A method, device and model construction method for hole 3D printing

Abstract

The present invention discloses a method, device and model construction method for hole 3D printing, and relates to the technical field of 3D printing. The present invention comprises the following steps: A100. Constructing a printing parameter group, wherein the parameter group is composed of a material type parameter, a nozzle temperature parameter, a printing speed parameter and a hole position parameter; A200. Generating slice data according to a sample model corresponding to each parameter group; A300. Printing and obtaining a plurality of sample models based on the generated slice data; A400. Comparing the hole data of each sample model with the designed hole data, and obtaining error parameters respectively; A500. Constructing a data set based on each printing parameter group and the corresponding error parameters; A600. Constructing a printing model by machine learning the data set; so as to achieve the purpose of accurately and quickly determining the printing model standard.

Description

Technical Field

[0001] The present invention relates to the field of 3D printing technology, and in particular to a method, a device and a model building method for hole 3D printing. Background Art

[0002] 3D printing is a process in which the filaments ejected from the nozzle of the nozzle are stacked and formed on the platform layer by layer during the continuous movement of the 3D printing platform and the nozzle. For models containing hole designs, such as circular holes and regular polygonal holes (such as triangular holes and hexagonal holes), when the central axis of the hole is not perpendicular to the base plate during the 3D printing process, problems such as deformation and large size scaling are prone to occur. If these holes are used for important purposes such as positioning and assembly, the problems caused by such deformation will seriously affect whether the final printed 3D model is usable. Summary of the invention

[0003] One of the purposes of the present invention is to provide a method for constructing a hole 3D printing model to achieve the purpose of accurately and quickly determining the printing model standard.

[0004] The second purpose of the present invention is to provide a method for 3D printing of holes to solve the problem that the existing 3D printing is prone to deformation and large size scaling when printing holes.

[0005] At the same time, the present invention also provides a device for hole 3D printing.

[0006] In order to achieve one of the purposes of the present invention, the present invention adopts the following technical means:

[0007] A method for constructing a hole 3D printing model comprises the following steps:

[0008] A100. Construct a printing parameter group, wherein the parameter group consists of a material type parameter, a nozzle temperature parameter, a printing speed parameter, and a hole position parameter;

[0009] A200. Generate slice data according to the sample model corresponding to each parameter group;

[0010] A300. Print several sample models based on the generated slice data;

[0011] A400. Compare the hole data of each sample model with the designed hole data to obtain error parameters respectively;

[0012] A500. Constructing a data set based on each printing parameter group and the corresponding error parameter;

[0013] A600. Constructing a printing model by machine learning the data set;

[0014] The nozzle temperature parameter = ; Among them, T min is the minimum temperature value of the corresponding material, T max is the maximum temperature value of the corresponding material, ∆T is the temperature interval value;

[0015] The printing speed parameter = ;

[0016] The hole position parameters are .

[0017] Preferably, the error parameters include circular hole error parameters, rectangular hole error parameters and polygonal error parameters;

[0018] The circular hole error parameters include:

[0019] Inner diameter error =

[0020] Roundness error = ;

[0021] The rectangular hole error parameters include:

[0022] Rectangle length error = ;

[0023] Rectangle side ratio error = ;

[0024] The polygon error parameters include:

[0025] Polygon length error =

[0026] Polygon side ratio error = .

[0027] Furthermore, the data set is

[0028] ;

[0029] Among them, MAT nm Indicates the material property code, CVS npi and CVP npi Respectively represent the inner diameter error and the roundness error, RVS npj and RVP npj Respectively represent the rectangle length error and the rectangle side length error, DVS npk and DVP npk They represent the polygon side length error and the polygon side length ratio error respectively, PC npi Represents the hole position parameter of the circular hole, PR npj Represents the hole position parameter of the rectangular hole, PDnpk represents the hole position parameter of the polygonal hole, ∆CR is PC npi -PR npj The angle difference between PC npi -PD npk The angle difference between PR and npj -PD npk The angle difference, T nt Indicates the nozzle temperature parameter, V nv represents the printing speed parameter, and G is the number of times the sample model is printed.

[0030] Furthermore, the step A600 includes:

[0031] A610. Randomly classify the data set into a training group and a prediction group;

[0032] A620. Train the neural network model using the training group, and verify the prediction accuracy of the trained neural model using the prediction group;

[0033] A630. Based on the verification results, the neural network model is screened and the printing model is obtained.

