CN109814817A - A kind of 3D printing training data base construction method based on artificial intelligence technology - Google Patents
A kind of 3D printing training data base construction method based on artificial intelligence technology Download PDFInfo
- Publication number
- CN109814817A CN109814817A CN201910085587.3A CN201910085587A CN109814817A CN 109814817 A CN109814817 A CN 109814817A CN 201910085587 A CN201910085587 A CN 201910085587A CN 109814817 A CN109814817 A CN 109814817A
- Authority
- CN
- China
- Prior art keywords
- printing
- parameter
- product
- artificial intelligence
- construction method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010146 3D printing Methods 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 20
- 238000005516 engineering process Methods 0.000 title claims abstract description 17
- 238000012549 training Methods 0.000 title claims abstract description 16
- 238000009411 base construction Methods 0.000 title claims abstract description 14
- 238000007639 printing Methods 0.000 claims abstract description 53
- 238000010276 construction Methods 0.000 claims abstract description 4
- 238000012797 qualification Methods 0.000 claims description 15
- 239000000463 material Substances 0.000 claims description 7
- 235000013399 edible fruits Nutrition 0.000 claims 2
- 239000000047 product Substances 0.000 description 41
- 235000013569 fruit product Nutrition 0.000 description 8
- 238000000465 moulding Methods 0.000 description 6
- 238000004519 manufacturing process Methods 0.000 description 4
- 239000000843 powder Substances 0.000 description 3
- 239000002184 metal Substances 0.000 description 2
- 238000005507 spraying Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 239000006227 byproduct Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010891 electric arc Methods 0.000 description 1
- 238000005323 electroforming Methods 0.000 description 1
- 239000011888 foil Substances 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- 238000010309 melting process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000008188 pellet Substances 0.000 description 1
- 238000000110 selective laser sintering Methods 0.000 description 1
- 238000007740 vapor deposition Methods 0.000 description 1
Abstract
The present invention provides a kind of 3D printing training data base construction method based on artificial intelligence technology, comprising: first setting input parameter step S1, in the system of 3D printer；Step S2, after running 3D printer, the result parameter of printing is generated；Step S3, expert's assessment is carried out to the result parameter, generates assessment numerical value；Step S4, by the input parameter, result parameter and assessment records of values to tranining database.The present invention has the advantage that the construction method of artificial intelligence 3D printing tranining database of the invention passes through the parameter for largely acquiring the 3D printing pre-entered, after 3D printing operation, expert's assessment is carried out to print result, then tranining database is constructed, help is provided for the parameter decision of engineer, the quality and efficiency of 3D printing are improved, and 3D printing system realizes intelligent parameter selection with the 3D printing experience recorded in tranining database.
Description
Technical field
The present invention relates to material increasing fields, more particularly to a kind of building side of artificial intelligence 3D printing tranining database
Method.
Background technique
3D printer is also known as three-dimensional printer, is a kind of increases material manufacturing technology, it is adopted based on digital model file
With moulding material, three-dimensional entity is constructed by layer-by-layer printing.Before printing, it needs to utilize computer modeling software
Modeling, forms 3D model to be printed, then the 3D model " subregion " built up is sliced, at layer-by-layer section to instruct 3D
Printer successively prints.
At present typical 3D printing technique include foil technique for sticking, Stereolithography technique, fused glass pellet technique,
Powder selective laser sintering technique, 3-D spraying binding moulding process, 3 D-printing moulding process, metal powder selectively swash
Light is melted and molded technique, metal powder high energy beam current melting and coating process, electric arc spraying moulding process and vapor deposition molding, electroforming
The techniques such as molding.
The setting of current 3D printing technique depends on engineer experience, and engineer is according to selected increasing material manufacturing technique
Technology carries out lift height to the basic manufacturing process of model data objects and fabrication orientation analysis design and optimization, technique is joined
Several formulations whether the superiority and inferiority of engineer's technique initialization, finally affects the quality of 3D printing.For example, support parameters, filling it is close
The improper product that will lead to printing of the setting such as degree, print temperature is coarse or even printing fails.Engineer is in face of a large amount of in practice
3D printing parameter, can inevitably calculate mistake, can thus reduce the quality and efficiency of 3D printing.
Summary of the invention
The technical problem to be solved in the present invention is to provide a kind of building side of artificial intelligence 3D printing tranining database
Method constructs tranining database, provides ginseng for the parameter decision of engineer by largely acquiring the parameter of 3D printing and being assessed
It examines, improves the quality and efficiency of 3D printing.
