A process parameter optimization method for large-tow carbon fiber pre-oxidation process

Abstract

The application provides a process parameter optimization method for large-tow carbon fiber pre-oxidation process, and relates to the field of computer simulation. The method comprises the following steps: a fluid domain model corresponding to a pre-oxidation furnace is established through a three-dimensional modeling technology, and a plurality of fiber tows running side by side are defined as porous medium areas; a multi-physical field cooperative strong coupling model is constructed based on the fluid domain model; process parameters in the carbon fiber pre-oxidation process are acquired, and based on the process parameters and a multi-physical process state field set, a sensitivity analysis method is used to reduce the optimization dimension to obtain target key variables; an improved non-dominated sorting genetic algorithm is used to construct a plurality of optimization objectives, and a process parameter optimization scheme is generated by defining preset constraint conditions; and based on the process parameter optimization scheme, the fluid domain model is corrected online through digital twinning technology. The application solves the problem that the existing carbon fiber pre-oxidation process control system cannot accurately reflect the real temperature distribution inside the tow.

Description

Technical Field

[0001] This application relates to the field of computer simulation, and in particular to a method for optimizing process parameters for the pre-oxidation process of large-tow carbon fibers. Background Technology

[0002] Currently, over 90% of commercially available carbon fibers worldwide are produced using polyacrylonitrile (PAN) as a precursor. The production process of PAN-based carbon fibers mainly includes precursor preparation, pre-oxidation, low-temperature carbonization, high-temperature carbonization, and surface treatment. Among these, the pre-oxidation process is widely recognized as the most critical, time-consuming, and energy-intensive stage in carbon fiber production, accounting for approximately 18%-20% of the total energy consumption of the production line. In this stage, the PAN precursor fibers undergo heat treatment in an oxygen-containing atmosphere at a temperature range of 180°C to 300°C for 30 to 90 minutes. During this period, the linear PAN molecular chains undergo three complex chemical reactions: cyclization, dehydrogenation, and oxidation, transforming into a heat-resistant trapezoidal structure, thus preventing melting and fiber condensation during the subsequent high-temperature carbonization process.

[0003] Currently, most existing carbon fiber pre-oxidation process control systems adjust process parameters through trial and error or single-factor experiments. This method first conducts production tests by setting process conditions such as temperature zones, linear velocity, or circulating air volume. Then, based on the measured product density or structural performance results, the process parameters are manually corrected to gradually approach a combination of process parameters that meets product quality requirements. However, because this method relies primarily on experience and single-variable adjustment, it ignores the significant impact of exothermic chemical reactions on the local temperature field. Therefore, during the pre-oxidation process parameter control, it is difficult to accurately reflect the true temperature distribution and reaction state changes within the fiber bundle. This can lead to significant differences in the core-sheath structure within the fiber, thereby reducing the stable operation of the carbon fiber production line and the quality of the finished product.

[0004] Therefore, there is an urgent need for a method to optimize the process parameters of the pre-oxidation process for large-tow carbon fibers. Summary of the Invention

[0005] This application provides a method for optimizing process parameters in the pre-oxidation process of large-tow carbon fibers. It solves the problem that existing carbon fiber pre-oxidation process control systems cannot accurately reflect the actual temperature distribution and reaction state changes inside the tow during the pre-oxidation process parameter adjustment, which leads to obvious differences in the core-sheath structure inside the fiber, thereby reducing the stable operation of the carbon fiber production line and the quality of the finished product.

[0006] The first aspect of this application provides a method for optimizing process parameters in the pre-oxidation process of large-tow carbon fibers. The method includes: establishing a fluid domain model corresponding to the pre-oxidation furnace using 3D modeling technology, and defining multiple parallel-running fiber bundles as porous media regions; constructing a multi-physics field synergistic strong coupling model based on the fluid domain model, which is used to couple and solve the gas flow process, heat transfer process, and fiber pre-oxidation reaction process, and outputting a set of multi-physics process state fields; obtaining process parameters in the carbon fiber pre-oxidation process, and reducing the optimization dimension based on the process parameters and the set of multi-physics process state fields to obtain target key variables through sensitivity analysis; constructing multiple optimization objectives using an improved non-dominated sorting genetic algorithm, and generating process parameter optimization schemes by defining preset constraints; the optimization objectives include temperature non-uniformity coefficient optimization objectives, unit energy consumption optimization objectives, and core-sheath structure difference optimization objectives; the constraints include fiber center temperature constraints, outlet average density constraints, and unit residence time constraints; and correcting the fluid domain model online using digital twin technology based on the process parameter optimization scheme.

[0007] Optionally, a fluid domain model corresponding to the pre-oxidation furnace is established using 3D modeling technology, and multiple parallel-running fiber bundles are defined as porous media regions. Specifically, this includes: establishing a fluid domain model based on the structural parameters of the pre-oxidation furnace using 3D modeling technology; the fluid domain model includes an air inlet, an air outlet, a rectifier plate, a heater, and a fiber feeding channel; defining multiple parallel-running fiber bundles as porous media regions based on the specifications and arrangement of the fiber bundles, and describing the pressure drop when the airflow passes through the fiber bundles using a preset law; the geometric parameters corresponding to the porous media are consistent with the width and thickness of the fiber bundle sheets in actual production.

[0008] Optionally, the gas flow process, heat transfer process, and fiber pre-oxidation reaction process are coupled and solved to output a set of multi-physical process state fields: a turbulence model is constructed by solving the flow field control equations, and the boundary layer flow state of the furnace wall surface and fiber bundle surface in the pre-oxidation furnace is captured based on the turbulence model to obtain the gas flow state field inside the furnace; a heat transfer solution model is constructed by solving the energy equation containing the chemical internal heat source, and the heat transfer state between the air and fiber bundle inside the pre-oxidation furnace is captured based on the heat transfer solution model to obtain the temperature distribution state field inside the furnace; a reaction kinetic model is constructed by solving the fiber pre-oxidation reaction kinetic equations, and the reaction transformation state in the fiber pre-oxidation reaction process is captured based on the reaction kinetic model to obtain the fiber reaction transformation state field; a component transport model is constructed by solving the oxygen component transport equations, and the diffusion and consumption state of oxygen in the porous medium region is captured based on the component transport model to output the oxygen concentration distribution state field; the gas flow state field inside the furnace, the temperature distribution state field inside the furnace, the fiber reaction transformation state field, and the oxygen concentration distribution state field are used as a set of multi-physical process state fields.

