CN110815812A

CN110815812A - Parallel printing method of multi-nozzle 3D printer

Info

Publication number: CN110815812A
Application number: CN201810893948.2A
Authority: CN
Inventors: 严铜
Original assignee: Individual
Current assignee: Individual
Priority date: 2018-08-08
Filing date: 2018-08-08
Publication date: 2020-02-21

Abstract

The invention discloses a method for parallel printing of a multi-nozzle 3D printer, which realizes efficient collaborative printing of the multi-nozzle 3D printer. The method comprises the following steps: analyzing the model file, adding support, slicing, generating a printing section, partitioning the section, planning a printing path, and printing according to the planned path. According to the invention, by adopting the method of partitioning based on the printing section and combining the printing time estimation algorithm, the printing capacity of each spray head assembly is fully utilized, and the efficiency of large-size 3D printing is greatly improved.

Description

Parallel printing method of multi-nozzle 3D printer

Technical Field

The invention belongs to the technical field of 3D printers, and particularly relates to a parallel printing method of a multi-nozzle 3D printer.

Background

The 3D printing technology is a fast forming technology, which is a technology for forming an object by using forming materials such as metal, plastic, photosensitive resin and the like in a layer-by-layer printing mode on the basis of a digital three-dimensional model file, and belongs to additive manufacturing. At present, a 3D printer based on Fused Deposition Modeling (FDM) principle has become a 3D printer with the highest popularity due to the advantages of simple structure, rich types of applicable materials, low cost of equipment and consumables and the like.

The FDM printer is formed by melting and extruding printing consumables into filament, and then slowly stacking and depositing from point to line to face, so compared with the 3D printing technology of the face forming process, the printing efficiency is lower, especially when printing a large-size model, it often takes tens of hours to complete, thus limiting the application of the technology to large-size printers. The current common method for improving the printing efficiency is to increase the thickness of a single layer, and the increase of the thickness of the layer leads to lower printing precision, so that the surface of the model becomes rough. Increasing the efficiency by simply increasing the layer thickness is therefore not an ideal option.

The present invention provides a structure and a method for improving printing efficiency through multi-nozzle parallel printing, for example, an invention patent 201510706198.X, which is a multi-nozzle 3D printer and a collaborative printing method thereof, and provides a method for improving printing efficiency through multi-nozzle parallel printing, wherein the method is that a printing platform of the multi-nozzle 3D printer is averagely divided into N printing subareas according to the number of nozzles, then model slicing and path planning are performed according to a slicing algorithm of a single nozzle, paths falling into different subareas are re-planned, and in order to avoid interference generated in the parallel printing process of two adjacent nozzles, each printing subarea is divided into two parts to obtain two left and right subareas. Then each printing head synchronously prints the left sub-subarea or the right sub-subarea when printing, and after all the printing heads are completely finished, the printing heads are switched to the next subarea together.

The scheme has the advantages that the idea of multi-nozzle parallel printing is adopted, so that the printing speed is improved, the adopted slicing algorithm is basically consistent with that of a single nozzle, secondary distribution is carried out on paths falling into different partitions only after the integral slicing is completed, the method for avoiding the interference of adjacent nozzles is to further equally divide the partitions, the synchronization of the printing actions of all the nozzles is realized, the algorithm is simple, and the realization is easy.

However, the method of directly performing average partition on the printing platform is also very disadvantageous, and although the method is easy to operate, the actual situation of each cross section of the model to be printed cannot be considered, so that the printing resources of multiple nozzles cannot be fully used, and the printing efficiency is reduced. For example, a model with a diamond cross section is printed, as shown in fig. 1, according to the method described in the above, assuming that the diamond spans 3 partitions, namely, the 1 st partition 11, the 2 nd partition 12, and the 3 rd partition 13, each partition is further divided into two left and right sub-partitions, and three nozzles print the 1 st left sub-partition 111, the 2 nd left sub-partition 121, and the 3 rd left sub-partition 131, respectively, during printing. The FDM printer has a positive correlation between the single-layer printing time and the cross-sectional area of the printing region when the packing density is the same, which can be understood as an approximate proportional relationship. It will be understood that the 1 st left sub-partition 111 will be printed first, followed by the 3 rd sub-partition 131, and finally the 2 nd sub-partition 121. In order to avoid interference, the printer needs to start to print the three right sub-partitions after the three left sub-partitions are all printed, and the three right sub-partitions cannot be printed at the same time. It can be seen that two of the three nozzles are frequently idle, and the efficiency of parallel printing cannot be fully exerted.

