Intelligent optimization device for food 3D printing material

Abstract

The invention belongs to the technical field of food processing equipment, and particularly relates to an intelligent optimization device for a food 3D printing material. According to the device, a multi-mode sensing technology is adopted to automatically monitor mechanical properties and form information of a printing base material, high-resolution image data of extrusion deformation and extrusion form of the material are obtained through an image acquisition device, and key mechanical parameters such as hardness and resilience are rapidly predicted in combination with principal component analysis and a machine learning algorithm; and dynamic matching is conducted according to the real-time monitoring data and a preset printing quality threshold value, and when it is detected that the material characteristics or the printing effect deviates from the optimal interval, the particle swarm optimization algorithm is automatically triggered to optimize the printing parameters, and the execution module is driven to execute the adjustment instruction. The online characterization and closed-loop regulation and control functions of the device replace the subjectivity of tedious processes and manual experience parameter adjustment, the 3D printing process is promoted to span from manual trial and error to data-driven intelligentization, and an efficient and reliable technical scheme is provided for precise additive manufacturing of complex materials such as food and composite materials.

Description

Technical Field

[0001] The present invention belongs to the technical field of food processing equipment, and specifically relates to an intelligent optimization device for food 3D printing materials, which is suitable for parameter optimization of 3D printing of meat materials. Background Art

[0002] Food 3D printing technology has developed rapidly in recent years, showing great potential in the application of meat materials. Due to its advantages in rapid prototyping and the production of complex structures, food 3D printing technology has been widely used in aerospace, medical equipment, and automotive manufacturing.

[0003] However, the 3D printing process for meat materials faces challenges such as difficulty in parameter adaptation and imperfect printability assessment. Traditional methods rely on manual experience to adjust printing parameters, which lacks scientific basis, is inefficient, and makes real-time monitoring and dynamic optimization difficult. Furthermore, the mechanical property characterization cycle for food printing materials is long, making it difficult to dynamically monitor the printing process and achieve comprehensive optimization of the printing effect. To address these issues, an intelligent device that can automatically monitor the printing process and optimize parameters is urgently needed. Summary of the Invention

[0004] In light of the above issues, the present invention aims to overcome the shortcomings of the existing technology by providing an intelligent optimization device for 3D printing processes based on rapid characterization of mechanical properties. This device integrates multi-sensor technology to collect key parameters during the printing process in real time and utilizes intelligent optimization algorithms to rapidly characterize the mechanical properties of materials, thereby improving printing efficiency and product quality.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] An intelligent optimization device for food 3D printing materials, comprising a computer, a control module, two sets of symmetrically arranged motion components, and a support frame for fixing the motion components;

[0007] The motion assembly includes a stepper motor, a motor connecting plate, a ball screw, a bearing plate, a slider, a coupling, a bearing steel linear rail and a limit switch;

[0008] The bearing steel linear rail is fixed by a support frame; bearing plates are provided at both ends of the bearing steel linear rail; the ball screw is arranged parallel to the bearing steel linear rail, and the two ends of the ball screw are correspondingly embedded in the holes of the bearing plates at both ends of the bearing steel linear rail to achieve fixation; and a motor connecting plate is provided at one end of the bearing steel linear rail, and the motor connecting plate is connected to the stepper motor; at the same time, a coupling is provided on one side of the motor connecting plate, and one end of the ball screw is connected to the output shaft of the stepper motor through the coupling;

[0009] The slider is sleeved on the ball screw and slides by cooperating with the ball screw through the internal nut of the slider; at the same time, a limit switch is provided on the support frame on one side of the slider, which is fixed by a right-angle bracket II to limit the stroke of the slider.

[0010] An image acquisition support frame is provided in the middle of the two sets of motion components, and an image acquisition device I is provided on the image acquisition support frame; a movable plate is provided below the image acquisition device I, and the movable plate is made of a transparent material, and one end of the movable plate is fixedly connected to the slider via a right-angle bracket I; the other end of the movable plate is symmetrically arranged, and a right-angle bracket is also provided to connect the slider at the symmetrical end, and the movable plate is driven by the slider to achieve up and down movement; an extrusion push rod is provided on the lower surface of the movable plate, and a carrier plate is provided below the extrusion push rod, and a printing mechanism is provided below the carrier plate;

[0011] The printing mechanism includes a syringe, a needle nozzle, a circular turntable and an image acquisition device II; the lower end of the syringe is connected to the needle nozzle, and a circular turntable is provided below the needle nozzle for receiving the printed material; at the same time, an image acquisition device II is also provided above the circular turntable for real-time acquisition of the shape of the printed material.

