Optimization design method of nozzle for CFRP-FDM printer

Abstract

The invention relates to an optimal design method of a nozzle for a CFRP-FDM printer, which comprises the following steps: according to the printing performance requirement of the printer, backstepping to determine the process parameter limit value of the printer and the necessary size of a melt co-extrusion runner of the nozzle, and establishing the geometric configuration of the melt co-extrusion runner; carrying out parametric modeling on the size to be designed; according to the melt extrusion process of the printer, establishing a fluid mechanics numerical simulation model, training the fluid mechanics numerical simulation model, and optimizing and searching the optimal geometric solution set of the melt co-extrusion runner by using a genetic algorithm, so as to obtain the optimal size of the melt co-extrusion runner in the nozzle. According to the optimization design method, the nozzle can be customized according to the printing performance requirement, the design size of the nozzle does not need to be obtained through a large number of simulation or experimental tests, and time and cost are saved.

Description

Technical Field

[0001] The present invention relates to the technical field of 3D printing, and in particular to an optimization design method for a nozzle of a CFRP-FDM printer. Background Art

[0002] Additive manufacturing (AM), also known as 3D printing, is a manufacturing technology based on computer-aided design (CAD) models that manufactures complex parts by slicing and stacking materials layer by layer. 3D printing technology can efficiently form parts with complex shapes at a low cost, and is widely used in aviation, automobile manufacturing, biomedicine and other fields. Common additive manufacturing technologies include photopolymerization, material extrusion, binding jetting, material jetting, sheet lamination, powder bed fusion and direct energy deposition, etc. Among these additive manufacturing technologies, fused deposition modeling (FDM), or fused filament fabrication (FFF), melts thermoplastic materials and extrudes them and stacks them layer by layer to form objects. Due to its simple structure, low cost, low difficulty in use and high printing freedom, it has become one of the most widely used AM technologies. On this basis, fused deposition modeling (CFRP-FDM) based on continuous fiber reinforced composites introduces high-strength continuous fibers, such as carbon or glass fibers, to enhance the mechanical properties of products and expand the application field. This innovative process combines the flexibility of FDM with the excellent performance of continuous fibers and is widely used in aviation, automobile and other fields, providing lightweight, high-strength customized solutions.

[0003] However, the continuous fiber FDM process still has problems such as low fiber content, difficult to control nozzle extrusion of resin, and slow extrusion printing speed, which limit the application and development of the process. The CFRP-FDM printing process is generally that the resin enters the flow channel from the inlet and is gradually heated and melted, while the fiber enters the flow channel from another inlet and is wrapped by the molten resin, and then is extruded by the co-extrusion nozzle. Obviously, the resistance that needs to be overcome during the entire process of the resin entering the flow channel and being melted and extruded (i.e., the feed force of the resin wire), the efficiency of the resin being heated and melted, and the resistance of the fiber in the flow channel all significantly affect the printing performance of the fiber nozzle. Among them, the smaller the resistance, the higher the heating efficiency, and the smaller the nozzle outlet size, the higher the CFRP-FDM printing speed and resolution upper limit, and these performances are almost affected by the nozzle flow channel geometry, and the traditional CFRP-FDM nozzle flow channel design method is mainly obtained through a large number of experimental tests, which undoubtedly costs a lot of time and cost.

[0004] Therefore, seeking an efficient and optimized manufacturing design method for CFRP-FDM printer nozzles customized for printing performance requirements may be an urgent problem to be solved by people in this field.

[0005] It should be stated that the above background technology content is only used to assist in understanding the inventive concept and technical solution of this patent application. It does not necessarily belong to the prior art of this patent application. In the absence of clear evidence that the above content has been disclosed before the filing date of this patent application, the above background technology should not be used to evaluate the novelty and creativity of this application. Summary of the invention

[0006] The purpose of the present invention is to overcome the defects of the prior art and provide a method for optimizing the design of a nozzle for a CFRP-FDM printer.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is:

[0008] A method for optimizing the design of a nozzle for a CFRP-FDM printer comprises the following steps:

[0009] Step S1, according to the printing performance requirements of the printer, reversely determine the process parameter limit values ​​of the printer and the necessary size of the melt co-extrusion flow channel of the nozzle;