The invention patent application with application number 201410534951.7, namely a control method and a printing method of a 3D printer with at least two printing heads, and other invention patent applications with application numbers 201410445615.5, 201410445661.5, 201410445664.9 and 201410445665.3 filed by the same applicant and based on the same idea, show another control method for parallel printing. The core thought of the scheme is that the printing section is divided into partitions with 2 times of the number of the printing heads in an equal area. Taking the double printing heads as an example, the double printing heads divide 4 areas in a shape of Chinese character tian, the circular area at the junction position of the centers of the 4 areas is the conflict area, the two printing heads respectively print corresponding non-conflict areas, and then one printing head prints the middle circular conflict area, so that the problem of multi-nozzle interference is solved.

The biggest defect of the scheme is that the scheme only considers the physical size of the spray head, the X, Y shaft motion mechanism for driving the spray head to move is idealized into a line with the width of 0, the conflict area possibly caused by the motion mechanism is not considered in the partition scheme, therefore, the scheme can only be implemented by a structure similar to a serial mechanical arm, and the arrangement position has a plurality of limits. Therefore, the realization is complicated, the cost is high, and the practicability is not ideal.

From the above analysis, it can be seen that the prior art scheme of multi-nozzle parallel printing is not very ideal.

Disclosure of Invention

In order to overcome the defects of the prior art, the invention provides a method for parallel printing of a multi-nozzle 3D printer, which enables the printing capacity of each nozzle assembly to be fully utilized by adopting a method for dividing partitions based on a printing section and combining a printing time estimation algorithm and a partition algorithm, and greatly improves the efficiency of large-size 3D printing.

In order to achieve the purpose, the invention adopts the following technical scheme:

a method for parallel printing of a multi-nozzle 3D printer is provided, and comprises the following steps:

(1) and acquiring and analyzing the model file, and adding a support structure at a required position.

(2) The method comprises the steps of slicing a model file and generating a section of each printing layer, distinguishing the sections according to different types, filling modes, filling rates, materials, printing silk widths or printing speeds, and recording the corresponding sections respectively, wherein the type refers to that the area is a model part or a supporting part, the filling mode refers to an internal filling mode, common filling modes include linear filling, square filling, triangular filling, hexagonal filling, three-dimensional filling and the like, and the filling rate refers to the ratio of the area occupied by the filling materials in a certain area to the area of the area.

(3) The cross section is integrally divided into K subareas, the K subareas are sequentially arranged along the X-axis direction, K nozzle assemblies are selected, the printing path of each nozzle assembly is planned respectively, so that interference among the nozzle assemblies cannot occur in the printing process, the K is not more than T, the K is a positive integer, and the T is the number of the nozzle assemblies. The partitioning method in the prior art is area-averaged partitioning.

(4) And the K spray head assemblies print simultaneously according to the respective planned paths.

It should be noted that, the partitions are sequentially arranged along the X-axis direction for convenience of description, and the partitions are not necessarily arranged along the X-axis direction in the embodiment based on the multi-head printer structure.

Preferably, the method for dividing the whole cross section into K subareas is to calculate the estimated printing time or the relative printing time of each subarea, wherein the relative printing time is obtained by using a parameter value which is proportional or approximately proportional to the printing time as the printing time, and if the printing speeds are the same, the printing time is approximately proportional to the printing path length, so that the relation between the printing times can be indirectly known by comparing the path lengths. By utilizing the estimated printing time or the relative printing time of each subarea, the total printing time of which subarea mode among different subarea modes is the least can be judged and compared.

