Skin folding system for moon cake production

Abstract

The present invention discloses a mooncake production skin folding system, and relates to the technical field of food processing, the system comprises: a dough initial information acquisition module, which acquires information of a dough image, dough humidity and hardness and a previous process production speed in real time by deploying a camera and an intelligent sensor, and transmits the information to a data processing center after encoding, constructing a dough characteristic database; through highly integrated and intelligent design, the automation level and the product quality control capability of moon cake production are remarkably improved, a dough initial information acquisition module is arranged in the system, advanced camera and intelligent sensor technologies are utilized, and the automation level and the product quality control capability of moon cake production are greatly improved. According to the method, the image, the humidity, the hardness and the key information of the production speed of the previous procedure of the dough can be accurately acquired in real time, and the dough characteristic database is constructed, so that reliable data support is provided for the subsequent procedure, and the production process can be adaptively adjusted according to the change of the dough characteristics; therefore, the production efficiency and the product consistency are greatly improved.

Description

Technical Field

[0001] The invention relates to the technical field of food processing, in particular to a mooncake production wrapper folding system. Background Art

[0002] Mooncakes are an important food in the traditional Chinese Mid-Autumn Festival. Their production process has been inherited and developed for thousands of years and has become a part of Chinese culture. With the rapid development of the food industry and the continuous improvement of consumers' requirements for food safety and quality, the mooncake production industry is facing the pressure of transformation and upgrading. Although the traditional handmade method contains rich cultural connotations, it has obvious deficiencies in production efficiency, product consistency and food safety control. The introduction of automated and intelligent production lines has become the key to improving the competitiveness of the mooncake production industry and meeting market demand. In this context, the mooncake production folding system is an important link in the mooncake production process. Its technological innovation is of great significance to improving the efficiency of the entire production line and product quality.

[0003] The traditional mooncake production and folding process mostly relies on manual operation, which is not only labor-intensive and low in production efficiency, but also difficult to ensure the consistency of dough thickness and shape indicators. Manual operation is also easily affected by the operator's skill level and fatigue level, resulting in large fluctuations in product quality. In addition, traditional technology lacks effective automation means in dough information collection, dough positioning, dough shaping and quality inspection, making it difficult to achieve precise control of dough characteristics and real-time monitoring of product quality. These problems not only limit the expansion of mooncake production scale, but also restrict the further improvement of product quality, making it difficult to meet the needs of modern consumers for food safety, quality and diversity.

[0004] Therefore, the development of a mooncake production and folding system not only significantly improves the production efficiency and product quality of mooncake folding, but also reduces production costs and defective rates, providing strong support for the intelligent upgrade of the mooncake production industry. Summary of the Invention

[0005] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide a mooncake production and folding system. By integrating the modules of initial dough information collection, regular positioning, strip dough shaping, quantitative cutting and segmentation, dough press forming and real-time detection of mooncake quality, the automation and intelligence of the mooncake production process are realized. The system uses advanced sensors and algorithms to accurately control the dough processing process, improve production efficiency and product consistency, and ensure the quality of mooncake dough through the quality scoring entropy weight model and blockchain technology, providing data support for process optimization.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a mooncake production and folding system, the system comprising:

[0007] Dough initial information collection module: By deploying cameras and intelligent sensors, it collects real-time dough images, dough moisture and hardness, and information on the production speed of the previous process. After encoding, it is transmitted to the data processing center to build a dough characteristics database;

[0008] Dough Smoothing and Positioning Module: This module receives dough images, uses algorithms to extract contour features and calculate the center of mass, thins the dough using a set of parallel rollers and controls thickness uniformity. It also uses a prediction model to predict the position and angle at which the dough enters the folding mechanism, and uses an adjustable baffle to adjust its direction of travel.

[0009] Strip dough shaping module: The folding mechanism performs multi-layer rolling and path guidance on irregular strip dough to form regular strip dough with consistent cross-section.

[0010] Quantitative cutting and segmentation module: Based on the preset mooncake specifications, the servo-driven blade is used to cut regular strips of dough into fixed lengths. The visual sensor monitors the dough conveying speed and dynamically adjusts the cutting frequency.