[0010] Step S2, establishing the geometric configuration of the melt co-extrusion flow channel, and performing parameterized modeling on the dimensions to be designed;

[0011] Step S3, determining the design constraint conditions of the geometric configuration, and performing geometric sampling using a Latin hypercube algorithm combined with geometric constraints to obtain a small number of uniformly distributed geometric samples;

[0012] Step S4, establishing a fluid mechanics numerical simulation model according to the melt extrusion process of the printer;

[0013] Step S5, defining performance indicators of the melt co-extrusion flow channel that meet design expectations;

[0014] Step S6, substituting the geometric samples obtained in step S3 into the fluid mechanics numerical simulation model established in step S4 for calculation, obtaining the performance index value corresponding to each sample, and establishing a model training data set;

[0015] Step S7, using the small amount of geometric samples obtained in step S3 as input data, using the performance index value in step S6 as output data, training a proxy model for the entire design domain, and using a genetic algorithm to optimize the search for the optimal geometric solution set of the melt co-extrusion flow channel.

[0016] Preferably, in step S1, the necessary dimensions of the melt co-extrusion flow channel include: a diameter D of an inlet for the resin filament to enter, and a diameter d of an outlet for the molten material to be co-extruded;

[0017] The process parameter limits of the printer include: the maximum heating temperature T of the nozzle, the maximum speed u of the resin wire feeding, and the maximum speed u of the fiber wire feeding. f .

[0018] In some embodiments, in step S1, the maximum printing speed V, the resolution range, and the highest melting point Tm, the resin filament diameter Dp, and the fiber filament diameter d of the printer are first determined. f , where the resolution range includes layer height h and track width b;

[0019] Then, the necessary size of the melt co-extrusion flow channel and the process parameter limit of the printer are determined by reverse calculation, wherein the diameter D of the inlet is D p +0.2mm; the diameter d of the outlet satisfies d f <a≤b; the maximum heating temperature T of the nozzle is T=T m +50℃; Maximum feeding speed of resin wire Maximum fiber feeding speed u f =V.

[0020] In some embodiments, in step S3, when the Latin hypercube algorithm combined with geometric constraints is used for geometric sampling, the geometric constraints considered include: the spatial constraints of the melt co-extrusion flow channel, the continuity constraints of the melt co-extrusion flow channel, the size of the inlet for the resin filament to enter, and the size of the outlet for the molten material to be extruded.

[0021] In some embodiments, in step S4, the fluid mechanics numerical simulation model is established based on Fluent or OpenFOAM simulation software.

[0022] In some embodiments, in step S4, the fluid mechanics numerical simulation model is established based on the mass conservation equation, the momentum conservation equation, the energy conservation equation, and the boundary conditions and material properties, and is discretized using a finite volume algorithm. The Gauss elimination method is then used to iteratively calculate and solve the obtained numerical model, wherein the boundary conditions are determined according to the process parameter limit values ​​of the printer in step S1.

[0023] In some embodiments, in step S5, the performance indicators of the melt co-extrusion runner include at least: the wire feeding force F of the resin wire, the fiber wire feeding resistance F f , and extrusion temperature.

[0024] In some embodiments, the wire feeding force F of the resin wire material and the fiber wire feeding resistance F f According to the fluid mechanics numerical simulation model in step S4, the wire feeding force F of the resin wire is directly obtained: The fiber feeding resistance F f F f =∫ A τdA,

[0025] Where p is the pressure, obtained from simulation calculation; d f is the fiber diameter; τ is the shear force distribution of the fiber, which comes from simulation calculation; A is the area of ​​the fiber in the flow channel.

[0026] In some embodiments, in step S7, the proxy model is a Gaussian regression model, in which the input variables are the sample size set of the geometric configuration, and the output variables are the wire feeding force of the resin wire, the fiber wire feeding resistance, and the extrusion temperature.

