Curtain type membrane element preparation method and system based on artificial intelligence

Abstract

The invention provides a curtain type membrane element preparation method and system based on artificial intelligence, and the method comprises the steps: controlling a sieving device to sieve a material based on sieving parameters to obtain a dispersed material; controlling mixing equipment to mix the dispersed material and a lubricant based on the mixing parameters to obtain a mixed material; according to the screening parameters and the material mixing parameters, forming parameters of the mixed materials are determined based on a pre-trained first prediction model; controlling extrusion equipment to perform pre-forming treatment on the mixed material according to the forming parameters to obtain an extrusion-formed hollow fiber membrane; according to the forming parameters, on the basis of a pre-trained second prediction model, sintering parameters of the hollow fiber membrane are determined; and controlling sintering equipment to perform sintering treatment on the hollow fiber membrane based on the sintering parameters to obtain the hollow fiber curtain type membrane element. The invention relates to the technical field of intelligent manufacturing, and the preparation efficiency of the curtain type membrane element can be improved.

Description

Technical Field

[0001] This application relates to the field of intelligent manufacturing technology, and in particular, to a method and system for preparing a curtain membrane element based on artificial intelligence. Background Art

[0002] At present, with the development of data science, some enterprises or organizations tend to improve productivity and the efficiency of production equipment maintenance through data analysis in the production environment. In the process of preparing a curtain membrane element, a screening device and a mixing device are usually used to preprocess the materials, and an extrusion device is used to process the preprocessed materials to obtain a hollow fiber curtain membrane element. In order to improve the preparation efficiency, a preparation strategy is usually formulated based on digital control technology at present. However, this method has insufficient adaptability to the dynamically changing production parameters during the preparation process, and is likely to cause the problem of reduced preparation efficiency. Summary of the Invention

[0003] In view of the above, it is necessary to propose a method and system for preparing a curtain membrane element based on artificial intelligence to solve the technical problems of low finished product consistency and low preparation efficiency of the hollow fiber curtain membrane element.

[0004] This application provides a method for preparing a curtain membrane element based on artificial intelligence, which is applied to an electronic device. The electronic device is communicatively connected to a screening device, a mixing device, an extrusion device, and a sintering device. The method includes: controlling the screening device to perform a screening process on the materials based on preset screening parameters to obtain dispersed materials; controlling the mixing device to perform a mixing process on the dispersed materials and a lubricant based on preset mixing parameters to obtain mixed materials; determining forming parameters of the mixed materials based on a pre-trained first prediction model according to the screening parameters and the mixing parameters; controlling the extrusion device to perform a pre-forming process on the mixed materials according to the forming parameters to obtain an extruded hollow fiber membrane; determining sintering parameters of the hollow fiber membrane based on a pre-trained second prediction model according to the forming parameters; and controlling the sintering device to perform a sintering process on the hollow fiber membrane based on the sintering parameters to obtain a hollow fiber curtain membrane element.

[0005] In some embodiments, the method further includes training the first prediction model, and the training of the first prediction model includes: obtaining multiple batches of first production data, second production data, and first label data; wherein, the first production data includes the sieving temperature of the sieving device, the mesh aperture of the sieving device, the mixing temperature of the mixing device, the aging temperature of the mixing device, and the mixing rotation speed of the mixing device; the second production data includes the extrusion temperature of the extrusion device and the extrusion speed of the extrusion device, and the first label data includes the cone angle, compression ratio, aspect ratio, and porosity of the hollow fiber membrane; the first label data is used to indicate the performance of the hollow fiber membrane corresponding to the second production data of each batch; determining a first reward value corresponding to the second production data of any batch according to the first label data of any batch; inputting the first production data of any batch into a pre-constructed first initial prediction model to obtain predicted production data output by the first initial prediction model; determining a first loss value of the first initial prediction model based on the first reward value, the predicted production data, and the second production data; updating the first prediction model based on the backpropagation algorithm until the first loss value meets a preset condition, and stopping updating the first prediction model to obtain the first prediction model trained to a convergent state.

[0006] In some embodiments, the determining a first reward value corresponding to the second production data of any batch according to the first label data of any batch includes: performing normalization processing on the first label data of the multiple batches to obtain the normalized first label data of any batch; wherein, the normalized first label data includes the normalized cone angle, normalized compression ratio, normalized aspect ratio, and normalized porosity; determining a first difference between the normalized first label data and a preset first standard data; wherein, the first difference includes the cone angle difference, compression ratio difference, aspect ratio difference, and porosity difference; determining the first reward value corresponding to the second production data of any batch according to a preset first weight parameter and the first difference.

[0007] In some embodiments, the determining the first reward value corresponding to the second production data of any batch according to a preset first weight parameter and the first difference includes: Among them, Reward1 represents the first reward value corresponding to the second production data of any one batch; α represents the first weight in the first weight parameter, and the first weight is used to characterize the influence degree of the cone angle on the first reward value; β represents the second weight in the first weight parameter, and the second weight is used to characterize the influence degree of the compression ratio on the first reward value; γ represents the third weight in the first weight parameter, and the third weight is used to characterize the influence degree of the aspect ratio on the first reward value; θ represents the fourth weight in the first weight parameter, and the fourth weight is used to characterize the influence degree of the porosity on the first reward value.

[0008] In some embodiments, determining the first loss value of the first initial prediction model based on the first reward value, the predicted production data, and the second production data includes: Among them, Loss1 represents the first loss value of the first prediction model; i represents the index of the batch of the second production data, m represents the number of batches of the second production data; j represents the index of the dimension in the second production data, k represents the number of dimensions of the second production data; e represents the natural constant; represents the first reward value corresponding to the second production data of the i-th batch; A ij represents the value of the j-th dimension in the second production data of the i-th batch; B ij represents the value of the j-th dimension in the predicted production data of the i-th batch.