[0012] A through hole is provided on the surface of the carrier plate, and the syringe is embedded in the through hole of the carrier plate. The syringe is a cylindrical cavity structure, and the extrusion push rod corresponds to the top of the opening, so that the extrusion push rod is embedded in the cavity of the syringe when it moves downward, so as to extrude the material in the syringe;

[0013] One side of the support frame of the device is a control module, which is a development board. It is electrically connected to the stepper motor, image acquisition device I, image acquisition device II and a computer, and is used to control the stepper motor and realize data transmission; at the same time, the stepper motor, image acquisition device I and image acquisition device II are also electrically connected to the computer to realize data acquisition and control.

[0014] Preferably, the support frame is made of aluminum profile material, and an anti-slip pad is provided at the bottom end of the support frame;

[0015] Preferably, both the image acquisition device I and the image acquisition device II are cameras; the image acquisition device II is tilted and forms an angle of 45° with the horizontal line of the upper surface of the circular turntable;

[0016] Preferably, the movable plate is an acrylic plate, and the loading plate is a stainless steel plate; the movable plate is 24 cm long and 12 cm wide; the loading plate is 36 cm long and 12 cm wide.

[0017] Preferably, the right-angle code I is made of hot-dip galvanized material with an aperture of 3 mm and a thickness of 3 mm.

[0018] Preferably, the right-angle code II hot-dip galvanized material has a pore diameter of 3 mm and a thickness of 1.8 mm.

[0019] Preferably, the stepper motor model is 57BYG250A.

[0020] Preferably, the development board model is STM32F103ZET6.

[0021] A method for operating an intelligent optimization device for food 3D printing materials, comprising the following steps:

[0022] S1. Preparation of printing substrate: First, fresh meat was cut into blocks and then ground in a meat grinder to obtain minced meat; mixed with distilled water and konjac gum and placed in a silicone mold to form a meat gel sample;

[0023] Preferably, in step S1, the amount of distilled water is 20% of the mass of the minced meat, and the mass amount of konjac gum is 3-12% of the mass of the minced meat;

[0024] S2. Device Initialization and Meat Paste Testing Preparation: Divide the meat gel sample prepared in S1 into two portions. Place one portion in the middle of the carrier plate, away from the syringe; the other portion is placed inside the syringe. Then, adjust the needle tip downward vertically, 10-15 mm from the surface of the circular turntable. Align Image Capture Device I vertically with the carrier plate; and Image Capture Device II with the circular turntable.

[0025] Preferably, the image acquisition device I and the image acquisition device II are set to have a resolution of 720p and a frame rate of 25fps; the ball screw has a movement speed of 25mm / s

[0026] Start the stepper motor so that the slider drives the moving plate back to its initial position. Set the ball screw speed and calibrate the motor pulse frequency using the development board. The development board establishes communication with the computer via the USART serial port to transmit motor torque and slider displacement data. Confirm that the delay of the acquisition device's synchronous trigger signal is ≤50ms. Insert the syringe into the through-hole of the carrier plate with a gap of ≤0.1mm. Synchronize the circular turntable speed with the extrusion speed. Transmit the information collected by the image acquisition device and stepper motor back to the computer, and generate an initialization report on the computer containing the camera's field of view calibration diagram, motor motion curve, and sensor connection status.

[0027] S3. Deformation image and mechanical data acquisition: Turn on the stepper motor, driving the ball screw to move the slider downward, which in turn drives the movable plate downward, causing the movable plate to squeeze the fresh meat gel sample on the sample carrier. Simultaneously, as the extrusion push rod moves downward, it enters the cavity of the syringe and squeezes the fresh meat gel sample in the syringe. The circular turntable rotates to receive the meat paste in the needle nozzle; image acquisition device I captures the deformation image during the extrusion process, obtaining deformation image data 1; image acquisition device II captures the shape of the meat paste extruded on the circular turntable, obtaining deformation image data 2; the development board collects the stepper motor's motion data in real time, and transmits the motion data and the captured image data to the computer;

[0028] S4. Rapid Characterization of Mechanical Properties: Computers preprocess the collected deformation images using image processing techniques to extract 10 characteristic parameters: area change rate, absolute area change, contour convex hull area ratio, Euler number, first-order derivative mean, grayscale mean, first-order derivative variance, grayscale variance, perimeter change, and first-order derivative maximum. Principal component analysis (PCA) is then used to reduce the dimensionality of the mechanical characteristic parameters hardness, resilience, adhesion, complex viscosity, and deformation characteristic parameters. A mechanical property prediction model is established using random forest and decision tree regression models, capable of characterizing the key mechanical parameters of fresh meat gel.

[0029] Preferably, when characterizing the key mechanical parameters of the fresh meat gel in step S4, the single characterization time is 40 seconds.