[0009] Optionally, process parameters during the carbon fiber pre-oxidation process are obtained, and based on the process parameters and the set of multiple physical process state fields, the optimization dimensionality is reduced through sensitivity analysis to obtain the target key variables. Specifically, this includes: obtaining process parameters during the carbon fiber pre-oxidation process, including the temperature zone set temperature, circulating fan frequency, exhaust damper opening, and draw ratio; and reducing the optimization dimensionality through sensitivity analysis based on the process parameters and the set of multiple physical process state fields to obtain the target key variables, including the standard deviation of fiber density, the highest center temperature, and energy consumption.

[0010] Optionally, multiple optimization objectives are constructed using an improved non-dominated sorting genetic algorithm, specifically including: optimizing the objective by constructing a first objective function and optimizing the temperature non-uniformity coefficient. ; in, Denotes the first objective function. This represents minimizing the temperature non-uniformity coefficient. Indicates the target set temperature. Represents the spatial temperature distribution function. Indicates the volume of the control volume. Let the volume element be ; the objective is optimized by constructing a second objective function and optimizing the unit energy consumption: ; in, Describes the second objective function. This indicates minimizing the difference in skin-core structure. Indicates surface conversion rate, Indicates core conversion rate. This represents the average conversion rate; the objective is optimized by constructing a third objective function and optimizing the differences in the core-skin structure. ; in, This represents the third objective function. This represents minimizing unit energy consumption. Indicates heater power. This indicates the power of the fan.

[0011] Optionally, based on the process parameter optimization scheme, the fluid domain model is corrected online using digital twin technology. Specifically, this includes: constructing a preset proxy model based on the big data generated by offline CFD; embedding the preset proxy model into the PLC control system and correcting it based on the acquired real-time monitoring data; and correcting the fluid domain model online based on the real-time monitoring data and the process parameter optimization scheme.

[0012] Optionally, the fluid domain model is corrected online, specifically including: predicting the internal reaction state of the pre-oxidation furnace based on a preset surrogate model within milliseconds; and adjusting the fan speed and heating power based on the internal reaction state.

[0013] A second aspect of this application provides a process parameter optimization device for the pre-oxidation process of large-tow carbon fibers. The device includes an acquisition module and a processing module, wherein... The acquisition module is used to establish a fluid domain model corresponding to the pre-oxidation furnace through 3D modeling technology, and to define multiple parallel-running fiber bundles as porous media regions. Based on the fluid domain model, a multi-physics field synergistic strong coupling model is constructed. The multi-physics field synergistic strong coupling model is used to couple and solve the gas flow process, heat transfer process and fiber pre-oxidation reaction process, and output the multi-physics process state field set.

[0014] The processing module acquires process parameters during the carbon fiber pre-oxidation process and, based on these parameters and a multi-physical process state field set, reduces the optimization dimensionality through sensitivity analysis to obtain key target variables. It then constructs multiple optimization objectives using an improved non-dominated sorting genetic algorithm and generates process parameter optimization schemes by defining preset constraints. The optimization objectives include temperature non-uniformity coefficient optimization, unit energy consumption optimization, and core-sheath structure difference optimization. The constraints include fiber center temperature constraints, outlet average density constraints, and unit residence time constraints. Based on the process parameter optimization schemes, the fluid domain model is corrected online using digital twin technology.

[0015] A third aspect of this application provides an electronic device including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described above.

[0016] A fourth aspect of this application provides a non-transitory computer-readable storage medium storing a computer program, the computer program being executed by a processor using any of the methods described above.

[0017] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. A fluid domain model corresponding to the pre-oxidation furnace is established using 3D modeling technology, and multiple parallel-running fiber bundles are defined as porous media regions. A multi-physics field collaborative strong coupling model is constructed based on the fluid domain model. Process parameters in the carbon fiber pre-oxidation process are obtained, and based on the process parameters and the set of multi-physics process state fields, the optimization dimensionality is reduced through sensitivity analysis to obtain the target key variables. Multiple optimization objectives are constructed using an improved non-dominated sorting genetic algorithm, and process parameter optimization schemes are generated by defining preset constraints. Based on the process parameter optimization schemes, the fluid domain model is corrected online using digital twin technology, thereby improving the accuracy of process parameter control in the pre-oxidation process, reducing the difference in fiber sheath-core structure, and improving the stability of production line operation and the quality of finished products.

[0018] 2. A turbulence model is constructed by solving the flow field control equations, and the boundary layer flow state on the furnace wall surface and fiber bundle surface in the pre-oxidation furnace is captured based on the turbulence model to obtain the gas flow state field inside the furnace; a heat transfer solution model is constructed by solving the energy equation containing the chemical internal heat source, and the heat transfer state between the air and fiber bundle inside the pre-oxidation furnace is captured based on the heat transfer solution model to obtain the temperature distribution state field inside the furnace; a reaction kinetic model is constructed by solving the fiber pre-oxidation reaction kinetic equations, and the reaction transformation state during the fiber pre-oxidation reaction process is captured based on the reaction kinetic model to obtain the fiber reaction transformation state field; a component transport model is constructed by solving the oxygen component transport equations, and the diffusion and consumption state of oxygen in the porous medium region is captured based on the component transport model to output the oxygen concentration distribution state field; the gas flow state field, temperature distribution state field, fiber reaction transformation state field, and oxygen concentration distribution state field inside the furnace are combined as a set of multi-physical process state fields to obtain multi-physical field state information that can characterize the gas flow behavior, temperature evolution characteristics, and fiber reaction process inside the pre-oxidation furnace, providing a basic calculation basis for subsequent process parameter analysis and optimization.

[0019] 3. Multiple optimization objectives are constructed using an improved non-dominated sorting genetic algorithm, and process parameter optimization schemes are generated by defining preset constraints. The optimization objectives include the optimization objectives of temperature non-uniformity coefficient, unit energy consumption, and core-sheath structure difference. The constraints include fiber center temperature constraint, outlet average density constraint, and unit residence time constraint, thereby achieving multi-objective collaborative optimization of pre-oxidation process parameters. Attached Figure Description

[0020] Figure 1 This is a schematic flowchart of a method for optimizing process parameters in the pre-oxidation process of large-tow carbon fibers, provided in an embodiment of this application. Figure 2 This is a schematic diagram of a fluid domain model set structure provided in an embodiment of this application; Figure 3 This is a schematic diagram comparing the surface temperature distribution cloud map of the wire bundle in the furnace before and after optimization, provided in an embodiment of this application. Figure 4 This application provides an embodiment of a Pareto optimal frontier curve generated based on the NSGA-II algorithm; Figure 5 This is a schematic diagram of a process parameter optimization device for the pre-oxidation process of large tow carbon fibers provided in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0021] Explanation of reference numerals in the attached figures: 51, acquisition module; 52, processing module; 601, processor; 602, communication bus; 603, user interface; 604, network interface; 605, memory. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0023] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items.

