Solidification forming process parameter optimization method and device, electronic equipment and medium

Abstract

The invention relates to a coagulation forming process parameter optimization method, which comprises the following steps of: constructing a coagulation forming process optimization model by taking a stock solution DMSO (Dimethylsulfoxide) concentration, an air layer height and a coagulation bath draw ratio in a coagulation forming process as decision variables and taking an absolute error between a nascent fiber DMSO residual volume and a nascent fiber radius and an expected radius value as an optimization target; based on the environment change, determining a plurality of environment windows, each environment window comprising a preset number of continuous dynamic environments; and based on a solidification forming process parameter time domain robust optimization algorithm, solving the solidification forming process optimization model in each environment window to obtain a deployment solution of each environment window. According to the method, the solidification forming process model is solved in the environment window, multiple environments in the same environment window multiplex deployment solutions to reduce the switching frequency of the solutions, theoretical support is provided for solidification forming process parameter setting in actual carbon fiber production, and the quality and performance of carbon fiber precursors are improved.

Description

Technical Field

[0001] This disclosure relates to the field of carbon fiber production technology, and in particular to a method, apparatus, electronic device and medium for optimizing solidification forming process parameters. Background Technology

[0002] Carbon fiber, a high-performance fiber material with a carbon content exceeding 90%, possesses high strength, high modulus, high temperature resistance, corrosion resistance, and lightweight properties, making it an indispensable strategic material in modern industry. The preparation process of carbon fiber mainly includes carbon fiber precursor production, pre-oxidation, carbonization, and graphitization. Among these, carbon fiber precursor is the key material in the production process; its quality directly affects the mechanical properties, structural stability, and processing performance of the final carbon fiber, thus determining its overall performance. The solidification and forming process, as a crucial step in carbon fiber precursor preparation, directly determines the quality of the nascent fiber through the setting of its process parameters, ultimately affecting the mechanical properties and stability of the carbon fiber. Therefore, obtaining high-quality carbon fiber precursor is a prerequisite for preparing high-performance carbon fibers.

[0003] The solidification process, a crucial step in carbon fiber precursor preparation, directly determines the quality of the nascent fibers, ultimately affecting their mechanical properties and stability. During solidification, the residual DMSO content in the nascent fibers and the absolute error between the nascent fiber radius and the desired radius are two key performance indicators for evaluating the quality of the solidification process. Excessive residual DMSO content in the nascent fibers will damage the cross-sectional shape and surface morphology of the carbon fiber precursor; an excessively large nascent fiber radius will lead to insufficient drawing force and reduced molecular chain orientation during subsequent drawing; and an excessively small nascent fiber radius may result in over-drawing or even fiber breakage during drawing. The concentration of DMSO in the dosing solution, the air layer height, and the drawing ratio of the solidification bath are the core process parameters affecting these two key indicators.

[0004] Currently, the relationship between solidification process parameters and nascent fiber performance indicators mainly relies on manual experience, lacking systematic theoretical guidance and failing to meet the needs of intelligent production. Modeling and optimizing the solidification process would allow for precise control of process parameters, thereby improving the quality and performance of nascent fibers. Therefore, there is an urgent need for an algorithm that can efficiently solve the solidification process parameter optimization problem, while also possessing stronger robustness and faster convergence speed. Summary of the Invention

[0005] To overcome the problems existing in related technologies, this disclosure provides a method, apparatus, electronic device and medium for optimizing solidification forming process parameters.

[0006] According to a first aspect of the present disclosure, a method for optimizing solidification forming process parameters is provided, the optimization method comprising: The concentration of DMSO in the raw solution, the height of the air layer, and the stretching ratio of the coagulation bath during the solidification process are used as decision variables, and the residual amount of DMSO in the nascent fiber and the absolute error between the nascent fiber radius and the desired radius value are used as optimization objectives to construct an optimization model for the solidification process. Based on environmental changes, multiple environmental windows are determined, and each environmental window includes a preset number of continuous dynamic environments. Based on the time-domain robust optimization algorithm for solidification forming process parameters, the optimization model of the solidification forming process is solved within each environment window to obtain the deployment solution for each environment window.

[0007] In some embodiments, determining multiple environment windows based on environmental changes includes: Based on different environmental conditions, environmental changes are determined, wherein the environmental conditions and the environmental changes correspond one-to-one; Based on the aforementioned environmental changes, determine the number of environmental changes; Based on the environmental change data, the plurality of environmental windows are determined.

[0008] In some embodiments, the time-domain robust optimization algorithm based on solidification process parameters solves the solidification process optimization model within each environment window to obtain the deployment solution for each environment window, including: Determine the number of decision variables, the range of constraints, and the objective function; When the current environment window is the first environment window or the second environment window, an initial population is generated based on the number of decision variables and the range of constraints; Based on the initial population, genetic iterative optimization is performed on all environments within the current environment window to obtain a robust solution set for the current environment window; Based on the robust solution set of the current environment window, the deployment solution of the current environment window is determined.

[0009] In some embodiments, the step of solving the solidification process optimization model within each environment window using a time-domain robust optimization algorithm based on solidification process parameters to obtain a deployment solution for each environment window further includes: When the current environment window is an environment window other than the first environment window and the second environment window, a high-quality initial population is generated based on the superior population generation strategy; Based on the high-quality initial population, genetic iterative optimization is performed on all environments within the current environment window to obtain a robust solution set for the current environment window; Based on the robust solution set of the current environment window, the deployment solution of the current environment window is determined.

[0010] In some embodiments, generating a high-quality initial population based on a superior population generation strategy includes: Based on the population center points of the robust solution sets of the two adjacent environmental windows of the current environmental window, predict the population center of the current environmental window and generate the predicted population. Random uniform individuals are generated based on Latin hypercube sampling; Based on the predicted population and the random uniform individuals, a high-quality initial population for the current environment window is obtained.

[0011] In some embodiments, the genetic iterative optimization includes: A subpopulation is obtained based on the population corresponding to the current environment window. When the current environment window is the first environment window or the second environment window, the population is the initial population. When the current environment window is any other environment window besides the first environment window or the second environment window, the population is the high-quality initial population. A temporary population is obtained based on the population and subpopulation corresponding to the current environment window; Based on the temporary population, an initial robust solution set is obtained; In response to the fact that the number of iterations has not reached the termination condition, the temporary population is determined as the population for the next iteration, and the initial robust solution set is updated; In response to the termination condition being reached after a certain number of iterations, the current initial robust solution set is determined as the robust solution set of the current environment window.