[0009] In some embodiments, the method further includes training the second prediction model. Training the second prediction model includes: obtaining multiple batches of second production data, historical sintering data, and second label data; wherein, the second production data includes the extrusion temperature of the extrusion equipment and the extrusion speed of the extrusion equipment, and the historical sintering data includes the sintering temperature and the sintering time; the second label data includes the filament diameter, flux, pressure difference, and tensile strength of the hollow fiber curtain membrane element; the second label data is used to indicate the performance of the curtain membrane element corresponding to the second production data of each batch; determining the second reward value corresponding to the second production data of any one batch according to the second label data of any one batch; inputting the second production data of any one batch into a pre-constructed second initial prediction model to obtain the predicted sintering data output by the second initial prediction model; determining the second loss value of the second initial prediction model based on the second reward value, the predicted sintering data, and the historical sintering data; updating the second prediction model based on the backpropagation algorithm until the second loss value meets a preset condition, and stopping updating the second prediction model to obtain the second prediction model trained to a converged state.

[0010] In some embodiments, determining the second reward value corresponding to the second production data of any batch according to the second label data of any batch includes: performing normalization processing on the second label data of multiple batches to obtain the normalized second label data of any batch; wherein, the normalized second label data includes normalized membrane filament diameter, normalized flux, normalized pressure difference, and normalized tensile strength; determining a second difference between the normalized second label data and preset second standard data; wherein, the second difference includes diameter difference, flux difference, pressure difference, and tensile strength difference; determining the second reward value corresponding to the second production data of any batch according to the preset second weight parameter and the second difference.

[0011] In some embodiments, determining the second reward value corresponding to the second production data of any batch according to the preset second weight parameter and the second difference includes: Wherein, Reward2 represents the second reward value corresponding to the second production data of any batch; a represents the first weight in the second weight parameter, which is used to indicate the influence degree of the membrane filament diameter on the second reward value; b represents the second weight in the second weight parameter, which is used to characterize the influence degree of the flux on the second reward value; c represents the third weight in the second weight parameter, which is used to characterize the influence degree of the pressure difference on the second reward value; d represents the fourth weight in the second weight parameter, which is used to characterize the influence degree of the tensile strength on the second reward value.

[0012] In some embodiments, determining the second loss value of the second initial prediction model based on the second reward value, the predicted sintering data, and the historical sintering data includes: Wherein, Loss2 represents the second loss value of the second prediction model; i represents the index of the batch of the second production data, and m represents the number of batches of the second production data; x represents the index of the dimension in the historical sintering data, and y represents the number of dimensions of the historical sintering data; e represents the natural constant; represents the second reward value corresponding to the second production data of the i-th batch; U ix represents the value of the x-th dimension in the historical sintering data of the i-th batch; V ix represents the value of the x-th dimension in the predicted sintering data of the i-th batch.

[0013] The embodiment of the present application also provides an artificial intelligence-based preparation system for curtain membrane elements. The system includes an electronic device, which is communicatively connected to a sieving device, a mixing device, an extrusion device, and a sintering device. The electronic device is configured to control the sieving device to perform sieving on materials based on preset sieving parameters to obtain dispersed materials. The electronic device is further configured to control the mixing device to mix the dispersed materials and lubricants based on preset mixing parameters to obtain mixed materials. The electronic device is further configured to determine the forming parameters of the mixed materials based on the preset sieving parameters and mixing parameters and a pre-trained first prediction model. The electronic device is further configured to control the extrusion device to perform pre-forming on the mixed materials according to the forming parameters to obtain an extruded hollow fiber membrane. The electronic device is further configured to determine the sintering parameters of the hollow fiber membrane based on the forming parameters and a pre-trained second prediction model. The electronic device is further configured to control the sintering device to sinter the hollow fiber membrane based on the sintering parameters to obtain a hollow fiber curtain membrane element.

[0014] It can be seen from the above technical solutions that in the embodiment of the present application, the pre-trained first prediction model learns the complex non-linear relationship between the sieving parameters, mixing parameters, and extrusion forming parameters, and the pre-trained second prediction model learns the correlation between the extrusion forming parameters and the sintering parameters, so as to achieve accurate prediction of process parameters. Based on a large amount of historical data and real-time data, the dependence on human experience and the trial-and-error method is reduced, and the accuracy of parameters is improved. It can ensure that the process parameters remain relatively stable and consistent under different batches and production conditions, thereby reducing the performance differences between products and improving the stability and reliability of products. It can also realize real-time monitoring and adjustment of the entire preparation process. When the parameters of a certain step fluctuate, the parameters of subsequent steps can be quickly predicted and adjusted to maintain the stability of the entire process. It can reduce the testing and adjustment process, thereby shortening the production cycle and improving production efficiency. At the same time, due to the improvement of the stability of product performance, the scrap rate and rework rate can be reduced, further reducing costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 FIG. is an application scenario diagram of an artificial intelligence-based preparation method for curtain membrane elements provided by an embodiment of the present application.

[0016] Figure 2 FIG. is a flowchart of an artificial intelligence-based preparation method for curtain membrane elements provided by an embodiment of the present application.

[0017] Figure 3 FIG. is a functional module diagram of an artificial intelligence-based preparation system for curtain membrane elements provided by an embodiment of the present application.

[0018] Figure 4 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0019] In order to more clearly understand the purpose, features and advantages of the present application, the present application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments may be combined with each other. Many specific details are set forth in the following description in order to fully understand the present application. The described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments.