[0024] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0025] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0026] Please refer to Figure 1The diagram shows a process parameter optimization method for the pre-oxidation process of large tow carbon fiber provided in this application embodiment. The flowchart mainly includes the following steps: S101 to S105.

[0027] Step S101: Establish the fluid domain model corresponding to the pre-oxidation furnace through three-dimensional modeling technology, and define multiple parallel-running fiber bundles as porous media regions.

[0028] Specifically, the pre-oxidation reaction, especially the nitrile cyclization reaction, is a strongly exothermic reaction, with the heat released reaching over 2000 kJ / kg. Under industrial production conditions, to improve efficiency, large tow structures or high packing density arrangements are typically used, such as 24k, 48k, or 50k tows. This high-density arrangement significantly increases the number of fibers per unit cross-section, causing the heat released during the reaction to accumulate rapidly within the tow. This internal heat is difficult to transfer to the external environment in a timely manner through air convection, resulting in a temperature in the central region of the tow that is significantly higher than the external ambient temperature. This overheating in the central region further intensifies the radial reaction rate difference within the fiber, leading to a distinct core-sheath structure. When heat continues to accumulate and cannot dissipate effectively, the local temperature may continue to rise and approach the fiber's thermal runaway threshold. In severe cases, this can even cause fiber breakage or fire, significantly impacting production stability and product quality.

[0029] Meanwhile, existing industrial pre-oxidation furnaces typically use hot air circulation heating for heat transfer. Due to the large overall size of these furnaces, with widths often reaching 3 to 4 meters and complex internal air duct structures, coupled with the significant airflow obstruction caused by the fiber bundles during operation, it is difficult to maintain uniform gas flow and temperature distribution across the furnace cross-section. Airflow dead zones or velocity attenuation areas form in certain regions, resulting in uneven temperature and flow field distribution within the furnace. This spatial non-uniformity leads to differences in the thermal history experienced by fibers at different locations, causing dispersion in the degree of fiber reaction and structural transformation, ultimately affecting the consistency of the final carbon fiber product's performance. Current production processes typically rely on thermocouples placed on the furnace wall or in localized locations for temperature monitoring. However, these sensors only acquire localized ambient temperature information and cannot directly reflect the true temperature distribution and reaction state within the fiber bundles, making it difficult to accurately assess the overall thermal state within the furnace.

[0030] In actual production operations, adjustments to process parameters typically rely on experience or single-factor experiments. When product density or performance indicators fail to meet requirements, operators often manually correct them by increasing furnace temperature, decreasing linear velocity, or adjusting airflow. However, the pre-oxidation process involves multiple physical processes, including gas flow, heat transfer, and fiber chemical reactions, and these processes exhibit significant nonlinear coupling relationships. For example, while increasing temperature can accelerate the reaction rate, it also simultaneously increases the rate of exothermic reaction, thereby exacerbating heat accumulation within the fiber bundle and potentially further deteriorating the core-sheath structure and reducing the mechanical properties of the carbon fiber. Traditional control methods struggle to accurately describe this complex coupling relationship. Existing control systems typically employ simple feedback control strategies such as PID, which can only adjust single variables and lack the ability to predict the multi-physical coupling behavior during pre-oxidation. Therefore, they often exhibit strong blindness and lag in process control.

[0031] To address the aforementioned issues, computational fluid dynamics (CFD) methods were introduced to simulate and analyze the gas flow and heat transfer behavior inside the pre-oxidation furnace. Combined with fiber pre-oxidation reaction kinetics, the internal reaction process of the fibers was described, thus constructing a multiphysics coupled computational framework capable of characterizing the interactions between gas flow, heat transfer, and chemical reactions. Within this framework, a fluid domain model corresponding to the pre-oxidation furnace was established using 3D modeling techniques. Multiple parallel-running fiber bundles were defined as porous media regions to describe the gas flow path, heat transfer behavior, and flow resistance characteristics generated when the gas passes through the fiber bundles within the computational space. This modeling approach allows for the reproduction of the flow field structure and temperature distribution characteristics under actual production conditions in the computational environment, providing a fundamental computational model for further analysis of the exothermic reaction behavior inside the fibers and its coupling relationship with gas flow and heat transfer processes.

[0032] In one possible implementation, step S101 further includes: establishing a fluid domain model based on the structural parameters of the pre-oxidation furnace using three-dimensional modeling technology; the fluid domain model includes an air inlet, an air outlet, a rectifier plate, a heater, and a fiber feeding channel; based on the specifications and arrangement of the fiber bundles, defining multiple parallel-running fiber bundles as porous media regions, and describing the pressure drop when the airflow passes through the fiber bundles using a preset law; the geometric parameters corresponding to the porous media are consistent with the width and thickness of the fiber bundle sheets in actual production.

[0033] For details, please refer to Figure 2The document presents a schematic diagram of a fluid domain model assembly structure provided in an embodiment of this application. During the construction of a parametric model of the pre-oxidation furnace and fiber bundles, a fluid domain model of the pre-oxidation furnace is established using 3D modeling software to describe the spatial range of gas flow and heat transfer within the pre-oxidation furnace in the computational space. The fluid domain model includes structural units such as air inlets, air outlets, rectifiers, heaters, and fiber-feeding channels. The air inlet describes the inlet boundary where hot air enters the furnace, the air outlet describes the outlet boundary where gas is discharged or recirculated, the rectifier adjusts the airflow distribution within the furnace, the heater provides heat source boundary conditions, and the fiber-feeding channel describes the running path of the fiber bundles within the furnace body. This allows the established fluid domain model to reflect the true structural characteristics and gas flow path of the pre-oxidation furnace.

[0034] In the modeling of the fiber region, a large number of parallel-running fiber bundles are treated as a porous media region, rather than as simple solid walls. The geometric parameters of the porous media region include length, width, and thickness, and these parameters are set to match the width and thickness of the fiber bundles in actual production. This allows the computational model to reflect the obstruction effect of the fiber bundles on gas flow on a macroscopic scale. This modeling approach enables the description of gas flow within and around the fiber bundles in the computational space, providing a structural basis for subsequent coupled calculations of gas flow, heat transfer, and reaction processes.