[0012] In some embodiments, determining the deployment solution for each environment window based on the robust solution set for each environment window includes: Based on the robust solution set of the current environment window, the optimization objective value of all individuals is normalized. Determine the ideal solution based on the normalized optimization objective value; Calculate the Euclidean distance between each individual in the robust solution set and the ideal solution, and select the individual with the smallest Euclidean distance as the deployment solution for the current environment window.

[0013] According to a second aspect of the present disclosure, a solidification forming process parameter optimization apparatus is provided, the optimization apparatus comprising: The module is configured to use the concentration of DMSO in the raw solution, the height of the air layer, and the stretching ratio of the coagulation bath as decision variables during the solidification process, and the residual amount of DMSO in the nascent fiber and the absolute error between the nascent fiber radius and the desired radius value as optimization objectives to build an optimization model for the solidification process. The determination module is configured to determine multiple environment windows based on environmental changes, each of which includes a preset number of continuous dynamic environments; The solution module is configured to solve the solidification process optimization model within each environment window using a time-domain robust optimization algorithm based on solidification process parameters, thereby obtaining the solidification process deployment solution for each environment window.

[0014] According to a third aspect of the present disclosure, an electronic device is provided, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to perform the optimization method as described in the first aspect of this disclosure.

[0015] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium is provided, wherein when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the optimization method as described in the first aspect of the present disclosure.

[0016] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects: The solidification forming process parameter optimization method disclosed herein uses the concentration of DMSO in the raw solution, the air layer height, and the coagulation bath draw ratio as decision variables, and the residual DMSO content in the nascent fiber and the absolute error between the nascent fiber radius and the desired radius as optimization objectives to construct a solidification forming process optimization model. Based on environmental changes, multiple environmental windows are determined, each including a preset number of continuous dynamic environments. Based on a time-domain robust optimization algorithm for solidification forming process parameters, the solidification forming process optimization model is solved within each environmental window to obtain the deployment solution for each environmental window. Multiple environments within the same environmental window reuse the deployment solution to reduce the number of solution switching times. This provides theoretical support for setting solidification forming process parameters in actual carbon fiber production, ensures stable and efficient production of carbon fiber solidification forming technology, and is beneficial for improving the quality and performance of carbon fiber precursor.

[0017] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0019] Figure 1 This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment.

[0020] Figure 2 This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment.

[0021] Figure 3 This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment.

[0022] Figure 4 This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment.

[0023] Figure 5 This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment.

[0024] Figure 6 This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment.

[0025] Figure 7 This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment.

[0026] Figure 8 This is a schematic diagram of an iterative algorithm for time-domain robust optimization of solidification process parameters and a schematic diagram of an environment window deployment solution selection scheme, according to an exemplary embodiment.

[0027] Figure 9a and 9b These are flowcharts illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment.

[0028] Figure 10 This is a block diagram illustrating a solidification forming process parameter optimization apparatus according to an exemplary embodiment.

[0029] Figure 11 This is a block diagram of an electronic device according to an exemplary embodiment. Detailed Implementation

[0030] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0031] In the solidification process, the residual DMSO content in nascent fibers refers to the amount of dimethyl sulfoxide (DMSO) remaining inside the nascent fibers after solidification. The nascent fiber radius refers to the actual radius of the fiber immediately after solidification, before subsequent stretching or other treatments. It is a key parameter reflecting the fiber's physical morphology and directly relates to the fiber's uniformity of thickness and mechanical properties. The desired radius value is a pre-set target radius value that the nascent fibers should achieve based on product design requirements or process standards. It serves as a benchmark for process optimization. By comparing the actual nascent fiber radius with this desired radius, the accuracy of the forming process can be measured, ensuring that the fiber size meets production requirements. Excessive residual DMSO content in nascent fibers will damage the cross-sectional shape and surface morphology of the carbon fiber filament. An excessively large nascent fiber radius will lead to insufficient stretching force and reduced molecular chain orientation during subsequent stretching. An excessively small nascent fiber radius may cause overstretching or even fiber breakage during stretching. Therefore, the residual DMSO content in nascent fibers and the absolute error between the nascent fiber radius and the desired radius value are two key performance indicators for evaluating the quality of the solidification process. The concentration of the original DMSO solution, the height of the air layer, and the stretching ratio of the coagulation bath are the core process parameters that affect these two key indicators.

[0032] In actual production processes, the lack of modeling and optimization of the solidification process makes it impossible to precisely control process parameters. Current parameter optimization research mostly employs traditional dynamic multi-objective optimization methods to solve this problem. Once the solidification bath environment changes, the optimization process is restarted to find a new optimal solution. Time-domain robust optimization algorithms are rarely used, failing to utilize effective information from historical populations and thus failing to generate superior initial populations. Therefore, there is an urgent need for an algorithm that can efficiently solve the solidification process parameter optimization problem while possessing stronger robustness and faster convergence speed.

[0033] To address the aforementioned issues, this disclosure provides a method for optimizing solidification forming process parameters. By solving the solidification forming process optimization model, multiple environments within the same environment window are reused to reduce the number of solution switching times. This provides theoretical support for setting solidification forming process parameters in actual carbon fiber production, ensures stable and efficient production of carbon fiber solidification forming technology, and helps improve the quality and performance of carbon fiber precursor.

[0034] The solidification forming process parameter optimization method provided in this embodiment is executed by an electronic device, which can be a mobile phone, tablet computer, laptop, smart robot, smart wearable device, or other smart device. In addition, the electronic device is equipped with various hardware resources and an energy storage device that provides electrical energy for the operation of these hardware resources.

[0035] Figure 1This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment, executed by an electronic device. See also... Figure 1 The method includes the following steps: Step S110: Using the concentration of DMSO in the raw solution, the height of the air layer, and the stretching ratio of the coagulation bath as decision variables during the solidification process, and the residual amount of DMSO in the nascent fiber and the absolute error between the nascent fiber radius and the desired radius value as optimization objectives, an optimization model for the solidification process is constructed.

[0036] Step S120: Based on environmental changes, determine multiple environment windows, each environment window including a preset number of continuous dynamic environments.

