A multi-device collaborative operation control system for 3D printing of jewelry

Abstract

The present invention relates to the field of 3D printing technology, and discloses a multi-device collaborative operation control system for 3D printing of jewelry, including an analysis and parameter adjustment module, a model manufacturing module, a scanning and inspection module, a data sharing module, a mold construction module, a finished product casting module, a finished product quality inspection module, and a recycling module; the analysis and parameter adjustment module obtains and sorts triangular patches for the complex structure of the jewelry STL model, and adaptively generates a layer parameter table; the model manufacturing module prints a jewelry resin model based on the layer parameter table; the scanning and inspection module inspects the model, uses a defect recognition network to generate problem labels for unqualified models and autonomously selects adjustment solutions, improving the stability and automation of the printing process; the finished product quality inspection module monitors the quality of jewelry finished products by combining image recognition technology and traditional measurement technology, melts defective products through the recycling module, and re-obtains precious metal raw materials, ensuring the quality of jewelry finished products and reducing the cost of the lost wax casting process.

Description

Technical Field

[0001] The present invention relates to the technical field of 3D printing, and particularly to a multi-device collaborative operation control system for 3D printing of jewelry. Background Art

[0002] 3D printing technology is a technology for manufacturing solid parts by gradually accumulating materials. It is a "bottom-up" manufacturing method that can solve the forming of many complex structural parts that were difficult to manufacture in the past and greatly reduce the processing procedures. With the increasing maturity of 3D printing technology, it has been applied to various industries, such as jewelry production.

[0003] Currently, there are mainly two technical solutions for 3D printing applied to the jewelry industry. One is laser sintering technology, that is, directly printing precious metal entities, which is less used. The other is to use a 3D printed resin mold and adopt the lost-wax casting method to pour finished products. This solution can achieve rapid mass production and is the mainstream application solution.

[0004] The existing Chinese patent application with the publication number CN118144276A proposes a 3D printing system. The server automatically switches the consumable type and process parameters according to the resin tank model data, generates printing data, and the device control board is used to send the resin tank model data provided by the control board to the server, and control the printing module to execute the printing action according to the received printing data, improving the intelligence and convenience of the 3D printing system.

[0005] However, jewelry often has a relatively precise and complex structure. Using the existing method of generating printing data based on resin materials will result in the loss of the precise structure of the jewelry or the phenomenon of stepped curved surfaces, seriously affecting the appearance of the jewelry. It is necessary to adaptively generate printing and curing data for the complex structure of a single piece of jewelry. Moreover, the jewelry 3D printing process involves multiple steps such as model printing and lost-wax casting. The existing technology lacks an effective monitoring mechanism to automatically detect defects and make adjustments. Summary of the Invention

[0006] The purpose of the present invention is to provide a multi-device collaborative operation control system for 3D printing of jewelry, so as to provide an adaptive layer-by-layer printing and curing method for the complex structure of jewelry, and at the same time, through model inspection and finished product quality inspection feedback results, realize the adjustment of the printing plan and the recycling of precious metal raw materials.

[0007] The technical solution to achieve the purpose of the present invention is as follows:

[0008] A multi-device collaborative operation control system for 3D printing of jewelry includes an analysis and parameter adjustment module, a model manufacturing module, a scanning and inspection module, a data sharing module, a mold construction module, a finished product casting module, a finished product quality inspection module, and a recycling module:

[0009] The analysis and parameter adjustment module exports the jewelry STL model, obtains and sorts triangular patches, and adaptively generates a layer-by-layer parameter table;

[0010] The model manufacturing module prints and performs photocuring on each single slice layer in sequence according to the preset printing parameters and the layer-by-layer parameter table to obtain a jewelry resin model;

[0011] The scanning and inspection module analyzes the jewelry resin model, screens qualified jewelry resin models, assigns problem labels to unqualified jewelry resin models based on a defect recognition network, generates adjustment instructions and transmits them to the model manufacturing module;

[0012] The data sharing module has a built-in label correspondence table, stores and shares the uploaded data of the other modules within the system;

[0013] The mold construction module uses the qualified jewelry resin model to manufacture a jewelry mold;

[0014] The finished product casting module pours precious metal raw materials into the jewelry mold. After the precious metal raw materials solidify, they are taken out and processed mechanically to obtain the jewelry finished product;

[0015] The finished product quality inspection module inspects the jewelry finished product and determines excellent and defective products;

[0016] The recycling module melts the defective products to obtain precious metal raw materials again.

[0017] Furthermore, the analysis and parameter adjustment module includes a parsing unit, an adaptive layering unit, an orbit determination unit, a light adjustment unit, and a first communication unit;

[0018] The parsing unit imports the jewelry STL model in a spatial rectangular coordinate system to obtain triangular patches, sorts the triangular patches to construct an adjacent undirected graph ;

[0019] The adaptive layering unit adaptively adjusts the layering thickness to divide the jewelry STL model into slice layers;

[0020] The orbit determination unit uses NURBS curve fitting to obtain the contour of a single slice layer;

[0021] The light adjustment unit adaptively adjusts the photocuring parameters of a single slice layer;

[0022] The first communication unit correspondingly stores the layering thickness 、the slice layer 's NURBS curve equation of order 、the exposure duration and the light intensity distribution Construct a hierarchical parameter table and upload the jewelry STL model and the adjacency matrix and the hierarchical parameter table to the data sharing module.