[0035] In describing gas flow resistance, the pressure drop generated when airflow passes through a porous medium region is characterized by a pre-defined law, the expression of which is: Wherein, ΔP represents the pressure drop, characterizing the pressure difference of the fluid before and after passing through the fiber bundle. This parameter reflects the magnitude of the resistance that the gas needs to overcome to penetrate the fiber layer; L represents the thickness, indicating the physical thickness of the fiber bundle layer in the direction of gas flow; μ represents the dynamic viscosity, characterizing the viscous characteristics of air, and its value changes with gas temperature; D represents the viscous drag coefficient, characterizing the resistance effect of porous media on the viscous flow of gas; v represents the gas velocity, indicating the flow velocity of the circulating hot air inside the pre-oxidation furnace; ρ represents the gas density, characterizing the density characteristics of the hot air inside the furnace; and C represents the inertial drag coefficient, describing the kinetic energy loss caused by the gas impacting the fibers during high-speed movement. The above expressions simultaneously consider the combined effects of gas viscous drag and inertial drag, thus describing the pressure loss characteristics of gas flowing inside the fiber bundle.

[0036] Since fiber bundles are typically arranged in a unidirectional direction, their gas permeability varies significantly in different directions. Therefore, a permeability tensor is further introduced into the model to describe the anisotropic characteristics of the porous media region. Specifically, by setting the permeability tensor K, such that... Where X represents the fiber's direction of travel, Y represents the direction of transverse penetration through the fiber arrangement, and Z represents the direction of perpendicular penetration through the fiber layer. This method reflects the actual physical characteristics of gas sliding and flowing along the fiber direction in the fiber gaps, rather than simply passing through the fiber layer in a uniform manner, thus more accurately describing the interaction between gas flow and fiber structure inside the pre-oxidation furnace.

[0037] in, This represents axial permeability, which characterizes the ability of gas to permeate through the fiber bundle along the fiber length direction. Since gas can flow along the gaps between fibers, this direction usually has a higher permeability. Radial permeability is used to characterize the ability of gas to permeate through the fiber arrangement direction in a horizontal transverse direction. The normal permeability is used to characterize the ability of gas to penetrate the fiber layer in a vertical direction. By setting the anisotropic permeability, the established porous media model can more realistically reflect the flow characteristics of gas inside the fiber bundle, thereby improving the accuracy of subsequent multiphysics coupling calculations.

[0038] Step S102: Construct a multi-physics field collaborative strong coupling model based on the fluid domain model.

[0039] Specifically, the multi-physics field synergistic strong coupling model is used to couple and solve the gas flow process, heat transfer process and fiber pre-oxidation reaction process, and output the multi-physics process state field set.

[0040] In one possible implementation, step S102 further includes: constructing a turbulence model by solving the flow field control equations, and capturing the boundary layer flow state of the furnace wall surface and fiber bundle surface in the pre-oxidation furnace based on the turbulence model to obtain the gas flow state field inside the furnace; constructing a heat transfer solution model by solving the energy equation containing the chemical internal heat source, and capturing the heat transfer state between the air and fiber bundle inside the pre-oxidation furnace based on the heat transfer solution model to obtain the temperature distribution state field inside the furnace; constructing a reaction kinetic model by solving the fiber pre-oxidation reaction kinetic equations, and capturing the reaction transformation state during the fiber pre-oxidation reaction process based on the reaction kinetic model to obtain the fiber reaction transformation state field; constructing a component transport model by solving the oxygen component transport equations, and capturing the diffusion and consumption state of oxygen in the porous medium region based on the component transport model to output the oxygen concentration distribution state field; and using the gas flow state field inside the furnace, the temperature distribution state field inside the furnace, the fiber reaction transformation state field, and the oxygen concentration distribution state field as a set of multi-physical process state fields.

[0041] Specifically, the gas flow behavior inside the pre-oxidation furnace is described by constructing flow field control equations to obtain the velocity and pressure distribution of the gas inside the furnace. Based on this, a turbulence model is established to characterize the turbulent flow characteristics of the gas under the complex structural conditions inside the furnace. In the solution process, a turbulence model that can simultaneously consider the calculation accuracy of the near-wall region and the mainstream region is adopted, thereby effectively capturing the boundary layer flow state near the furnace wall surface and the fiber bundle surface, so that the obtained gas flow state field inside the furnace can reflect the actual flow path and velocity distribution of hot air inside the furnace.

[0042] In the heat transfer calculation, an energy equation incorporating an internal chemical heat source is established to describe the heat transfer process inside the pre-oxidation furnace. In the fiber porous media region, the heat of chemical reaction released during the fiber pre-oxidation reaction is introduced as a volumetric heat source term into the energy equation, allowing gas flow, heat transfer, and reaction exothermic processes to be coupled within the same computational framework. The expression is as follows:

[0043] Wherein, ρ represents fluid density, which is used to characterize the mass density characteristics of air inside the furnace; It represents the specific heat capacity of a fluid at constant pressure, and is used to characterize the heat capacity characteristics of air when it absorbs or releases heat under constant pressure conditions; This represents the gas velocity vector, used to describe the flow direction and velocity of hot air inside the furnace. Represents the temperature gradient, used to describe the rate of temperature change in space; It represents the effective thermal conductivity, used to characterize the heat conduction capacity under the combined effect of gas and fiber structure; The term representing the exothermic source of the chemical reaction is used to characterize the heat released during the fiber pre-oxidation process. This energy equation describes the evolution of the temperature field within the furnace by simultaneously considering three heat transfer mechanisms: convective heat transfer, conductive heat transfer, and exothermic reaction.

[0044] The exothermic chemical reaction term is used to characterize the heat of chemical reaction released during the pre-oxidation reaction of PAN fibers; based on the heat transfer solution model, the heat transfer state between the air and fiber bundles inside the pre-oxidation furnace is captured, and the temperature distribution state field inside the furnace is obtained, the expression of which is: in, It represents the packing density of fiber bundles and is used to describe the mass of fiber material per unit volume; It represents the total enthalpy of the pre-oxidation reaction, used to characterize the total heat released per unit mass of fiber during the pre-oxidation reaction; This represents the reaction conversion rate, with a value ranging from 0 to 1, and is used to characterize the degree of completion of the fiber pre-oxidation reaction; This represents the reaction rate, used to describe how quickly the reaction conversion rate changes over time. This expression couples the exothermic behavior during the fiber reaction process with the temperature field calculation, allowing the exothermic reaction to directly affect the temperature distribution within the furnace.