[0030] S5. Printing Parameter Optimization: Based on the characterized mechanical property parameters, a response surface model was constructed, with printing speed, temperature, and nozzle diameter as independent variables and the printability score as the response variable. The printability score was calculated by averaging the extrudability score, dimensional accuracy score, and stability score, and ultimately used as the evaluation metric.

[0031] The extrusion performance score is calculated by calculating the continuity (number of breaks) and diameter uniformity of the extruded wire (the deviation between the measured value and the theoretical value is ≤5% for passing), according to the formula J1 = 1-0.2N 断裂 Quantification, the full score is 1.0; J1 is the extrudability score result (between 0 and 1.0, the closer to 1, the more continuous and uniform the extruded filament is), N 断裂 It is the number of times the meat paste breaks when it is squeezed out of the syringe and falls on the circular turntable.

[0032] The dimensional accuracy score is based on the width error of the printed model. and height retention The average value is calculated, with a full score of 1.0; Z1 is the width accuracy score (0 to 1.0, the closer to 1, the smaller the width error); W 实测W is the actual printed model width; 理论 The model width required during design.

[0033] Z2 is the height retention rate score (0 to 1.0, the closer to 1, the closer the height is to the theoretical value). 实测 H is the actual height of the printed model. 理论 is the model height required during design; the actual model width and actual model height here are the models formed after the meat paste is squeezed out through the syringe and falls on the circular turntable, and then transmitted to the computer through the image acquisition device II via electrical signals for recognition.

[0034] The stability score is a measure of the height change at 0 minutes and 40 seconds after printing. The average value is taken as the stability index, with a full score of 1.0; P1 is the height retention rate at 0 minutes after printing (the moment the product is obtained after printing is recorded as 0 minutes) (0 to 1.0, the closer to 1, the more accurate the height is right after printing). H0 is the height of the model measured at the moment of printing completion. 理论 The model height required during design. P2 is the height retention rate 40 seconds after printing (0-1.0, the closer to 1, the smaller the height change after 1 hour, and the better the stability). 40 H is the height of the model measured 40 seconds after printing is completed. 理论 The model height required during design.

[0035] Beneficial effects:

[0036] This device uses multimodal sensing technology to automatically monitor the mechanical properties and morphological information of the printed substrate. It uses an image acquisition device to capture high-resolution image data of the material's extrusion deformation and morphology. It then combines principal component analysis with machine learning algorithms to rapidly predict key mechanical parameters such as hardness and resilience. It dynamically matches real-time monitoring data with preset print quality thresholds. When it detects that material properties or printing results deviate from the optimal range, it automatically triggers a particle swarm algorithm to optimize printing parameters and drives the execution module to execute adjustment instructions. The device's online characterization and closed-loop control capabilities replace the cumbersome processes and subjectivity of manual parameter adjustment. The time required for a single mechanical property test has been reduced from over 20 minutes to 40 seconds, while printing parameter optimization efficiency has increased by over 50%. Overall printing performance scores have increased by up to 12.13%, and dimensional accuracy errors have been kept within 5%. This advances the 3D printing process from "manual trial and error" to "data-driven intelligence," providing an efficient and reliable technical solution for the precise additive manufacturing of complex materials such as food and composite materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a 45° oblique three-dimensional wire drawing of the device of the present invention;

[0038] Figure 2 It is a schematic diagram of the rear of the device of the present invention;

[0039] Figure 3 This is the effect diagram before optimization, where (a)-(e) correspond to konjac gum with a weight percentage of 0%, 3%, 6%, 9%, and 12%, respectively;

[0040] Figure 4 This is the effect diagram after optimization, where the weight percentages of konjac gum corresponding to (a)-(e) are 0%, 3%, 6%, 9%, and 12%, respectively;

[0041] Figure 5 This is a physical picture of the intelligent optimization device;

[0042] Figure 6 This is a physical picture of the support frame and motion components;

[0043] The accompanying drawings are marked as follows: 1-computer; 2-development board; 3-anti-slip mat; 4-circular turntable; 5-syringe; 6-needle nozzle; 7-angle code; 8-support frame; 10-stepping motor; 11-motor connecting plate; 12-ball screw; 13-bearing plate; 14-slider; 15-coupling; 16-bearing steel linear rail; 18-screw I; 20-camera I; 21-screw II; 23-stainless steel plate; 24-extrusion push rod; 25-acrylic plate; 26-right angle code I; 27-camera II; 28-right angle code II; 29-limit switch. DETAILED DESCRIPTION

[0044] To make the purpose, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0045] Example 1:

[0046] like Figure 1-2 As shown, an intelligent optimization device for food 3D printing materials includes two sets of symmetrically arranged motion components and a support frame, wherein the motion components are fixed by the support frame; Figure 5-6 This is a physical picture of the device;