[0023] Furthermore, the parsing unit constructs an undirected adjacency graph including:

[0024] Sort triangular facets to construct a point set , construct a dimensional all-zero matrix, defined as the adjacency matrix ;

[0025] Lock the triangular facet , obtain the triangular facet with two identical vertex coordinates , define the triangular facet and the triangular facet are adjacent and have an edge relationship , set the adjacency matrix 's th row and th column element and the th row and th column element to 1;

[0026] Traverse the triangular facets to obtain all edge relationships and construct an edge set , and establish an undirected adjacency graph in the form of the adjacency matrix .

[0027] Furthermore, the adaptive layering unit adaptively divides slicing layers including:

[0028] Obtain the minimum printing thickness and the maximum printing thickness of the model manufacturing module, obtain the total height of the jewelry STL model , set plane as the starting cutting plane , and translate the starting cutting plane along the axis positive direction times to obtain the actual slicing layer ;

[0029] Judge times the remaining model height is less than the maximum printing thickness ;

[0030] If ​Secondary remaining model height Greater than or equal to the maximum printing thickness , translate along the positive direction of the axis by the maximum printing thickness to obtain the initial slice layer , and obtain the triangular facet set intersecting with the initial slice layer

[0031] Traverse the individual triangular facets in the triangular facet set , calculate the angle between the normal vector of each individual triangular facet and the layering direction , and obtain the minimum angle and calculate the theoretical layering thickness ;

[0032] Select the larger value between the theoretical layering thickness and the minimum printing thickness as the actual layering thickness , and obtain the actual slice layer ;

[0033] If the secondary remaining model height is less than the maximum printing thickness and greater than or equal to the minimum printing thickness , then set the actual layering thickness to be the secondary remaining model height and obtain the actual slice layer , and define ;

[0034] If the secondary remaining model height is less than the minimum printing thickness , then cancel the actual slice layer , obtain the secondary remaining model height , set the actual layering thickness to be the minimum printing thickness , and take the rest as the actual slice layer , and define .

[0035] Further, the orbit determination unit obtains the layer profile of a single slice layer by NURBS curve fitting, including:

[0036] Obtain the set of triangular patches intersecting with the actual slice layer and calculate the intersection coordinates of intersection points. Select the starting triangular patch based on the adjacency matrix to construct a triangular patch chain of the set of triangular patches, select the traversal direction, and sort the intersection points ; ;

[0037] Select order NURBS curve, and determine the knot intervals of intersection points based on the recurrence formula of the basis function . Add the pre-supplementary knots and post-supplementary knots to construct the knot vector of the actual slice layer ; ;

[0038] Determine the values of the knots in the knot vector of the actual slice layer , and obtain the order NURBS curve equation of the actual slice layer as the layer profile. ; ;

[0039] Further, determining the values of the knots in the knot vector of the actual slice layer includes: Assign the pre-supplementary knots and post-supplementary knots to 0 and 1 respectively;

[0040] ;

[0041] Obtain the total length of the broken line formed by the intersection points ; ;

[0042] Calculate the intersection points that control the segment of the curve constructed, and the length of the segment of the broken line ;

[0043] If the knot interval of the segment of the curve is , then the value of the knot is the previous Ratio of the total length of the segmented polyline to the total length of the polyline , the node takes a value of the ratio of the total length of the previous segmented polyline to the total length of the polyline .

[0044] Furthermore, the light curing parameters of a single slice layer are adaptively adjusted by the dimming unit, including:

[0045] Set the maximum printing thickness , and the corresponding exposure duration is ;

[0046] According to the ratio of the actual layer thickness to the maximum printing thickness , obtain the exposure duration ;

[0047] Further obtain the actual layer thickness , calculate the curing depth ;

[0048] Based on the curing depth , calculate the light intensity distribution .

[0049] Furthermore, the model manufacturing module includes a printing unit, a light curing unit, an adjustment unit, and a second communication unit;

[0050] The printing unit obtains the actual layer thickness and the slice layer of the nth-order NURBS curve equation and respectively use them as the additive thickness and boundary contour of the nozzle to perform 3D printing with preset printing parameters;

[0051] The light curing unit obtains the exposure duration and the light intensity distribution , and sets the light curing parameters of the ultraviolet laser to perform light curing;

[0052] The adjustment unit locks the abnormal slice layer based on the adjustment instruction and adjusts the preset printing parameters and / or exposure duration of the abnormal slice layer;

[0053] The second communication unit is used to receive the adjustment instruction.

[0054] Further, the scanning and inspection module includes a scanning unit, a detection and analysis unit, a text generation unit, and a third communication unit;

[0055] The scanning unit obtains a resin STL model by scanning a jewelry resin model;

[0056] The detection and analysis unit determines whether there is an abnormality in the resin STL model. If there is no abnormality, it is determined as a qualified jewelry resin model. If there is an abnormality, a problem label is generated based on the defect recognition network;

[0057] The text generation unit searches the label correspondence table, obtains the adjustment plan corresponding to the problem label, and generates an adjustment instruction based on the abnormal sequence to obtain the abnormal slice layer;

[0058] The third communication unit calls the jewelry STL model and the label correspondence table, and feeds back the adjustment instruction to the model manufacturing module.