[0045] Regarding the description of the fiber reaction process, the pre-oxidation reaction rate of fibers was calculated by establishing a reaction kinetic model. This model provides a unified description of the fiber cyclization and oxidation reactions through a multi-step series and parallel reaction mechanism, and its expression is as follows:

[0046] in, Indicates the reaction rate, used to describe the rate of change of reaction conversion; The temperature-dependent reaction rate constant is used to characterize the effect of temperature on the reaction rate, and its expression is: Where A represents the pre-exponential factor, used to characterize the collision frequency of reactant molecules; R represents the activation energy of the reaction, used to describe the energy barrier that must be overcome for a reaction to occur; R represents the gas constant; T represents the absolute temperature. (Function) Used to describe the effect of reaction conversion rate on reaction rate, it can be expressed in the form of: Where n represents the reaction order, Represents the autocatalytic reaction coefficient; function It is used to describe the effect of oxygen concentration on the reaction rate, thus reflecting the limiting effect of oxygen diffusion on the reaction process.

[0047] A component transport model is constructed by solving the oxygen component transport equation. During the gas component transport calculation, the diffusion and consumption process of oxygen within the furnace body and the fibrous porous media region is described by establishing the oxygen component transport equation, the expression of which is: in, The convective transport term for oxygen is used to describe the mass transfer process of oxygen as it flows with the gas. This represents the oxygen diffusion and transport term, used to describe the diffusion process of oxygen under the influence of a concentration gradient; Indicates fluid density; Represents the gas velocity vector; It represents the mass fraction of oxygen and is used to characterize the concentration ratio of oxygen in a gas mixture; It represents the effective diffusion coefficient, which describes the ability of oxygen to diffuse in porous media environments; Represents the oxygen concentration gradient; This indicates the oxygen consumption rate, a parameter related to the fiber pre-oxidation reaction rate. Phase coupling is used to characterize the amount of oxygen consumed during a reaction.

[0048] Through the synergistic coupling of the aforementioned flow field calculations, heat transfer calculations, reaction kinetics calculations, and component transport calculations, a dynamic feedback relationship is established between gas flow, temperature evolution, reaction rate, and oxygen diffusion. As temperature increases, the reaction rate increases, leading to an increase in the heat released during the reaction; this heat release further increases the fiber temperature, causing the reaction rate to accelerate even further. Simultaneously, oxygen is continuously consumed during the reaction, causing the local oxygen concentration to gradually decrease. When the oxygen concentration drops to a certain level, the reaction rate is then limited by the diffusion process. This bidirectional coupling relationship between the temperature field, reaction field, and component transport allows for a more accurate description of the fiber reaction behavior and its heat accumulation process within the pre-oxidation furnace, thus providing a reliable computational basis for predicting fiber core-sheath structure formation and the risk of thermal runaway.

[0049] Step S103: Obtain the process parameters in the carbon fiber pre-oxidation process, and based on the process parameters and the set of multi-physical process state fields, reduce the optimization dimension through sensitivity analysis to obtain the target key variables.

[0050] Specifically, by acquiring process parameters such as the set temperature of the pre-oxidation furnace, the frequency of the circulating fan, the opening degree of the exhaust damper, and the draw ratio during the operation of the pre-oxidation furnace, and combining them with the multi-physical process state field set obtained in step S102, the correlation between the operating state of the pre-oxidation furnace and the fiber reaction state under different combinations of process parameters is analyzed. The influence of each process parameter on the system operating state and performance indicators is evaluated by using sensitivity analysis, thereby screening out the target key variables that have a significant impact on the pre-oxidation process, so as to reduce the calculation dimension of the subsequent process parameter optimization process.

[0051] In one possible implementation, step S103 further includes: obtaining process parameters during the carbon fiber pre-oxidation process, including temperature zone setting temperature, circulating fan frequency, exhaust damper opening and draw ratio; and based on the process parameters and the set of multi-physical process state fields, reducing the optimization dimension through sensitivity analysis to obtain target key variables, including fiber density standard deviation, maximum center temperature and energy consumption.

[0052] Specifically, before performing full parameter optimization, key variables affecting the pre-oxidation process are screened using a global sensitivity analysis method to reduce the dimensionality of subsequent optimization calculations. During the analysis, the set temperature of each temperature zone, the frequency of the circulating fan, the opening degree of the exhaust damper, and the draw ratio are used as input variables. The correlation between the input variables and the system response is established by combining the multi-physical process state field set obtained in step S102. Specifically, the set temperature of each temperature zone characterizes the temperature control level of different heating areas, the frequency of the circulating fan characterizes the intensity of the circulating airflow in the furnace, the opening degree of the exhaust damper characterizes the gas renewal and emission capacity, and the draw ratio characterizes the tension state and linear velocity changes of the fiber during the pre-oxidation process.

[0053] Regarding output response, response indicators reflecting the system's operating state and product quality characteristics are extracted through a multi-physical process state field set. These indicators include the standard deviation of fiber bulk density, the highest center temperature, and energy consumption. Specifically, the standard deviation of fiber bulk density characterizes the uniformity of fiber structure, the highest center temperature characterizes the degree of heat accumulation within the fiber bundle and production safety risks, and energy consumption characterizes the energy consumption level during the pre-oxidation process.

[0054] In the specific calculation process, sample points with different combinations of process parameters are generated using the Latin hypercube sampling method. These sample points are then calculated based on the established multi-physics synergistic strong coupling model to obtain the corresponding system response results. Furthermore, by calculating the first-order effect index and the total effect index of each input variable, the influence of different process parameters on the system response is quantitatively analyzed. Parameters whose influence on the objective function is below a preset threshold are eliminated, thereby identifying the core control variables that significantly affect the pre-oxidation process. For example, sensitivity analysis results show that temperatures in certain temperature zones and the intensity of circulating airflow have a high degree of influence on the internal reaction behavior of the fiber and the formation of the core-sheath structure, thus making them key control variables in subsequent process optimization.

[0055] Step S104: Construct multiple optimization objectives using an improved non-dominated sorting genetic algorithm, and generate process parameter optimization schemes by defining preset constraints.

[0056] Specifically, after obtaining the key target variables, they are input into a multi-objective optimization framework as optimization variables. Multiple objective functions are constructed to collaboratively optimize the temperature distribution, fiber structure uniformity, and energy consumption levels during the pre-oxidation furnace operation. The optimization objectives include optimization of the temperature non-uniformity coefficient, unit energy consumption, and core-sheath structure differences. Simultaneously, constraints such as fiber center temperature, outlet average density, and unit residence time are incorporated to limit the feasible solution space during the optimization process, thereby achieving comprehensive optimization of process parameters while ensuring production safety and product quality.