[0037] Step S130: Based on the time-domain robust optimization algorithm for solidification process parameters, solve the optimization model of solidification process within each environment window to obtain the deployment solution for each environment window.

[0038] In step S110, an optimization model for the solidification process is constructed based on the dual diffusion model, phase separation model, mass conservation model, and force balance model in the solidification process.

[0039] When the fiber stream enters the coagulation bath, a double diffusion phenomenon occurs due to the concentration difference between the fiber stream and the coagulation bath. During this process, the concentration of dimethyl sulfoxide within the fiber stream... and coagulant concentration The dual-diffusion model is in a dynamic state of flux and is as follows:

[0040] Where z is the axial direction, representing the direction along the length of the fiber stream (i.e., the direction in which the fiber stream flows in the coagulation bath), used to describe the dimethyl sulfoxide concentration at different length positions of the fiber stream. and coagulant concentration The distribution of methyl sulfoxide (DMSO). r represents the radial direction, extending from the center of the fiber stream to its edge, with the central axis of the fiber stream as the origin. It describes the concentration of DMSO at different radial locations within the fiber stream. and coagulant concentration Distribution of [something].

[0041] As the fiber stream undergoes double diffusion in the coagulation bath, the concentrations of various components within the fiber continuously change, and the fiber state gradually transforms from an initial liquid state to a solid-liquid mixed state until it is finally completely solidified. This transformation process is known as phase separation, and the phase separation model is as follows:

[0042] in, The concentration curve is for phase separation. This refers to the solidification bath temperature. Let z be the radius of the fiber stream at the axial position z. Let z be the concentration of the internal components of the fiber at the axial position z.

[0043] The flow rate of raw liquid delivered by the metering pump per unit time is constant. Based on this, the fiber radius R can be calculated, and the mass conservation model is as follows:

[0044] in, Let be the radius of the fiber-solid-liquid interface. At the spinneret location, since the fiber has not yet begun to solidify, When the fiber streams have completely solidified, , The initial fiber traction speed is when r=1.

[0045] During fiber solidification, the fiber is subjected to various forces, including rheological forces, inertial forces, and frictional forces. These forces remain in equilibrium during the solidification process. The velocity of the fiber along the z-axis can be determined using a force equilibrium model, as shown below:

[0046] in, For the initial rheological force, For viscous resistance, For osmotic pressure, is the viscosity coefficient of the fiber in the liquid state.

[0047] The symbols used in the four models above are shown in Table 1, and some fixed parameters and their specific values ​​are shown in Table 2.

[0048] Table 1. Model Symbols and Meanings

[0049] Table 2 Key Parameter Values

[0050] Air layer height refers to the vertical distance of the air zone that the molten liquid passes through after being extruded from the spinneret and before entering the coagulation bath. The air layer height directly affects the pre-forming state of the molten liquid. If the air layer height is too short, the molten liquid will not be sufficiently regularized before entering the coagulation bath, which can easily lead to wrinkles on the surface of the nascent fibers and uneven diameter (increasing radius error). If the air layer height is too long, the molten liquid is easily affected by gravity or airflow and will sag, which will also destroy the forming stability. At the same time, excessive solvent evaporation may indirectly affect the subsequent coagulation process, making it more difficult to control the residual amount of DMSO.

[0051] The coagulation bath draw ratio refers to the ratio of the length of the fiber after being drawn to its initial length before drawing, after applying tension to the nascent fiber that has just partially solidified in the coagulation bath. The draw ratio affects the structural stability of the fiber by adjusting the orientation of the polymer chains within the fiber. A reasonable draw ratio can reduce internal porosity and indirectly reduce DMSO residue (pores easily trap solvent). Since the fiber shrinks radially during drawing, precise control of the draw ratio can directly regulate the final radius of the nascent fiber, reducing the absolute error from the desired radius and ensuring that the dimensions meet design requirements.

[0052] Based on the four models mentioned above, this embodiment selects the air layer height during the solidification process as... The coagulation bath draw ratio is The concentration of the original DMSO solution is As the decision variable, the residual amount of DMSO in the nascent fiber Elementary fiber radius The absolute error between the radius and the desired radius is set as the optimization objective, and the function is used to optimize it. The solidification process optimization model, representing the relationship between the three decision variables and the two optimization objectives, is as follows:

[0053] in, As the desired radius value, this value is set to [value] in this embodiment. .

[0054] In step S120, the environmental change represents the change in the concentration of DMOS in the coagulation bath. In the actual carbon fiber coagulation process, the concentration of DMOS in the coagulation bath fluctuates within a certain range depending on environmental conditions, thus causing changes in the model parameters. Each change in the concentration of DMOS in the coagulation bath with environmental conditions is considered a new environment. Therefore, the change in the concentration of DMOS in the coagulation bath with changing environmental conditions can be regarded as a continuous dynamic environmental change. This continuous dynamic environmental change is divided into multiple environmental windows, so that each environmental window includes a preset number of continuous dynamic environments. For example, when the preset number is 2, each divided environmental window includes 2 dynamic environments.

[0055] In step S130, each environment window includes a preset number of continuous dynamic environments. Based on the solidification forming process parameter time-domain robust optimization algorithm, all environments included in each environment window are simultaneously optimized and solved to obtain the deployment solution for each environment window. Multiple dynamic environments in each environment window share a single deployment solution, which can reduce the number of solution switching caused by frequent changes in the environment, improve the generalization of the method to the environment, and thus reduce the switching cost, such as the cost of adjusting production parameters.

[0056] The solidification forming process parameter optimization method disclosed in this paper uses the concentration of DMSO in the raw solution, the air layer height, and the coagulation bath draw ratio as decision variables, and the residual DMSO content in the nascent fiber and the absolute error between the nascent fiber radius and the desired radius as optimization objectives to construct a solidification forming process optimization model. Based on environmental changes, multiple environmental windows are determined, each including a preset number of continuous dynamic environments. Based on a time-domain robust optimization algorithm for solidification forming process parameters, the solidification forming process optimization model is solved within each environmental window to obtain the deployment solution for each environmental window. This disclosure, by solving the solidification forming process optimization model, reuses deployment solutions for different environments within the same environmental window to reduce the number of solution switching times. This provides theoretical support for setting solidification forming process parameters in actual carbon fiber production, ensures stable and efficient production of carbon fiber solidification forming technology, and is beneficial for improving the quality and performance of carbon fiber precursor.