[0059] Furthermore, the detection and analysis unit determining whether there is an abnormality in the resin STL model includes:

[0060] Obtaining the resin STL model in the same way as the analysis and parameter adjustment module triangular facets and sorting them;

[0061] Traversing the individual triangular facets of the resin STL model and the jewelry STL model according to the sorting order;

[0062] Obtaining the th triangular facet of the resin STL model and getting the three vertex coordinates of the th triangular facet of the jewelry STL model, calculating and judging whether the deviation degree of the th triangular facet is greater than the deviation threshold. If so, record the triangular facet number in the abnormal sequence ;

[0063] After the traversal is completed, count the number of numbers in the abnormal sequence , and if the ratio of the number of numbers to the number of triangular facets is less than or equal to the ratio threshold, it is determined that the resin STL model is a qualified jewelry resin model, otherwise it is determined that there is an abnormality.

[0064] Furthermore, the detection and analysis unit assigning problem labels through the defect recognition network includes:

[0065] Converting the jewelry STL model and the resin STL model into a first image and a second image;

[0066] Input the first image and the second image into the downsampling sub-network, reduce the dimension through pooling, and further extract deep features through multiple convolutions, and output the first feature and the second feature;

[0067] Input the first feature and the second feature into the residual encoding sub-network, generate the first reconstructed feature and the second reconstructed feature through the attention mechanism, and generate the first code and the second code through the superposition of the residual structure;

[0068] Input the first code and the second code into the linear sub-network, and generate the aggregated code through the combination of the linear fully connected layer and the ReLU activation function;

[0069] Input the aggregated code into the upsampling sub-network, and generate the dimension-expanded feature through multiple transposed convolutions;

[0070] Input the dimension-expanded feature into the multi-layer perceptron sub-network, and use the softmax function to non-linearly activate the multiple linear modulations of the dimension-expanded feature to generate a probability vector, and select the problem label to assign to the resin STL model.

[0071] Furthermore, the data sharing module includes a storage unit and a fourth communication unit. The storage unit pre-stores the label correspondence table and stores the uploaded data of the other units. The fourth communication unit receives the uploaded data of the other units and transmits the data called by the other units.

[0072] Furthermore, the mold construction module fixes the qualified jewelry resin model in the medium, and the qualified jewelry resin model must be fully immersed in the medium, sets the heating temperature and heating duration, and heats to obtain the jewelry mold.

[0073] Furthermore, the finished product quality inspection module includes an image acquisition unit, an image analysis unit, a parameter measurement unit, and an evaluation unit;

[0074] The image acquisition unit takes pictures of the surface image of the jewelry finished product;

[0075] The image analysis unit obtains the surface image of the jewelry finished product, and determines whether there are defects on the surface of the jewelry finished product through image recognition technology. If there are no defects, it transmits the surface qualified label to the evaluation unit;

[0076] The parameter measurement unit measures the size and quality of the jewelry finished product and calculates the size error and quality error. If the sum of the size error and the quality error is less than or equal to the preset error threshold, it transmits the parameter compliance label to the evaluation unit;

[0077] After the evaluation unit obtains the surface qualified label and the parameter compliance label of the same jewelry finished product, it determines it as a high-quality product, and the rest are determined as defective products.

[0078] Compared with the prior art, the remarkable advantages of the present invention are:

[0079] 1. For the complex structure of the jewelry STL model, obtain and sort the triangular facets, establish the adjacent matrix of triangular facets, and adopt adaptive hierarchical adjustment of the single slice layer thickness to avoid the staircase effect during printing;

[0080] 2. For a single slice layer, use NURBS curve fitting to generate an accurate layer profile, and at the same time, adjust the stereolithography parameters in real time based on the thickness of the single slice layer to achieve fine printing of the jewelry resin model;

[0081] 3. Analyze the jewelry resin model and screen qualified ones, assign problem labels to unqualified jewelry resin models based on the defect recognition network, and select corresponding adjustment solutions based on the problem labels to improve the stability of the process flow for printing jewelry resin models;

[0082] 4. Combine image recognition technology and traditional measurement technology to determine the quality of finished jewelry, ensure the quality of finished jewelry, and at the same time recycle and melt inferior products to obtain precious metal raw materials again, reducing the cost of the lost-wax casting process for jewelry. Description of the Drawings

[0083] Figure 1 It is a schematic diagram of a multi-device collaborative operation control system for 3D printing of jewelry;

[0084] Figure 2 It is a flow chart of adaptively dividing slice layers in the present invention;

[0085] Figure 3 It is a flow chart of obtaining the layer profile of a single slice layer in the present invention;

[0086] Figure 4 It is a model diagram of the defect recognition network in the present invention. Detailed Embodiments

[0087] The following further describes the present invention in detail with reference to the drawings and embodiments.