[0034] Furthermore, the step A610 includes shuffling the data set in the row direction, taking the first g rows as the training group, g<the number of times the sample model is printed, and the rest as the prediction group.

[0035] Furthermore, the present invention corrects the printing results by establishing different error parameters based on different holes. At the same time, it combines different material type parameters, nozzle temperature parameters, printing speed parameters and hole position parameters to quickly build a standard printing model, providing an accurate data basis for subsequent 3D printing of holes.

[0036] In order to achieve the second purpose of the present invention, the present invention adopts the following technical means:

[0037] A method for 3D printing of holes, comprising constructing the aforementioned printing model, and further comprising the following steps:

[0038] B100. Receive target model file;

[0039] B200. Determine printing parameters of the 3D printing device based on the printing model and the target model;

[0040] B300. Perform 3D printing on the target model based on the determined printing parameters.

[0041] Preferably, the B200 comprises:

[0042] B210. Input the printing data of the target model through the interactive medium, wherein the printing data includes printing materials and error requirements;

[0043] B220. Input the printing data into the printing model to obtain nozzle temperature parameters, printing speed parameters and hole position parameters corresponding to the target model;

[0044] B230. Adjust the angle of the target model in the 3D printing device according to the hole position parameters to determine the target slice data;

[0045] The printing parameters include nozzle temperature parameters, printing speed parameters and target slice data.

[0046] Furthermore, the present invention also relates to a device for hole 3D printing, which is used for 3D printing using the aforementioned method for hole 3D printing; comprising:

[0047] An input data acquisition module is used to acquire printing input data of a to-be-printed three-dimensional model, wherein the printing input data includes: printing materials and error requirements of a target object in the to-be-printed three-dimensional model;

[0048] A model processing module, used for inputting the printing input data into the printing model to obtain the nozzle temperature parameters, printing speed parameters and hole position parameters of the target object output by the printing model;

[0049] An angle adjustment module, used to adjust the angle of the to-be-printed three-dimensional model according to the placement angle of the target object, to obtain an adjusted three-dimensional model to be printed;

[0050] A slicing module, used for slicing the adjusted three-dimensional model to be printed to obtain slicing data;

[0051] The printing module is used to print the slice data according to the printing material, the nozzle temperature parameter and the printing speed parameter to obtain the printing result of the three-dimensional model to be printed.

[0052] Also included is a pre-training module for building the printing model.

[0053] The method disclosed in the present invention has the following beneficial effects: the 3D printing method for holes disclosed in the present invention can analyze the relationship between the printing parameters of the 3D printing device and the actual printing effect of the model containing the hole design (including the size error and accuracy of various types of holes), and establishes a prediction model of the material, hole size error and accuracy and printing speed, nozzle temperature and hole placement angle based on a neural network by adopting a machine learning method. The prediction model can be used as built-in software of the 3D printing device to predict the printing quality of the model containing the hole design, thereby improving the success rate of printing the finished product and reducing unnecessary material loss and time waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A schematic diagram of key dimensions for determining error parameters of the present invention. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0056] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0058] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0059] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inside", "outside", etc. indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, or the positions or positional relationships in which the product of the invention is usually placed when in use, or the positions or positional relationships commonly understood by those skilled in the art, which are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0060] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0061] A 3D printing method, comprising:

[0062] Acquire printing input data of a three-dimensional model to be printed, wherein the printing input data includes: printing materials and error requirements of a target object in the three-dimensional model to be printed;

[0063] Inputting the print input data into a print model to obtain the placement angle, printer nozzle temperature and print speed of the target object output by the print model;

[0064] According to the placement angle of the target object, the angle of the three-dimensional model to be printed is adjusted to obtain an adjusted three-dimensional model to be printed;

[0065] Slicing the adjusted three-dimensional model to be printed to obtain slice data;

[0066] The slice data is printed according to the printing material, the temperature of the printer nozzle and the printing speed to obtain a printing result of the three-dimensional model to be printed.

[0067] Furthermore, the aforementioned printing model is trained in the following way:

[0068] Acquire a plurality of sample models, wherein each of the plurality of sample models includes at least one type of hole group, and each sample model having the same type of hole group has the hole groups of the same type placed at different angles, and each type of hole group includes holes of at least one size;

[0069] Print each sample model multiple times using different printing materials, different printing speeds, and different nozzle temperatures to obtain a printing result of each printing of each sample model;

[0070] Measuring the critical dimensions of all hole groups included in each sample model and the printing results thereof each time;

[0071] According to the key dimensions of all hole groups included in each sample model and the printing results of each printing, the error of each type of hole group at each placement angle, each printing material, each printing speed, and each nozzle temperature is obtained;

[0072] The printing material and error of each type of hole group are used as training input data, and the corresponding placement angle, printing speed, and nozzle temperature are used as training output data to perform model training to obtain the printing model.