[0047] The support frame is made of aluminum profile material, and an anti-slip pad 3 is provided at the bottom end of the support frame;

[0048] The motion assembly includes a stepper motor 10, a motor connecting plate 11, a ball screw 12, a bearing plate 13, a slider 14, a coupling 15, a bearing steel linear rail 16 and a limit switch 29; wherein the model of the stepper motor 10 is 57BYG250A;

[0049] The upper end of the bearing steel linear rail 16 is fixed to the support frame by a screw II21, and the lower end is fixed by an angle code 7; bearing plates 13 are provided at both ends of the bearing steel linear rail 16; the ball screw 12 is arranged parallel to the bearing steel linear rail 16, and the two ends of the ball screw 12 are correspondingly embedded in the holes of the bearing plates 13 at both ends of the bearing steel linear rail 16 to achieve fixation; and a motor connecting plate 11 is provided at one end of the bearing steel linear rail 16, and the motor connecting plate 11 is connected to the stepper motor 10; at the same time, a coupling 15 is provided on one side of the motor connecting plate 11, and one end of the ball screw 12 is connected to the output shaft of the stepper motor 10 through the coupling 15;

[0050] The slider 14 is sleeved on the ball screw 12, and sliding is achieved by cooperating with the ball screw 12 through the internal nut of the slider 14; at the same time, a limit switch 29 is provided on the support frame on one side of the slider 14, which is fixed by a right-angle code II28 to limit the stroke of the slider.

[0051] An image acquisition support frame is also provided in the middle position of the two groups of motion components, and an image acquisition device, which is a camera I 20, is provided on the image acquisition support frame; an acrylic plate 25 with a length of 24 cm and a width of 12 cm is provided below the camera I 20. The acrylic plate is made of transparent material, and one end of the acrylic plate 25 is fixedly connected to the slider 14 by a right-angle bracket I 26; the other end of the acrylic plate 25 is symmetrically arranged and is also connected to the slider at the symmetrical end by a right-angle bracket, and the slider 14 drives the acrylic plate 25 to move up and down; an extrusion push rod 24 is provided on the lower surface of the acrylic plate 25, and a stainless steel plate 23 with a length of 36 cm and a width of 12 cm is provided below the extrusion push rod 24, and a printing mechanism is provided below the stainless steel plate 23;

[0052] The printing mechanism includes a syringe 5, a needle nozzle 6, a circular turntable 4 and a camera II 27; the lower end of the syringe 5 is connected to the needle nozzle 6, and a circular turntable 4 is provided below the needle nozzle 6 for receiving the printed material; at the same time, a camera II is also provided above the circular turntable 4. The camera II 27 is tilted and arranged at a 45° angle to the horizontal line of the upper surface of the circular turntable 4 for real-time capture of the shape of the printed material.

[0053] A through hole is provided on the surface of the stainless steel plate 23, and the syringe 5 is embedded in the through hole of the stainless steel plate 23 of the carrier plate. The syringe 5 is a cylindrical cavity structure, and the extrusion push rod 24 corresponds to the opening just above. The extrusion push rod 24 is also a cylindrical structure, and its inner diameter is smaller than the inner diameter of the syringe 5, so that the extrusion push rod 24 is just embedded in the cavity of the syringe 5 when it moves downward, and is used to extrude the material in the syringe 5; the support frame is also provided with a development board 2, model STM32F103ZET6, which is electrically connected to the stepper motor 10, camera I 20, camera II 27 and computer 1 for driving the motor 10; at the same time, the stepper motor 10, camera I 20, camera II 27 are also electrically connected to the computer 1 to realize data acquisition and control.

[0054] This example uses chicken breast gel as the research object to prepare printing substrates with different konjac gum concentrations (0%, 3%, 6%, 9%, and 12%). The specific steps are as follows:

[0055] (1) Preparation of printing materials: First, prepare 500g / group of fresh chicken breasts, remove connective tissue and fat, and cut into 1cm pieces. 3 Small piece, grinds 5 minutes to non-granular pulverized meat with rotating speed 1500r / min meat grinder; Take konjac glucomannan and 100g distilled water (20% pulverized meat weight) of edible grade purity 〉=95% by formula, wherein konjac glucomannan weight is respectively 0g (0%), 15g (3%), 30g (6%), 45g (9%), 60g (12%), and total mixture weight is followed successively by 600g, 615g, 630g, 645g, 660g.After adding the manual preliminary mixing of distilled water in pulverized meat, divide and sprinkle into konjac glucomannan powder for 3 times, 3%, 6% concentration sample stirred 2 minutes with 2000r / min with homogenizer, and 9%, 12% concentration extends to 5 minutes to fully dispersing. Mixture is poured into the 4cm × 3cm × 3cm silica gel mold of inwall coating release coating, after scraping off the surface, in 25 ℃, humidity 60% environment, left standstill 6 hours molding, obtain meat gel sample;

