers describing the components, number of attributes are used, it being understood that such numbers being used in the description of embodiments are modified in some examples by the modifier "about," approximately, "or" substantially. Unless otherwise indicated, "about," "approximately," or "substantially" indicate that the number allows for a 20% variation. Accordingly, in some embodiments, numerical parameters set forth in the specification and claims are approximations that may vary depending upon the desired properties sought to be obtained by the individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and employ a method for preserving the general number of digits. Although the numerical ranges and parameters set forth herein are approximations that may be employed in some embodiments to confirm the breadth of the range, in particular embodiments, the setting of such numerical values is as precise as possible.

Each patent, patent application publication, and other material, such as articles, books, specifications, publications, documents, etc., referred to in this specification is incorporated herein by reference in its entirety. Except for application history documents that are inconsistent or conflicting with the content of this specification, documents that are currently or later attached to this specification in which the broadest scope of the claims to this specification is limited are also. It is noted that, if the description, definition, and/or use of a term in an attached material in this specification does not conform to or conflict with what is described in this specification, the description, definition, and/or use of the term in this specification controls.

Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments of this specification. Other variations are possible within the scope of this description. Thus, by way of example, and not limitation, alternative configurations of embodiments of the present specification may be considered as consistent with the teachings of the present specification. Accordingly, the embodiments of the present specification are not limited to only the embodiments explicitly described and depicted in the present specification.

Claims

1. A method of producing a multifunctional filter material, the method comprising:

Acquiring stock solution parameters, spinning parameters and solid parameters of the filter material; the method comprises the following steps:

acquiring the stock solution parameters, preset spinning parameters and preset solid type parameters of the filter material;

according to the preset spinning parameters and the preset solidifying parameters, predicting the predicted solidifying time of the filter material in the solidification process based on a prediction model;

the prediction model comprises a first embedded layer, a second embedded layer, a third embedded layer and a second prediction layer; the input of the first embedded layer is the original liquid parameter, and the output is the original liquid characteristic; the input of the second embedded layer is the characteristics of the stock solution and the preset spinning parameters, and the output is the spinning characteristics; the input of the third embedded layer is the characteristics of the stock solution, the spinning characteristics and the preset solid type parameters, and the output is the solid type characteristics; the input of the second prediction layer comprises the spinning characteristic and the solid type characteristic, and the output is the predicted solid type duration; the first embedded layer, the second embedded layer and the third embedded layer are feature extraction models, and the second prediction layer is a feature selection model;

judging whether the predicted type fixing time length meets a preset type fixing condition or not;

If yes, taking the preset spinning parameters and the preset shaping parameters which meet the preset shaping conditions as the spinning parameters and the shaping parameters;

if not, determining a parameter adjustment value of the preset spinning parameter and/or the preset die-fixing parameter according to the predicted die-fixing time length and the preset die-fixing condition, and adjusting the preset spinning parameter and/or the preset die-fixing parameter according to the parameter adjustment value to determine the spinning parameter and the die-fixing parameter;

wherein the determining the preset spinning parameters and/or the parameter adjustment values of the preset die fixing parameters according to the predicted die fixing time length and the preset die fixing conditions includes:

determining a fixed time length difference value according to a preset fixed time length range in the preset fixed conditions and the predicted fixed time length;

determining the preset spinning parameters and/or the predicted adjustment values of the preset solid type parameters according to the solid type duration difference value, and updating the preset spinning parameters and the preset solid type parameters;

inputting the updated preset spinning parameters and the preset fixed type parameters into the prediction model, determining the updated predicted fixed type duration based on the output result of the prediction model, and judging whether the updated predicted fixed type duration meets the preset fixed type condition;

Repeating the steps until the updated predicted shaping time meets the preset shaping condition, and determining the parameter adjustment value according to the predicted adjustment value meeting the preset shaping condition;

determining a predicted fiber quality of the filter material according to the stock solution parameter, the spinning parameter and the solid-state parameter;

determining a pass rate of the filter material based on the predicted fiber quality;

and generating early warning information based on the qualification rate so as to remind a user to carry out parameter adjustment.

2. The method of claim 1, wherein said determining the predicted fiber quality of the filter material based on the dope parameters, the spinneret parameters, and the die set parameters comprises:

and processing the stock solution parameter, the spinning parameter and the solid type parameter based on the prediction model to determine the predicted fiber quality of the filter material, wherein the prediction model is a machine learning model.

3. A system for producing a multi-functional filter material, the system comprising:

the parameter acquisition module is used for acquiring stock solution parameters, spinning parameters and solid parameters of the filter material; the method comprises the following steps:

the prediction model comprises a first embedded layer, a second embedded layer, a third embedded layer and a second prediction layer; the input of the first embedded layer is the original liquid parameter, and the output is the original liquid characteristic;

the input of the second embedded layer is the characteristics of the stock solution and the preset spinning parameters, and the output is the spinning characteristics; the input of the third embedded layer is the characteristics of the stock solution, the spinning characteristics and the preset solid type parameters, and the output is the solid type characteristics; the inputs to the second predictive layer include the spin signature and

the shape fixing characteristic is output as a predicted shape fixing time length; the first embedded layer, the second embedded layer,

The third embedded layer is a feature extraction model, and the second prediction layer is a feature selection model;

the fiber quality determining module is used for determining the predicted fiber quality of the filter material according to the stock solution parameter, the spinning parameter and the solid-state parameter;

a pass rate determination module for determining a pass rate of the filter material based on the predicted fiber quality;

and the early warning module is used for generating early warning information based on the qualification rate so as to remind a user to carry out parameter adjustment.

4. The system of claim 3, wherein the fiber quality determination module is further to:

and processing the stock solution parameter, the spinning parameter and the solid type parameter based on a prediction model, and determining the predicted fiber quality of the filter material, wherein the prediction model is a machine learning model.

5. A production apparatus for a multifunctional filter material, the apparatus comprising a processor and a memory; the memory is configured to store instructions that, when executed by the processor, cause the apparatus to perform operations corresponding to the method of producing a multifunctional filter material according to any one of claims 1 to 2.

6. A computer readable storage medium storing computer instructions which, when read by a computer in the storage medium, operate the method of producing a multifunctional filter material according to any one of claims 1 to 2.