[0073] Furthermore, each sample model includes three types of hole groups placed at the same angle, and the three types of hole groups are respectively a circular hole group, a rectangular hole group and a regular hexagonal hole group. The circular hole group includes circular holes arranged from small to large in size, the rectangular hole group includes rectangular holes arranged from large to small in size, and the regular hexagonal hole group includes regular hexagonal holes arranged from small to large in size.

[0074] Furthermore, according to the sizes of all hole groups included in each sample model and the printing results thereof each time, the error of each type of hole group at each placement angle, each printing material, each printing speed, and each nozzle temperature is obtained, including:

[0075] Obtaining the error of a circular hole group under a placement angle, a printing material, a printing speed, and a nozzle temperature, including: obtaining the inner diameter design value of circular holes of various sizes in a sample model of the circular hole group with the placement angle, and the inner diameter actual measured value, the maximum inner diameter actual measured value, and the minimum inner diameter actual measured value of circular holes of various sizes in a print result obtained by printing using the printing material, printing speed, and nozzle temperature, and obtaining the inner diameter error of the circular hole group according to the ratio of the absolute value of the difference between the inner diameter actual measured value and the inner diameter design value of the circular holes of various sizes to the inner diameter design value, and obtaining the roundness error of the circular hole group according to the ratio of the difference between the maximum inner diameter actual measured value and the minimum inner diameter actual measured value of the circular holes of various sizes to the inner diameter design value;

[0076] Obtaining the error of the rectangular hole group under a placement angle, a printing material, a printing speed, and a nozzle temperature includes: obtaining the design value of the length and the design value of the aspect ratio of the rectangular holes of each size in the sample model of the rectangular hole group with the placement angle, and the measured value of the length and the measured value of the aspect ratio of the rectangular holes of each size in the printing results obtained by printing with the printing material, printing speed, and nozzle temperature, and obtaining the length error of the rectangular hole group according to the ratio of the absolute value of the difference between the measured value of the length and the design value of the length of the rectangular holes of each size to the design value of the length, and obtaining the aspect ratio error of the rectangular hole group according to the ratio of the measured value of the aspect ratio of the rectangular holes of each size to the design value of the aspect ratio;

[0077] Obtaining the error of a regular hexagonal hole group under a placement angle, a printing material, a printing speed, and a nozzle temperature, including: obtaining the design value of the side length of regular hexagonal holes of each size in a sample model of the regular hexagonal hole group with the placement angle, and the measured value of each side length, the measured shortest side length value, and the measured longest side length value of the regular hexagonal holes of each size in the print results obtained by printing with the printing material, printing speed, and nozzle temperature, and obtaining the side length error of the hexagonal hole group according to the ratio of the absolute value of the difference between the measured value of each side length of the hexagonal holes of each size and the design side length value to the design side length value, and obtaining the side length ratio error of the hexagonal hole group according to the ratio of the measured shortest side length value to the measured longest side length value of the hexagonal holes of each size.

[0078] Furthermore, the step of using the printing material and error of each type of hole group as training input data, and using the corresponding placement angle, printing speed, and nozzle temperature as training output data to perform model training to obtain the printing model includes:

[0079] The printing material, the inner diameter error of the circular hole group, the roundness error of the circular hole group, the length error of the rectangular hole group, the aspect ratio error of the rectangular hole group, the side length error of the hexagonal hole group, the side length ratio error of the hexagonal hole group, the placement angle difference between the circular hole group and the rectangular hole group, the placement angle difference between the circular hole group and the hexagonal hole group, and the placement angle difference between the rectangular hole group and the hexagonal hole group are used as input data, and the corresponding placement angle of the circular hole group, the placement angle of the rectangular hole group, the placement angle of the hexagonal hole group, the printing speed, and the nozzle temperature are used as training output data for model training to obtain the printing model.