[0088] As Figure 1 shown, the embodiment provided by the present invention: A multi-device collaborative operation control system for 3D printing of jewelry includes an analysis and parameter adjustment module, a model manufacturing module, a scanning and inspection module, a data sharing module, a mold construction module, a finished product casting module, a finished product quality inspection module, and a recycling module:

[0089] The analysis and parameter adjustment module exports the jewelry STL model, obtains and sorts the triangular facets, and adaptively generates a hierarchical parameter table based on the triangular facets;

[0090] The model manufacturing module prints and stereolithographs each single slice layer in sequence according to the preset printing parameters and the hierarchical parameter table to obtain the jewelry resin model;

[0091] The scanning inspection module analyzes the jewelry resin model, screens qualified jewelry resin models, assigns problem tags to unqualified jewelry resin models based on the defect recognition network, generates adjustment instructions and transmits them to the model manufacturing module;

[0092] The data sharing module has a label correspondence table built-in, receives and stores the uploaded data of the other modules, responds to the data calls of the other modules at any time, and realizes the sharing of the uploaded data among the units in the system;

[0093] The mold construction module manufactures jewelry molds using qualified jewelry resin models;

[0094] The finished product casting module pours precious metal raw materials into the jewelry mold. After the precious metal raw materials solidify, they are taken out and processed mechanically to obtain jewelry finished products;

[0095] The finished product quality inspection module inspects the jewelry finished products and determines excellent products and defective products;

[0096] The recycling module melts the defective products to obtain precious metal raw materials again.

[0097] Furthermore, the analysis and parameter adjustment module includes an analysis unit, an adaptive layering unit, an orbit determination unit, a light adjustment unit and a first communication unit;

[0098] The analysis unit imports the jewelry STL model in a spatial rectangular coordinate system to obtain triangular patches, sorts the triangular patches to construct an adjacent undirected graph ;

[0099] The adaptive layering unit adaptively adjusts the layering thickness to divide the jewelry STL model into slicing layers;

[0100] The orbit determination unit uses NURBS curve fitting to obtain the layer profile of a single slicing layer;

[0101] The light adjustment unit adaptively adjusts the light curing parameters based on the layering thickness of a single slicing layer;

[0102] The first communication unit constructs a layering parameter table and uploads the jewelry STL model, the adjacency matrix and the layering parameter table to the data sharing module.

[0103] Even further, the analysis unit constructs an adjacent undirected graph including:

[0104] Sort the triangular patches to construct a point set , where, represents the th triangular patch;

[0105] Lock the triangular patch , determine the triangular facet whether it is adjacent to the triangular facet . If so, define that there is an edge relationship between the triangular facet ; ;

[0106] Traverse the triangular facets through recursive search to obtain all edge relationships and construct an edge set , and establish an undirected adjacency graph in the form of an adjacency matrix .

[0107] Furthermore, whether the parsing unit determines that the triangular facet is adjacent to the triangular facet can be determined by the vertex coordinates of the triangular facet and the triangular facet in the space rectangular coordinate system. If two vertex coordinates of the triangular facet coincide with those of the triangular facet , then the triangular facet is adjacent to the triangular facet .

[0108] Furthermore, the adjacency matrix is an -dimensional matrix. The element in the th row and th column of the adjacency matrix indicates whether there is an edge relationship between the triangular facet . If so, , otherwise, . Since a triangular facet cannot be adjacent to itself, the adjacency matrix is a symmetric matrix with diagonal elements all being 0. ;

[0109] As shown in Figure 2 , furthermore, the adaptive hierarchical unit adaptively divides slicing layers, including:

[0110] Obtain the minimum printing thickness and the maximum printing thickness of the model manufacturing module, obtain the total height of the jewelry STL model, set plane as the starting cutting plane , and translate the starting cutting plane times along the positive direction of the axis to obtain the actual slicing layer ;

[0111] Judge The height of the remaining model at this time Is it less than the maximum printing thickness , where Is the actual Slice layer The actual Layer thickness;

[0112] If The height of the remaining model at this time Is greater than or equal to the maximum printing thickness , that is , along the Positive direction of the axis, translate by the maximum printing thickness Obtain the initial Slice layer , obtain the one that intersects with the initial Slice layer Set of triangular patches ;

[0113] Traverse the single triangular patches in the set of triangular patches, and calculate the angle between the normal vector of the single triangular patch and the layer direction respectively, and obtain the Minimum angle ; ;

[0114] Calculate the theoretical Layer thickness , and further judge whether the theoretical Layer thickness Is less than the minimum printing thickness ;

[0115] If the theoretical Layer thickness Is less than the minimum printing thickness , then the actual Layer thickness , otherwise, the actual Layer thickness , obtain the actual Slice layer ;

[0116] If The height of the remaining model at this time Is less than the maximum printing thickness And greater than or equal to the minimum printing thickness , that is , then the actual Layer thickness , obtain the actual Slice layer , and define at this time ;

[0117] If the height of the remaining model for the th time is less than the minimum printing thickness , then obtain the height of the remaining model for the th time. It is easy to know that the height of the remaining model for the th time satisfies . Adjust the actual layer thickness , then the actual layer thickness . Respectively obtain the actual slice layer and the actual slice layer , and define at this time .

[0118] As Figure 3 shown, further, the fixed orbit unit uses NURBS curve fitting to obtain the profile of a single slice layer, including:

[0119] Determine the actual slice layer , confirm the set of triangular facets intersecting with the actual slice layer and calculate the intersection point coordinates. Assume that a total of intersection points are obtained. These intersection points control the profile shape of the actual slice layer ;

[0120] Sort the intersection points, defined as ;

[0121] Select -order NURBS curve. Based on the intersection points, confirm the knot intervals, and add pre-supplementary knots and post-supplementary knots to construct the knot vector of the actual slice layer ;

[0122] Determine the values of the knots in the knot vector of the actual slice layer , and obtain the -order NURBS curve equation of the actual slice layer ​ , and used as the layer contour.