[0057] Figure 2 This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment, executed by an electronic device. See also... Figure 2 Based on environmental changes, multiple environmental windows are determined, including: S210. Based on different environmental conditions, determine the environmental changes, wherein the environmental conditions and environmental changes correspond one-to-one.

[0058] S220. Determine the number of environmental changes based on different environmental changes; S230. Based on the number of environmental changes, determine multiple environmental windows.

[0059] The dynamic environment design in this embodiment is as follows:

[0060] in, Indicates the environment a The concentration of DMSO in the coagulation bath is [value missing]. Nearby fluctuations, For the environment The concentration of DMSO in the lower coagulation bath is constant and is taken as 55%. Indicates the range of change, with values ​​ranging from 1 to 2. , The mean is Standard deviation is The value here is a normal distribution. , .

[0061] In step S210, different environmental conditions represent the environment. a Different values ​​of indicate changes in the concentration of DMOS in the coagulation bath, as the environment changes. Different values ​​correspond to different DMOS concentrations in the coagulation bath, and each environmental condition... For each DMOS concentration in a coagulation bath, the corresponding DMOS concentration value represents an environmental change. For example, in an environment... When taking 1, the result is and The difference indicates a change in the DMOS concentration in the coagulation bath under environment 1, thus determining the environmental changes under environment 1; in environment... When taking 2, the result is and The difference indicates a change in the DMOS concentration in the coagulation bath under environment 2, thus determining the environmental change under environment 2; in environment When taking 3, the result is and The difference indicates that the concentration of DMOS in the coagulation bath changes under environment 3, thus determining the environmental changes under environment 3.

[0062] In step S220, one change in the DMOS concentration in the coagulation bath corresponds to one environmental change, and one environmental change corresponds to an environmental change count of 1. Therefore, the number of times the DMOS concentration in the coagulation bath changes corresponds to the number of environmental changes. For example, in the environment... By selecting values ​​of 1, 2, 3, 4…100, 100 continuously varying DMOS concentrations in the coagulation bath can be obtained, i.e. , , , … These 100 coagulation bath DMOS concentrations represent 100 changes in the coagulation bath DMOS concentration, corresponding to 100 environmental changes, with the number of determined environmental changes being 100.

[0063] In step S230, the number of environmental changes is divided into groups based on the set number of environments, resulting in multiple environment windows. For example... Figure 8 As shown, environment When the range is [1, 100], for 100 continuously dynamically changing environments, the 100 continuously dynamically changing environments are divided into groups of two environments, resulting in 50 environment windows, and each environment window includes two continuously dynamic environments. Using the solidification forming process parameter time-domain robust optimization algorithm, the preset number of environments in each environment window can be optimized and solved simultaneously, avoiding restarting the optimization process to find a new optimal solution whenever the environment changes, reducing the number of solution switching times, and thus reducing the switching cost.

[0064] Figure 3 This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment, executed by an electronic device. See also... Figure 3Based on a time-domain robust optimization algorithm for solidification process parameters, the solidification optimization model is solved within each environment window to obtain the deployment solution for each environment window, including: S310. Determine the number of decision variables, the range of constraints, and the objective function.

[0065] S320. When the current environment window is the first or second environment window, generate an initial population based on the number of decision variables and the range of constraints.

[0066] S330. Based on the initial population, perform genetic iterative optimization on all environments within the current environment window to obtain a robust solution set for the current environment window.

[0067] S340. Based on the robust solution set of the current environment window, determine the deployment solution of the current environment window.

[0068] In step S310, to ensure the stability of the carbon fiber solidification process, three decision variables are used: the concentration of the original solution DMSO (… ), air layer height ( ), coagulation bath draw ratio ( It needs to be strictly controlled within a specific range, and the corresponding constraint range is:

[0069] The objective function represents the need to clearly define two key performance indicators in the solidification process parameter optimization model within the time-domain robust optimization algorithm. These indicators provide a clear and quantifiable evaluation benchmark for the subsequent robust solution. The two key performance indicators are the residual DMSO content in the nascent fiber and the absolute error between the nascent fiber radius and the desired radius. The residual DMSO content reflects the amount of residual solvent in the fiber, affecting the fiber cross-sectional shape and surface quality. The absolute error between the nascent fiber radius and the desired radius reflects the fiber dimensional stability, preventing breakage or insufficient molecular chain orientation during the drawing process.

[0070] In step S320, the dimensional structure of each initial individual is determined according to the number of decision variables, ensuring that each individual contains all decision variables to form a complete potential solution. Then, within the constraints corresponding to each decision variable, values ​​are assigned to each dimension through uniform random sampling, ultimately generating an initial population of a specified size. This ensures that all individuals in the initial population are valid solutions that meet the requirements of the actual problem, avoiding the algorithm from wasting computational power in the invalid solution space. Furthermore, a reasonable sampling method can achieve a uniform distribution of the initial population in the solution space, providing sufficient population diversity for subsequent genetic operations such as crossover and mutation. This effectively reduces the risk of the algorithm getting trapped in local optima and ensures the effectiveness and efficiency of the optimization iteration.

[0071] The initial population represents the set of solutions to the solidification process parameters, based on the concentration of the original solution DMSO ( ), air layer height ( ), coagulation bath draw ratio ( ) and the constraint boundaries that it needs to satisfy, and randomly generate a ) based on satisfying the above constraint boundaries. , and Sets, for example, generating an initial population containing 100 solutions, that is, 100 sets. , and The numerical values ​​are determined based on the objective function after generating the initial population. Specifically, the target value for each individual in the initial population is determined, namely, the residual amount of primary fiber DMSO and the absolute error between the primary fiber radius and the desired radius. This provides an evaluation basis for subsequent genetic operations, ensuring that the entire optimization process proceeds around the preset optimization objective and selecting solutions that meet the performance requirements.

[0072] In step S330, since the initial population is randomly generated, it may contain fragile solutions that cannot withstand fluctuations in the actual environment. Therefore, for all environments included in the current environment window, the initial population is subjected to genetic iterative optimization to obtain the robust solution set (RPS) for the current environment window. When the current environment window is the first environment window, the robust solution set for the first environment window can be obtained based on the initial population; when the current environment window is the second environment window, the robust solution set for the second environment window can be obtained based on the initial population. By generating the initial population, robust solution sets corresponding to the first and second environment windows are obtained respectively, providing preliminary data support and optimization foundation for generating high-quality initial populations for subsequent environment windows. Figure 8 As shown, an initial population is randomly generated, and genetic iteration optimization is performed on the two environments included in environment window 1 to obtain the robust solution set of the first environment window. Then, the initial population is used to perform genetic iteration optimization on the two environments included in environment window 2 to obtain the robust solution set of the second environment window.