Ideally, the partition mode with the minimum total printing time is that all the nozzle assemblies are utilized, and each nozzle assembly can realize full-load work, but in practical situations, when the partition is too small, interference may frequently occur between adjacent nozzle assemblies, and in order to avoid interference, a plurality of nozzle assemblies need to print conflict areas in a time-sharing manner, so that the printing capacity of the nozzle assemblies is idle, and the total printing time is prolonged. Therefore, the determination of the partition number K needs to be determined according to the actual situation of the printing section, and the optimal scheme in various partition modes when different partition numbers are adopted is found out.

The most accurate method for calculating the estimated printing time is to plan the printing path, and then calculate the time required for walking the whole path according to the set parameters of printing speed, printer acceleration and the like and by combining the motion control algorithm of the main board.

Preferably, the slicing time becomes very long because the above-mentioned calculation method is computationally intensive. To simplify the calculation, the printing time can be estimated by calculating a parameter that is approximately linearly proportional to the printing time. When the acceleration and deceleration in the printing operation are omitted and each movement path is idealized to be a uniform movement, it can be found that the printing time t is proportional to the path length L and inversely proportional to the movement speed v, that is, t = L/v. The actual print area S in the subarea can be found by combining the print wire width b, the filling rate A and the subarea area S_pSince = S · a = L · b, t = (S · a)/(v · b) can be inferred.

Generally, in order to consider both printing quality and printing efficiency, the printing speed of the model is lower than the supporting printing speed, the printing speed of the outer ring of the model is lower than the internal printing speed, and the filling of the model material and the supporting material is generally performedThe fill rates are also different, and in addition, generally the print filament width b is positively correlated with the nozzle diameter, and the filament printed with the larger nozzle diameter is wider, so further considering that the fill rate a, the print speed v and the print filament width b of different sub-regions may be different in the whole sub-region, a more generalized formula is that

In the formula, the partition is divided into N sub-areas according to the difference of the three parameters A, v and b, so that the three parameters in each sub-area are the same, and the printing time required by each of the N sub-areas is accumulated and summed to obtain the estimated printing time of the whole partition.

Preferably, when the parameters v and b are the same or similar, or the difference part is smaller, the two parameters in all the subareas can be considered to be approximately the same, and at this time, the formula can be combined to find that the printing time is in direct proportion to the actual printing area of the cross section, so that the scheme that the printing time is the shortest in different subarea modes can be judged by calculating the actual printing area value of each subarea as the relative printing time and also by the relative printing time. The actual printing area is calculated by the formula

And splitting the partition into M sub-areas according to different filling rates A.

Preferably, to avoid interference between the multiple nozzle assemblies, it is necessary to prevent the adjacent nozzle assemblies from appearing in a collision region at the same time during path planning, and in order to simplify a path planning algorithm, each partition may be further divided into a left sub-partition and a right sub-partition arranged along the X-axis direction, and the method for respectively planning the printing paths of the K nozzle assemblies is to plan all the nozzle assemblies to print the sub-partitions on the same side first, and then to synchronously start printing the sub-partitions on the other side after all the nozzle assemblies complete the printing of the sub-partitions on the one side. For example, the left sub-partitions are printed at the same time, and after all the left sub-partitions are printed, the right sub-partitions are printed together. The idea of the mode is that different subareas synchronously print the subareas on the same side in the respective subareas, and the subareas on the other side are used for separating the adjacent nozzle assemblies, so that no conflict area exists between the adjacent nozzle assemblies when the subareas on each side print.

In order to meet the condition that when the same-side subarea is printed, no conflict area exists, the subarea number K meets the condition of W_minD is more than or equal to d. The W is_minThe minimum value of the projection widths W of all the other sub-subareas except for two sub-subareas at two ends on the X axis is defined, wherein W is an area which is not overlapped with other adjacent sub-subareas in the projection of the corresponding sub-subarea on the X axis, and d is the maximum value of the projection distances of all the nozzles used for printing the respective sub-subareas when the two nozzle assemblies are closest to each other on the X axis. It should be noted that when the cross-sectional area is very small so that K =1, since printing is performed directly in the single head mode by using only one head assembly, it is not necessary to calculate W_min. It is easy to understand that d is the minimum distance between two nozzle assemblies that do not have a collision zone when printing, and for the sake of calculation, it can be simply considered that all the nozzles in the nozzle assemblies are used, and then d becomes a constant completely determined by the mechanical structure of the printer.