[0123] Furthermore, the sorting of the intersections includes:

[0124] Select the triangular facet with the smallest sort in the set of triangular facets , assumed to be ;

[0125] Based on the adjacency matrix , find the two triangular facets adjacent to the triangular facet in the set of triangular facets , assumed to be and , and construct a triangular facet chain ;

[0126] Further based on the adjacency matrix , find the two triangular facets adjacent to the triangular facet in the set of triangular facets , one of which is the known triangular facet , and the other is assumed to be , and expand the triangular facet chain ;

[0127] Until all the triangular facets in the set of triangular facets are retrieved. At this time, the triangular facet chain must form a closed loop;

[0128] Select a traversal direction. Taking the clockwise direction as an example, use the triangular facet as the starting point, and the two intersections of the triangular facet are sorted in sequence as and , and then search along the triangular facet chain to the triangular facet . One of the two intersections of the triangular facet must be , and the other is sorted as ;

[0129] Until the triangular facet chain searches to the triangular facet . One of the two intersections of the triangular facet must be , and the other is sorted as .

[0130] Furthermore, construct the node vector of the actual slice layer including:

[0131] Set the actual slice layer The order of the NURBS curve , so the actual slice layer Each segment of the NURBS curve of intersection points for control, that is, the segment of the curve is controlled by the intersection points . It can be deduced that the actual slice layer The number of curve segments of the NURBS curve is ;

[0132] Since the actual slice layer of the order NURBS curve equation , where the recurrence formula of the basis function is as follows:

[0133] ,

[0134] ,

[0135] If , then the th intersection point in the knot vector has a knot interval of . If , the th intersection point in the knot vector has a knot interval of ;

[0136] To ensure that the th segment of the curve is only controlled by the intersection points and not affected by the other intersection points, the knot intervals of the th segment of the curve's intersection points must not intersect with the union of the knot intervals of the other intersection points. Obtain the union of the knot intervals of the intersection point and the intersection point . Therefore, the knot interval of the intersection point of the th segment of the curve in the knot vector is ;

[0137] Obtain the knot intervals of the segments of the curve in sequence, and a total of knots are obtained. To ensure that the first segment of the curve and the th segment of the curve are exactly connected without overlapping, it is necessary to add a pre-supplementary node and a post-supplementary node ;

[0138] Since , thus, construct the node vector .

[0139] Furthermore, determine the node vector of the actual slice layer , and the values of the nodes in it include: Assign the pre-supplementary node to 0, that is

[0140] , and assign the post-supplementary node to 1, that is ;

[0141] Obtain the total length of the polyline formed by intersection points, where is the Euclidean distance between the -th intersection point and the -th intersection point, which can be calculated by the coordinates of the -th intersection point and the -th intersection point;

[0142] Calculate the length of the polyline segment constructed by the intersection points that control the -th segment of the curve ; ;

[0143] The intersection points of the -th segment of the curve are in the node interval of the node vector , then the value of the node is the ratio of the total length of the first polyline segments to the total length of the polyline , that is , and the value of the node is the ratio of the total length of the first polyline segments to the total length of the polyline , that is .

[0144] Furthermore, the adaptive adjustment of the light curing parameters of a single slice layer by the dimming unit includes:

[0145] Set the exposure duration corresponding to the maximum printing thickness to be ;

[0146] Determine the actual Slice layer The actual Layer thickness , obtain the Slice layer The Exposure duration ;

[0147] Further determine the actual Slice layer The actual Layer thickness , obtain the actual Slice layer The Curing depth ;

[0148] Based on the Curing depth Calculate the actual Slice layer The Light intensity distribution , where, and Are the curing threshold and the longitudinal curing coefficient respectively, which are only related to the photosensitive resin used in printing and can be regarded as fixed values.

[0149] Furthermore, the first communication unit constructs a layer parameter table including:

[0150] Determine the actual Slice layer The actual Layer thickness ;

[0151] Obtain the actual Slice layer The Order NURBS curve equation ;

[0152] Obtain the Exposure duration and the Light intensity distribution ;

[0153] Correspondingly store the actual Layer thickness , the Slice layer The Order NURBS curve equation , the Exposure duration and the Light intensity distribution To the layering parameter table.

[0154] Furthermore, the model manufacturing module includes a printing unit, a photocuring unit, an adjustment unit, and a second communication unit;

[0155] The printing unit locks the actual slicing layer , obtains the actual layer thickness and the slicing layer of order NURBS curve equation , and respectively uses them as the additive thickness and the boundary contour of the nozzle, and performs 3D printing with preset printing parameters, where the preset printing parameters include the nozzle heating temperature and the nozzle moving speed;

[0156] The photocuring unit locks the actual slicing layer , obtains the exposure duration and the light intensity distribution , sets the photocuring parameters of the ultraviolet laser, and performs photocuring after the actual slicing layer is printed;

[0157] The adjustment unit locks the abnormal slicing layer based on the adjustment instruction, and adjusts the preset printing parameters and / or the exposure duration of the abnormal slicing layer based on the adjustment scheme in the adjustment instruction;

[0158] The second communication unit receives the adjustment instruction fed back by the scanning inspection module.