[0011] Dough skin pressurization module: A hydraulic cylinder applies adjustable pressure to the cut dough skin blocks. The pressure is adjusted according to the dough hardness and the speed of pressure is adjusted according to the dough moisture to press the dough into square dough skins. Abnormal fluctuations are monitored by pressure sensors during the pressurization process.

[0012] Real-time crust quality detection module: An industrial camera is used to capture images of the crust after pressurization, and the crust after pressurization is evaluated using a quality scoring entropy weight model. Based on the scoring results, a feedback mechanism is triggered to adjust the cutting frequency and eliminate unqualified products. At the same time, the data is stored in the blockchain database for periodic process optimization analysis.

[0013] Furthermore, the camera and intelligent sensor used in the dough initial information collection module are:

[0014] Camera: A linear array camera is used and installed above the dough conveyor belt;

[0015] Capacitive humidity sensor: installed above or on the side of the dough conveyor belt;

[0016] Pressure-sensitive hardness sensor: Installed above or on the side of the dough conveyor belt,

[0017] Speed ​​encoder: installed on the power transmission shaft of the preceding process equipment.

[0018] Furthermore, the formula for extracting contour features in the dough regularization and positioning module is: Among them, φ is the comprehensive value of the contour feature, i is the pixel index, n is the total number of pixels, μ i (P i ) is the fuzzy membership of the gray value of the i-th pixel, and the calculation formula is: P i The grayscale value of the i-th pixel, is the average value of pixel grayscale value, k is the sensitivity coefficient, δ i is the pixel grayscale deviation activation value, and the calculation formula is: sgn(·) is a sign function used to determine the direction of grayscale value deviation, λ is a weight coefficient that adjusts the proportion of fuzzy edge detection results in contour features, Edge Fuzzy (I) is the edge detection result based on fuzzy C-means clustering, and I is the input image.

[0019] Furthermore, the dough center of mass in the dough regularization and positioning module is calculated using the formula: Among them, (x, y) is the coordinate of the center of mass of the dough, j is the feature point index, ω j The coordinates of the jth feature point are calculated as follows: Φ j The contour eigenvalue of the jth feature point, m represents the total number of feature points involved in the centroid calculation, σ(·) is the Sigmoid function, which normalizes the eigenvalue to the range of 0-1, dist j is the Euclidean distance from feature point j to the geometric center of the dough, L is the estimated length of the dough, max(Φ) represents the maximum value of the contour eigenvalue Φ of all feature points, and min(Φ) represents the minimum value of the contour eigenvalue Φ of all feature points.

[0020] Furthermore, the prediction model in the dough regularization and positioning module is constructed using the formula: Among them, T is the predicted position offset or angle deviation of the dough entering the folding mechanism, α and β are model parameters that adjust the weight of each factor on the prediction result, and φ sum is the sum of the profile eigenvectors, d centroid is the distance from the center of mass to the image origin, representing the dough position offset, and the calculation formula is: t is the square root number, v pre Production speed of the preceding process equipment, ΔT AR(2) It is an autoregressive term based on the trajectory deviation of the previous two moments (t-1, t-2) and is used to capture the trajectory change trend.

[0021] Furthermore, the folding mechanism in the strip noodle shaping module is composed of a conveyor belt, a first-level rolling assembly, a second-level rolling assembly and a third-level rolling assembly, wherein the first-level rolling assembly, the second-level rolling assembly and the third-level rolling assembly are installed on the conveyor belt in sequence.

[0022] Furthermore, the working principle of the dough folding mechanism in the strip dough shaping module is as follows: the dough folding mechanism drives the irregular strip dough forward through the conveyor belt, cuts, rolls and aligns the left side thereof by the first-level rolling assembly, cuts, rolls and aligns the right side thereof by the second-level rolling assembly, and finally cuts, rolls and aligns the left side again by the third-level rolling assembly. Through the coordination of the third-level rolling assembly and the guide device, the irregular strip dough is gradually integrated into a regular strip dough with consistent cross-section.