[0057] Regarding temperature uniformity optimization, a first objective function is constructed to evaluate the uniformity of temperature distribution within the furnace, and its expression is as follows: in, This represents the first objective function, used to measure the spatial uniformity of the temperature field inside the furnace; This indicates an optimization aimed at minimizing the temperature non-uniformity coefficient. Indicates the preset target temperature; A function representing the temperature distribution at any location within the furnace body; Indicates the total volume of the control volume; This represents a volumetric element. This function allows for the overall measurement of the deviation between the furnace temperature distribution and the target temperature, thereby optimizing temperature uniformity.

[0058] Regarding the optimization of fiber structure uniformity, a second objective function is constructed to evaluate the differences between the core and sheath structure, and its expression is as follows: in, This represents the second objective function, used to measure the difference in reaction conversion rates between the fiber sheath and the core. This indicates minimizing the difference in skin-core structure; This indicates the reaction conversion rate of the fiber surface area; This indicates the reaction conversion rate in the fiber core region; This represents the overall average conversion rate of the fiber. This function allows for a quantitative evaluation of the uniformity of the radial reaction of the fiber, thereby reducing the inhomogeneity of the core-sheath structure.

[0059] Regarding energy consumption optimization, a third objective function is constructed to evaluate the energy consumption during the operation of the pre-oxidation furnace, and its expression is as follows: in, This represents the third objective function, used to characterize the energy consumption level of the system per unit time. This represents minimizing unit energy consumption; Indicates heater power; This represents the power of the circulating fan. This function allows for optimization of system energy consumption while ensuring product quality and safety constraints.

[0060] During the optimization process, the feasible solution space is restricted by setting preset constraints. Among them, the highest temperature at the fiber center satisfies... This is to avoid the risk of thermal runaway or combustion of fibers during the pre-oxidation process; the average density at the outlet meets the requirements. This is used to ensure that the product density meets the process quality standards; the residence time in a single zone meets the requirements. This is used to ensure that the fibers have sufficient reaction time in each temperature zone.

[0061] In the specific optimization process, multiple sets of process parameter combinations are randomly generated as the initial population, and a multiphysics collaborative strong coupling model is invoked to calculate the system response corresponding to each set of parameters, thereby obtaining the fitness values ​​of each objective function. Subsequently, the population is stratified according to the non-dominated sorting principle, and the diversity of the solution set is maintained through crowding calculation, retaining the better-performing individuals in the selection operation. New combinations of process parameters are generated through crossover and mutation operations, and the population structure is continuously updated iteratively. The optimization calculation stops when the Pareto front no longer changes significantly.

[0062] Step S105: Based on the process parameter optimization scheme, the fluid domain model is corrected online using digital twin technology.

[0063] Specifically, in order to overcome the drawback of long offline simulation time, this invention also proposes an online reduced-order model (ROM) scheme: using the big data generated by offline CFD, a deep neural network (DNN) or Gaussian process regression (Kriging) model is trained as a surrogate model; in actual production, the surrogate model can be embedded in the PLC control system, and predict the current internal reaction state in milliseconds based on the real-time monitored ambient temperature and feeding parameters, and fine-tune the fan speed and heating power.

[0064] In one possible implementation, step S105 further includes: constructing a preset agent model based on the big data generated by offline CFD; embedding the preset agent model into the PLC control system and obtaining real-time monitoring data; and performing online correction of the fluid domain model based on the real-time monitoring data and process parameter optimization scheme, specifically including: predicting the internal reaction state of the pre-oxidation furnace based on the preset agent model within milliseconds; and adjusting the fan speed and heating power based on the internal reaction state.

[0065] Specifically, after completing the multi-objective optimization calculation, in order to reduce the computational overhead caused by directly calling the multi-physics field collaborative strongly coupled model for real-time calculation, a surrogate model is constructed using a large amount of historical sample data generated during the offline CFD simulation calculation. The historical sample data contains multi-physics process state field information such as flow field distribution, temperature distribution, reaction conversion rate, and oxygen concentration distribution under different combinations of process parameters. By training on the above data, a preset surrogate model that can characterize the mapping relationship between process parameters and system operating state can be established.

[0066] During system operation, a pre-set proxy model is embedded into the PLC control system, and real-time monitoring data of the pre-oxidation furnace is acquired in conjunction with on-site monitoring equipment to achieve rapid assessment of the current process status. Real-time monitoring data can include temperature monitoring data, airflow status data, and equipment operating parameters. By inputting the real-time monitoring data into the pre-set proxy model, the reaction state and temperature evolution trend inside the pre-oxidation furnace can be predicted on a millisecond timescale.

[0067] After obtaining the predicted results of the internal reaction state, and combining them with the optimized process parameters obtained in step S104, the circulating fan speed and heating power are dynamically adjusted to achieve real-time control of the gas flow intensity and heat input level within the furnace. This method ensures that the actual operating state continuously approaches the target operating state corresponding to the optimized scheme, and the fluid domain model is corrected online based on real-time monitoring results. This achieves dynamic updating and closed-loop control of the pre-oxidation furnace's operating state, improving the response speed and stability of process control. Please refer to [reference needed]. Figure 3The document presents a comparative schematic diagram of the surface temperature distribution of the wire bundle in the furnace before and after optimization, as provided in an embodiment of this application. Please refer to the provided diagram. Figure 4 The document presents a Pareto optimal frontier curve generated based on the NSGA-II algorithm, as provided in an embodiment of this application.

[0068] Please refer to Figure 5 This illustration shows a schematic diagram of a process parameter optimization device for the pre-oxidation process of large-tow carbon fibers provided in an embodiment of this application. The device includes an acquisition module 51 and a processing module 52, wherein... The acquisition module 51 is used to establish a fluid domain model corresponding to the pre-oxidation furnace through three-dimensional modeling technology, and to define multiple parallel-running fiber bundles as porous media regions; based on the fluid domain model, a multi-physics field synergistic strong coupling model is constructed. The multi-physics field synergistic strong coupling model is used to couple and solve the gas flow process, heat transfer process and fiber pre-oxidation reaction process, and output the multi-physics process state field set.

[0069] Processing module 52 is used to acquire process parameters during the carbon fiber pre-oxidation process, and based on the process parameters and the set of multi-physical process state fields, it reduces the optimization dimensionality through sensitivity analysis to obtain the target key variables; it constructs multiple optimization objectives using an improved non-dominated sorting genetic algorithm, and generates process parameter optimization schemes by defining preset constraints; the optimization objectives include temperature non-uniformity coefficient optimization objectives, unit energy consumption optimization objectives, and core-sheath structure difference optimization objectives; the constraints include fiber center temperature constraints, outlet average density constraints, and unit residence time constraints; based on the process parameter optimization schemes, the fluid domain model is corrected online using digital twin technology.