[0056] (2) Device initialization: Start the stepper motor 10, drive the slider 14 to drive the acrylic plate 25 to reset to the initial position, calibrate the position through the limit switch 29, set the movement speed of the ball screw 12 to 25 mm / s, and calibrate the motor pulse frequency through the development board 2; electrically connect the camera to the computer 1 through the USB cable, control the camera through the image acquisition software built into the computer 1, and establish communication between the development board 2 and the computer 1 through the USART serial port to transmit data such as motor torque and slider 14 displacement.

[0057] (3) Preparation of the device for testing: The prepared meat gel sample was divided into two parts, one of which was placed in the middle of the surface of the stainless steel plate 23, not in contact with the syringe 5; the other was placed in the syringe 5; then the needle mouth 6 was adjusted to be vertically downward and 12 mm away from the surface of the circular turntable 4; the camera I 20 was vertically aligned with the transparent acrylic plate 23; and the camera II 27 was aligned with the circular turntable 4;

[0058] The parameters of camera I 20 are set to 720p resolution and 25fps frame rate; camera II 27 is aimed at the circular turntable 4 at an inclined angle, with the angle between its inclined central axis and the horizontal line of the upper surface of the circular turntable 4 being 45°, and the focus is adjusted to obtain a clear image; confirm that the camera synchronization trigger signal delay is ≤50ms, the fitting clearance between the syringe 5 and the circular hole of the stainless steel plate 23 is ≤0.1mm, and the rotation speed of the circular turntable 4 is synchronized with the extrusion speed (25mm / s); generate an initialization report on the computer 1 terminal, and if any abnormality is detected, the system will automatically alarm and prompt calibration.

[0059] (4) Deformation image and mechanical data acquisition: Start the stepper motor 10, drive the ball screw 12 to drive the slider 14 to move, and then drive the acrylic plate 25 to move towards the stainless steel plate 23 at a speed of 25 mm / s, and squeeze the meat gel sample on the stainless steel plate 23. At the same time, when the extrusion push rod 24 moves downward, it will enter the cavity of the syringe 5 and squeeze the meat gel sample in the syringe 5. The circular turntable 4 slowly rotates to receive the meat paste in the needle nozzle 6. The camera I 20 collects the deformation image during the extrusion process in real time (a total of 800 frames), and the camera II 27 photographs the shape of the meat paste extruded on the circular turntable 4; the development board 2 collects the motion data, extrusion pressure and other parameters of the stepper motor 10 in real time, and transmits them to the computer 1 terminal through the USART serial port to form a synchronous data set containing a timestamp.

[0060] (5) Rapid characterization of mechanical properties: Computer 1 uses the built-in OpenCV library to preprocess the deformation image and extract 10 characteristic parameters, including area change rate, absolute area change, contour convex hull area ratio, Euler number, first-order derivative mean, grayscale mean, first-order derivative variance, grayscale variance, perimeter change, and first-order derivative maximum value; Computer 1 uses principal component analysis (PCA) to reduce the dimension of the four mechanical parameters of hardness, resilience, adhesion, and complex viscosity and the deformation characteristic parameters, and combines random forest and decision tree regression models to establish a mechanical property prediction model to quickly characterize the key mechanical parameters of fresh meat gel.

[0061] (6) Based on the characterized mechanical property parameters, a response surface model was constructed, with printing speed, temperature, and nozzle diameter as independent variables and printability score as the response variable; the printability score was obtained by adding and averaging the extrudability score, dimensional accuracy score, and stability score, and finally the printability score was used as the evaluation index;

[0062] The extrusion performance score is calculated by calculating the continuity (number of breaks) and diameter uniformity of the extruded wire (the deviation between the measured value and the theoretical value is ≤5% for passing), according to the formula J1 = 1-0.2N 断裂 Quantification, the full score is 1.0; J1 is the extrudability score result (between 0 and 1.0, the closer to 1, the more continuous and uniform the extruded filament is), N 断裂 It is the number of times the meat paste breaks when it is squeezed out through the syringe 5 and falls on the circular turntable 4.

[0063] The dimensional accuracy score is based on the width error of the printed model. and height retention The average value is calculated, with a full score of 1.0; Z1 is the width accuracy score (0 to 1.0, the closer to 1, the smaller the width error); W 实测 W is the actual width of the printed model (for example, the theoretical width is 5mm, but the actual width is 4.8mm); 理论 The model width required during design (such as 5mm set in the 3D model file).