[0023] Furthermore, the calculation formula for the comprehensive quality score in the real-time crust quality detection module is: Among them, Q cake skin comprehensive quality score, S i It is a single score of folding effect, damage, and thickness uniformity. i is the weight of the i-th indicator, and the calculation formula is: e i is the information entropy of the i-th indicator, reflecting the disorder of the scoring data, and the calculation formula is: v is the total number of samples, S ij is the score value of the jth sample in the i-th indicator, when Q threshold When the quality feedback is triggered, Q threshold is the preset quality threshold.

[0024] Furthermore, the basis for quality scoring in terms of folding effect, breakage, and thickness uniformity in the real-time crust quality detection module is as follows:

[0025] The folding effect is scored based on: the accuracy of the number of layers, the clarity of the layers, and the regularity of the folding;

[0026] The damage scoring is based on: percentage of damaged area, damage location and damage depth;

[0027] The thickness uniformity scoring is based on: thickness deviation range, deviation between average thickness and standard thickness, and thickness distribution uniformity.

[0028] Furthermore, the cutting frequency is adjusted by a cutting frequency adjustment algorithm in the crust quality real-time detection module. Assuming the current cutting frequency is f and the adjusted cutting frequency is f′, the cutting frequency adjustment formula is: Among them, ΔQ is the sum of the comprehensive quality score Q and the preset quality threshold Q threshold The difference is calculated as follows: ΔQ = QQ threshold , Q threshold It is the preset quality threshold used to judge whether the crust is qualified. t is the adjustment factor, ranging from 0.1 to 1. α is the attenuation coefficient, ranging from 0.8 to 0.95.

[0029] Compared with the existing technology, the mooncake production folding system has the following beneficial effects:​

[0030] 1. The present invention significantly improves the automation level and product quality control capability of mooncake production through its highly integrated and intelligent design. The system's built-in dough initial information acquisition module, using advanced camera and intelligent sensor technology, can accurately collect key information such as the image, humidity, hardness and production speed of the previous process of the dough in real time, and build a dough property database, which not only provides reliable data support for subsequent processes, but also enables the production process to be adaptively adjusted according to changes in dough properties, thereby greatly improving production efficiency and product consistency. Secondly, the dough regularization and positioning module uses advanced algorithms to extract dough contour features and calculate the center of mass, and combines the prediction model to predict the position and angle of the dough entering the folding mechanism, thereby achieving precise positioning of the dough and adjustment of its travel direction, effectively solving the problems of inaccurate dough positioning and unstable travel direction in traditional production, improving the accuracy and efficiency of the folding process, and laying a solid foundation for subsequent strip dough shaping and quantitative cutting and segmentation.

[0031] 2. The present invention uses an industrial camera to collect images of the dough after pressurization, and uses a quality scoring entropy weight model to comprehensively evaluate the quality of the dough. It not only realizes real-time monitoring of the quality of the dough, but also triggers a feedback mechanism based on the scoring results, dynamically adjusts the cutting frequency and eliminates unqualified products, thereby ensuring the high-quality production of mooncake dough. At the same time, the system stores the detection data in the blockchain database, providing valuable data resources for periodic process optimization analysis, which not only improves the intelligent level of product quality control, but also provides strong support for enterprises to continuously improve production processes and enhance product competitiveness. In addition, the system realizes quantitative cutting and segmentation through servo-driven blades, and combines visual sensors to monitor the dough conveying speed and dynamically adjust the cutting frequency, further improving production efficiency and the accuracy of product specifications.

[0032] Other advantages, objects and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be learned from the practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0034] Figure 1 Flowchart of the mooncake production folding system;

[0035] Figure 2 It is the front view of the folding mechanism;

[0036] Figure 3 It is a top view of the folding mechanism.

[0037] In the figure: 1. Conveyor belt; 2. Primary roller assembly; 3. Secondary roller assembly; 4. Tertiary roller assembly. DETAILED DESCRIPTION

[0038] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0039] Example 1:

[0040] Automated folding production line for Cantonese-style mooncakes.