[0027] In some embodiments, step S7 includes:

[0028] S71, substituting the model training data set obtained in step S6 into the Gaussian regression model for training;

[0029] S72, establish a data verification set, wherein the Latin hypercube algorithm combined with geometric constraints is used again to randomly generate N groups of spatially evenly distributed sample points as the data verification set of the Gaussian regression model, substitute them into the Gaussian regression model for verification, and obtain the prediction accuracy value R of the Gaussian regression model 2 , if the prediction accuracy value R 2If the preset value is not reached, return to step S71; if the prediction accuracy value R 2 When the preset value is reached, it indicates that the Gaussian regression model meets expectations, and the training of the Gaussian regression model is stopped, where N is a positive integer.

[0030] In some embodiments, in step S7, a multi-objective optimization problem is constructed with the wire feeding force of the resin wire of the nozzle, the fiber wire feeding resistance, and the extrusion temperature as targets, and a multi-objective optimization method based on an evolutionary algorithm is used to obtain the optimal geometric solution set of the melt co-extrusion flow channel.

[0031] In some embodiments, the multi-objective optimization method based on evolutionary algorithm is one of the multi-objective genetic algorithm NSGA-II, the multi-objective particle swarm algorithm MOPSO, the multi-objective artificial hummingbird algorithm MOAHA, and the multi-objective grey wolf optimization algorithm MOGWO.

[0032] In some embodiments, the design method further includes:

[0033] Step S8, combining design requirements and processing technology restrictions, selecting the optimal flow channel geometry from the optimal geometric solution set obtained in step S7, and obtaining the parts drawing of the nozzle accordingly.

[0034] Due to the application of the above technical solution, the present invention has the following advantages compared with the prior art: the optimization design method of the nozzle for the CFRP-FDM printer of the present invention can customize the nozzle according to the printing performance requirements, and does not need to obtain the design size of the nozzle through a large number of simulations or experimental tests, thus saving time and cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Attached Figure 1 This is a schematic diagram of the working principle of the nozzle used in the CFRP-FDM printer;

[0036] Attached Figure 2 It is a three-dimensional schematic diagram of the geometric configuration of the melt co-extrusion flow channel in a specific embodiment of the present invention;

[0037] Attached Figure 3 For attachment Figure 2 A longitudinal cross-sectional schematic diagram of a melt co-extrusion flow channel, wherein the determined dimensions are marked;

[0038] Attached Figure 4 For attachment Figure 2 A longitudinal cross-sectional schematic diagram of a melt coextrusion flow channel, wherein the geometric dimensions to be optimized are marked;

[0039] Attached Figure 5 To Attachment Fig.14 Schematic diagram of other geometric configurations of melt co-extrusion flow channels;

[0040] Attached Fig.15 Flowchart of Latin hypercube algorithm combined with geometric constraints;

[0041] Attached Fig.16 For attachment Figure 2 A schematic longitudinal cross-sectional view of the melt coextrusion flow channel of an embodiment after optimization. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art. Obviously, the embodiments described in this application are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of this application.

[0043] The present invention is directed to a nozzle optimization design method, which is directed to a nozzle for a CFRP-FDM printer, wherein the working principle of the nozzle is as follows: Figure 1 As shown, specifically, the nozzle has a melt co-extrusion flow channel, which has an inlet for resin filament 1 to enter, a fiber inlet for fiber 2 to enter, and an outlet for molten material to be co-extruded. During the printing process of the CFRP-FDM printer, the resin filament 1 enters the melt co-extrusion flow channel from the inlet and is gradually heated and melted, and the fiber 2 enters the melt co-extrusion flow channel from the fiber inlet and is wrapped by the molten resin, and then is co-extruded from the outlet. During the printing process, the size parameters of the melt co-extrusion flow channel play a key role in the printing performance of the nozzle. Therefore, the core of the design of the nozzle lies in the design of the melt co-extrusion flow channel.

[0044] The optimization design method of the nozzle for the CFRP-FDM printer of the present invention specifically comprises the following steps:

[0045] Step S1:

[0046] According to the printing performance requirements of the printer, the process parameter limit values ​​of the printer and the necessary dimensions of the melt co-extrusion flow channel of the nozzle are determined by reverse deduction.