[0080] Furthermore, a three-dimensional printing device comprises:

[0081] An input data acquisition module is used to acquire printing input data of a to-be-printed three-dimensional model, wherein the printing input data includes: printing materials and error requirements of a target object in the to-be-printed three-dimensional model;

[0082] A model processing module, used for inputting the print input data into a print model, and obtaining the placement angle, printer nozzle temperature and print speed of the target object output by the print model;

[0083] An angle adjustment module, used to adjust the angle of the to-be-printed three-dimensional model according to the placement angle of the target object, to obtain an adjusted three-dimensional model to be printed;

[0084] A slicing module, used for slicing the adjusted three-dimensional model to be printed to obtain slicing data;

[0085] The printing module is used to print the slice data according to the printing material, the temperature of the printer nozzle and the printing speed to obtain the printing result of the three-dimensional model to be printed.

[0086] Also included are pre-trained modules for:

[0087] Acquire a plurality of sample models, wherein each of the plurality of sample models includes at least one type of hole group, and each sample model having the same type of hole group has the hole groups of the same type placed at different angles, and each type of hole group includes holes of at least one size;

[0088] Print each sample model multiple times using different printing materials, different printing speeds, and different nozzle temperatures to obtain a printing result of each printing of each sample model;

[0089] Measuring the sizes of all hole groups included in each sample model and the printing results thereof each time;

[0090] According to the sizes of all hole groups included in each sample model and the printing results thereof each time, the error of each type of hole group at each placement angle, each printing material, each printing speed, and each nozzle temperature is obtained;

[0091] The printing material and error of each type of hole group are used as training input data, and the corresponding placement angle, printing speed, and nozzle temperature are used as training output data to perform model training to obtain the printing model.

[0092] Furthermore, each sample model includes three types of hole groups with the same placement angles, and the three types of hole groups are respectively a circular hole group, a rectangular hole group and a regular hexagonal hole group. The circular hole group includes circular holes arranged from small to large in size, the rectangular hole group includes rectangular holes arranged from large to small in size, and the regular hexagonal hole group includes regular hexagonal holes arranged from small to large in size.

[0093] The following is a more specific description of the implementation method:

[0094] First, construct the printing model using the following steps:

[0095] Step A100: Set printing parameters

[0096] Printing parameters include at least printing speed, nozzle temperature, material, and hole placement angle.

[0097] The material can be one of PLA / ABS / PETG / Nylon / TPU.

[0098] If the nozzle temperature is too low, the viscosity of the material increases, the extrusion speed is slow, the filling is insufficient, and the 3D printed model becomes loose; if it is too high, the material tends to be liquid, the viscosity coefficient decreases, the fluidity increases, and the extrusion is too fast to form a precisely controllable filament. If the nozzle temperature is too high or too low, the final effect on the 3D printed model is a large error. The melting temperature range of different materials is different. For example, the melting temperature range of PLA is 180 degrees to 230 degrees, the melting temperature range of ABS is 210 degrees to 250 degrees, the melting temperature range of PETG is 220 degrees to 250 degrees, the melting temperature range of nylon is 240 degrees to 260 degrees, and the melting temperature range of TPU is 210 degrees to 230 degrees. The nozzle temperature at which the best printing effect can be obtained is not only related to the material, but also varies between different individuals for the same model of 3D printing equipment.

[0099] The printing speed can be any value between 30mm / s and 100mm / s. When the printing speed is too fast, the nozzle will drag the extruded filament, causing the 3D printed model to become loose, especially for the boundary part of the hole. This looseness eventually leads to errors in the hole size. When the printing speed is too slow, the printing time is too long.

[0100] The hole placement angle can be any value between 0 and π / 2. According to general experience, when the angle is 0, the accuracy of the printed holes is higher, and when the angle is π / 2, the accuracy of the printed holes is lower.

[0101] The accuracy of a hole refers to the similarity between the shape of the hole in the 3D printed sample model and the shape of the hole in the calibration sample model. It is a quantifiable parameter. For example, the accuracy of a circular hole can be expressed by roundness, the accuracy of a rectangular hole can be expressed by the error of the length-to-width ratio of the rectangular hole, and the accuracy of a regular polygonal hole can be expressed by the error of the ratio of the longest side to the shortest side.

[0102] In one implementation, the parameter values ​​of the four printing parameters may be set in the following manner:

[0103] The material parameter value can be PLA, ABS, PETG, nylon or TPU, and can be represented by the material property code, such as 1 for PLA, 2 for ABS, 3 for PETG, 4 for nylon, and 5 for TPU. In actual application, it is set according to the types of materials that the 3D printing device can support.