[0073] In step S340, the deployment solution corresponding to the current environment window is determined based on the robust solution set of the current environment window. The parameters during the solidification process are adjusted to ensure that the deployment solution has robustness to adapt to all environments within the current environment window, thus avoiding system performance failure due to environmental fluctuations.

[0074] Figure 4 This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment, executed by an electronic device. See also... Figure 4 Based on a time-domain robust optimization algorithm for solidification process parameters, the optimization model is solved within each environment window to obtain the deployment solution for each environment window. This also includes: S410. When the current environment window is any other than the first environment window or the second environment window, a high-quality initial population is generated based on the superior population generation strategy.

[0075] S420. Based on a high-quality initial population, perform genetic iterative optimization on all environments within the current environment window to obtain a robust solution set for the current environment window.

[0076] S430. Based on the robust solution set of the current environment window, determine the deployment solution of the current environment window.

[0077] In step S410, when the current environment is an environment window other than the first environment window and the second environment window, a high-quality initial population is generated based on the excellent population generation strategy. The generated high-quality initial population provides a high-quality optimization starting point for determining robust solution sets for other environment windows. This can shorten the convergence time of genetic iteration optimization, improve the generation efficiency of robust solution sets, and ensure the quality of robust solution sets, thereby making the subsequently determined deployment solution more in line with the dynamic characteristics of the environment window.

[0078] like Figure 8 As shown, when the current environment window is environment window 3, a high-quality initial population for environment window 3 is obtained by using a superior population generation strategy based on the robust solution set t1 of environment window 1 and the robust solution set t2 of environment window 2. Similarly, when the current environment window is environment window 4, a high-quality initial population for environment window 4 is obtained by using a superior population generation strategy based on the robust solution set t2 of environment window 2 and the robust solution set t3 of environment window 3. Therefore, high-quality initial populations for each environment window in environment windows 5-50 can be obtained in the same way.

[0079] It is understood that steps S420 and S430 are the same as steps S330 and S340, and will not be explained here.

[0080] Figure 5 This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment, executed by an electronic device. See also... Figure 5 Based on superior population generation strategies, a high-quality initial population is generated, including: S510. Based on the population center points of the robust solution sets of the two adjacent environment windows of the current environment window, predict the population center of the current environment window and generate the predicted population.

[0081] S520, generating random uniform individuals based on Latin hypercube sampling.

[0082] S530. Based on the predicted population and random uniform individuals, obtain a high-quality initial population for the current environment window.

[0083] In step S510, when determining the high-quality initial population corresponding to the current environment window, the population center points of the robust solution sets of two environment windows located before and adjacent to the current environment window are selected to predict the population center of the current environment window and generate the predicted population. For example, if the current environment window is m, the robust solution sets of two environment windows, m-1 and m-2, are selected, and the evolutionary direction between the (m-2)th and (m-1)th environment windows is calculated. and step length The process is as follows:

[0084] in, The population center is located at the (m-1)th environmental window. The population center is located at the (m-2)th environmental window. Let represent the evolutionary direction between the (m-2)th and (m-1)th environmental windows, indicating the change in the population center between the two environmental windows. This reflects the shifting trend of the "population center" between the two environmental windows. Evolutionary step size is a quantification of the degree of change in the population center from the m-2 to the m-1 environmental window.

[0085] The center point of the k-th environment window The calculation method is as follows:

[0086] In the formula, This represents the cardinality of the k-th environment window RPS. For example, if the number of individuals in the RPS is 100, express The i-th individual in the group.

[0087] Through the center point of the kth environmental window mentioned above The calculation method can yield the following results. and .

[0088] And in 100 individuals Based on this, generate 100 new individuals. The calculation method is as follows:

[0089] in, This is a Gaussian perturbation term used to handle uncertainties and biases in the prediction process, improving the algorithm's convergence performance while avoiding getting trapped in local optima. 100 new individuals are generated. This represents the predicted population corresponding to the current environmental window.

[0090] In step S520, Latin Hypercube Sampling (LHS) is used to generate random uniform individuals. This involves uniformly dividing each dimension of the decision vector into several intervals and randomly selecting sample points within each interval, thereby randomly generating uniformly distributed individuals in the decision vector space. For example, LHS can be used to generate 100 random uniform individuals.

[0091] In step S530, the optimization objective value corresponding to each individual in the predicted population and the randomly uniform individuals is determined according to the objective function. The generated predicted population and the generated randomly uniform individuals are merged. Based on the optimization objective value of each individual, a certain number of individuals are selected as a high-quality initial population for the current environment window using non-dominated sorting and crowding distance, which is then used to optimize the current environment window. For example, the generated predicted population includes 100 individuals, and the randomly uniform individuals include 100 individuals. After merging, 200 individuals are obtained. 100 excellent individuals are selected using non-dominated sorting and crowding distance, which are then used as a high-quality initial population for the current environment window, which is then used to optimize the current environment window.

[0092] This embodiment combines effective information from historical populations with randomly uniform individuals generated by Latin hypercube sampling to obtain a high-quality initial population. This avoids starting optimization from random initialization, reduces computational resource consumption, accelerates algorithm convergence, and effectively copes with frequent environmental changes.

[0093] Figure 6 This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment, executed by an electronic device. See also... Figure 6 Genetic iterative optimization includes: S610. Based on the population corresponding to the current environment window, obtain a subpopulation. When the current environment window is the first environment window or the second environment window, the population is the initial population. When the current environment window is another environment window besides the first environment window or the second environment window, the population is a high-quality initial population.

[0094] S620. Obtain a temporary population based on the population and subpopulation corresponding to the current environment window.

[0095] S630. Obtain the initial robust solution set based on the temporary population.

[0096] S640. In response to the fact that the number of iterations has not reached the termination condition, determine the temporary population as the initial population for the next iteration and update the initial robust solution set.

[0097] S650. In response to the termination condition being reached by the number of iterations, determine the current initial robust solution set as the robust solution set of the current environment window.