[0020] In addition, the terms "first" and "second" are only used for descriptive purposes, and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present application, "a plurality of" means two or more, unless otherwise specifically defined.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used in the description of this application herein are only for the purpose of describing specific embodiments, and are not intended to limit this application. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0022] An embodiment of the present application provides a method for preparing a curtain membrane element based on artificial intelligence, which can be applied to one or more electronic devices. An electronic device is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes, but is not limited to, a microprocessor, an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), an embedded device, etc.

[0023] An electronic device can be any electronic product that can perform human-computer interaction with a customer. For example, a personal computer, a tablet computer, a smart phone, a personal digital assistant (PDA), a game console, an Internet Protocol Television (IPTV), a smart wearable device, etc.

[0024] The electronic device may further include a network device and / or a client device. Among them, the network device includes, but is not limited to, a single network server, a server group composed of multiple network servers, or a cloud composed of a large number of hosts or network servers based on Cloud Computing.

[0025] The network where the electronic device is located includes, but is not limited to, the Internet, wide area network, metropolitan area network, local area network, Virtual Private Network (VPN), etc.

[0026] As Figure 1 shown is an application scenario diagram of a method for preparing a curtain - type membrane element based on artificial intelligence provided by an embodiment of the present application. A method for preparing a curtain - type membrane element based on artificial intelligence provided by the present application can be applied to the electronic device 100. Among them, the electronic device 100 is communicatively connected to a sieving device 200, a mixing device 300, an extrusion device 400, and a sintering device 500. Specifically, the electronic device 100 is used to control the sieving device 200 to perform sieving treatment on the material based on preset sieving parameters to obtain dispersed material. The electronic device 100 is further used to control the mixing device 300 to perform mixing treatment on the dispersed material and the lubricant based on preset mixing parameters to obtain a mixed material. The electronic device 100 is further used to determine the forming parameters of the mixed material based on a pre - trained first prediction model according to the sieving parameters and the mixing parameters. The electronic device 100 is further used to control the extrusion device 400 to perform pre - forming treatment on the mixed material according to the forming parameters to obtain an extruded hollow fiber membrane. The electronic device 100 is further used to determine the sintering parameters of the hollow fiber membrane based on a pre - trained second prediction model according to the forming parameters. The electronic device 100 is further used to control the sintering device 500 to perform sintering treatment on the hollow fiber membrane based on the sintering parameters to obtain a hollow fiber curtain - type membrane element. In this way, it can ensure that the membrane filaments of the hollow fiber curtain - type membrane element are neatly arranged, thereby improving the pollution resistance of the curtain - type membrane element.

[0027] As Figure 2 shown is a flowchart of a method for preparing a curtain - type membrane element based on artificial intelligence provided by an embodiment of the present application. According to different requirements, the order of steps in this flowchart can be changed, and some steps can be omitted. The method for preparing a curtain - type membrane element based on artificial intelligence provided by the embodiment of the present application includes the following steps.

[0028] S20, controlling the sieving device to perform sieving treatment on the material based on preset sieving parameters to obtain dispersed material.

[0029] In some embodiments, the material can be polytetrafluoroethylene, and the auxiliary oil can be paraffin oil. Among them, polytetrafluoroethylene is a polymer with low friction coefficient, non-adhesiveness, high temperature resistance and corrosion resistance. Specifically, polytetrafluoroethylene can be applied to the treatment process of highly polluted industrial wastewater containing oil, solvents, etc. For example, polytetrafluoroethylene can be used for landfill leachate treatment, can also be used for water reuse in the steel industry, can also be used for wastewater treatment in the printing and dyeing and electroplating industries, and can also be used for wastewater treatment in the chemical industry. Among them, polytetrafluoroethylene can solve problems such as the oil intolerance and insufficient anti-pollution performance of the membrane bioreactor products made of polyvinylidene fluoride, and can improve the efficiency of industrial sewage treatment. Specifically, compared with polyvinylidene fluoride, polytetrafluoroethylene has more stable hydrophilicity; the curtain membrane element made of polytetrafluoroethylene is manufactured through a stretching process, so that the pore opening rate of the membrane filaments of the curtain membrane element is higher than that of the membrane made of polyvinylidene fluoride, so that the flux of the curtain membrane element is higher than that of the membrane made of polyvinylidene fluoride, and the strength of the membrane filaments is high; and polytetrafluoroethylene can be stored dry during shutdown. Compared with polyvinylidene fluoride stored by wet method, polytetrafluoroethylene is convenient for storage.

[0030] In some embodiments, in the process flow of preparing hollow fiber curtain membrane elements using polytetrafluoroethylene (PTFE), since the polytetrafluoroethylene dispersion resin may form agglomerates or have overly large particles during storage and transportation due to factors such as moisture absorption, pressure, or temperature changes, it may affect the fluidity and dispersibility of the material during subsequent processing, and thus affect the quality of the final product. Therefore, the screening device can be controlled based on preset screening parameters to screen the material, obtaining dispersed material. Among them, the screening parameters include the screening temperature and the screen aperture of the screening device. For example, the polytetrafluoroethylene dispersion resin can be screened using a screen with an aperture of about 2 mm under the condition that the screening temperature is 19 degrees Celsius. Through the screening treatment, the uniformity and consistency of the resin particles can be ensured, providing high-quality raw materials for subsequent processing. The dispersed material after the screening treatment has better fluidity and dispersibility and can be more easily mixed evenly with other additives such as lubricants. And the screening treatment can remove impurities and unqualified particles in the resin, reducing defects in the final product.

[0031] S21, controlling the mixing device to perform a mixing process on the dispersed material and the lubricant based on preset mixing parameters to obtain a mixed material.