[0047] In step S1, the necessary dimensions of the melt co-extrusion flow channel include: the diameter D of the inlet for the resin filament to enter, and the diameter d of the outlet for the molten material to be co-extruded; the process parameter limit values ​​of the printer include: the maximum heating temperature T of the nozzle, the maximum speed u of the resin filament feeding, and the maximum speed u of the fiber feeding. f .

[0048] Specifically, the following performance requirements of the printer can be determined first, such as the maximum printing speed V, the resolution range (including layer height h and track width b), and the highest melting point Tm, resin filament diameter Dp, and fiber filament diameter d in the expected printing series of resin materials. f Then, the necessary dimensions of the melt co-extrusion flow channel and the process parameter limit values ​​of the printer are determined by reverse deduction. In some embodiments, these necessary dimensions and process parameter limit values ​​are optionally as follows:

[0049] The diameter of the inlet is D p +0.2mm;

[0050] The diameter d of the outlet satisfies d f <a≤b;

[0051] The maximum heating temperature T of the nozzle is T = T m +50℃;

[0052] Maximum feeding speed of resin wire

[0053] Maximum fiber feeding speed u f =V.

[0054] Step S2:

[0055] The geometric configuration of the melt co-extrusion flow channel is established, and its dimensions to be designed are parametrically modeled.

[0056] Specifically, based on Figure 1 The working principle of the nozzle is established, and the geometric configuration of the melt co-extrusion flow channel is established. The geometric configuration can be one or more. For a specific geometric configuration, its size to be designed is determined, and the size to be designed is parameterized and modeled, such as Figure 2 , Figure 3 shown.

[0057] Step S3:

[0058] Determine the design constraints of the geometric configuration in step S2, and use the Latin hypercube algorithm combined with the geometric constraints to perform geometric sampling to obtain a small number of uniformly distributed geometric samples.

[0059] In step S3, when the Latin hypercube algorithm combined with geometric constraints is used for geometric sampling, the geometric constraints considered include: the spatial constraint conditions of the melt co-extrusion flow channel, the continuity constraint conditions of the melt co-extrusion flow channel, the size of the inlet for the resin filament to enter, and the size of the outlet for the molten material to be extruded,

[0060] The size of the inlet for the resin filament to enter and the size of the outlet for the molten material to extrude can be determined by the diameter D of the inlet and the diameter d of the outlet in the necessary dimensions obtained in step S1.

[0061] Step S4:

[0062] According to the melt extrusion process of the printer, a fluid mechanics numerical simulation model is established.

[0063] In step S4, the fluid mechanics numerical simulation model is established based on Fluent or Open FOAM simulation software. The establishment of the fluid mechanics numerical simulation model is based on mass, momentum, and energy conservation equations, and is established in combination with boundary conditions and material properties, and is discretized using a finite volume algorithm, and then the Gauss elimination method is used to iteratively calculate and solve the obtained numerical model, wherein the mass conservation equation, momentum conservation equation, and energy conservation equation are specifically selected from the simulation software Model format, which specifically includes:

[0064]

[0065]

[0066]

[0067] In the above conservation equations, ρ is density, u is fluid velocity vector, t is time, p is pressure; τ is fluid stress, g is gravitational acceleration, T is temperature; Cp is specific heat capacity, k is thermal conductivity;

[0068] The boundary conditions are determined according to the process parameter limit values ​​of the printer in step S1, namely, the maximum heating temperature T of the nozzle, the maximum speed u of the resin wire feeding, and the maximum speed u of the fiber wire feeding. f .

[0069] Step S5:

[0070] Define the performance indicators of melt coextrusion flow channels that meet design expectations.

[0071] In this step, the performance indicators of the melt co-extrusion flow channel include at least: the wire feeding force F of the resin wire, the fiber wire feeding resistance F f , and extrusion temperature, where the resin wire feeding force F and the fiber wire feeding resistance F f According to the fluid mechanics numerical simulation model in step S4, the wire feeding force F of the resin wire is directly obtained as follows: Fiber feeding resistance F f F f =∫ A τdA, where p is the pressure; d f is the fiber diameter; τ is the fluid stress, d is the outlet diameter, and A is the cross-sectional area of ​​the fiber.