[0098] In step S610, genetic operations such as selection, crossover, and mutation are performed on the population corresponding to the current environment window to generate a subpopulation. For example, if the population corresponding to the current environment window contains 100 individuals, genetic operations such as selection, crossover, and mutation are performed on these 100 individuals to generate a subpopulation with 100 individuals.

[0099] In one example, when the current environment window is the first environment window or the second environment window, the corresponding population is a randomly generated initial population.

[0100] In another example, when the current environment window is an environment window other than the first and second environment windows, such as the third environment window, the fourth environment window, etc., the corresponding initial population is a high-quality initial population generated based on the superior population generation strategy and corresponding to the current environment window.

[0101] In step S620, the population and subpopulations are merged to obtain a temporary population. For example, if the population contains 100 individuals and the subpopulation contains 100 individuals, the merged population will have 200 individuals.

[0102] In step S630, the individuals in the temporary population are sorted in ascending order of non-dominance level and descending order of crowding distance (NSGAII, Pareto front). The top 100 individuals are selected from the temporary population to update the population, and then the robust solution set is updated to obtain the initial robust solution set. For example, if 200 individuals in the temporary population are sorted in ascending order of non-dominance level and descending order of crowding distance, the top 100 individuals are selected from the temporary population to update the population, and then the robust solution set is updated, the updated robust solution set includes the top 100 individuals selected from the temporary population, and the resulting initial robust solution set includes these 100 individuals.

[0103] In step S640, the termination condition represents the set number of iterations that must be satisfied. If the termination condition is not met at the current iteration count, the updated population is used as the initial population for a new round, and steps S610-S630 are repeated. This avoids the obtained initial robust solution set remaining at a "locally good" rather than a "globally optimal" state, while enhancing the stability and adaptability of the initial robust solution set. For example, setting the number of iterations to 10,000, after multiple iterations, the final initial robust solution set is more likely to approach the global optimum and has stronger stability and adaptability in all environments within the current environment window.

[0104] In step S650, when it is determined that the current iteration number meets the termination condition, the currently obtained initial robust solution set is used as the robust solution set of the current environment window.

[0105] In this embodiment, when the current environment window is the first or second environment window, the initial population is used to perform genetic iterative optimization on all environments within the first or second environment window to determine the robust solution set of the first or second environment window. When the current environment window is another environment window (the third, fourth, etc.), a high-quality initial population corresponding to the current environment window is obtained through an excellent population generation strategy. Genetic iterative optimization is then performed on all environments within the current environment window to determine the robust solution set of the current environment window. This solves the problems of high cost and slow response speed in real-time switching of solidification forming process parameters, improves the convergence speed of the time-domain robust optimization algorithm, and provides theoretical support for setting solidification forming process parameters in actual carbon fiber production.

[0106] Figure 7 This is a flowchart illustrating a method for optimizing solidification forming process parameters according to an exemplary embodiment, executed by an electronic device. See also... Figure 7 Based on the robust solution set for each environment window, the deployment solution for each environment window is determined, including: S710. Based on the robust solution set of the current environment window, normalize the optimization objective value of all individuals.

[0107] S720. Determine the ideal solution based on the normalized optimization objective value.

[0108] S730. Calculate the Euclidean distance between each individual in the robust solution set and the ideal solution, and select the individual with the smallest Euclidean distance as the deployment solution for the current environment window.

[0109] In step S710, the robust solution set of the current environment window includes multiple individuals. The optimization objective value for each individual is determined according to the objective function, and the optimization objective value for each individual is normalized to obtain the normalized optimization objective value for each individual, which is then represented in coordinate form. For example... Figure 8 As shown, if the RPS of the current environment window includes 100 individuals, the optimization objective values ​​corresponding to these 100 individuals are normalized to obtain 100 coordinate values, where the coordinates of the optimization status of the i-th individual are... , ,in, This represents the normalized value of the i-th individual on the target amount of DMSO residue in primary fibers. This represents the normalized value of the i-th individual on the target absolute error between the nascent fiber radius and the desired radius.

[0110] In step S720, the normalized optimization objective values ​​of all individuals are compared. The ideal solution is formed when all optimization objective values ​​are minimized. In this embodiment, the optimization objectives are the residual DMSO content of the nascent fiber and the absolute error between the nascent fiber radius and the desired radius. The minimum values ​​of the two optimization objectives are determined from the normalized optimization objective values, and these two minimum values ​​constitute the ideal solution. Figure 8 As shown, This indicates all elements within the current environment window. The minimum value in the middle. This indicates all elements within the current environment window. The minimum value is found, and the ideal solution at this point is... .

[0111] In step S730, when the robust solution set includes 100 individuals, the Euclidean distance between the normalized optimization objective value and the ideal solution is calculated for each individual. The ideal solution is the one where both the i-th individual and the objective value are minimized. The Euclidean distance is:

[0112] Comparing all individuals , and select The smallest individual is used as the deployment solution for the current environment window.

[0113] For ease of understanding, a specific embodiment is given below to describe the solidification forming process parameter optimization method of this disclosure, such as... Figure 9a and 9b As shown, the method includes the following steps: S901. Using the concentration of DMSO in the raw solution, the height of the air layer, and the stretching ratio of the coagulation bath as decision variables during the solidification process, and the residual amount of DMSO in the nascent fiber and the absolute error between the nascent fiber radius and the desired radius value as optimization objectives, an optimization model for the solidification process is constructed.

[0114] S902. Based on different environmental conditions, determine the environmental changes, wherein the environmental conditions and environmental changes correspond one-to-one.

[0115] S903. Determine the number of environmental changes based on environmental changes.

[0116] S904. Based on the number of environmental changes, determine multiple environmental windows.

[0117] S905. Determine the number of decision variables, the range of constraints, and the objective function.

[0118] S906. Generate an initial population based on the number of decision variables and the range of constraints.

[0119] S907. When the current environment window is the first environment window, based on the initial population, perform genetic iterative optimization on all environments within the first environment window to obtain the robust solution set of the first environment window.

[0120] Genetic iteration includes: Based on the population corresponding to the current environment window, obtain the subpopulation; A temporary population is obtained based on the population and subpopulation corresponding to the current environment window; Based on the temporary population, obtain the initial robust solution set; If the number of iterations has not reached the termination condition, a temporary population is determined as the population for the next iteration, and the initial robust solution set is updated. In response to the termination condition being reached, the current initial robust solution set is determined as the robust solution set of the current environment window.