[0032] In an embodiment of the present application, in the process of preparing a hollow fiber curtain membrane element using polytetrafluoroethylene (PTFE), in order to improve the uniformity and consistency of the materials, ensure that the PTFE resin and other auxiliary materials such as lubricants are fully mixed during the mixing process, and ensure that the materials have uniform fluidity and dispersibility during subsequent processing. The mixing equipment can be controlled based on preset mixing parameters to mix the dispersed materials and lubricants to obtain a mixed material. Among them, the mixing parameters include the mixing temperature of the mixing equipment, the aging temperature of the mixing equipment, and the mixing speed of the mixing equipment. For example, the PTFE resin and the lubricant can be mixed in a certain proportion under the temperature condition that the mixing temperature is below 19 degrees Celsius. Manual shaking or a double-roll rolling stirrer can be used for mixing during mixing to ensure the uniformity of mixing. And after mixing, in order to promote the penetration of the lubricant, the mixed material is placed in an environment with an aging temperature of 35 degrees Celsius for 10 hours, so that the lubricant gradually penetrates into the interior of the PTFE resin, improving the lubricity between resin particles and reducing friction and wear during the processing. Through the mixing process, good combination between the PTFE resin and other auxiliary materials such as lubricants can be ensured, thereby improving the quality and performance of the final product.

[0033] In an embodiment of the present application, adding a lubricant during the mixing process can reduce the friction and wear between the PTFE resin and the processing equipment, protecting the equipment from damage. The types of lubricants include liquid lubricants and solid lubricants. Liquid lubricants include mineral oil, sodium-based lubricating oil, synthetic oil, etc. They can be used to reduce the friction coefficient, lower the wear rate, and protect mechanical equipment. They can penetrate between the PTFE resins, improve the fluidity of the materials, and reduce friction and heat generation during processing. Solid lubricants include graphite, molybdenum disulfide, polytetrafluoroethylene (PTFE) micro-powder, etc., which are used to fill the micropores on the surface of PTFE to reduce friction and improve wear resistance. They can form a lubricating film on the surface of the PTFE resin, reduce the friction force between resin particles, and improve the surface finish and mechanical properties of the product.

[0034] S22. According to the sieving parameters and the mixing parameters, based on a pre-trained first prediction model, determine the forming parameters of the mixed material.

[0035] In one embodiment of the present application, in order to improve the performance of the hollow fiber membrane obtained by processing the mixed material using the extrusion equipment, the forming parameters of the mixed material can be determined based on the pre-trained first prediction model according to the sieving parameters and the mixing parameters. In this way, during the preparation of the flat-sheet membrane element, the degree of association between each process step can be improved, so as to determine appropriate forming parameters according to the preset sieving parameters and mixing parameters, and thus the performance of the hollow fiber membrane generated by controlling the extrusion equipment according to the forming parameters can be improved. Among them, the forming parameters include the extrusion temperature of the extrusion equipment and the extrusion speed of the extrusion equipment. Specifically, training the first prediction model includes: obtaining multiple batches of first production data, second production data, and first label data; wherein, the first production data includes the sieving temperature of the sieving equipment, the screen aperture of the sieving equipment, the mixing temperature of the mixing equipment, the aging temperature of the mixing equipment, and the mixing speed of the mixing equipment; the second production data includes the extrusion temperature of the extrusion equipment and the extrusion speed of the extrusion equipment, and the first label data includes the cone angle, compression ratio, aspect ratio, and porosity of the hollow fiber membrane; the first label data is used to indicate the performance of the hollow fiber membrane corresponding to the second production data of each batch; determining a first reward value corresponding to the second production data of any batch according to the first label data of any batch; inputting the first production data of any batch into the pre-constructed first initial prediction model to obtain the predicted production data output by the first initial prediction model; determining a first loss value of the first initial prediction model based on the first reward value, the predicted production data, and the second production data; updating the first prediction model based on the backpropagation algorithm until the first loss value meets the preset condition, and stopping updating the first prediction model to obtain the first prediction model trained to the convergence state.

[0036] In one embodiment of the present application, determining the first reward value corresponding to the second production data of any batch according to the first label data of any batch includes: normalizing the first label data of the multiple batches to obtain the normalized first label data of any batch; wherein, the normalized first label data includes the normalized cone angle, normalized compression ratio, normalized aspect ratio, and normalized porosity; determining a first difference between the normalized first label data and the preset first standard data; wherein, the first difference includes the cone angle difference, compression ratio difference, aspect ratio difference, and porosity difference; determining the first reward value corresponding to the second production data of any batch according to the preset first weight parameter and the first difference.

[0037] In one embodiment of the present application, determining the first reward value corresponding to the second production data of any batch according to the preset first weight parameter and the first difference includes: Among them, Reward1 represents the first reward value corresponding to the second production data of any one batch; α represents the first weight in the first weight parameter, and the first weight is used to characterize the influence degree of the cone angle on the first reward value; β represents the second weight in the first weight parameter, and the second weight is used to characterize the influence degree of the compression ratio on the first reward value; γ represents the third weight in the first weight parameter, and the third weight is used to characterize the influence degree of the aspect ratio on the first reward value; θ represents the fourth weight in the first weight parameter, and the fourth weight is used to characterize the influence degree of the porosity on the first reward value.

[0038] In an embodiment of the present application, determining the first loss value of the first initial prediction model based on the first reward value, the predicted production data, and the second production data includes: Among them, Loss1 represents the first loss value of the first prediction model; i represents the index of the batch of the second production data, m represents the number of batches of the second production data; j represents the index of the dimension in the second production data, k represents the number of dimensions of the second production data; e represents the natural constant; represents the first reward value corresponding to the second production data of the i-th batch; A ij represents the value of the j-th dimension in the second production data of the i-th batch; B ij represents the value of the j-th dimension in the predicted production data of the i-th batch.