[0070] In one possible implementation, the acquisition module 51 is used to establish a fluid domain model corresponding to the pre-oxidation furnace using three-dimensional modeling technology, and to define multiple parallel-running fiber bundles as porous media regions. Specifically, this includes: establishing a fluid domain model based on the structural parameters of the pre-oxidation furnace using three-dimensional modeling technology; the fluid domain model includes an air inlet, an air outlet, a rectifier plate, a heater, and a fiber feeding channel; defining multiple parallel-running fiber bundles as porous media regions based on the specifications and arrangement of the fiber bundles, and describing the pressure drop when the airflow passes through the fiber bundles using a preset law; the geometric parameters corresponding to the porous media are consistent with the width and thickness of the fiber bundle sheets in actual production.

[0071] In one possible implementation, the acquisition module 51 is used to couple and solve the gas flow process, heat transfer process, and fiber pre-oxidation reaction process, and output a set of multi-physical process state fields: A turbulence model is constructed by solving the flow field control equations, and the boundary layer flow state of the furnace wall surface and fiber bundle surface in the pre-oxidation furnace is captured based on the turbulence model to obtain the gas flow state field inside the furnace; a heat transfer solution model is constructed by solving the energy equation containing the chemical internal heat source, and the heat transfer state between the air and fiber bundle inside the pre-oxidation furnace is captured based on the heat transfer solution model to obtain the temperature distribution state field inside the furnace; a reaction kinetic model is constructed by solving the fiber pre-oxidation reaction kinetic equations, and the reaction transformation state in the fiber pre-oxidation reaction process is captured based on the reaction kinetic model to obtain the fiber reaction transformation state field; a component transport model is constructed by solving the oxygen component transport equations, and the diffusion and consumption state of oxygen in the porous medium region is captured based on the component transport model to output the oxygen concentration distribution state field; the gas flow state field inside the furnace, the temperature distribution state field inside the furnace, the fiber reaction transformation state field, and the oxygen concentration distribution state field are used as a set of multi-physical process state fields.

[0072] In one possible implementation, the processing module 52 is used to acquire process parameters during the carbon fiber pre-oxidation process, and based on the process parameters and the set of multi-physical process state fields, to reduce the optimization dimensionality through sensitivity analysis to obtain target key variables. Specifically, this includes: acquiring process parameters during the carbon fiber pre-oxidation process, including temperature zone set temperature, circulating fan frequency, exhaust damper opening, and draw ratio; and based on the process parameters and the set of multi-physical process state fields, reducing the optimization dimensionality through sensitivity analysis to obtain target key variables, including fiber density standard deviation, maximum center temperature, and energy consumption.

[0073] In one possible implementation, the processing module 52 is used to construct multiple optimization objectives using an improved non-dominated sorting genetic algorithm, specifically including: optimizing the objectives by constructing a first objective function and optimizing the temperature non-uniformity coefficient. ; in, Denotes the first objective function. This represents minimizing the temperature non-uniformity coefficient. Indicates the target set temperature. Represents the spatial temperature distribution function. Indicates the volume of the control volume. Let the volume element be ; the objective is optimized by constructing a second objective function and optimizing the unit energy consumption: ; in, Describes the second objective function. This indicates minimizing the difference in skin-core structure. Indicates surface conversion rate, Indicates core conversion rate. This represents the average conversion rate; the objective is optimized by constructing a third objective function and optimizing the differences in the core-skin structure. ; in, This represents the third objective function. This represents minimizing unit energy consumption. Indicates heater power. This indicates the power of the fan.

[0074] In one possible implementation, the processing module 52 is used to perform online correction of the fluid domain model based on the process parameter optimization scheme using digital twin technology. Specifically, this includes: constructing a preset proxy model based on the big data generated by offline CFD; embedding the preset proxy model into the PLC control system and performing online correction of the fluid domain model based on the acquired real-time monitoring data; and performing online correction of the fluid domain model based on the real-time monitoring data and the process parameter optimization scheme.

[0075] In one possible implementation, the processing module 52 is used to perform online correction of the fluid domain model, specifically including: predicting the internal reaction state of the pre-oxidation furnace based on a preset surrogate model within milliseconds; and adjusting the fan speed and heating power based on the internal reaction state.

[0076] It should be noted that the above embodiments of the apparatus are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0077] This application also provides an electronic device. (See reference...) Figure 6 , Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: at least one processor 601, at least one communication bus 602, a user interface 603, at least one network interface 604, and a memory 605.

[0078] The communication bus 602 is used to enable communication between these components.

[0079] The user interface 603 may include a display screen and a camera. Optionally, the user interface 603 may also include a standard wired interface and a wireless interface.

[0080] The network interface 604 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0081] The processor 601 may include one or more processing cores. The processor 601 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 605, and by calling data stored in the memory 605. Optionally, the processor 601 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 601 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 601 and may be implemented as a separate chip.

[0082] The memory 605 may include random access memory (RAM) or read-only memory. Optionally, the memory 605 may include a non-transitory computer-readable storage medium. The memory 605 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 605 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 605 may also be at least one storage device located remotely from the aforementioned processor 601. (Refer to...) Figure 6The memory 605, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for optimizing process parameters for the pre-oxidation process of large-tow carbon fibers.

[0083] exist Figure 6 In the illustrated electronic device, the user interface 603 is primarily used to provide an input interface for the user and acquire user input data; while the processor 601 can be used to call the process parameter optimization application stored in the memory 605 for the pre-oxidation process of large-tow carbon fibers. When executed by one or more processors 601, the electronic device performs one or more of the methods described in the above embodiments. It should be noted that, for the foregoing method embodiments, for the sake of simplicity, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0084] This application also provides a non-transitory computer-readable storage medium storing instructions. When executed by one or more processors, these instructions cause an electronic device to perform one or more of the methods described in the above embodiments.

[0085] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0086] In the various embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between apparatuses or units may be electrical or other forms.

[0087] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0088] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0089] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0090] The above description is merely an exemplary embodiment disclosed in this application and should not be construed as limiting the scope of this application. Any equivalent changes and modifications made in accordance with the teachings of this application shall still fall within the scope of this application.

[0091] This application is intended to cover any variations, uses, or adaptations disclosed herein that follow the general principles disclosed herein and include common knowledge or customary technical means in the art that are not described in this application.