[0121] S908. Determine the deployment solution for the first environment window.

[0122] Among them, the optimization objective value of all individuals in the robust solution set based on the current environment window is normalized. Determine the ideal solution based on the normalized optimization objective value; Calculate the Euclidean distance between each individual in the robust solution set and the ideal solution, and select the individual with the smallest Euclidean distance as the deployment solution for the current environment window.

[0123] S909. When the current environment window is the second environment window, based on the initial population, perform genetic iterative optimization on all environments within the second environment window to obtain the robust solution set of the second environment window.

[0124] S910. Determine the deployment solution for the second environment window.

[0125] For the specific process of steps S909 to S910, please refer to steps S907 to S908.

[0126] S911. When the current environment window is the third environment window, a high-quality initial population of the third environment window is obtained based on the superior population generation strategy.

[0127] Among them, based on superior population generation strategies, a high-quality initial population is generated, including: Based on the population center points of the robust solution sets of the two adjacent environmental windows of the current environmental window, predict the population center of the current environmental window and generate the predicted population. Random uniform individuals are generated based on Latin hypercube sampling; Based on the predicted population and random uniform individuals, a high-quality initial population is obtained for the current environmental window.

[0128] High-quality populations that have obtained the third environmental window include: Based on the population center points of the robust solution sets of the first and second environment windows, predict the population center of the third environment window and generate the predicted population. Random uniform individuals are generated based on Latin hypercube sampling; A high-quality initial population for the third environmental window is obtained based on the predicted population and random uniform individuals.

[0129] S912. Based on the high-quality initial population of the third environment window, perform genetic iteration optimization on all environments within the third environment window to obtain the robust solution set of the third environment window.

[0130] S913, Determine the deployment solution for the third environment window.

[0131] For the specific steps to determine the deployment solution of the third environment window, please refer to step S908.

[0132] Repeat steps S911 to S913 above to determine the corresponding deployment solution when the current environment window is the fourth, fifth...fiftieth environment window in turn.

[0133] Figure 10 This is a flowchart illustrating a solidification forming process parameter optimization apparatus according to an exemplary embodiment, see [link to flowchart]. Figure 10 The device includes: Module 1010 is configured to use the concentration of DMSO in the raw solution, the height of the air layer, and the stretching ratio of the coagulation bath as decision variables during the solidification process, and the residual amount of DMSO in the nascent fiber and the absolute error between the nascent fiber radius and the desired radius value as optimization objectives to construct an optimization model for the solidification process.

[0134] The determination module 1020 is configured to determine multiple environment windows based on environmental changes, each environment window including a preset number of continuous dynamic environments.

[0135] The solver module 1030 is configured to solve the solidification process optimization model within each environment window using a time-domain robust optimization algorithm based on solidification process parameters, thereby obtaining the deployment solution for each environment window.

[0136] In some embodiments, the determining module 1020 is further configured to: Based on different environmental conditions, environmental changes are determined, where there is a one-to-one correspondence between environmental conditions and environmental changes; Determine the number of environmental changes based on environmental changes; Based on the number of environmental changes, multiple environmental windows are determined.

[0137] In some embodiments, the solver module 1030 is further configured to: Determine the number of decision variables, the range of constraints, and the objective function; When the current environment window is the first environment window or the second environment window, an initial population is generated based on the number of decision variables and the range of constraints; Based on the initial population, genetic iteration optimization is performed on all environments within the current environment window to obtain a robust solution set for the current environment window; Based on the robust solution set of the current environment window, determine the deployment solution for the current environment window.

[0138] In some embodiments, the solver module 1030 is further configured to: When the current environment window is any other than the first environment window or the second environment window, a high-quality initial population is generated based on the superior population generation strategy. Based on a high-quality initial population, genetic iteration optimization is performed on all environments within the current environment window to obtain a robust solution set for the current environment window; Based on the robust solution set of the current environment window, determine the deployment solution for the current environment window.

[0139] In some embodiments, the solver module 1030 is further configured to: Based on the population center points of the robust solution sets of the two adjacent environmental windows of the current environmental window, predict the population center of the current environmental window and generate the predicted population. Random uniform individuals are generated based on Latin hypercube sampling; Based on the predicted population and random uniform individuals, a high-quality initial population is obtained for the current environmental window.

[0140] In some embodiments, the solver module 1030 is further configured to: Based on the population corresponding to the current environment window, a subpopulation is obtained. When the current environment window is the first environment window or the second environment window, the population is the initial population. When the current environment window is any other environment window besides the first or second environment window, the population is a high-quality initial population. A temporary population is obtained based on the population and subpopulation corresponding to the current environment window; Based on the temporary population, obtain the initial robust solution set; If the number of iterations has not reached the termination condition, a temporary population is determined as the population for the next iteration, and the initial robust solution set is updated. In response to the termination condition being reached, the current initial robust solution set is determined as the robust solution set of the current environment window.

[0141] In some embodiments, the solver module 1030 is further configured to: Based on the robust solution set of the current environment window, the optimization objective value of all individuals is normalized. Determine the ideal solution based on the normalized optimization objective value; Calculate the Euclidean distance between each individual in the robust solution set and the ideal solution, and select the individual with the smallest Euclidean distance as the deployment solution for the current environment window.

[0142] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0143] Each module in the aforementioned solidification forming process parameter optimization device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0144] In one exemplary embodiment, an electronic device is provided, including: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to execute the solidification forming process parameter optimization method in the above embodiments.

[0145] Figure 11 This is a block diagram of an electronic device 1100 according to an exemplary embodiment.

[0146] Electronic device 1100 may include one or more of the following components: processing component 1102, memory 1104, power supply component 1106, multimedia component 1108, audio component 1110, input / output (I / O) interface 1112, sensor component 1114, and communication component 1116.

[0147] Processing component 1102 typically controls the overall operation of electronic device 1100, such as operations associated with display, telephone calls, data communication, camera operation, and recording operations. Processing component 1102 may include one or more processors 1120 to execute instructions to perform all or part of the steps of the methods described above. Furthermore, processing component 1102 may include one or more modules to facilitate interaction between processing component 1102 and other components. For example, processing component 1102 may include a multimedia module to facilitate interaction between multimedia component 1108 and processing component 1102.