[0039] S23. Control the extrusion equipment to perform preforming treatment on the mixed material according to the forming parameters, and obtain a hollow fiber membrane formed by extrusion.

[0040] In an embodiment of the present application, in the process flow of preparing a hollow fiber curtain membrane element by using polytetrafluoroethylene (PTFE), preforming the mixed material can reduce the volume of the mixed powder. The blank after preforming treatment has better fluidity and dispersibility, preparing for the subsequent pressing process. During the preforming process, by controlling the blanking speed and pressure, the mixed material is subjected to uniform shear force in the material cavity of the extrusion equipment, so that the resin particles are fibrillated. The fibrillated resin particles have better fluidity and dispersibility. In addition, by controlling the forming parameters in the preforming process (for example, the extrusion temperature of the extrusion equipment, the extrusion speed of the extrusion equipment, etc.), performance parameters such as the pore size and porosity of the final product can be regulated. Ensure that the product after preforming treatment has better mechanical properties such as tensile strength and elongation at break, as well as higher dimensional accuracy and stability.

[0041] S24. Based on the pre-trained second prediction model and according to the forming parameters, determine the sintering parameters of the hollow fiber membrane.

[0042] In an embodiment of the present application, in order to improve the performance of the curtain membrane element obtained by processing the hollow fiber membrane using a sintering device, the sintering parameters of the hollow fiber membrane can be determined based on the pre-trained second prediction model and according to the forming parameters. Among them, the sintering parameters include the sintering temperature and sintering time of the sintering device. Specifically, training the second prediction model includes: obtaining multiple batches of second production data, historical sintering data, and second label data; where the second production data includes the extrusion temperature of the extrusion device and the extrusion speed of the extrusion device, and the historical sintering data includes the sintering temperature and sintering time; the second label data includes the filament diameter, flux, differential pressure, and tensile strength of the hollow fiber curtain membrane element; the second label data is used to indicate the performance of the curtain membrane element corresponding to each batch of second production data; determining the second reward value corresponding to the second production data of any batch according to the second label data of any batch; inputting the second production data of any batch into the pre-constructed second initial prediction model to obtain the predicted sintering data output by the second initial prediction model; determining the second loss value of the second initial prediction model based on the second reward value, the predicted sintering data, and the historical sintering data; updating the second prediction model based on the backpropagation algorithm until the second loss value meets a preset condition, and stopping updating the second prediction model to obtain the second prediction model trained to a convergent state.

[0043] In an embodiment of the present application, the determining the second reward value corresponding to the second production data of any batch according to the second label data of any batch includes: performing normalization processing on the second label data of the multiple batches to obtain the normalized second label data of any batch; where the normalized second label data includes the normalized filament diameter, normalized flux, normalized differential pressure, and normalized tensile strength; determining the second difference between the normalized second label data and the preset second standard data; where the second difference includes the diameter difference, flux difference, differential pressure difference, and tensile strength difference; determining the second reward value corresponding to the second production data of any batch according to the preset second weight parameter and the second difference.

[0044] In an embodiment of the present application, the determining the second reward value corresponding to the second production data of any batch according to the preset second weight parameter and the second difference includes: Among them, Reward2 represents the second reward value corresponding to the second production data of any one batch; a represents the first weight in the second weight parameter, which is used to indicate the influence degree of the membrane filament diameter on the second reward value; b represents the second weight in the second weight parameter, which is used to characterize the influence degree of the flux on the second reward value; c represents the third weight in the second weight parameter, which is used to characterize the influence degree of the pressure difference on the second reward value; d represents the fourth weight in the second weight parameter, which is used to characterize the influence degree of the tensile strength on the second reward value.

[0045] In an embodiment of the present application, determining the second loss value of the second initial prediction model based on the second reward value, the predicted sintering data, and the historical sintering data includes: Among them, Loss2 represents the second loss value of the second prediction model; i represents the index of the batch of the second production data, and m represents the number of batches of the second production data; x represents the index of the dimension in the historical sintering data, and y represents the number of dimensions of the historical sintering data; e represents the natural constant; represents the second reward value corresponding to the second production data of the i-th batch; U ix represents the value of the x-th dimension in the historical sintering data of the i-th batch; V ix represents the value of the x-th dimension in the predicted sintering data of the i-th batch.

[0046] S25. Control the sintering equipment to perform sintering treatment on the hollow fiber membrane based on the sintering parameters to obtain a hollow fiber curtain membrane element.

[0047] In one embodiment of the present application, in the process of using polytetrafluoroethylene (PTFE) to prepare a hollow fiber curtain membrane element, performing sintering and shaping treatment can cause the resin particles in the preformed blank to melt and fuse into a uniform structure. For example, the preformed blank is heated to a temperature above the melting point of polytetrafluoroethylene (327 °C) to melt the resin particles. In the molten state, the interfaces between the resin particles disappear, forming a uniform structure. Through sintering and shaping treatment, the resin particles in the blank are fully melted and fused, eliminating the boundaries between the particles and forming a dense and uniform structure. This significantly improves the density, hardness, and mechanical properties of the product. Sintering and shaping treatment can also eliminate the internal stress and defects in the blank, improving the dimensional accuracy and stability of the product. During the sintering process, the blank is subjected to high temperature, and the internal stress and defects are released and repaired. At the same time, by controlling process parameters such as the heating rate, sintering temperature, and holding time, it can be ensured that the blank is uniformly heated, avoiding local overheating or overcooling. After sintering and shaping treatment, the internal stress and defects in the blank are effectively eliminated, and the dimensional accuracy and stability of the product are improved. By controlling the process parameters during sintering, performance parameters such as the pore size and porosity of the product can be adjusted. Process parameters such as the sintering temperature, holding time, and heating rate have a significant impact on the pore size and porosity of the product. By optimizing these parameters, products with different pore sizes and porosities can be prepared to meet the requirements of different application scenarios. By regulating the process parameters during sintering, performance parameters such as the pore size and porosity of the product can be precisely controlled. This enables the PTFE hollow fiber curtain membrane element to be optimized according to different usage requirements, improving its applicability and performance.