Since the sub-segments at both ends have adjacent sub-segments on only one side, they do not need to be used to isolate adjacent showerhead modules and thus their width W is not limited.

Preferably, to achieve the highest printing efficiency, the best condition is that the printing of the sub-partitions on the same side is completed simultaneously, and then the printing of the sub-partitions on the other side is started together. However, to satisfy the minimum width requirement of the sub-partitions, the print time of all the sub-partitions is not always evenly divided, and to minimize the total print time, the sum of the maximum value of the estimated print times of all the left sub-partitions and the maximum value of the estimated print times of all the right sub-partitions needs to be minimized. Therefore, all possible partition modes can be listed according to a certain step value, and the partition mode with the minimum sum of the two partition modes can be found.

The calculation amount of exhausting all the partition modes and comparing is large, in order to improve the efficiency, the partition comparison in the full range can be firstly carried out according to a large step value, the approximate position of the optimal partition mode is found, and then the optimal partition mode is further searched near the current partition position according to a small step value.

Although the optimal partition mode can be found very accurately without being influenced by the specific conditions of the printing section by exhausting all partition modes, the mode needs to carry out a large amount of operation, when the platform operation speed is insufficient, the partition can be divided averagely according to the estimated printing time, and K partitions are divided into a left sub-partition and a right sub-partition, so that the printing time of the sub-partitions on the same side is equal, and the requirement of W_minD is more than or equal to d; the partition number K is the maximum integer value satisfying the above condition and K is less than or equal to T.

Compared with the mode of exhaustively dividing all the subareas according to a certain step value, the calculation amount of the mode is greatly reduced, but the mode simply adopts the method of averagely dividing according to the estimated printing time, and when the subarea number K is measured, W is measured_min>d, if K at this time<T, the capacity of the showerhead assembly cannot be brought to its limit. At this point, the partition sizes may be compressed such that W_minD, printing the excessive area by the K +1 th spray head assembly, so that the capacity of the printer can be exerted to the limit. In actual operation, the partition number K is obtained according to the method, and whether K is met is judged<T and W_min>d, if yes, partitioning according to K +1 to enable W_minD, the printing time of all the other subareas except the two subareas at the two ends is the same, and the estimated printing time of the two subareas at the two ends is less than or equal to the estimated printing time of the other subareas. When the two sub-areas at two ends do not divide the sub-areas if the width W before the two sub-areas divide the sub-areas is less than or equal to d, for example, when the width W of the right-most sub-area is less than or equal to d, the whole right-most sub-area can be regarded as the left sub-area and the right sub-area is 0.

Preferably, in order to improve the overlapping strength at the boundary between the adjacent partitions or sub-partitions, the boundary between the adjacent partitions or sub-partitions is a curved line or a broken line such that the two partitions are staggered with each other at the interface to improve the overlapping strength.

Furthermore, the boundaries of the partitions or sub-partitions between adjacent printing layers are not overlapped on the projection plane in the vertical direction, that is, after the cross sections of the adjacent printing layers are vertically projected onto the same plane, the respective boundaries of the adjacent printing layers are staggered, so that the boundaries of the layers of the model are distributed in a certain range instead of being concentrated together, and the overall strength of the printing model is improved.

It should be noted that, in the above method, for example, average partition, minimum printing time, and equal printing time are theoretical optimal solutions of the method, in practical operation, because of the limitation of computing platform computing capability, it is not always necessary to find absolute average, minimum, or equal, for example, when a partition mode makes the maximum deviation between the estimated printing times of the partitions less than 1%, they may be considered approximately equal, because it takes a lot of computing power to eliminate 1% of deviation, but the final printing efficiency is improved very little, so it is not economical to find an absolute optimal solution. In operation, the maximum allowable deviation may be set with the target of the optimal solution described in the method, and when the maximum allowable deviation is smaller, the optimal solution is considered to have been achieved.