[0041] Dough initial information collection module:

[0042] Sensor deployment: A linear array camera is installed directly above the conveyor belt to capture the dough conveying scene at a frequency of 20 frames per second, capturing image information of the dough surface texture and edge morphology; a capacitive humidity sensor is embedded on the side of the conveyor belt to monitor the humidity fluctuation caused by the syrup content in real time by sensing the changes in the dielectric constant of the dough surface; a pressure-sensitive hardness sensor is installed above the conveyor belt to detect the uniformity of the softness and hardness of the oil dough after kneading by touching and pressing the dough surface; a speed encoder is installed on the power shaft of the upstream dough mixer to monitor the dough output rate.

[0043] Data processing: All sensor data are encoded and transmitted to the data processing center to build a dough characteristics database, and record the parameters of humidity, hardness, and image characteristics by timestamp for subsequent modules to call, such as Figure 1 shown.

[0044] Dough regularization and positioning module:

[0045] Contour feature extraction: The image captured by the camera is analyzed based on the contour feature extraction algorithm. By calculating the fuzzy membership and deviation activation value of the pixel grayscale value, the irregular protrusions or depressions on the edge of the dough are identified, and a comprehensive contour feature value is generated to reflect the overall shape regularity. The formula is: Among them, φ is the comprehensive value of the contour feature, i is the pixel index, n is the total number of pixels, μ i (P i ) is the fuzzy membership of the gray value of the i-th pixel, and the calculation formula is: P i The grayscale value of the i-th pixel, is the average value of pixel grayscale value, k is the sensitivity coefficient, δ iis the pixel grayscale deviation activation value, and the calculation formula is: sgn(·) is a sign function used to determine the direction of grayscale value deviation, λ is a weight coefficient that adjusts the proportion of fuzzy edge detection results in contour features, Edge Fuzzy (I) is the edge detection result based on fuzzy C-means clustering, and I is the input image.

[0046] Centroid calculation: Using the centroid coordinate calculation model, weights are assigned to pixels with significant contour features (such as edge inflection points) to calculate the dough centroid coordinates (x, y). The formula is: Among them, (x, y) is the coordinate of the center of mass of the dough, j is the feature point index, ω j The coordinates of the jth feature point are calculated as follows: φ j The contour eigenvalue of the jth feature point, m represents the total number of feature points involved in the centroid calculation, σ(·) is the Sigmoid function, which normalizes the eigenvalue to the range of 0-1, dist j is the Euclidean distance from feature point j to the geometric center of the dough, L is the estimated length of the dough, max(φ) represents the maximum value of the contour eigenvalue Φ of all feature points, and min(φ) represents the minimum value of the contour eigenvalue Φ of all feature points to determine the overall offset direction.

[0047] Pre-adjustment mechanism: The parallel roller group adjusts the roller spacing according to the preset thickness to perform the initial thinning of the dough; the prediction model combines the profile feature comprehensive value Φ, the center of mass position d centroid and the preceding production speed v pre , predict the angle deviation of the dough entering the folding mechanism, the formula is: Among them, T is the predicted position offset or angle deviation of the dough entering the folding mechanism, α and β are model parameters that adjust the weight of each factor on the prediction result, and φ sum is the sum of the profile eigenvectors, d centroid is the distance from the center of mass to the image origin, representing the dough position offset, and the calculation formula is: t is the square root number, v pre Production speed of the preceding process equipment, ΔT AR(2) It is an autoregressive term based on the trajectory deviation of the previous two moments (t-1, t-2), which is used to capture the trajectory change trend. The adjustable baffle is used to fine-tune the travel direction 20 cm in advance to ensure that the center axis of the dough is aligned with the rolling assembly.

[0048] Strip skin shaping module:

[0049] The dough folding mechanism operates as follows: the conveyor belt 1 transports the dough at a speed of 0.5 m / s. It first passes through the first-stage rolling assembly 2, which cuts and rolls the excess dough on the left side so that the left edge is perpendicular to the central axis. It then passes through the second-stage rolling assembly 3, which cuts and rolls the right edge to ensure left-right symmetry. Finally, it passes through the third-stage rolling assembly 4, which integrates the dough into a strip of dough with a width of 10 cm and uniform thickness.