[0064] Z2 is the height retention rate score (0 to 1.0, the closer to 1, the closer the height is to the theoretical value). 实测 The actual height of the printed model (for example, the theoretical setting is to print 10mm high, but the actual measured height is 9.5mm). 理论 The actual model width and actual model height are the model formed after the meat paste is squeezed out through the syringe 5 and falls on the circular turntable 4, and then transmitted to the computer 1 through the camera II27 via electrical signals for recognition.

[0065] The stability score is a measure of the height change at 0 minutes and 40 seconds after printing. The average value is taken as the stability index, with a full score of 1.0; P1 is the height retention rate at 0 minutes after printing (the moment the product is obtained after printing is recorded as 0 minutes) (0 to 1.0, the closer to 1, the more accurate the height is right after printing). H0 is the height of the model measured at the moment of printing completion (for example, the theoretical value is 10mm, and the measured value is 9.8mm right after printing). 理论 The model height required during design (same as above, for example, 10mm). P2 is the height retention rate 40 seconds after printing (0 to 1.0, the closer to 1, the smaller the height change after 1 hour, and the better the stability). 40 H is the model height measured 40 seconds after printing is completed (for example, 9.6mm after 40 seconds). 理论The actual model width and actual model height P1 and P2 are the models formed 1 minute and 5 minutes after the meat paste is squeezed out by the syringe 5 and falls on the circular turntable 4, and then the models are recognized by the camera II (27) through the electrical signal transmitted to the computer 1.

[0066] Samples with five concentrations (0%, 3%, 6%, 9%, and 12%) were selected for comparative testing before and after optimization. Before optimization, the default printing parameters were used: temperature 25°C, speed 25 mm / s, and nozzle diameter 0.84 mm. The comparison data after optimization are shown in Table 1.

[0067] Table 1 Printing parameter optimization and optimization effect

[0068]

[0069] Images were transferred to a computer using an image acquisition device and automatically scored using Python-based QT software. Pre-optimization scores ranged from 0.80 to 0.89, showing an initial upward and then downward trend. The 6% concentration sample had the highest initial score of 0.89. After optimization, scores generally exceeded 0.92, with the 6% concentration sample scoring 0.98, demonstrating significant results. The optimization effect was measured by the ratio of pre-optimization score increases. The 0% concentration sample saw the largest improvement (16.24%), and the overall printing quality improved by approximately 12.13% compared to pre-optimization. This demonstrates that the intelligent optimization strategy based on this device significantly improves the printability of chicken breast gels at different concentrations.

[0070] according to Figure 3 It can be concluded that when the default printing parameters (temperature 25°C, speed 25mm / s, nozzle diameter 0.84mm) are used before optimization, the 0% concentration chicken breast gel Figure 3 (a) The printing score is only 0.81, and the extruded wire has problems such as breakage and large diameter fluctuation. 3% concentration chicken breast gel Figure 3 (b) Although the stability is improved, the dimensional accuracy error is 15%. Figure 3 The dimensional accuracy errors of (c), (d), and (e) also exceed 10%, indicating that traditional parameter settings are difficult to adapt to the rheological properties of materials with different concentrations.

[0071] according to Figure 4 It can be concluded that after optimization, the parameters are dynamically adjusted by the particle swarm algorithm, such as 0% concentration chicken breast gel Figure 4 (a) Corresponding to the temperature of 35.34℃ and the speed of 23.53mm / s, the image was transferred to the computer 1 through the image acquisition device and automatically scored in the QT program software written in Python. The printing scores of the samples of each concentration were all improved to above 0.92. Figure 4(b)-(e) also have excellent printing results, with 6% chicken breast gel Figure 4 (c) scored as high as 0.98; the edge uniformity of the printed object was improved by 30%, the error in the extruded filament diameter was controlled within 5%, and the meat paste on the circular turntable was evenly arranged to form a standard ring, verifying that the closed-loop control strategy based on mechanical property characterization can significantly improve printing stability and accuracy, achieving a leap from "manual trial and error" to "data-driven optimization."

[0072] Note: The above embodiments are only used to illustrate the present invention and are not intended to limit the technical solutions described in the present invention. Therefore, although this specification has described the present invention in detail with reference to the above embodiments, it should be understood by those skilled in the art that the present invention may still be modified or replaced by equivalents. All technical solutions and improvements that do not depart from the spirit and scope of the present invention should be included within the scope of the claims of the present invention.