[0104] The parameter values ​​of nozzle temperature include , where ∆T is the temperature interval value, which can be set to 3 degrees; T min is the minimum temperature, T max is the maximum temperature, T max and T min Set according to the material parameter value. When the material is PLA, T min =180 degrees, T max =230 degrees, when the material is ABS, T min =210 degrees, T max =250 degrees, when the material is PETG, T min =220 degrees, T max =250 degrees, when the material is nylon, T min =240 degrees, T max =260 degrees, when the material is TPU, T min =210 degrees, T max =230 degrees.

[0105] The parameters of printing speed include

[0106] The parameter values ​​for hole placement angles include

[0107] In one implementation, the setting values ​​of the printing parameters may be selected by referring to the following method:

[0108] For each material, the values ​​of nozzle temperature, printing speed and hole placement angle are arranged and combined to obtain N t *N v *N p Combination settings, where N t is the total number of possible values ​​of the nozzle temperature, N v is the total number of possible values ​​of the printing speed, N p is the total number of possible values ​​for the hole placement angle, so there are a total of N m *N t *N v *N p Combination settings, N m is the number of material types. During the entire calibration process, the 3D printing equipment needs to perform N calibration on the sample model. m *N t *Nv *N p Print again.

[0109] In another embodiment, the setting value of the printing parameter can be selected by referring to the following method:

[0110] For one material, first make the following settings: the hole placement angle is set to 0, the printing speed is set to a moderate speed, such as 50mm / s, and the nozzle temperature is set to:

[0111] ;

[0112] At this time, the total number of printing parameters is N t Combination settings, 3D printing equipment here t The target calibration model is printed under various combination settings, and N t The first 3D printed sample model was obtained by measuring and calculating its key dimensions. t The key dimensions of the first 3D printed sample model are obtained, and the nozzle temperature corresponding to the first 3D printed sample model with the best key dimensions is recorded as T opt The optimal key dimension refers to the key dimension that is closest to the theoretical design value of the calibration sample model. Next, the nozzle temperature is taken as T opt , the values ​​of printing speed and hole placement angle are arranged and combined to obtain N v *N p Do the same for other materials. With this setting method, a material has a total of N t +N v *N p There are N combinations of settings. m *(N t +N v *N p ) combinations of settings. During the entire calibration process, the 3D printing device needs to perform N calibration on the sample model. m *(N t +N v *N p ) printing, which greatly shortens the cycle of the entire calibration process.

[0113] Step A200: The main control system slices the calibration sample model according to the printing parameters to generate slice data.

[0114] The main control system is pre-installed with 3D printing slicing software, which slices the calibration sample model according to the printing parameters and generates slicing data.

[0115] Step A300. The main control system controls the 3D printing device to perform a 3D printing operation of the calibration sample model based on the slice data of step A200 to obtain a 3D printed sample model.

[0116] Step A400: The clamping module obtains the sample model, and the measuring module measures and calculates the key dimensions of the sample model.

[0117] The key dimensions of the 3D printed sample model include dimensional error and shape accuracy. For example, for a circular hole, the dimensional error is the inner diameter error of the circular hole, and the accuracy is the roundness error. For a rectangular hole, the dimensional error is the length error, and the accuracy is the aspect ratio error. For a regular polygon hole, the dimensional error is the side length error of the polygon, and the accuracy is the side length ratio error. For a cylinder, the dimensional error is the outer diameter error, and the accuracy is the cylindricality. For a rectangular column, the dimensional error is the length error and the width error, and the accuracy is the aspect ratio error. Refer to Figure 1.

[0118] Figure 1 -a is a schematic diagram of the key dimensions of a circular hole. The inner diameter error and roundness error of a circular hole can be calculated using the following method:

[0119]

[0120] Among them, the measured value of the inner diameter is the diameter closest to circle A2, such as Figure 1 -a, A1 represents the boundary line of the circular hole of the 3D printed sample model, A2 is the closest circle to the boundary line A1, O1 is the center of the circle closest to the circle A2, A3 is the smallest circle with O1 as the center that includes the boundary line A1, and A4 is the largest circle with O1 as the center that does not include the boundary line A1. The maximum inner diameter value is the inner diameter of the A3 circle, and the minimum inner diameter value is the inner diameter of the A4 circle. The inner diameter design value is the theoretical design value of the inner diameter of the cylindrical hole of the calibration sample model. The larger the calculated value of the roundness, the lower the accuracy.