[0050] Quantitative cutting and segmentation module:

[0051] The servo-driven blade sets the cutting length according to the circumference of the dough, and the visual sensor monitors the dough conveying speed, triggering dynamic cutting frequency adjustment: if the speed increases, the blade reciprocating frequency increases, and if the speed slows down, the frequency decreases.

[0052] Dough skin pressure forming module:

[0053] Hydraulic cylinder control logic: Receives dough hardness data. If it detects soft dough with high humidity, it automatically reduces the pressure by 20% to avoid crushing the dough layers. It also adjusts the pressurization speed based on the dough humidity data. The pressure sensor monitors the pressurization process in real time. If the pressure fluctuation exceeds 15%, it immediately suspends pressurization and issues an alarm.

[0054] Forming effect: Within 10 seconds, the dough piece is pressed into a strip with a diameter of 8cm and a thickness of 2mm, with smooth edges and no jagged edges.

[0055] Real-time detection module for crust quality:

[0056] Industrial camera inspection: The formed dough is photographed at a frequency of 5 frames per second. The quality score is analyzed using the entropy weight model to calculate the comprehensive quality score for the pressed dough. The formula is: Among them, Q cake skin comprehensive quality score, S i It is a single score of folding effect, damage, and thickness uniformity. i is the weight of the i-th indicator, and the calculation formula is: e i is the information entropy of the i-th indicator, reflecting the disorder of the scoring data, and the calculation formula is: v is the total number of samples, S ij It is the score value of the jth sample in the i-th indicator. Folding effect evaluation: identifies the number of layers, layer clarity, and folding regularity; damage evaluation: detects the proportion, location, and depth of the damaged area; thickness uniformity evaluation: scans the thickness of 5 key points of the skin, calculates the deviation range and the difference from the standard value.

[0057] Based on the collected information, the comprehensive quality score is calculated by the quality score entropy weight model to evaluate the skin after compression. The formula is: Among them, Q cake skin comprehensive quality score, S i It is a single score of folding effect, damage, and thickness uniformity.i is the weight of the i-th indicator, and the calculation formula is: e i is the information entropy of the i-th indicator, reflecting the disorder of the scoring data, and the calculation formula is: v is the total number of samples, S ij is the score value of the jth sample in the i-th indicator.

[0058] Feedback mechanism: If the comprehensive score Q threshold When the quality feedback is triggered, Q threshold It is a preset quality threshold, and the system automatically performs:

[0059] The cutting frequency is adjusted by the cutting frequency adjustment formula. Assume that the current cutting frequency is f and the adjusted cutting frequency is f'. The cutting frequency adjustment formula is: Among them, ΔQ is the sum of the comprehensive quality score Q and the preset quality threshold Q threshold The difference is calculated as follows: ΔQ = QQ threshold , Q threshold is a preset quality threshold used to determine whether the crust is qualified, t is an adjustment factor with a value range of 0.1-1, α is an attenuation coefficient with a value range of 0.8-0.95, and the speed of the conveyor belt (1) is reduced by 10% to extend the cutting time; and the quality data is stored in the blockchain database for weekly analysis by the process team.

[0060] In summary, the Cantonese mooncake automated folding production line achieves efficient production through the collaboration of multiple modules. The dough initial information acquisition module uses a linear array camera, humidity and hardness sensors to obtain data in real time, and builds a characteristic database to provide a basis for subsequent processes. The dough regularization and positioning module uses the contour feature extraction algorithm, the center of mass calculation model and the prediction model to ensure the precise positioning and pre-adjustment of the dough. The strip dough shaping module is gradually integrated into regular strip dough through the three-level roller pressing component 4. The quantitative cutting and dough press forming module is dynamically adjusted according to the preset specifications to ensure that the dough size and shape meet the requirements. The real-time quality detection module uses the quality scoring entropy weight model to realize the rejection of unqualified products and data traceability. This production line is suitable for high-humidity syrup oil dough, has a high degree of automation, and effectively improves the dough quality and production efficiency.

[0061] Example 2:

[0062] Mechanized folding production line for Beijing-style mooncakes.