Claims

1. An intelligent optimization device for food 3D printing materials, characterized in that: It comprises a computer (1), a control module, two sets of symmetrically arranged motion components, and a support frame for fixing the motion components; The motion assembly includes a stepper motor (10), a motor connecting plate (11), a ball screw (12), a bearing plate (13), a slider (14), a coupling (15), a bearing steel linear rail (16) and a limit switch (29); The bearing steel linear rail (16) is fixed by a support frame; bearing plates (13) are provided at both ends of the bearing steel linear rail (16); the ball screw (12) is arranged in parallel with the bearing steel linear rail (16), and the two ends of the ball screw (12) are correspondingly embedded in the holes of the bearing plates (13) at both ends of the bearing steel linear rail (16) to achieve fixation; and a motor connecting plate (11) is provided at one end of the bearing steel linear rail (16), and the motor connecting plate (11) is connected to the stepper motor (10); at the same time, a coupling (15) is provided on one side of the motor connecting plate (11), and one end of the ball screw (12) is connected to the output shaft of the stepper motor (10) through the coupling (15); The slider (14) is sleeved on the ball screw (12), and the slider (14) is slidable by cooperating with the ball screw (12) through the internal nut of the slider (14); a limit switch (29) is also provided on the support frame on one side of the slider (14), fixed by a right-angle code II (28), and used to limit the stroke of the slider; An image acquisition support frame is also provided in the middle position of the two groups of motion components, and an image acquisition device I (20) is provided on the image acquisition support frame; a movable plate (25) is provided below the image acquisition device I (20), the movable plate (25) is made of a transparent material, and one end of the movable plate (25) is fixedly connected to the slider (14) through a right-angle code I (26); the other end of the movable plate (25) is symmetrically arranged, and a right-angle code is also provided to connect the sliders at the symmetrical ends, and the movable plate (25) is driven by the slider to realize up and down movement; an extrusion push rod (24) is provided on the lower surface of the movable plate (25), a carrier plate (23) is provided below the extrusion push rod (24), and a printing mechanism is provided below the carrier plate (23); The printing mechanism comprises a syringe (5), a needle nozzle (6), a circular turntable (4) and an image acquisition device II (27); the lower end of the syringe (5) is connected to the needle nozzle (6), and a circular turntable (4) is provided below the needle nozzle (6) for receiving printed matter; and an image acquisition device II (27) is also provided above the circular turntable (4) for real-time acquisition of the shape of the printed matter; A through hole is provided on the surface of the carrier plate (23), and the syringe (5) is embedded in the through hole of the carrier plate (23). The syringe (5) is a cylindrical cavity structure, and the extrusion propulsion rod (24) corresponds to the top of the opening, so that the extrusion propulsion rod (24) is embedded in the cavity of the syringe (5) when it moves downward, and is used to extrude the material in the syringe (5); One side of the support frame of the device is a control module, which is a development board (2) and is electrically connected to the stepper motor (10), the image acquisition device I (20), the image acquisition device II (27) and the computer (1) for controlling the stepper motor (10) and realizing data transmission; at the same time, the stepper motor (10), the image acquisition device I (20) and the image acquisition device II (27) are also electrically connected to the computer (1) to realize data acquisition and control.

2. The intelligent optimization device for food 3D printing materials according to claim 1, characterized in that: The support frame is made of aluminum profile material, and an anti-slip pad (3) is correspondingly provided at the bottom end of the support frame.

3. The intelligent optimization device for food 3D printing materials according to claim 1, characterized in that: The image acquisition device I (20) and the image acquisition device II (27) are both cameras; the image acquisition device II (27) is tilted and forms an angle of 45° with the horizontal line of the upper surface of the circular turntable (4).

4. The intelligent optimization device for food 3D printing materials according to claim 1, characterized in that: The right-angle bracket I (26) is made of hot-dip galvanized material, with an aperture of 3 mm and a thickness of 3 mm; the right-angle bracket II (28) is made of hot-dip galvanized material, with an aperture of 3 mm and a thickness of 1.8 mm.

5. The intelligent optimization device for food 3D printing materials according to claim 1, characterized in that: The stepper motor (10) is of model 57BYG250A.

6. The intelligent optimization device for food 3D printing materials according to claim 1, characterized in that: The development board (2) is of model STM32F103ZET6.