[0072] Step S6:

[0073] Substitute the geometric samples obtained in step S3 into the fluid mechanics numerical simulation model established in step S4 for calculation, obtain the performance index value corresponding to each sample, and establish a model training data set.

[0074] Step S7:

[0075] A small amount of geometric samples obtained in step S3 are used as input data, and the performance index value in step S6 is used as output data to train a proxy model for the entire design domain, and a genetic algorithm is used to optimize the search for the optimal geometric solution set of the melt co-extrusion flow channel.

[0076] In this step, the proxy model adopts a Gaussian regression model. In the Gaussian regression model, the input variable is the sample size set of the geometric configuration in step S2, and the output variable is the wire feeding force, fiber wire feeding resistance, and extrusion temperature of the resin wire. The number of samples in the sample size set should be no less than 256 samples.

[0077] In this step, in order to make the response surface model obtained by the Gaussian regression process more accurate, two fitting optimizations were performed using local addition of points based on the predicted minimum value.

[0078] The model training data set obtained in step S6 is substituted into the Gaussian regression model for training, and then a data verification set is established. Specifically, the Latin hypercube algorithm with geometric constraints is recombined to randomly generate N groups of spatially evenly distributed sample points as the data verification set of the Gaussian regression model, and substitute them into the Gaussian regression model for verification, where N is a positive integer, preferably greater than 10, and here N is 25.

[0079] During verification, obtain the prediction accuracy value R of the Gaussian regression model 2 , if the prediction accuracy value R 2 If the preset value is not reached, return to continue training the Gaussian model; if the prediction accuracy value R 2 If the preset value is reached, it indicates that the Gaussian regression model has met expectations, and the training of the Gaussian regression model is stopped. 2 It is set to 0.98, indicating that the Gaussian regression model has high accuracy.

[0080] In this step, when a genetic algorithm is used to optimize and search for the optimal geometric solution set of the melt co-extrusion flow channel, a multi-objective optimization problem is constructed with the wire feeding force of the resin filament of the nozzle, the fiber wire feeding resistance, and the extrusion temperature as the objectives, and a multi-objective optimization method based on an evolutionary algorithm is used to obtain the optimal geometric solution set of the melt co-extrusion flow channel. The multi-objective optimization method based on an evolutionary algorithm is preferably a multi-objective genetic algorithm NSGA-II, and can also be one of a multi-objective particle swarm algorithm MOPSO, a multi-objective artificial hummingbird algorithm MOAHA, and a multi-objective grey wolf optimization algorithm MOGWO.

[0081] Step S8:

[0082] In combination with design requirements and processing technology restrictions, the optimal flow channel geometry is selected from the optimal geometric solution set obtained in step S7, and the nozzle part drawing is obtained accordingly.

[0083] In this step, the design requirements mainly include the following factors: the overall size of the printer, whether the printer needs to print curved parts (the nozzle end hits the part surface), etc.; the processing technology limitations generally refer to the difficulty of drilling and cutting (drilling with an angle is not as easy as vertical drilling, etc.), so that the designed parts are easy to manufacture.

[0084] The following is a description of a specific implementation plan:

[0085] It is necessary to design a nozzle for a printer, where the printing performance requirements of the printer are as follows: maximum printing speed V = 30mm / s (1800mm / min), printing layer height h*pass width d is 0.3*2mm, and the expected printing material is nylon 66 and epoxy resin impregnated fiber filament. It is known that the diameter of nylon 66 filament Dp = 1.75mm, and the fiber filament diameter d f =0.35mm, the highest melting point T of nylon 66 is 260℃.

[0086] The necessary dimensions of the melt co-extrusion flow channel of the nozzle are determined by reverse calculation as follows: the diameter D of the inlet in the melt co-extrusion flow channel is designed to be 2 mm, and the diameter of the outlet d is designed to be 0.5 mm;

[0087] The process parameters of the printer are determined by reverse deduction as follows:

[0088] The heating temperature of the nozzle is 260~310℃;

[0089] The maximum speed of resin wire feeding is

[0090] Maximum fiber feeding speed u f It is 30mm / s.