[0063] Dough initial information collection module:

[0064] Sensor deployment and data application: Linear array cameras capture images of dough at 18 frames per second to identify surface cracks and burrs, focusing on crack length and burr area ratio.

[0065] ​The capacitive humidity sensor monitors the dough humidity in real time. If the humidity is detected to be lower than the threshold, it is marked as a "dry batch" and synchronized to the dough characteristics database, and the spray humidification device is activated in conjunction with the previous process.

[0066] The speed encoder is installed on the power shaft of the preceding noodle press to monitor the dough output speed and notify the subsequent roller pressing module to adjust the spacing between the parallel rollers in advance.

[0067] Dough regularization and positioning module:

[0068] Contour feature and centroid processing: The contour feature extraction algorithm focuses on edge smoothness. By calculating the fuzzy membership and deviation activation value of the pixel grayscale value, it identifies the edge burr area (for example, the burr on the left accounts for more than 3%) and generates a contour feature comprehensive value φ to evaluate edge quality. The formula is: The centroid coordinate calculation model takes the geometric center of the dough as the reference and calculates the centroid coordinates (x, y) by weighting the feature points in the middle area. The formula is: If the center of mass deviation is greater than 1.5 mm, the prediction model is used to predict the position and angle at which the dough enters the folding mechanism. The formula is: Adjustable baffles correct the direction during transportation to ensure that the dough enters the roller assembly in the center.

[0069] Pre-adjustment mechanism: For dry batches of dough, the initial roller spacing of the parallel roller group is increased by 0.3mm, allowing for initial thinning with lower pressure to reduce the risk of cracking.

[0070] Strip skin shaping module:

[0071] Customized operation of the dough folding mechanism: the dough folding mechanism drives the irregular strip dough forward through the conveyor belt 1, and the first-level rolling component 2 cuts, rolls and aligns the left side to align the left side; the second-level rolling component 3 cuts, rolls and aligns the right side; the third-level rolling component 4 cuts, rolls and aligns the left side again, and through the coordination of the third-level rolling component 4 and the guide device, the irregular strip dough is gradually integrated into regular strip dough with consistent cross-section.

[0072] Quantitative cutting and segmentation module:

[0073] Servo blade dynamic adaptation: The cutting length is set according to the circumference of the circular dough skin. The visual sensor monitors the dough skin conveying status in real time. If the stretching rate is detected to be greater than 5%, the blade reciprocating frequency is reduced through dynamic cutting frequency adjustment to ensure that the dough skin shrinkage after cutting is controlled within 0.5mm.

[0074] The blade adopts a serrated design to reduce the adhesion of dough during cutting, and is equipped with a negative pressure adsorption device to fix the dough to prevent displacement.

[0075] Dough skin pressurizing and forming module: A hydraulic cylinder is used to apply adjustable pressure to the cut dough skin pieces. The pressure is adjusted according to the hardness of the dough, and the pressurizing speed is adjusted according to the humidity of the dough skin to press it into dough skin. Abnormal fluctuations are monitored through pressure sensors during the pressurization process.

[0076] Real-time crust quality detection module: The crust image after pressurization is collected by an industrial camera, and the comprehensive score is calculated and evaluated using the quality scoring entropy weight model. The formula is: Among them, Q cake skin comprehensive quality score, S i It is a single score of folding effect, damage, and thickness uniformity. i is the weight of the i-th indicator, and the calculation formula is: e i is the information entropy of the i-th indicator, reflecting the disorder of the scoring data, and the calculation formula is: v is the total number of samples, S ij is the score value of the jth sample in the i-th indicator, when Q threshold When , the quality feedback is triggered, and the cutting frequency is adjusted by the cutting frequency adjustment algorithm. Let the current cutting frequency be f, and the adjusted cutting frequency be f′. The cutting frequency adjustment formula is:

[0077] Feedback optimization mechanism: If the surface smoothness does not meet the standard, the system calls the cutting frequency adjustment algorithm to adjust the cutting frequency, reducing the roller speed by 5% to extend the action time of the heating roller; the data of unqualified products is stored in the blockchain database, and the humidity data of the drying batch is traced to optimize the pre-sequence humidification frequency.