[0121] The same type of holes on the calibration sample model are set with multiple different sizes. The calibration sample model is set with four sizes of circular holes. When performing the above error calculation, the inner diameter error and roundness error of these four circular holes of different sizes are calculated respectively, and the average value is taken as the final inner diameter error and roundness error; similarly, when calculating the errors of rectangular holes and regular polygonal holes, the same method is used.

[0122] Figure 1 -b is a schematic diagram of the key dimensions of a rectangular hole. The length error and aspect ratio error of a rectangular hole can be calculated as follows:

[0123]

[0124] Figure 1-b, CL1, CL2, CL3, CL4 represent the closest line segments of the four sides of the rectangular hole of the 3D printed sample model, respectively. The measured length value is the longest value of the CL1, CL2, CL3, CL4 line segments, and the measured aspect ratio value is the ratio of the longest line segment to the shortest line segment among the CL1, CL2, CL3, CL4 line segments. The length design value and aspect ratio design value are the theoretical design values ​​of the length of the rectangular hole of the calibration sample model and the theoretical design values ​​of the aspect ratio, respectively.

[0125] Figure 1 -c is a schematic diagram of the key dimensions of a regular hexagonal hole. The side length error and side length ratio error of a regular polygonal hole can be calculated as follows:

[0126]

[0127] Where N is the number of sides of the polygon, Figure 1 -c, N is 6. EL1, EL2, EL3, EL4, EL5, and EL6 represent the boundary segments of the regular hexagonal holes of the 3D printed sample model, respectively. The measured values ​​of the side lengths are the lengths of the six boundary segments, the shortest measured side length is the length of the shortest boundary segment, and the longest measured side length is the length of the longest boundary segment. The side length design value is the theoretical design value of the side length of the rectangular hole of the calibration sample model.

[0128] Step A500. Repeat the operation process from step A100 to step A400 to obtain multiple sets of data sets containing printing parameters and key dimensions of the 3D target printing model, and generate a database containing input data and output data.

[0129] For different materials, the data sets of the key dimensions of the corresponding 3D target printing model are fused as input data. The input data includes material property code, circular hole roundness error, circular hole inner diameter error, rectangular hole length error, rectangular hole aspect ratio error, regular polygon hole side length error, regular polygon hole side length ratio error. The data sets of printing parameters including nozzle temperature, printing speed and hole column placement angle (to obtain the final theoretical model flip angle or placement angle) are fused as output data. The database is in the following format:

[0130] ;

[0131] Specifically, the database data set is a Gx15-order matrix, where each row represents a data sample, the first 10 columns of each row represent a data sample of the input data set, and the last 5 columns represent a data sample of the output data set. nm Indicates the material property code, CVS npi and CVP npi Respectively represent the inner diameter error of the circular hole and the roundness error of the circular hole, RVSnpj and RVP npj They represent the rectangular hole length error and the rectangular hole aspect ratio error, respectively. npk and DVP npk They represent the regular polygon hole side length error and regular polygon hole side length ratio error, PC npi Indicates the hole placement angle of the circular hole, PR npj Indicates the hole placement angle of the rectangular hole, PD npk represents the hole placement angle of the regular polygonal hole, ∆CR is PC npi -PR npj , which means the difference between the hole placement angle of the npi-th circular hole and the hole placement angle of the npj-th rectangular hole; ∆CD is the PC npi -PD npk , which means the difference between the hole placement angle of the npi-th circular hole and the hole placement angle of the npk-th regular polygonal hole; ∆RD is PR npj -PD npk , which means the difference between the hole placement angle of the npjth rectangular hole and the hole placement angle of the npkth regular polygonal hole; T nt Indicates the nozzle temperature, V nv Indicates the printing speed. nm is 1:N m , that is, the total number of materials is N m , nm represents the 1st to Nth m One of the materials (nm-th). npi takes the value of 1:N p , that is, the total number of possible values ​​of the hole placement angle is N p , then npi represents the 1st to Nth p One of the hole placement angles (the npith), with npi as the subscript, represents a value at the npith hole placement angle, such as CVS npi Indicates the inner diameter error of the circular hole at the npith hole placement angle; npj takes the value of 1:N p , npj represents the 1st to Nth p One of the hole placement angles (npjth), with npj as the subscript, represents a value at the npjth hole placement angle, such as RVS npj Represents the rectangular hole length error at the npjth hole placement angle; npk takes the value of 1:N p , npk represents the 1st to Nth p One of the hole placement angles (npkth), with npk as the subscript, represents a value at the npkth hole placement angle, such as DVS npk Indicates the error in the side length of the regular polygon hole at the npkth hole placement angle; nt takes the value of 1:Nt , the total number of available values for the nozzle temperature is N t , then nt represents one of the nozzle temperatures from the 1st to the Nth (the nt-th); nv takes values from 1 to N t ; nv takes values from 1 to N v , the total number of available values for the printing speed is N v , then nv represents one of the printing speeds from the 1st to the Nth (the nv-th); G represents the number of rows of the constructed dataset, with a value of N v ; G represents the number of rows of the constructed dataset, with a value of N m *N t *N v *N p *N p *N p .