Compared with the prior art, the invention realizes the optimized distribution of the workload of each spray head component after the partition by adopting the partition method based on the printing section, and further provides the method for calculating the estimated printing time of each partition, so that the total printing time after the partition is minimum, the capacity of the printer is fully utilized, and the overall printing efficiency of the printer is improved. Meanwhile, a printing time pre-estimation algorithm and a partition algorithm are provided, and the calculation of parameters such as section filling rate, printing speed, printing silk width and the like is added, so that the printing condition of various speeds and various filling rates in the section can be adapted, the partition is more accurate, and the printing efficiency is higher.

Drawings

FIG. 1 is a schematic diagram of a prior art parallel printing partition.

Fig. 2 is a schematic diagram of a movement structure of the multi-nozzle 3D printer.

Fig. 3 is a three-dimensional view of a print model.

Fig. 4 is a front view of the print module with added support.

FIG. 5 is a cross-section of one of the print layers after the print model has been sliced.

FIG. 6 is a schematic representation of a print after a section of the model is sectioned.

Fig. 7 is a front view schematic diagram of a nozzle part of a multi-nozzle 3D printer.

FIG. 8 is a schematic view of a fully filled rectangular cross-section partition.

FIG. 9 is a schematic representation of several forms of a cross-sectional zoning boundary.

Detailed Description

The method for parallel printing by a multi-nozzle 3D printer according to the present invention is further described below with reference to the accompanying drawings and the detailed description, so as to more clearly understand the technical idea claimed in the present invention.

The embodiment of the parallel printing method of the multi-nozzle 3D printer comprises the following steps:

fig. 2 shows a schematic diagram of a movement structure of a plurality of independent nozzle assemblies, a plurality of Y-axis guide rails 22 are disposed on an X-axis guide rail 21, a plurality of nozzle assemblies 23 are mounted on respective corresponding sliders of the Y-axis guide rails 22, and the plurality of Y-axis guide rails 22 can move independently on the X-axis guide rail 21.

The method for multi-nozzle parallel printing provided by the embodiment comprises the following steps:

(1) obtaining a model file, analyzing the model file, and adding a corresponding support structure at a place needing to be supported;

(2) and slicing the model file to generate a section of each printing layer. Dividing the section of each printing layer into a plurality of areas according to different parameters such as types, filling rates, materials, printing silk widths or printing speeds and the like, and independently recording each area;

(3) according to the actual size of the printing section, the printing section is integrally divided into K subareas, the K subareas are sequentially arranged along the X-axis direction, K nozzle assemblies are selected, and the printing path of each nozzle assembly is respectively planned, so that the nozzle assemblies cannot interfere with each other in the printing process, and the number K of the subareas cannot exceed the number T of the nozzle assemblies;

The details of the implementation of each step will be described below by taking the model shown in fig. 3 as an example.

First, a model file is obtained and analyzed, as shown in fig. 4, the model is an inverted U-shaped structure, and a suspended structure is arranged below a top beam of the model 31, so that a support structure 32 needs to be added below the top beam during printing.

After the support structure is added to the printing model, slicing can be performed. FIG. 5 is a cross-section of a printed layer of one of the layers after the model has been sliced. The section can be divided into model parts (including a model wall outer ring 311, a model wall inner ring 312 and a model filling 313) and a support part 321 according to types; the filling rate can be divided into 100% filling part (comprising a model wall outer ring 311 and a model wall inner ring 312), 50% filling part (comprising a model inner filling 313) and 10% filling part (comprising a supporting part 321); since only one model material and one support material are used in this example, the result of the division by material is the same as the division by type; in the embodiment, the printing silk widths of all the areas are 0.4mm, so that the printing silk widths do not need to be divided; the printing speed is divided into a 40mm/s printing speed part (comprising a model wall outer ring 311), a 70mm/s printing speed part (comprising a model wall inner ring 312 and a model inner filling 313) and an 80mm/s printing speed part (comprising a supporting part 321), the dividing results of the parameters are integrated to ensure that each parameter in each area is the same, and the section can be finally divided into four areas of the model wall outer ring 311, the model wall inner ring 312, the model inner filling 313 and the supporting part 321.