[0078] In summary, the mechanized folding production line for Beijing-style mooncakes is designed for the dry climate in the north and the characteristics of high-gluten dough. Efficient production is achieved through the collaboration of multiple modules. The dough initial information acquisition module uses linear array cameras and humidity sensors to monitor cracks and humidity in real time, and links with pre-humidification to ensure the dough state. The dough regularization and positioning module uses contour feature extraction algorithms and center of mass calculation models to accurately adjust the dough position and roller spacing to reduce the risk of drying and cracking. The strip dough shaping module adopts blunt knife trimming and heated roller smoothing technology to adapt to the toughness requirements of high-gluten dough. The quantitative cutting and pressurization forming module ensures the accuracy of dough size and smooth surface by dynamically adjusting the cutting frequency and stepped pressurization. The quality inspection module takes surface smoothness and toughness as core indicators, combined with the blockchain database optimization process. This production line effectively responds to the challenges of dry environment and improves the quality of Beijing-style mooncake dough and production efficiency.

[0079] ​The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as above in terms of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can, without departing from the scope of the technical solution of the present invention, make some changes or modifications to equivalent embodiments using the technical contents disclosed above. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. Mooncake production folding system, characterized by: The system includes: Dough initial information collection module: By deploying cameras and intelligent sensors, it collects real-time dough images, dough moisture and hardness, and information on the production speed of the previous process. After encoding, it is transmitted to the data processing center to build a dough characteristics database; Dough Smoothing and Positioning Module: This module receives dough images, uses algorithms to extract contour features and calculate the center of mass, thins the dough using a set of parallel rollers and controls thickness uniformity. It also uses a prediction model to predict the position and angle at which the dough enters the folding mechanism, and uses an adjustable baffle to adjust its direction of travel. Strip dough shaping module: The folding mechanism performs multi-layer rolling and path guidance on irregular strip dough to form regular strip dough with consistent cross-section. Quantitative cutting and segmentation module: Based on the preset mooncake specifications, the servo-driven blade is used to cut regular strips of dough into fixed lengths. The visual sensor monitors the dough conveying speed and dynamically adjusts the cutting frequency. Dough skin pressurization module: A hydraulic cylinder applies adjustable pressure to the cut dough skin blocks. The pressure is adjusted according to the dough hardness and the speed of pressure is adjusted according to the dough moisture to press the dough into square dough skins. Abnormal fluctuations are monitored by pressure sensors during the pressurization process. Real-time crust quality detection module: An industrial camera is used to capture images of the crust after pressurization, and the crust after pressurization is evaluated using a quality scoring entropy weight model. Based on the scoring results, a feedback mechanism is triggered to adjust the cutting frequency and eliminate unqualified products. At the same time, the data is stored in the blockchain database for periodic process optimization analysis.

2. The mooncake production and folding system according to claim 1, characterized in that: The cameras and smart sensors used in the dough initial information acquisition module are: Camera: A linear array camera is used and installed above the dough conveyor belt; Capacitive humidity sensor: installed above or on the side of the dough conveyor belt; Pressure-sensitive hardness sensor: Installed above or on the side of the dough conveyor belt, Speed ​​encoder: installed on the power transmission shaft of the preceding process equipment.

3. The mooncake production and folding system according to claim 1, characterized in that: The formula for extracting contour features in the dough regularization and positioning module is: Among them, Φ is the comprehensive value of the contour feature, i is the pixel index, n is the total number of pixels, μ i (P i ) is the fuzzy membership of the gray value of the i-th pixel, and the calculation formula is: P i The grayscale value of the i-th pixel, is the average value of pixel grayscale value, k is the sensitivity coefficient, δ i is the pixel grayscale deviation activation value, and the calculation formula is: sgn(·) is a sign function used to determine the direction of grayscale value deviation, λ is a weight coefficient that adjusts the proportion of fuzzy edge detection results in contour features, Edge Fuzzy (I) is the edge detection result based on fuzzy C-means clustering, and I is the input image.