[0091] Use pycharm to call SCDM software for parametric modeling, and the established geometric model is as follows Figure 2 As shown, the dimensions determined are as follows Figure 3 As shown in , the geometric dimensions to be optimized are Figure 4 Of course, the geometric model can also be Figures 5 to 14 Any one of .

[0092] Design constraints for the above geometric configuration, that is, determine the size range of each geometric dimension to be optimized, as well as the relationship between the various dimensions;

[0093] Then, the Latin hypercube algorithm combined with geometric constraints is used for geometric sampling to obtain a small number of uniformly distributed geometric samples. The process of the Latin hypercube algorithm combined with geometric constraints is as follows: Fig.15 As shown;

[0094] Call pyFluent software in pycharm to establish a fluid mechanics numerical simulation model based on the melt extrusion process of the printer;

[0095] According to step S5, the performance index of the melt co-extrusion flow channel that meets the design expectation is defined, and the wire feeding force, fiber wire feeding resistance, and extrusion temperature of the resin wire are calculated;

[0096] Substituting the sampled geometric samples into the fluid mechanics numerical simulation model to calculate and obtain a data set;

[0097] Then, according to the method of step S7, the proxy model is used for fitting optimization, and the optimal solution set is obtained after optimization, such as Fig.16 shown.

[0098] Finally, combining the design requirements and processing technology limitations, according to the optimal flow channel geometry, the nozzle parts drawing is obtained, and the drawing is generated. After processing and assembly, a nozzle that meets the expected printing performance is obtained.

[0099] In summary, the optimization design method for the nozzle of the CFRP-FDM printer of the present invention can customize the nozzle according to the printing performance requirements, and does not need to obtain the design size of the nozzle through a large number of simulations or experimental tests, thus saving time and cost.

[0100] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable people familiar with this technology to understand the content of the present invention and implement it. They cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made according to the spirit of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing the design of a nozzle for a CFRP-FDM printer, characterized in that: The following steps are involved: Step S1, according to the printing performance requirements of the printer, reversely determine the process parameter limit values ​​of the printer and the necessary size of the melt co-extrusion flow channel of the nozzle; Step S2, establishing the geometric configuration of the melt co-extrusion flow channel, and performing parameterized modeling on the dimensions to be designed; Step S3, determining the design constraint conditions of the geometric configuration, and performing geometric sampling using a Latin hypercube algorithm combined with geometric constraints to obtain a small number of uniformly distributed geometric samples; Step S4, establishing a fluid mechanics numerical simulation model according to the melt extrusion process of the printer; Step S5, defining performance indicators of the melt co-extrusion flow channel that meet design expectations; Step S6, substituting the geometric samples obtained in step S3 into the fluid mechanics numerical simulation model established in step S4 for calculation, obtaining the performance index value corresponding to each sample, and establishing a model training data set; Step S7, using the small amount of geometric samples obtained in step S3 as input data, using the performance index value in step S6 as output data, training a proxy model for the entire design domain, and using a genetic algorithm to optimize the search for the optimal geometric solution set of the melt co-extrusion flow channel.

2. The optimization design method for a nozzle for a CFRP-FDM printer according to claim 1, characterized in that: In the step S1, the necessary dimensions of the melt co-extrusion flow channel include: a diameter D of an inlet for the resin filament to enter, and a diameter d of an outlet for the molten material to be co-extruded; The process parameter limits of the printer include: the maximum heating temperature T of the nozzle, the maximum speed u of the resin wire feeding, and the maximum speed u of the fiber wire feeding. f .

3. The optimization design method for a nozzle for a CFRP-FDM printer according to claim 2, characterized in that: In step S1, the maximum printing speed V, the resolution range, the highest melting point Tm, the resin filament diameter Dp, and the fiber filament diameter d of the printer are first determined. f , where the resolution range includes layer height h and track width b; Then, the necessary size of the melt co-extrusion flow channel and the process parameter limit of the printer are determined by reverse calculation, wherein the diameter D of the inlet is D p +0.2mm; the diameter d of the outlet satisfies d f <a≤b; the maximum heating temperature T of the nozzle is T=T m +50℃; Maximum feeding speed of resin wire Maximum fiber feeding speed u f =V.