[0048] As can be seen from the above technical solutions, in the embodiment of the present application, the first pre-trained prediction model learns the complex non-linear relationship between the sieving parameters, mixing parameters, and extrusion molding parameters, and the second pre-trained prediction model learns the correlation between the extrusion molding parameters and the sintering parameters, thereby realizing the accurate prediction of process parameters. Based on a large amount of historical data and real-time data, the dependence on human experience and the trial-and-error method is reduced, and the accuracy of the parameters is improved. It can ensure that the process parameters remain relatively stable and consistent under different batches and production conditions, thereby reducing the performance differences between products and improving the stability and reliability of the products. It can also realize the real-time monitoring and adjustment of the entire preparation process. When the parameters of a certain step fluctuate, the parameters of the subsequent steps can be quickly predicted and adjusted to maintain the stability of the entire process. It can reduce the testing and adjustment process, thereby shortening the production cycle and improving production efficiency. At the same time, due to the improvement of the product performance stability, the rejection rate and rework rate can be reduced, further reducing costs.

[0049] Please refer to Figure 3 , Figure 3It is a functional module diagram of a curtain membrane element preparation system based on artificial intelligence provided by an embodiment of the present application. The curtain membrane element preparation system 600 based on artificial intelligence includes an electronic device 100, and the electronic device 100 is communicatively connected to a screening device 200, a mixing device 300, an extrusion device 400, and a sintering device 500. The module / unit mentioned in the present application refers to a series of computer-readable instruction segments that can be executed by a processor 13 and can complete fixed functions, and are stored in a memory 12. In this embodiment, the functions of each module / unit will be described in detail in subsequent embodiments.

[0050] The electronic device 100 is configured to control the screening device 200 to perform a screening process on materials based on preset screening parameters to obtain dispersed materials.

[0051] The electronic device 100 is further configured to control the mixing device 300 to perform a mixing process on the dispersed materials and a lubricant based on preset mixing parameters to obtain mixed materials.

[0052] The electronic device 100 is further configured to determine forming parameters of the mixed materials based on a pre-trained first prediction model according to the screening parameters and the mixing parameters.

[0053] The electronic device 100 is further configured to control the extrusion device 400 to perform a pre-forming process on the mixed materials according to the forming parameters to obtain an extruded hollow fiber membrane.

[0054] The electronic device 100 is further configured to determine sintering parameters of the hollow fiber membrane based on a pre-trained second prediction model according to the forming parameters.

[0055] The electronic device 100 is further configured to control the sintering device 500 to perform a sintering process on the hollow fiber membrane based on the sintering parameters to obtain a hollow fiber curtain membrane element.

[0056] Please refer to Figure 4 , which is a schematic structural diagram of an electronic device provided by an embodiment of the present application. The electronic device 100 includes a memory 12 and a processor 13. The memory 12 is used to store computer-readable instructions, and the processor 13 is used to execute the computer-readable instructions stored in the memory to implement a method for preparing a curtain membrane element based on artificial intelligence as described in any of the above embodiments.

[0057] In an embodiment of the present application, the electronic device 100 further includes a bus and a computer program stored in the memory 12 and executable on the processor 13, such as a program for preparing a curtain membrane element based on artificial intelligence.

[0058] Figure 4Only the electronic device 100 with a memory 12 and a processor 13 is shown. Those skilled in the art can understand that Figure 4 the shown structure does not constitute a limitation on the electronic device 100, and it may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0059] In combination with Figure 2 , the memory 12 in the electronic device 100 stores a plurality of computer-readable instructions to implement the method for preparing a curtain film element based on artificial intelligence. The processor 13 can execute the plurality of instructions to achieve: controlling the screening device to perform a screening process on the material based on preset screening parameters to obtain a dispersed material; controlling the mixing device to perform a mixing process on the dispersed material and a lubricant based on preset mixing parameters to obtain a mixed material; determining forming parameters of the mixed material based on a pre-trained first prediction model according to the screening parameters and the mixing parameters; controlling the extrusion device to perform a pre-forming process on the mixed material according to the forming parameters to obtain an extruded hollow fiber membrane; determining sintering parameters of the hollow fiber membrane based on a pre-trained second prediction model according to the forming parameters; and controlling the sintering device to perform a sintering process on the hollow fiber membrane based on the sintering parameters to obtain a hollow fiber curtain film element.

[0060] Specifically, for the specific implementation method of the processor 13 for the above instructions, reference can be made to Figure 3 the description of the relevant steps in the corresponding embodiments, which will not be elaborated here.

[0061] Those skilled in the art can understand that the schematic diagram is only an example of the electronic device 100 and does not constitute a limitation on the electronic device 100. The electronic device 100 can be a bus structure or a star structure. The electronic device 100 may also include more or fewer other hardware or software than shown, or different component arrangements. For example, the electronic device 100 may also include input / output devices, network access devices, etc.

[0062] It should be noted that the electronic device 100 is only an example. Other existing or future electronic products that can be adapted to this application should also be included in the protection scope of this application and are included herein by reference.