To simplify the calculation, the parameter regions with small area ratio or small difference can be combined with the similar regions. If the outer ring 311 and the inner ring 312 of the model wall are only the difference in the printing speed, and the outer ring 311 of the model wall generally has only a single filament, the ratio of the whole cross section is small, so the outer ring 311 and the inner ring 312 of the model wall can be combined, and the final influence is small by calculating according to the parameters of the inner ring 312 of the model wall. If the area ratio of the model wall to the whole model is low, the outer ring 311, the inner ring 312 and the inner filling 313 of the model wall may not be distinguished, and the calculation is performed according to the parameter of the inner filling 313 with the largest area ratio.

And partitioning the sliced section, wherein the partition method with the minimum total printing time is to ensure that the estimated printing time of all the spray head assemblies after partitioning is equal, so that all the spray head assemblies are in a full-load working state in the whole printing process, and the whole capacity of the printer is highest. As shown in fig. 6, taking a 3-nozzle assembly printer as an example, the cross section is divided into 3 partitions, namely, a 1 st partition 41, a 2 nd partition 42, and a 3 rd partition 43, and the estimated printing times of the 3 partitions are made to be substantially equal. As shown in the figure, 411 is the outer ring of the 1 st partition model wall, 412 is the inner ring of the 1 st partition model wall, 413 is the inner filling of the 1 st partition model, 421 is the outer ring of the 2 nd partition model wall, 422 is the inner ring of the 2 nd partition model wall, 423 is the inner filling of the 2 nd partition model, 424 is the supporting part of the 2 nd partition, 431 is the outer ring of the 3 rd partition model wall, 432 is the inner ring of the 3 rd partition model wall, and 433 is the inner filling of the 3 rd partition model.

In this embodiment, the estimated printing time of each partition is calculated by the formula

Calculated as S in the formula_nIs the area of the corresponding region in the partition, A_nIs the fill factor, v, of the area_nAs the printing speed of the area, b_nFor the printing line width of the region, taking the 1 st partition 41 as an example, the predicted printing time t₄₁=(S₄₁₁·100%)/(40mm/s·0.4mm)+ (S₄₁₂·100%)/(70mm/s·0.4mm)+ (S₄₁₃·50%)/(70mm/s·0.4mm)。

When the partition is divided, under the condition that the calculation capability allows, the partition position can be gradually adjusted by adopting a dichotomy to approach an ideal partition state with zero deviation, so that the probability of idle generation of each spray head assembly is reduced to the minimum.

There are 3 kinds of printing speeds in total in the section, 40mm/s for the outer ring 311 of the mold wall, 70mm/s for the inner ring 312 of the mold wall and the inner filler 313 of the mold, and 80mm/s for the supporting portion 321. The area of 40mm/s is only printed for one circle of the outer circle of the model wallThe wire, the 80mm/s supporting portion 321 is relatively small in actual printing area because of its packing density of only 10%, and the difference between the actual printing area and the printing speed of the model main portion 70mm/s is not large, so when the calculation capability is not sufficient, the difference between the printing speeds of the two areas can be ignored, and the actual printing area is approximately considered to be proportional to the estimated printing time, so that the actual printing area can be directly used as the relative printing time value to participate in the subsequent calculation, for example, the required printing time is equal, which can be converted into the required actual printing area is equal. Can be expressed according to formulaThe actual print area of each partition is calculated. Also taking the 1 st partition 41 as an example, its actual print area S_{p_41}=(S₄₁₁+S₄₁₂)·100%+S₄₁₃·50%。

When 3 nozzle assemblies are directly used for printing 3 partitions in parallel, the problem of interference between adjacent nozzle assemblies needs to be considered, and path planning is complex. In order to simplify the path planning in parallel printing, the partitions can be further divided into left and right sub-partitions. As shown in fig. 6, the 1 st partition 41 is divided into a 1 st left sub-partition 41A and a 1 st right sub-partition 41B, and likewise, the 2 nd partition 42 is divided into a 2 nd left sub-partition 42A and a 2 nd right sub-partition 42B, and the 3 rd partition 43 is divided into a 3 rd left sub-partition 43A and a 3 rd right sub-partition 43B.

The sub-partitions are divided, and the printing time of the sub-partitions on the same side is required to be equal as much as possible, so that all the nozzle assemblies can finish printing the sub-partitions on the same side in approximately the same time. And after all the spray head assemblies finish printing, synchronously starting the printing of the sub-partition on the other side.