[0132] Step A600. Generate a prediction model for 3D printing calibration through a machine learning algorithm.

[0133] Randomly classify the database into two groups: a training group and a prediction group. For example, randomly shuffle the dataset represented by the matrix in step S6 in the row direction, take the first g rows as the training group, where g < G, and the remaining as the prediction group. Use the training group to train the neural network model, and use the prediction group to verify the prediction accuracy of the trained neural network model, and select the neural network model that meets the requirements, that is, obtain the prediction model for 3D printing calibration. Different random groupings will result in differences in the trained neural network models. Use the corresponding test group to test and verify the model. Generally, an accuracy rate of not less than 90% is appropriate.

[0134] Based on the prediction model for 3D printing calibration, only need to use the data containing the key dimension information of the material and the 3D model physical object as the input of this prediction model, and a new set of 3D printing setting parameters can be obtained. Save the prediction model for 3D printing calibration in the control system of the 3D printing device. According to the user's requirements for the 3D printed solid model, the system predicts the 3D printing setting parameters through the prediction model, and executes the 3D printing operation according to the predicted 3D printing setting parameters, or provides auxiliary reference for the user to preset the 3D printing setting parameters.

[0135] In addition, for the overall steps of 3D printing of holes, based on the previously constructed printing model, 3D printing is carried out through the following steps.

[0136] Step B100. The 3D printing device receives the 3D model file

[0137] You can obtain 3D model files from storage hardware (such as USB, mobile phone, computer, etc.) connected to the 3D printing device through the physical interface (such as USB, type-c) on the 3D printing device, or receive 3D model files through wireless communication (such as cloud transmission, WiFi), without limitation. The format of the 3D model file can be any of 3dm, 3ds, dxf, sldprt, step, stl, and stp.

[0138] Step B200: Calculating 3D printing setting parameters of a 3D printing device based on the prediction model and the 3D model file;

[0139] Specifically, the printing quality requirements can be specified by the user. For example, an interactive panel is provided on the 3D printing device. The interactive panel is used to display the current working status of the 3D printing device and receive the user's operating instructions to set some parameters of the 3D printing device. The user can select or set materials, printing quality requirements, etc. through the interactive panel. Printing quality requirements include the size error and accuracy of different types of holes. If the user does not specify, the control system will autonomously specify according to the 3D model. The control system analyzes the received 3D model file. If the 3D model has multiple types of holes, such as round holes, hexagonal holes and rectangular holes, the size error and accuracy of these holes are set to the optimal by default.

[0140] The control system of the 3D printing equipment analyzes the 3D model file to obtain the angle difference between the central axes of different types of holes. This angle difference is the ∆CR, ∆CD, and ∆RD in the input data.

[0141] After determining the material and printing quality requirements, the control system calls the corresponding 3D printing calibration prediction model to calculate and obtain the 3D printing setting parameters including printing speed, nozzle temperature and hole placement angle.

[0142] Step B300: Slice the 3D model according to the 3D printing setting parameters and execute the 3D printing operation.

[0143] The control system converts the 3D model file into a file in stl format. If the 3D model file to be printed is in stl format, no format conversion is required. If it is not in stl format, the control system calls the format conversion algorithm to convert the file into stl format. Then, the control system sets the placement of the 3D model on the base plate of the 3D printing device according to the 3D printing setting parameters, and calls the slice generation algorithm to produce a slice file. The 3D printing device performs 3D printing operations based on the slice file.

[0144] The method disclosed in the present invention analyzes the relationship between the preset printing parameters of the 3D printing device and the actual printing effect of the model containing the hole design (including the size error and accuracy of various types of holes), and establishes a prediction model of the material, hole size error and accuracy and printing speed, nozzle temperature and hole placement angle based on a neural network by adopting a machine learning method. The method can be used as a built-in software algorithm of the 3D printing device to predict the printing quality of the model containing the hole design, improve the success rate of printing the finished product, and reduce unnecessary material loss and time waste.