[0159] Furthermore, the scanning inspection module includes a scanning unit, a detection and analysis unit, a text generation unit, and a third communication unit;

[0160] The scanning unit obtains the resin STL model by scanning the jewelry resin model;

[0161] The detection and analysis unit determines whether there is an abnormality in the resin STL model. If there is no abnormality, it is determined as a qualified jewelry resin model. If there is an abnormality, a problem label is generated based on the defect recognition network. The problem label is used to indicate the defect type of the jewelry resin model, and the defect type includes but is not limited to incomplete fusion, pores, unfused particles, balling phenomenon, fatigue cracks, and solidification discontinuity;

[0162] The text generation unit searches the label correspondence table, obtains the adjustment scheme corresponding to the problem label, obtains the actual slicing layer where the triangular facets in the abnormal sequence are located, defines it as the abnormal slicing layer, and generates an adjustment instruction;

[0163] The third communication unit calls the jewelry STL model and the label correspondence table in the data sharing module, and feeds back an adjustment instruction to the model manufacturing module.

[0164] Furthermore, the detection and analysis unit determines whether there are abnormalities in the resin STL model, including:

[0165] Obtain the triangular facets of the resin STL model, and sort the triangular facets in the same way as the analysis and parameter adjustment module;

[0166] Obtain the triangular facets of the jewelry STL model, and traverse the single triangular facets according to the sorting order;

[0167] Obtain the th triangular facet of the resin STL model and obtain the th triangular facet of the jewelry STL model and calculate the triangular facet deviation degree , where is the Euclidean distance between vertex coordinates and vertex coordinates ;

[0168] Judge whether the triangular facet deviation degree is greater than the deviation threshold. If so, it is regarded as non-coincidence, and the triangular facet number is recorded in the abnormal sequence . Otherwise, it is regarded as the triangular facet and the triangular facet being coincident;

[0169] After the traversal of the triangular facets is completed, count the number of numbers in the abnormal sequence . If the ratio of the number of numbers to the number of triangular facets is less than or equal to the ratio threshold, it is regarded that there is no abnormality in the resin STL model, and it is determined as a qualified jewelry resin model. Otherwise, it is regarded that there is an abnormality in the resin STL model.

[0170] As Figure 4 shown, furthermore, the detection and analysis unit assigns problem labels through a defect recognition network, including:

[0171] Convert the jewelry STL model and the resin STL model into a first image and a second image with a specified pixel size;

[0172] Synchronously input the first image and the second image into the downsampling sub-network. Reduce the dimensions of the first image and the second image through pooling and increase the number of channels. Then, further extract the deep features in the first image and the second image through two convolutional operations, and output the first feature and the second feature.

[0173] Further input the first feature and the second feature into the residual encoding sub-network. Reconstruct them through the attention mechanism to obtain the first reconstructed feature and the second reconstructed feature. Then, stack the first feature and the second feature through the residual structure to generate the first code and the second code.

[0174] Input the first code and the second code into the linear sub-network, and generate the aggregated code through the combination of the linear fully connected layer and the ReLU activation function.

[0175] Input the aggregated code into the upsampling sub-network, and perform upsampling decoding through two transposed convolutional operations to generate the expanded-dimensional feature.

[0176] Input the expanded-dimensional feature into the multi-layer perceptron sub-network. Linearly modulate the expanded-dimensional feature through two layers of linear fully connected layers, and perform non-linear activation using the softmax function to generate the probability vector. Select the problem label with the highest probability in the probability vector and assign it to the resin STL model.

[0177] Furthermore, the data sharing module includes a storage unit and a fourth communication unit.

[0178] The storage unit pre-stores the label correspondence table and stores the data uploaded by the other units in real time.

[0179] The fourth communication unit receives the data uploaded by the other units and responds to the calls of the other units to transmit the required data.

[0180] Furthermore, the mold construction module fixes the qualified jewelry resin model in the medium, and the qualified jewelry resin model must be fully immersed in the medium. Set the heating temperature and heating duration such that the heating temperature is greater than the melting temperature of the photosensitive resin and the hardening temperature of the medium, and less than the melting temperature of the medium. Continuously heat until the jewelry resin model is completely melted and the medium is completely hardened to obtain the jewelry mold.

[0181] Furthermore, the finished product quality inspection module includes an image acquisition unit, an image analysis unit, a parameter measurement unit, and an evaluation unit.

[0182] The image acquisition unit captures the surface image of the jewelry finished product and transmits it to the image analysis unit.

[0183] The image analysis unit obtains the surface image of the jewelry finished product, and determines whether there are defects on the surface of the jewelry finished product through image recognition technology. If there are no defects, transmit the surface qualified label to the evaluation unit.

[0184] The parameter measurement unit measures the size and quality of the finished jewelry, compares them with the predefined size standard and quality standard of the finished product, calculates the size error and quality error. If the sum of the size error and quality error is less than or equal to the preset error threshold, it transmits a parameter compliance label to the evaluation unit;

[0185] Only after the evaluation unit obtains the surface compliance label and the parameter compliance label of the same finished jewelry, it is judged as a high-quality product, and in other cases, it is judged as a defective product.