The previous zoning method is applicable to the general situation, but when the printing section is small, whether the current zoning can satisfy the simultaneous printing of three spray head assemblies without interference needs to be considered.

Before the interference judgment, the minimum spacing distance d of the adjacent nozzle assemblies in the movement collision area is required to be firstly confirmed when printing. As shown in FIG. 7, the 1 st zone 41 only needs to print model material, so only the T11 nozzle is used, and the 2 nd zone 42 needs to print model and support two materials, so two nozzles T21 and T22 are used. As shown, when the two showerhead modules are closest together, the projected distance between the T11 nozzle and the T22 nozzle on the X-axis is the maximum d of the projected distances of all the nozzles used. In order to simplify the determination process, the maximum value of the projection distances of all the nozzles may be set as d regardless of whether or not the nozzles are used.

The number K of the subareas is determined by determining the total width W of the section_TAnd determining the number T and W of the nozzles_TAnd (2d) and taking K as the smaller value of the two to partition.

When in partitioning, the printing time is firstly averagely partitioned according to the estimated printing time, and then each partition is partitioned into a left sub-partition and a right sub-partition, so that the estimated printing time of all the same-side sub-partitions is equal, and W is the same_minMaximum, W_minThe minimum value of the widths of all the sub-partitions except for two sub-partitions at two ends is defined, and when the boundaries of each sub-partition and the adjacent sub-partitions are overlapped in projection on an X axis, the width of the non-overlapped area of each sub-partition is taken as the width W of the non-overlapped area of each sub-partition.

Then, W is judged_minIn relation to d, when W_minD or more, completing the partition, if W is greater than d_min<d, taking K = K-1, and repeating the partitioning operation until W_minD is more than or equal to d. When the printing section is small, if K =2 does not satisfy the above condition, K =1 is taken, and the printing is performed in the single-nozzle mode, so that it is not necessary to calculate W_min。

However, this method of determining the number of divisions K cannot find an optimal division method because it always requires that the divisions be divided equally according to the estimated printing time, and thus only W is the division method_minD is the optimal solution, so if W is partitioned according to the method_min>d, when there are spare nozzle assemblies, the number of partitions K can be increased by 1, and the partitions are re-partitioned to W_minD, except for the two subareas at the two ends, the other middle subareas are estimated to have the same printing time, and when the width W of the two subareas at the two ends before the subareas are divided is not more than d, the subareas are not divided, for example, when the width W of the rightmost subarea is not more than d<d, then the rightmost end can be partitioned into the wholeIt is considered as its left sub-partition and its right sub-partition has a width of 0.

For example, as shown in fig. 8, assuming that a 100% filled rectangular cross section is printed, the printing speed and the filament width are the same, it is easy to understand that the printing width is directly proportional to the printing time at this time, let d =100mm, when the total width of the cross section is 540mm, the method is used to divide the rectangular cross section into sections, K =2, the width of each sub-section is 135mm, and the width to be printed by each nozzle assembly is 270mm, refer to a division scheme a in the figure, where the shaded area is the left sub-section of each section, and the non-affected area is the right sub-section of each section; and if W is taken_minIf = d is divided, K =3 may be adopted, and the widths of the remaining sub-partitions are all 100mm except that the width of the right sub-partition on the rightmost side is 40mm, so that the maximum width required to be printed by a single nozzle assembly is 200mm, which is referred to as the B-partition scheme in the figure. It can be seen that the partition scheme B is the optimal partition mode for the section.

When the printing section is of a special-shaped structure or the filling density is very uneven, the partitioning method cannot necessarily find the optimal partitioning mode well, in this case, all the partitioning possibilities can be exhausted according to a certain step, and the partitioning mode with the minimum sum of the maximum value in the estimated printing time of all the left sub-partitions and the maximum value in the estimated printing time of all the right sub-partitions is found through calculation. During operation, the optimal partition mode can be found according to a larger step value, and then the optimal partition mode is further searched near the current partition position according to a smaller step value. This method can well realize the optimal partition of various sections, but has the disadvantage that the required calculation amount is very large, so that the partition method can be selected according to the actual situation.