[0145] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing a hole 3D printing model, characterized in that: The following steps are involved: A100. Construct a printing parameter group, wherein the parameter group consists of a material type parameter, a nozzle temperature parameter, a printing speed parameter, and a hole position parameter; A200. Generate slice data according to the sample model corresponding to each parameter group; A300. Print several sample models based on the generated slice data; A400. Compare the hole data of each sample model with the designed hole data to obtain error parameters respectively; A500. Constructing a data set based on each printing parameter group and the corresponding error parameter; A600. Constructing a printing model by machine learning the data set; The nozzle temperature parameter = , where T min is the minimum temperature value of the corresponding material, T max is the maximum temperature value of the corresponding material, ∆T is the temperature interval value; The printing speed parameter = ; The hole position parameters are ; Where i is the number of prints, N is a natural number, and α is the hole position parameter; The error parameters include circular hole error parameters, rectangular hole error parameters and polygonal error parameters; The circular hole error parameters include: Inner diameter error = ; Roundness error = ; The rectangular hole error parameters include: Rectangle length error = ; Rectangle side ratio error = ; The polygon error parameters include: Polygon length error = ; Polygon side ratio error = ; The dataset is: ; Among them, MAT nm Indicates the material property code, CVS npi and CVP npi Respectively represent the inner diameter error and the roundness error, RVS npj and RVP npj Denote the rectangle length error and the rectangle side length ratio error respectively, DVS npk and DVP npk They represent the polygon length error and the polygon side length ratio error respectively, PC npi Represents the hole position parameter of the circular hole, PR npj Represents the hole position parameter of the rectangular hole, PD npk represents the hole position parameter of the polygonal hole, ∆CR is PC npi -PR npj The angle difference between PC npi -PD npk The angle difference between PR and npj -PD npk The angle difference, T nt Indicates the nozzle temperature parameter, V nv represents the printing speed parameter, G is the number of times the sample model is printed; The dataset contains MAT nm 、CVS npi 、CVP npi 、RVS npj 、RVP npj 、DVS npk 、DVP npk , ∆CR, ∆CD, ∆RD are input data, PC npi , PR npj , PD npk , T nt 、V nv For output data.

2. A method for constructing a hole 3D printing model according to claim 1, characterized in that: The step A600 includes: A610. Randomly classify the data set into a training group and a prediction group; A620. Train the neural network model using the training group, and verify the prediction accuracy of the trained neural model using the prediction group; A630. Based on the verification results, the neural network model is screened and the printing model is obtained.

3. A method for constructing a hole 3D printing model according to claim 2, characterized in that: The step A610 includes shuffling the data set in the row direction, taking the first g rows as the training group, g < the number of times the sample model is printed, and the rest as the prediction group.

4. A method for 3D printing of holes, characterized in that: The following steps are involved: B100. Receive target model file; B200. Based on a method for constructing a 3D printing model of a hole according to any one of claims 1 to 3, combined with a target model, determining printing parameters of a 3D printing device; The B200 includes: B210. Input the printing data of the target model through the interactive medium, wherein the printing data includes printing materials and error requirements; B220. Input the printing data into the printing model to obtain nozzle temperature parameters, printing speed parameters and hole position parameters corresponding to the target model; B230. Adjust the angle of the target model in the 3D printing device according to the hole position parameters to determine the target slice data; The printing parameters include nozzle temperature parameters, printing speed parameters and target slice data; B300. Perform 3D printing on the target model based on the determined printing parameters.

5. A device for hole 3D printing method according to claim 4, characterized in that: include: An input data acquisition module is used to acquire printing input data of a to-be-printed three-dimensional model, wherein the printing input data includes: printing materials and error requirements of a target object in the to-be-printed three-dimensional model; A model processing module, used for inputting the printing input data into the printing model to obtain the nozzle temperature parameters, printing speed parameters and hole position parameters of the target object output by the printing model; An angle adjustment module, used to adjust the angle of the to-be-printed three-dimensional model according to the placement angle of the target object, to obtain an adjusted three-dimensional model to be printed; A slicing module, used for slicing the adjusted three-dimensional model to be printed to obtain slicing data; The printing module is used to print the slice data according to the printing material, the nozzle temperature parameter and the printing speed parameter to obtain the printing result of the three-dimensional model to be printed.

6. The device according to claim 5, characterized in that Also included is a pre-training module for building the printing model.

Images