[0186] The present invention discloses a multi-device collaborative operation control system for 3D printing of jewelry, including an analysis and parameter adjustment module, a model manufacturing module, a scanning and inspection module, a data sharing module, a mold construction module, a finished product casting module, a finished product quality inspection module, and a recycling module; the analysis and parameter adjustment module aims at the complex structure of the jewelry STL model, obtains and sorts the triangular patches, establishes an adjacency matrix of the triangular patches, and uses adaptive layer-by-layer adjustment to adjust the layer thickness of a single slice layer to avoid the staircase effect during printing. For a single slice layer, NURBS curve fitting is used to generate an accurate layer profile, and at the same time, the stereolithography parameters are adjusted in real time based on the layer thickness of the single slice layer to generate a layer parameter table; the model manufacturing module finely prints the jewelry resin model based on the layer parameter table, and generates problem labels of unqualified models through the defect recognition network of the scanning and inspection module, and autonomously adjusts the printing parameters to ensure the printing quality and improve the stability and automation of the process of printing the jewelry resin model; the finished product quality inspection module monitors the quality of the finished jewelry by combining image recognition technology and traditional measurement technology, melts the defective products through the recycling module, and re-obtains the precious metal raw materials, ensuring the quality of the 3D printed finished jewelry of the present invention and reducing the cost of the lost wax casting process.

[0187] The above are only the preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the concept of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A multi-device collaborative operation control system for 3D printing of jewelry, characterized in that, It includes an analysis and parameter adjustment module, a model manufacturing module, and a scanning and inspection module: The analysis and parameter adjustment module obtains and sorts the triangular facets of the jewelry STL model, adjusts the layer thickness based on the minimum angle between the normal vector of the triangular facets in a single slice layer and the layering direction, uses NURBS curves to fit the intersection points of the single slice layer and the triangular facets to obtain the layer contour, and sets the light curing parameters based on the layer thickness to construct a layer parameter table; The model manufacturing module obtains the layer parameter table, prints and light cures the single slice layer in sequence to obtain a jewelry resin model, and adjusts the preset printing parameters and / or exposure duration based on the adjustment instruction; The scanning and inspection module compares the triangular facets and counts the abnormal sequences to identify unqualified jewelry resin models, assigns problem labels to the unqualified jewelry resin models through a defect recognition network, and generates an adjustment instruction; The analysis and parameter adjustment module includes a parsing unit and an orbit determination unit; the parsing unit obtains all triangular patches based on the jewelry STL model, sorts the triangular patches, and constructs an adjacent undirected graph in the form of an adjacency matrix ; The orbit determination unit uses NURBS curve fitting to obtain the intersection points of a single slice layer and the triangular patches to obtain the layer contour; Among them, the parsing unit sorts triangular patches to construct a point set , constructs a matrix of all zeros, defined as the adjacency matrix , locks the -th triangular patch, and obtains the -th triangular patch that has two identical vertices with the -th triangular patch. It is defined that the -th triangular patch is adjacent to the -th triangular patch and there is an edge relationship. Set the element in the -th row and the -th column, and the element in the -th row and the -th column of the adjacency matrix to be 1. Traverse the triangular patches to obtain all edge relationships and construct an edge set , and establish an undirected adjacency graph in the form of the adjacency matrix ;​ The orbit determination unit obtains the set of triangular facets intersecting with the actual slice layer and calculates the intersection point coordinates, selects the starting triangular facet, and constructs a triangular facet chain based on the adjacency matrix, selects the traversal direction, and sorts the intersection points; selects order NURBS curve, determines the knot interval of the intersection points based on the recurrence formula of the basis function, adds pre-supplementary knots and post-supplementary knots, and constructs the knot vector of the actual slice layer; assigns the pre-supplementary knots and post-supplementary knots as 0 and 1 respectively, determines the values of the remaining knots in the knot vector based on the length of the polyline segment constructed by the intersection points, and obtains the order NURBS curve equation of the actual slice layer and uses it as the layer profile; Determining the values of the remaining nodes in the knot vector based on the lengths of the broken line segments constructed from the intersection points includes: obtaining the total length of the broken line segments formed by all the intersection points; calculating the length of the th broken line segment constructed from the intersection points that control the th curve segment; The value of the previous node in the node interval of the segment curve is the ratio of the total length of the polyline segment of the previous segment to the total length of the polyline segment. The value of the subsequent node in the node interval of the segment curve is the ratio of the total length of the polyline segment of the previous segment to the total length of the polyline segment.

2. The multi-device collaborative operation control system for 3D printing of jewelry according to claim 1, characterized in that, The analysis and parameter adjustment module further includes an adaptive layering unit, a light adjustment unit, and a first communication unit; The adaptive layering unit adjusts the layer thickness based on the minimum angle between the normal vector of the triangular facets in a single slice layer and the layering direction, and divides the slice layer; The light adjustment unit sets the light curing parameters of the single slice layer based on the layer thickness; The first communication unit constructs a layer parameter table based on the layer thickness, layer contour, and light curing parameters.