4. The optimization design method for a nozzle for a CFRP-FDM printer according to claim 1, characterized in that: In step S3, when the Latin hypercube algorithm combined with geometric constraints is used for geometric sampling, the geometric constraints considered include: the spatial constraint conditions of the melt co-extrusion flow channel, the continuity constraint conditions of the melt co-extrusion flow channel, the size of the inlet for the resin filament to enter, and the size of the outlet for the molten material to be extruded.

5. The method for optimizing the design of a nozzle for a CFRP-FDM printer according to claim 1, characterized in that: In step S4, the fluid mechanics numerical simulation model is established based on Fluent or Open FOAM simulation software.

6. The optimization design method for a nozzle for a CFRP-FDM printer according to claim 1, characterized in that: In step S4, the fluid mechanics numerical simulation model is established based on the mass conservation equation, momentum conservation equation, energy conservation equation, and boundary conditions and material properties, and is discretized using a finite volume algorithm. The Gauss elimination method is then used to iteratively calculate and solve the obtained numerical model, wherein the boundary conditions are determined according to the process parameter limit values ​​of the printer in step S1.

7. The method for optimizing the design of a nozzle for a CFRP-FDM printer according to claim 1, characterized in that: In step S5, the performance indicators of the melt co-extrusion flow channel include at least: the wire feeding force F of the resin wire, the fiber wire feeding resistance F f , and extrusion temperature.

8. The method for optimizing the design of a nozzle for a CFRP-FDM printer according to claim 7, characterized in that: The resin wire feeding force F and the fiber wire feeding resistance F f According to the fluid mechanics numerical simulation model in step S4, the wire feeding force F of the resin wire is directly obtained: The fiber feeding resistance F f F f =∫ A τdA, Where p is the pressure, obtained from simulation calculation; d f is the fiber diameter; τ is the shear force distribution of the fiber, which comes from simulation calculation; A is the area of ​​the fiber in the flow channel.

9. The method for optimizing the design of a nozzle for a CFRP-FDM printer according to claim 1, characterized in that: In step S7, the proxy model is a Gaussian regression model, in which the input variables are the sample size set of the geometric configuration, and the output variables are the wire feeding force of the resin wire, the fiber wire feeding resistance, and the extrusion temperature.

10. The optimization design method for a nozzle for a CFRP-FDM printer according to claim 9, characterized in that: The step S7 comprises: S71, substituting the model training data set obtained in step S6 into the Gaussian regression model for training; S72, establish a data verification set, wherein the Latin hypercube algorithm combined with geometric constraints is used again to randomly generate N groups of spatially evenly distributed sample points as the data verification set of the Gaussian regression model, substitute them into the Gaussian regression model for verification, and obtain the prediction accuracy value R of the Gaussian regression model 2 , if the prediction accuracy value R 2 If the preset value is not reached, return to step S71; if the prediction accuracy value R 2 When the preset value is reached, it indicates that the Gaussian regression model meets expectations, and the training of the Gaussian regression model is stopped, where N is a positive integer.

11. The optimization design method for a nozzle for a CFRP-FDM printer according to claim 1, characterized in that: In the step S7, a multi-objective optimization problem is constructed with the resin wire feeding force of the nozzle, the fiber wire feeding resistance, and the extrusion temperature as the targets, and a multi-objective optimization method based on an evolutionary algorithm is used to obtain the optimal geometric solution set of the melt co-extrusion flow channel.

12. The optimization design method for a nozzle for a CFRP-FDM printer according to claim 11, characterized in that: The multi-objective optimization method based on evolutionary algorithm is one of the multi-objective genetic algorithm NSGA-II, the multi-objective particle swarm algorithm MOPSO, the multi-objective artificial hummingbird algorithm MOAHA, and the multi-objective grey wolf optimization algorithm MOGWO.

13. The optimization design method for a nozzle for a CFRP-FDM printer according to claim 1, characterized in that: The design method further comprises: Step S8, combining design requirements and processing technology restrictions, selecting the optimal flow channel geometry from the optimal geometric solution set obtained in step S7, and obtaining the parts drawing of the nozzle accordingly.

Images