In order to enhance the overlapping strength at the boundary, the boundary between the sub-partitions or the partitions is designed into a mutually staggered curve or broken line form, as shown in fig. 9, several forms of the boundary using broken lines and curves are shown, wherein the form 1 and the form 2 are broken line boundaries, and the form 3 and the form 4 are curved boundaries, so that the curved boundaries are smoother as a whole, and stress concentration points are not easy to appear on the structure, therefore, the form 2 and the form 4 are preferably inverted structures from the view of splicing effect, and splicing of the two sub-partitions can be more reliable. The two partitions are embedded into each other to a certain depth, so that the connection strength at the boundary of the partitions or the sub-partitions is higher, and the strength of the printed model is not obviously reduced at the interface due to the partitions.

Meanwhile, the respective boundary lines of the upper layer and the lower layer are staggered with each other, so that the boundary lines of the layers are not completely overlapped, the weak points of the model cannot be concentrated together, and the overall strength of the split printed model can be approximately equivalent to that of the model printed by a conventional single nozzle.

Various other changes and modifications to the above-described embodiments and concepts will become apparent to those skilled in the art from the above description, and all such changes and modifications are intended to be included within the scope of the present invention as defined in the appended claims.

Claims

1. A parallel printing method of a multi-nozzle 3D printer is characterized by comprising the following steps:

(1) obtaining and analyzing a model file, and adding a supporting structure at a required place;

(2) slicing the model file and generating a section of each printing layer, distinguishing the sections according to different types, filling rates, materials, printing silk widths or printing speeds, and respectively recording the corresponding sections;

(3) dividing the whole cross section into K subareas, sequentially arranging the K subareas along the X-axis direction, selecting K nozzle assemblies, and respectively planning the printing path of each nozzle assembly, so that interference among the nozzle assemblies cannot occur in the printing process, wherein K is less than or equal to T and is a positive integer;

2. The method for multi-nozzle parallel printing according to claim 1, wherein the method for dividing the whole cross section into K subareas is to calculate and determine the subarea mode with the least total printing time by calculating the estimated printing time or relative printing time of each subarea.

3. The method for multi-nozzle parallel printing as claimed in claim 2, wherein the method for calculating the estimated printing time of each partition is calculated by formula

The estimated print time for each partition is calculated.

4. The method for multi-nozzle parallel printing according to claim 2, wherein the method for calculating the relative printing time of each partition is to use the actual printing area value of each partition as the relative printing time.

5. The method for multi-nozzle parallel printing according to claim 2, wherein each divided partition comprises a left sub-partition and a right sub-partition arranged along the X-axis direction, the method for respectively planning the printing path of each nozzle assembly comprises planning all the nozzle assemblies to print the sub-partitions on the same side first, and synchronously starting the sub-partitions on the other side to print after all the nozzle assemblies complete the sub-partitions on the side, and the number of the partitions K satisfies W_min≥d。

6. The method of claim 5, wherein the partitioning is performed according to the minimum sum of the maximum predicted print time of all left sub-partitions and the maximum predicted print time of all right sub-partitions.

7. The method of claim 5, wherein the partitioning method is to divide the K partitions into the left and right sub-partitions according to the estimated printing time, so that the printing time of the same sub-partition is equal to satisfy W_minD is more than or equal to d; the partition number K is the maximum integer value satisfying the above condition and K is less than or equal to T.

8. The method of multi-nozzle parallel printing as in claim 7, wherein when K is<T and W_min>When d, take K = K +1 and repartition partitions to satisfy:

W_min=d，

all the other subareas except the two subareas at the two ends are estimated to be equal in printing time,

the estimated printing time of two subareas at two ends is less than or equal to the estimated printing time of other subareas;

if the width W of the two sub-areas at the two ends is less than or equal to d before the sub-areas are divided, the sub-areas are not divided.

9. The method of multi-nozzle parallel printing according to claim 1, 2 or 5, wherein the dividing line between the partitions or sub-partitions is a curved line or a broken line.

10. The method of multi-nozzle parallel printing according to claim 1, 2 or 5, wherein the divisional boundary lines between adjacent print layers do not coincide on the vertical direction projection plane.