4. The mooncake production and folding system according to claim 1, characterized in that: The calculation formula for the dough center of mass in the dough regularization and positioning module is: Among them, (x, y) is the coordinate of the center of mass of the dough, j is the feature point index, ω j The coordinates of the jth feature point are calculated as follows: φ j The contour eigenvalue of the jth feature point, m represents the total number of feature points involved in the centroid calculation, σ(·) is the Sigmoid function, which normalizes the eigenvalue to the range of 0-1, dist j is the Euclidean distance from feature point j to the geometric center of the dough, L is the estimated length of the dough, max(φ) represents the maximum value of the contour eigenvalue Φ of all feature points, and min(φ) represents the minimum value of the contour eigenvalue Φ of all feature points.

5. The mooncake production and folding system according to claim 1, characterized in that: The construction of the prediction model in the dough regularization and positioning module is as follows: Among them, T is the predicted position offset or angle deviation of the dough entering the folding mechanism, α and β are model parameters that adjust the weight of each factor on the prediction result, and φ sum is the sum of the contour eigenvectors, d centroid is the distance from the center of mass to the image origin, representing the dough position offset, and the calculation formula is: t is the square root number, v pre Production speed of the preceding process equipment, ΔT AR(2) It is an autoregressive term based on the trajectory deviation of the previous two moments (t-1, t-2) and is used to capture the trajectory change trend.

6. The mooncake production and folding system according to claim 1, characterized in that: The folding mechanism in the strip noodle shaping module is composed of a conveyor belt (1), a primary roller pressing assembly (2), a secondary roller pressing assembly (3) and a tertiary roller pressing assembly (4), wherein the primary roller pressing assembly (2), the secondary roller pressing assembly (3) and the tertiary roller pressing assembly (4) are sequentially mounted on the conveyor belt (1).

7. The mooncake production and folding system according to claim 1, characterized in that: The working principle of the dough folding mechanism in the strip dough shaping module is as follows: the dough folding mechanism drives the irregular strip dough forward through the conveyor belt (1), cuts, rolls and aligns the left side of the dough through the first-level rolling assembly (2), cuts, rolls and aligns the right side of the dough through the second-level rolling assembly (3), and finally cuts, rolls and aligns the left side again through the third-level rolling assembly (4). Through the cooperation of the third-level rolling assembly (4) and the guide device, the irregular strip dough is gradually integrated into a regular strip dough with a consistent cross section.

8. The mooncake production and folding system according to claim 1, characterized in that: The calculation formula for the comprehensive quality score in the real-time crust quality detection module is: Among them, Q cake skin comprehensive quality score, S i It is a single score of folding effect, damage, and thickness uniformity, w i is the weight of the i-th indicator, and the calculation formula is: e i is the information entropy of the i-th indicator, reflecting the disorder of the scoring data, and the calculation formula is: v is the total number of samples, S ij is the score value of the jth sample in the i-th indicator, when Q threshold When the quality feedback is triggered, Q threshold is the preset quality threshold.​ 9. The mooncake production and folding system according to claim 6, characterized in that: The basis for quality scoring in the folding effect, breakage, and thickness uniformity in the real-time crust quality detection module is as follows: The folding effect is scored based on: the accuracy of the number of layers, the clarity of the layers, and the regularity of the folding; The damage scoring is based on: percentage of damaged area, damage location and damage depth; The thickness uniformity scoring is based on: thickness deviation range, deviation between average thickness and standard thickness, and thickness distribution uniformity.

10. The mooncake production and folding system according to claim 6, characterized in that: The cutting frequency adjustment algorithm is used in the crust quality real-time detection module to adjust the cutting frequency. Assuming the current cutting frequency is f and the adjusted cutting frequency is f′, the cutting frequency adjustment formula is: Among them, ΔQ is the sum of the comprehensive quality score Q and the preset quality threshold Q threshold The difference is calculated as follows: ΔQ = QQ threshold , Q threshold It is the preset quality threshold used to judge whether the crust is qualified. t is the adjustment factor, ranging from 0.1 to 1. α is the attenuation coefficient, ranging from 0.8 to 0.95.

Images