[0063] Among them, the memory 12 includes at least one type of readable storage medium, which can be non-volatile or volatile. The readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, optical disc, etc. The memory 12 can be an internal storage unit of the electronic device 100 in some embodiments, such as the mobile hard disk of the electronic device 100. The memory 12 can also be an external storage device of the electronic device 100 in other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the electronic device 100. The memory 12 can be used not only to store application software and various types of data installed in the electronic device 100, such as the code of a curtain film element preparation program based on artificial intelligence, etc., but also to temporarily store data that has been output or will be output.

[0064] The processor 13 can be composed of integrated circuits in some embodiments. For example, it can be composed of a single packaged integrated circuit, or can be composed of multiple integrated circuits with the same or different functions packaged, including a combination of one or more Central Processing Units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips, etc. The processor 13 is the control core (Control Unit) of the electronic device 100, connecting various components of the entire electronic device 100 through various interfaces and lines, and by running or executing programs or modules stored in the memory 12 (such as executing a curtain film element preparation program based on artificial intelligence, etc.), and calling data stored in the memory 12, to execute various functions of the electronic device 100 and process data.

[0065] The processor 13 executes the operating system of the electronic device 100 and various installed application programs. The processor 13 executes the application program to implement the steps in the above embodiments of the various curtain film element preparation methods based on artificial intelligence, such as Figure 2 the steps shown.

[0066] Exemplarily, the computer program can be divided into one or more modules / units, and the one or more modules / units are stored in the memory 12 and executed by the processor 13 to complete this application. The one or more modules / units can be a series of computer-readable instruction segments capable of completing specific functions, and these instruction segments are used to describe the execution process of the computer program in the electronic device 100.

[0067] The integrated units implemented in the form of software functional modules can be stored in a computer-readable storage medium. The above-mentioned software functional modules are stored in a storage medium and include several instructions for causing a computer device (which may be a personal computer, a computer device, or a network device, etc.) or a processor to execute a part of the method for preparing a curtain membrane element based on artificial intelligence described in various embodiments of the present application.

[0068] If the integrated module / unit of the electronic device 100 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-mentioned embodiment methods of the present application, it can also be completed by a computer program instructing relevant hardware devices. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned various method embodiments can be implemented.

[0069] Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disc, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory, and other memories, etc.

[0070] Furthermore, the computer-readable storage medium mainly includes a storage program area and a storage data area. Among them, the storage program area can store an operating system, application programs required for at least one function, etc.; the storage data area can store data created according to the use of the blockchain node, etc.

[0071] The bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, in Figure 4 only one arrow is used to represent it, but it does not mean that there is only one bus or one type of bus. The bus is set to realize the connection and communication between the memory 12 and at least one processor 13, etc.

[0072] The embodiments of the present application further provide a computer-readable storage medium (not shown in the figure). Computer-readable instructions are stored in the computer-readable storage medium, and the computer-readable instructions are executed by a processor in an electronic device to implement the method for preparing a curtain membrane element based on artificial intelligence described in any of the above embodiments.

[0073] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0074] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0075] In addition, in each embodiment of the present application, the functional modules can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware, or in the form of a combination of hardware and software functional modules.

[0076] In addition, obviously, the word "including" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units or devices described in the specification can also be implemented by one unit or device through software or hardware. The terms such as "first" and "second" are used to represent names and do not represent any specific order.

[0077] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit them. Although the present application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A preparation method of a curtain film element based on artificial intelligence, applied to an electronic device, the electronic device being communicatively connected to a sieving device, a mixing device, an extrusion device, and a sintering device, characterized in that, The method includes: Controlling the screening device to perform screening on the material based on preset screening parameters to obtain dispersed material; Controlling the mixing device to mix the dispersed material and the lubricant based on preset mixing parameters to obtain a mixed material; Determining the forming parameters of the mixed material based on a pre-trained first prediction model according to the screening parameters and the mixing parameters; Controlling the extrusion device to perform pre-forming on the mixed material according to the forming parameters to obtain a hollow fiber membrane formed by extrusion; Determining the sintering parameters of the hollow fiber membrane based on a pre-trained second prediction model according to the forming parameters; Controlling the sintering device to sinter the hollow fiber membrane based on the sintering parameters to obtain a hollow fiber curtain membrane element.

2. The method for preparing a curtain membrane element based on artificial intelligence according to claim 1, wherein, The method further includes training the first prediction model, and the training of the first prediction model includes: Obtaining first production data, second production data, and first label data of multiple batches; wherein, the first production data includes the screening temperature of the screening device, the screen aperture of the screening device, the mixing temperature of the mixing device, the aging temperature of the mixing device, and the mixing rotation speed of the mixing device; the second production data includes the extrusion temperature of the extrusion device and the extrusion speed of the extrusion device, and the first label data includes the cone angle, compression ratio, aspect ratio, and porosity of the hollow fiber membrane; the first label data is used to indicate the performance of the hollow fiber membrane corresponding to the second production data of each batch; Determining a first reward value corresponding to the second production data of any batch according to the first label data of any batch; Inputting the first production data of any batch into a pre-constructed first initial prediction model to obtain predicted production data output by the first initial prediction model; Determining a first loss value of the first initial prediction model based on the first reward value, the predicted production data, and the second production data; Updating the first prediction model based on the backpropagation algorithm until the first loss value meets a preset condition, and stopping updating the first prediction model to obtain the first prediction model trained to a convergent state.

3. The method for preparing a curtain membrane element based on artificial intelligence according to claim 2, wherein The determining the first reward value corresponding to the second production data of any batch according to the first label data of any batch includes: Performing normalization processing on the first label data of the multiple batches to obtain normalized first label data of any batch; wherein, the normalized first label data includes a normalized cone angle, a normalized compression ratio, a normalized aspect ratio, and a normalized porosity; Determining a first difference between the normalized first label data and preset first standard data; wherein, the first difference includes a cone angle difference, a compression ratio difference, an aspect ratio difference, and a porosity difference; Determining the first reward value corresponding to the second production data of any batch according to a preset first weight parameter and the first difference.