7. An operating method for an intelligent optimization device for food 3D printing materials, characterized in that: Here are the steps: S1. Preparation of printing substrate: First, fresh meat was cut into blocks and then ground in a meat grinder to obtain minced meat; mixed with distilled water and konjac gum and placed in a silicone mold to form a meat gel sample; S2. Device initialization and meat paste test preparation: Divide the meat gel sample prepared in S1 into two portions, one portion is placed in the middle of the surface of the carrier plate (23) without contacting the syringe (5); the other portion is placed in the syringe (5); then adjust the needle nozzle (6) vertically downward and 10-15 mm away from the surface of the circular turntable (4); the image acquisition device I (20) is vertically aligned with the carrier plate (23); the image acquisition device II (27) is aligned with the circular turntable (4); The stepper motor (10) is started to make the slider (14) drive the moving plate (25) to reset to the initial position, the movement speed of the ball screw (12) is set, and the motor pulse frequency is calibrated through the development board (2); the development board (2) establishes communication with the computer (1) through the USART serial port to transmit the motor torque and slider displacement data; at the same time, it is confirmed that the delay of the synchronous trigger signal of the acquisition device is ≤50ms; the syringe (5) is embedded in the through hole of the carrier plate (23), and the gap between the two is ≤0.1mm; the rotation speed of the circular turntable (4) is set to be synchronized with the extrusion speed, and the information collected by the image acquisition device and the stepper motor (10) is transmitted back to the computer (1), and an initialization report including a camera field of view calibration diagram, a motor motion curve, and a sensor connection status is generated on the computer side; S3. Deformation image and mechanical data acquisition: Turn on the stepper motor (10), drive the ball screw (12) to drive the slider (14) to move downward, and then drive the movable plate (25) to move downward, so that the movable plate (25) squeezes the fresh meat gel sample on the loading plate (23). At the same time, when the extrusion push rod (24) moves downward, it enters the cavity of the syringe (5) and squeezes the meat gel sample in the syringe (5). The circular turntable (4) rotates to receive the meat paste in the needle nozzle (6); The image acquisition device I (20) acquires the deformation image during the extrusion process to obtain deformation image data 1; the image acquisition device II (27) acquires the shape of the meat paste extruded on the circular turntable (4) to obtain deformation image data 2; the motion data of the stepping motor (10) is acquired in real time through the development board (2), and the motion data and the captured image data are transmitted to the computer (1); S4. Rapid characterization of mechanical properties: (1) Computer uses image processing technology to pre-process the collected deformation image and extract 10 characteristic parameters: area change rate, absolute area change, contour convex hull area ratio, Euler number, first-order derivative mean, grayscale mean, first-order derivative variance, grayscale variance, perimeter change, and first-order derivative maximum value; Computer (1) reduces the dimension of mechanical property parameters such as hardness, resilience, adhesiveness, complex viscosity and deformation characteristic parameters through principal component analysis (PCA), and establishes a mechanical property prediction model by combining random forest and decision tree regression models, which can characterize the key mechanical parameters of fresh meat gel; S5. Printing Parameter Optimization: Based on the characterized mechanical property parameters, a response surface model was constructed, with printing speed, temperature, and nozzle diameter as independent variables and the printability score as the response variable. The printability score was calculated by averaging the extrudability score, dimensional accuracy score, and stability score, and ultimately used as the evaluation metric. The extrudability score is calculated according to the formula J1 = 1-0.2N. 断裂 Quantification, the full score is 1.0; J1 is the extrudability score, ranging from 0 to 1.0, the closer to 1, the more continuous and uniform the extruded filament is, N 断裂 The number of times the meat paste breaks when it is squeezed out of the syringe and falls on the circular turntable; The dimensional accuracy score is based on the width error of the printed model. and height retention The average value is calculated, with a full score of 1.0; Z1 is the width accuracy score ranging from 0 to 1.0, and the closer it is to 1, the smaller the width error; W 实测 W is the actual printed model width; 理论 The model width required during design; Z2 is the height retention rate score, ranging from 0 to 1.

0. The closer to 1, the closer the height is to the theoretical value; H 实测 H is the actual height of the printed model. 理论 The model height required during design; The stability score is a measure of the height change at 0 minutes and 40 seconds after printing. The average value is taken as the stability index, with a full score of 1.0; P1 is the height retention rate at 0 minutes after printing, ranging from 0 to 1.0; H0 is the height of the model measured at the moment of printing completion, H 理论 The model height required during design; P2 is the height retention rate 40 seconds after printing, ranging from 0 to 1.0, and the closer to 1, the better the stability; H 40 H is the height of the model measured 40 seconds after printing is completed. 理论 The model height required during design.

8. The method for operating the intelligent optimization device for food 3D printing materials according to claim 7, characterized in that: In step S1, the amount of distilled water is 20% of the mass of the minced meat, and the amount of konjac gum is 3-12% of the mass of the minced meat.

9. The method for operating the intelligent optimization device for food 3D printing materials according to claim 7, characterized in that: The image acquisition device I (20) and the image acquisition device II (27) are set to have a resolution of 720p and a frame rate of 25fps; the movement speed of the ball screw (12) is 25mm / s.

10. The method for operating the intelligent optimization device for food 3D printing materials according to claim 7, characterized in that: When characterizing the key mechanical parameters of the fresh meat gel in step S4, the single characterization time is 40 seconds.

Images