[0148] Memory 1104 is configured to store various types of data to support the operation of electronic device 1100. Examples of such data include instructions for any application or method operating on electronic device 1100, contact data, phonebook data, messages, pictures, videos, etc. Memory 1104 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0149] Power supply component 1106 provides power to various components of electronic device 1100. Power supply component 1106 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 1100.

[0150] Multimedia component 1108 includes a screen that provides an output interface between the electronic device 1100 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 1108 includes a front-facing camera and / or a rear-facing camera. When the electronic device 1100 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0151] Audio component 1110 is configured to output and / or input audio signals. For example, audio component 1110 includes a microphone (MIC) configured to receive external audio signals when electronic device 1100 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 1104 or transmitted via communication component 1116. In some embodiments, audio component 1110 also includes a speaker for outputting audio signals.

[0152] I / O interface 1112 provides an interface between processing component 1102 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0153] Sensor assembly 1114 includes one or more sensors for providing state assessments of various aspects of electronic device 1100. For example, sensor assembly 1114 may detect the on / off state of electronic device 1100, the relative positioning of components such as the display and keypad of electronic device 1100, changes in position of electronic device 1100 or a component of electronic device 1100, the presence or absence of user contact with electronic device 1100, the orientation or acceleration / deceleration of electronic device 1100, and temperature changes of electronic device 1100. Sensor assembly 1114 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 1114 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 1114 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0154] Communication component 1116 is configured to facilitate wired or wireless communication between electronic device 1100 and other devices. Electronic device 1100 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 1116 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 1116 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0155] In an exemplary embodiment, the electronic device 1100 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0156] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 1104 including instructions, which can be executed by a processor 1120 of an electronic device 1100 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0157] A non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to perform the solidification forming process parameter optimization method provided by an exemplary embodiment of this disclosure.

[0158] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the solidification forming process parameter optimization method described in the above embodiments.

[0159] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0160] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for optimizing parameters in a solidification forming process, characterized in that, The optimization method includes: The concentration of DMSO in the raw solution, the height of the air layer, and the stretching ratio of the coagulation bath during the solidification process are used as decision variables, and the residual amount of DMSO in the nascent fiber and the absolute error between the nascent fiber radius and the desired radius value are used as optimization objectives to construct an optimization model for the solidification process. Based on environmental changes, multiple environmental windows are determined, and each environmental window includes a preset number of continuous dynamic environments. Based on the time-domain robust optimization algorithm for solidification forming process parameters, the optimization model of the solidification forming process is solved within each environment window to obtain the deployment solution for each environment window.

2. The method for optimizing solidification forming process parameters according to claim 1, characterized in that, The determination of multiple environmental windows based on environmental changes includes: Based on different environmental conditions, environmental changes are determined, wherein the environmental conditions and the environmental changes correspond one-to-one; Based on the aforementioned environmental changes, determine the number of environmental changes; Based on the environmental change data, the plurality of environmental windows are determined.

3. The method for optimizing solidification forming process parameters according to claim 1, characterized in that, The time-domain robust optimization algorithm based on solidification process parameters solves the solidification process optimization model within each environment window to obtain the deployment solution for each environment window, including: Determine the number of decision variables, the range of constraints, and the objective function; When the current environment window is the first environment window or the second environment window, an initial population is generated based on the number of decision variables and the range of constraints; Based on the initial population, genetic iterative optimization is performed on all environments within the current environment window to obtain a robust solution set for the current environment window; Based on the robust solution set of the current environment window, the deployment solution of the current environment window is determined.

4. The method for optimizing solidification forming process parameters according to claim 3, characterized in that, The time-domain robust optimization algorithm based on solidification process parameters solves the solidification process optimization model within each environment window to obtain the deployment solution for each environment window, and further includes: When the current environment window is an environment window other than the first environment window and the second environment window, a high-quality initial population is generated based on the superior population generation strategy; Based on the high-quality initial population, genetic iterative optimization is performed on all environments within the current environment window to obtain a robust solution set for the current environment window; Based on the robust solution set of the current environment window, the deployment solution of the current environment window is determined.

5. The method for optimizing solidification forming process parameters according to claim 4, characterized in that, The generation of a high-quality initial population based on an excellent population generation strategy includes: Based on the population center points of the robust solution sets of the two adjacent environmental windows of the current environmental window, predict the population center of the current environmental window and generate the predicted population. Random uniform individuals are generated based on Latin hypercube sampling; Based on the predicted population and the random uniform individuals, a high-quality initial population for the current environment window is obtained.

6. The method for optimizing solidification forming process parameters according to claim 4, characterized in that, The genetic iterative optimization includes: A subpopulation is obtained based on the population corresponding to the current environment window. When the current environment window is the first environment window or the second environment window, the population is the initial population. When the current environment window is any other environment window besides the first environment window or the second environment window, the population is the high-quality initial population. A temporary population is obtained based on the population and subpopulation corresponding to the current environment window; Based on the temporary population, an initial robust solution set is obtained; In response to the fact that the number of iterations has not reached the termination condition, the temporary population is determined as the population for the next iteration, and the initial robust solution set is updated; In response to the termination condition being reached after a certain number of iterations, the current initial robust solution set is determined as the robust solution set of the current environment window.

7. The method for optimizing solidification forming process parameters according to claim 4, characterized in that, The determination of the deployment solution for each environment window based on the robust solution set for each environment window includes: Based on the robust solution set of the current environment window, the optimization objective value of all individuals is normalized. Determine the ideal solution based on the normalized optimization objective value; Calculate the Euclidean distance between each individual in the robust solution set and the ideal solution, and select the individual with the smallest Euclidean distance as the deployment solution for the current environment window.

8. A device for optimizing parameters in a solidification forming process, characterized in that, The optimization device includes: The module is configured to use the concentration of DMSO in the raw solution, the height of the air layer, and the stretching ratio of the coagulation bath as decision variables during the solidification process, and the residual amount of DMSO in the nascent fiber and the absolute error between the nascent fiber radius and the desired radius value as optimization objectives to build an optimization model for the solidification process. The determination module is configured to determine multiple environment windows based on environmental changes, each of which includes a preset number of continuous dynamic environments; The solution module is configured to solve the solidification process optimization model within each environment window using a time-domain robust optimization algorithm based on solidification process parameters, thereby obtaining the deployment solution for each environment window.

9. An electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to perform the optimization method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device is able to perform the optimization method as described in any one of claims 1 to 7.

Images