3. The multi-device collaborative operation control system for 3D printing of jewelry according to claim 2, characterized in that The adaptive layering unit dividing the slice layer includes: Obtain the minimum printing thickness, the maximum printing thickness, and the total height of the jewelry STL model, and translate the starting cutting plane along the positive direction of the axis to obtain the actual slicing layer for the th time and determine whether the remaining model height for the If the height of the secondary remaining model is greater than or equal to the maximum printing thickness, translate the maximum printing thickness to obtain the initial slicing layer, and obtain the set of triangular patches intersecting with the initial slicing layer; Traverse a single triangular patch in the set of triangular patches, and calculate the theoretical layer thickness based on the minimum angle between the normal vector of the single triangular patch and the layering direction. Select the larger value between the theoretical layer thickness and the minimum printing thickness as the actual layer thickness, and obtain the actual slicing layer; If the height of the secondary remaining model is less than the maximum printing thickness and greater than or equal to the minimum printing thickness, then set the actual layer thickness to be the height of the secondary remaining model and obtain the actual sliced layer; If the height of the secondary remaining model is less than the minimum printing thickness, then separate a part from the actual layer thickness so that the actual layer thickness is the minimum printing thickness, and obtain the actual sliced layer and the actual sliced layer.

4. The multi-device collaborative operation control system for 3D printing of jewelry according to claim 2, wherein The light adjustment unit setting the light curing parameters of the single slice layer based on the layer thickness includes: Setting the exposure duration corresponding to the maximum printing thickness; According to the actual Set the ratio of the layer thickness to the maximum printing thickness Exposure duration; Further obtain the layer thickness and calculate the curing depth; Based on the curing depth setting light intensity distribution.

5. The multi-device collaborative operation control system for 3D printing of jewelry according to claim 1, wherein The scanning and inspection module includes a scanning unit, a detection and analysis unit, a text generation unit, and a third communication unit; The scanning unit obtains a resin STL model by scanning the jewelry resin model; The detection and analysis unit compares the triangular facets of the resin STL model and the jewelry STL model, counts the abnormal sequences, determines whether the resin STL model is abnormal based on the number of abnormal sequence numbers, and assigns a problem label based on the defect recognition network if it is abnormal; The text generation unit searches the label correspondence table, obtains the adjustment plan corresponding to the problem label, obtains the abnormal slice layer based on the abnormal sequence, and generates an adjustment instruction; The third communication unit calls the jewelry STL model and the label correspondence table, and feeds back the adjustment instruction to the model manufacturing module.

6. The multi-device collaborative operation control system for 3D printing of jewelry according to claim 5, wherein, The detection and analysis unit determining whether the resin STL model is abnormal includes: Obtaining all the triangular facets of the resin STL model and sorting them; Traversing the single triangular facets of the resin STL model and the jewelry STL model according to the sorting order; Obtain the three vertex coordinates of the th triangular facet of the resin STL model, and obtain the three vertex coordinates of the th triangular facet of the jewelry STL model, calculate and judge whether the deviation degree of the th triangular facet is greater than the deviation threshold. If so, record the triangular facet number in the abnormal sequence ; Count the number of abnormal sequences after traversal If the ratio of the number of numbers to the number of triangular patches is less than or equal to the ratio threshold, the resin STL model is determined to be a qualified jewelry resin model; otherwise, it is determined that there is an abnormality.

7. A multi-device collaborative operation control system for 3D printing of jewelry according to claim 5 or 6, characterized in that, The detection and analysis unit assigning a problem label through the defect recognition network includes: Converting the jewelry STL model and the resin STL model into a first image and a second image and inputting them into the downsampling sub-network, and outputting a first feature and a second feature through pooling and multiple convolutions; Inputting the first feature and the second feature into the residual encoding sub-network, and generating a first encoding and a second encoding through the superposition of the attention mechanism and the residual structure; Inputting the first encoding and the second encoding into the linear sub-network, and generating an aggregated encoding through the combination of the linear fully connected layer and the ReLU activation function; Inputting the aggregated encoding into the upsampling sub-network, and generating an expanded feature through multiple deconvolutions; Input the dimension-expanded feature into the multi-layer perceptron sub-network, and use the softmax function to non-linearly activate the linear modulation of the dimension-expanded feature to generate a probability vector, and select the problem label to assign to the resin STL model.

8. The multi-device collaborative operation control system for 3D printing of jewelry according to claim 1, characterized in that, The model manufacturing module includes a printing unit, a light curing unit, an adjustment unit, and a second communication unit; The printing unit sets the additive thickness and boundary contour of the nozzle based on the layer parameter table, and performs 3D printing with preset printing parameters; The light curing unit sets the light curing parameters of the ultraviolet laser based on the layer parameter table for light curing; The adjustment unit locks the abnormal sliced layer based on the adjustment instruction, and adjusts the preset printing parameters and / or exposure duration of the abnormal sliced layer; The second communication unit is used to receive the adjustment instruction.

9. The multi-device collaborative operation control system for 3D printing of jewelry according to claim 1, characterized in that It further includes a data sharing module, a mold construction module, a finished product casting module, a finished product quality inspection module, and a recycling module: The data sharing module has a built-in label correspondence table to store and share the uploaded data of each module; The mold construction module fixes the qualified jewelry resin model in the medium and heats it to manufacture the jewelry mold; The finished product casting module pours precious metal raw materials to obtain the jewelry finished product; The finished product quality inspection module obtains the surface image of the jewelry finished product, identifies defects based on image recognition technology, measures the size and quality of the jewelry finished product, and comprehensively determines whether the jewelry finished product is a high-quality product or a defective product; The recycling module melts the defective products to re-obtain precious metal raw materials.

Images