4. The method for preparing a curtain membrane element based on artificial intelligence according to claim 3, wherein, The determining the first reward value corresponding to the second production data of any batch according to a preset first weight parameter and the first difference includes: Among them, Reward1 represents the first reward value corresponding to the second production data of any one batch; α represents the first weight in the first weight parameter, and the first weight is used to characterize the influence degree of the cone angle on the first reward value; β represents the second weight in the first weight parameter, and the second weight is used to characterize the influence degree of the compression ratio on the first reward value; γ represents the third weight in the first weight parameter, and the third weight is used to characterize the influence degree of the aspect ratio on the first reward value; θ represents the fourth weight in the first weight parameter, and the fourth weight is used to characterize the influence degree of the porosity on the first reward value.

5. The method for preparing a curtain membrane element based on artificial intelligence according to claim 2, wherein, Determining the first loss value of the first initial prediction model based on the first reward value, the predicted production data, and the second production data includes: Among them, Loss1 represents the first loss value of the first prediction model; i represents the index of the batch of the second production data, m represents the number of batches of the second production data; j represents the index of the dimension in the second production data, k represents the number of dimensions of the second production data; e represents the natural constant; represents the first reward value corresponding to the second production data of the i-th batch; A ij represents the value of the j-th dimension in the second production data of the i-th batch; B ij represents the value of the j-th dimension in the predicted production data of the i-th batch.

6. The method for preparing a curtain membrane element based on artificial intelligence according to claim 1, wherein, The method further includes training the second prediction model, and training the second prediction model includes: Obtaining second production data, historical sintering data, and second label data of multiple batches; wherein, the second production data includes the extrusion temperature of the extrusion device and the extrusion speed of the extrusion device, and the historical sintering data includes the sintering temperature and the sintering time; the second label data includes the filament diameter, flux, differential pressure, and tensile strength of the hollow fiber curtain membrane element; the second label data is used to indicate the performance of the curtain membrane element corresponding to the second production data of each batch. Determining the second reward value corresponding to the second production data of any one batch according to the second label data of any one batch. Inputting the second production data of any one batch into a pre-constructed second initial prediction model to obtain predicted sintering data output by the second initial prediction model. Determining the second loss value of the second initial prediction model based on the second reward value, the predicted sintering data, and the historical sintering data. Updating the second prediction model based on the backpropagation algorithm until the second loss value meets a preset condition, and stopping updating the second prediction model to obtain a second prediction model trained to a convergent state.

7. The method for preparing a curtain film element based on artificial intelligence according to claim 6, characterized in that, The determining the second reward value corresponding to the second production data of any one batch according to the second label data of any one batch includes: Performing normalization processing on the second label data of the multiple batches to obtain normalized second label data of any one batch; wherein, the normalized second label data includes normalized filament diameter, normalized flux, normalized differential pressure, and normalized tensile strength. Determining a second difference between the normalized second label data and preset second standard data; wherein, the second difference includes a diameter difference, a flux difference, a differential pressure difference, and a tensile strength difference. Determining the second reward value corresponding to the second production data of any one batch according to a preset second weight parameter and the second difference.

8. The method for preparing a curtain membrane element based on artificial intelligence according to claim 7, wherein The determining the second reward value corresponding to the second production data of any one batch according to a preset second weight parameter and the second difference includes: Among them, Reward2 represents the second reward value corresponding to the second production data of any one batch; a represents the first weight in the second weight parameter, which is used to indicate the influence degree of the membrane filament diameter on the second reward value; b represents the second weight in the second weight parameter, which is used to characterize the influence degree of the flux on the second reward value; c represents the third weight in the second weight parameter, which is used to characterize the influence degree of the pressure difference on the second reward value; d represents the fourth weight in the second weight parameter, which is used to characterize the influence degree of the tensile strength on the second reward value.

9. The method for preparing a curtain-type membrane element based on artificial intelligence according to claim 6, wherein Determining the second loss value of the second initial prediction model based on the second reward value, the predicted sintering data, and the historical sintering data includes: Among them, Loss2 represents the second loss value of the second prediction model; i represents the index of the batch of the second production data, and m represents the number of batches of the second production data; x represents the index of the dimension in the historical sintering data, and y represents the number of dimensions of the historical sintering data; e represents the natural constant; represents the second reward value corresponding to the second production data of the i-th batch; U ix represents the value of the x-th dimension in the historical sintering data of the i-th batch; V ix represents the value of the x-th dimension in the predicted sintering data of the i-th batch.

10. A curtain membrane element preparation system based on artificial intelligence, characterized in that, The system includes an electronic device, and the electronic device is communicatively connected to a sieving device, a mixing device, an extrusion device, and a sintering device; The electronic device is configured to control the sieving device to sieve the material based on preset sieving parameters to obtain dispersed material; The electronic device is further configured to control the mixing device to mix the dispersed material and the lubricant based on preset mixing parameters to obtain a mixed material; The electronic device is further configured to determine the forming parameters of the mixed material based on the sieving parameters and the mixing parameters and based on a pre-trained first prediction model; The electronic device is further configured to control the extrusion device to perform pre-forming on the mixed material according to the forming parameters to obtain a hollow fiber membrane formed by extrusion; The electronic device is further configured to determine the sintering parameters of the hollow fiber membrane based on the forming parameters and based on a pre-trained second prediction model; The electronic device is further configured to control the sintering device to sinter the hollow fiber membrane based on the sintering parameters to obtain a hollow fiber curtain membrane element.

Images