Claims

1. A method for optimizing process parameters in the pre-oxidation process of large-tow carbon fibers, characterized in that, The method includes: A fluid domain model corresponding to the pre-oxidation furnace was established using 3D modeling technology, and multiple parallel-running fiber bundles were defined as porous media regions. Based on the fluid domain model, a multi-physics field synergistic strong coupling model is constructed. The multi-physics field synergistic strong coupling model is used to couple and solve the gas flow process, heat transfer process and fiber pre-oxidation reaction process, and output the multi-physics process state field set. The process parameters in the carbon fiber pre-oxidation process are obtained, and based on the process parameters and the set of multi-physical process state fields, the optimization dimension is reduced by sensitivity analysis to obtain the target key variables. An improved non-dominated sorting genetic algorithm is used to construct multiple optimization objectives, and process parameter optimization schemes are generated by defining preset constraints. The optimization objectives include temperature non-uniformity coefficient optimization objective, unit energy consumption optimization objective, and core-sheath structure difference optimization objective. The constraints include fiber center temperature constraint, outlet average density constraint, and unit residence time constraint. Based on the process parameter optimization scheme, the fluid domain model is corrected online using digital twin technology.

2. The method according to claim 1, characterized in that, The process involves establishing a fluid domain model corresponding to the pre-oxidation furnace using 3D modeling technology, and defining multiple parallel-running fiber bundles as porous media regions, specifically including: Based on the structural parameters of the pre-oxidation furnace, a fluid domain model is established using three-dimensional modeling technology; the fluid domain model includes an air inlet, an air return outlet, a rectifier plate, a heater, and a wire feeding channel; Based on the specifications and arrangement of the fiber bundles, multiple parallel-running fiber bundles are defined as the porous medium region, and the pressure drop when airflow passes through the fiber bundles is described by a preset law; the geometric parameters corresponding to the porous medium are consistent with the width and thickness of the fiber bundle sheet in actual production.

3. The method according to claim 1, characterized in that, The process involves coupled solutions to the gas flow, heat transfer, and fiber pre-oxidation reactions, outputting a multi-physics process state field set. A turbulence model is constructed by solving the flow field control equations, and the boundary layer flow state of the furnace wall surface and fiber bundle surface in the pre-oxidation furnace is captured based on the turbulence model to obtain the gas flow state field inside the furnace. A heat transfer solution model is constructed by solving the energy equation containing the chemical internal heat source, and the heat transfer state between the air inside the pre-oxidation furnace and the fiber bundle is captured based on the heat transfer solution model to obtain the temperature distribution state field inside the furnace. A reaction kinetic model is constructed by solving the fiber pre-oxidation reaction kinetic equation, and the reaction transformation state during the fiber pre-oxidation reaction process is captured based on the reaction kinetic model to obtain the fiber reaction transformation state field. A component transport model is constructed by solving the oxygen component transport equation, and the diffusion and consumption state of oxygen in the porous medium region is captured based on the component transport model to output the oxygen concentration distribution state field. The gas flow state field inside the furnace, the temperature distribution state field inside the furnace, the fiber reaction and transformation state field, and the oxygen concentration distribution state field are taken as the set of state fields of the multi-physical process.

4. The method according to claim 1, characterized in that, The process parameters for obtaining carbon fiber pre-oxidation are then used, and based on these parameters and the multi-physical process state field set, sensitivity analysis is employed to reduce the optimization dimensionality and obtain the target key variables. Specifically, this includes: The process parameters for the carbon fiber pre-oxidation process are obtained, including the set temperature of the temperature zone, the frequency of the circulating fan, the opening degree of the exhaust damper, and the draw ratio. Based on the process parameters and the set of multi-physical process state fields, the optimization dimensionality is reduced by sensitivity analysis to obtain the target key variables, which include the standard deviation of fiber density, the highest center temperature, and energy consumption.

5. The method according to claim 1, characterized in that, The improved non-dominated sorting genetic algorithm is used to construct multiple optimization objectives, specifically including: The objective is optimized by constructing a first objective function and optimizing the temperature non-uniformity coefficient: ； in, Let represent the first objective function. This represents minimizing the temperature non-uniformity coefficient. Indicates the target set temperature. Represents the spatial temperature distribution function. Indicates the volume of the control volume. Let the volume element be ; the second objective function is constructed and the unit energy consumption optimization objective is optimized as follows: ； in, This represents the second objective function. This indicates minimizing the difference in skin-core structure. Indicates surface conversion rate, Indicates core conversion rate. This represents the average conversion rate; the objective is optimized by constructing a third objective function and optimizing the differences in the core-shell structure: ； in, This represents the third objective function. This represents minimizing unit energy consumption. Indicates heater power. This indicates the power of the fan.

6. The method according to claim 1, characterized in that, The optimization scheme based on the process parameters, which uses digital twin technology to correct the fluid domain model online, specifically includes: A pre-defined agent model is built based on the big data generated by offline CFD. The preset agent model is embedded into the PLC control system, and real-time monitoring data is acquired. The fluid domain model is corrected online based on the real-time monitoring data and the process parameter optimization scheme.

7. The method according to claim 6, characterized in that, The online correction of the fluid domain model specifically includes: Within milliseconds, the internal reaction state of the pre-oxidation furnace is predicted based on the preset proxy model; The fan speed and heating power are adjusted based on the internal reaction state.

8. A device for optimizing process parameters in the pre-oxidation process of large-tow carbon fibers, characterized in that, The device includes an acquisition module and a processing module, wherein, The acquisition module is used to establish a fluid domain model corresponding to the pre-oxidation furnace through three-dimensional modeling technology, and define multiple parallel-running fiber bundles as porous media regions; based on the fluid domain model, a multi-physics field synergistic strong coupling model is constructed, which is used to couple and solve the gas flow process, heat transfer process and fiber pre-oxidation reaction process, and output a set of multi-physics process state fields; The processing module is used to acquire process parameters during the carbon fiber pre-oxidation process, and based on the process parameters and the multi-physical process state field set, to reduce the optimization dimensionality through sensitivity analysis to obtain the target key variables; to construct multiple optimization objectives using an improved non-dominated sorting genetic algorithm, and to generate process parameter optimization schemes by defining preset constraints; the optimization objectives include temperature non-uniformity coefficient optimization objectives, unit energy consumption optimization objectives, and core-sheath structure difference optimization objectives; the constraints include fiber center temperature constraints, outlet average density constraints, and unit residence time constraints; based on the process parameter optimization schemes, the fluid domain model is corrected online using digital twin technology.

9. An electronic device, characterized in that, The device includes a processor, a communication bus, a user interface, a network interface, and a memory. The memory is used to store instructions. The user interface and the network interface are both used to communicate with other devices. The communication bus is used to enable communication between the components within the electronic device. The processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1-7.

10. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores instructions that, when executed, perform the method as described in any one of claims 1 to 7.

Images