The present invention is implemented as follows: a kind of 3D printing training data base construction method based on artificial intelligence technology, packet
It includes:
Step S1, the first setting input parameter in the system of 3D printer；
Step S2, after running 3D printer, the result parameter of printing is generated；
Step S3, expert's assessment is carried out to the result parameter, generates assessment numerical value；
Step S4, by the input parameter, result parameter and assessment records of values to tranining database.
Further, the input parameter in the step S1 include 3D printer performance parameter, 3D printing model parameter with
3D printing technological parameter.
Further, the 3D printer parameter includes stamp with the size, equipment printing precision, thickness and print temperature.
Further, the 3D printing model parameter includes fixed-point number, number of edges, face number, volume, surface area and normal mistake
Number.
Further, the 3D printing technological parameter includes packed density, lift height, wall thickness, print speed, empty walking speed
Degree, support volume and tilt angle.
Further, the result parameter in the step S2 includes product printing precision, product print time and beats
Print consumption of materials.
Further, expert's assessment is carried out to the result parameter in the step S3 specifically: using assessment formula:
P=Ad+Bt+Cm；
Wherein, P is that expert assesses numerical value, and A, B and C are weight parameter, A ∈ [0,1], B ∈ [0,1], C ∈ [0,1], and
A+B+C=1, d are the scoring of product printing precision, and t is the scoring of product print time, and m is the scoring of product printing consumables amount.
Further, the assessment formula specifically:
P=0.34d+0.33t+0.33m.
Further, the product printing precision scoring, the scoring of product print time and the scoring of product printing consumables amount
It chooses are as follows:
Default default print precision, the product printing precision is compared with the default print precision, according to than
Compared with as a result, product printing precision scoring is chosen for " outstanding ", " qualification " and one of them in " failing " this three；
The default default print time, the product print time is compared with the default print time, according to than
Compared with as a result, product print time scoring is chosen for " outstanding ", " qualification " and one of them in " failing " this three；
Default default print consumable quantity, the product printing consumables amount is compared with the default print consumable quantity,
According to comparison result, product printing precision scoring is chosen for " outstanding ", " qualification " and its in " failing " this three
In one.
Further, described " outstanding " value is 100, and the value of " qualification " is 80, and described " failing " takes
Value is 60.
The present invention has the advantage that the construction method of artificial intelligence 3D printing tranining database of the invention passes through largely
The parameter for acquiring the 3D printing pre-entered carries out expert's assessment to print result, then building training after 3D printing operation
Database provides help for the parameter decision of engineer, improves the quality and efficiency of 3D printing, and 3D printing system is with instruction
Practice the 3D printing experience recorded in database, realizes intelligent parameter selection.
Detailed description of the invention
The present invention is further illustrated in conjunction with the embodiments with reference to the accompanying drawings.
Fig. 1 is the construction method execution flow chart of artificial intelligence 3D printing tranining database of the invention.
Specific embodiment
Refering to fig. 1, a kind of 3D printing training data base construction method based on artificial intelligence technology, comprising:
Step S1, the first setting input parameter in the system of 3D printer；
Input parameter in the step S1 includes 3D printer performance parameter, 3D printing model parameter and 3D printing technique
Parameter.
The 3D printer parameter includes stamp with the size, equipment printing precision, thickness and print temperature.
The 3D printing model parameter includes fixed-point number, number of edges, face number, volume, surface area and normal error number.
The 3D printing technological parameter includes that packed density, lift height, wall thickness, print speed, sky walk speed, supporter
Long-pending and tilt angle.
The above input parameter is pre-entered into the system of 3D printer, then carries out 3D printing.
Step S2, after running 3D printer, the result parameter of printing is generated；
Result parameter in the step S2 includes product printing precision, product print time and printed material consumption.
Step S3, expert's assessment is carried out to result parameter, generates assessment numerical value；
Expert's assessment is carried out to the result parameter in the step S3 specifically: using assessment formula:
P=Ad+Bt+Cm；
Wherein, P is that expert assesses numerical value, and A, B and C are weight parameter, A ∈ [0,1], B ∈ [0,1], C ∈ [0,1], and
A+B+C=1, d are the scoring of product printing precision, and t is the scoring of product print time, and m is the scoring of product printing consumables amount.
Preferably, the value 0.34 of A, the value 0.33 of B, the value 0.33 of C；To which the assessment formula is specially P=
0.34d+0.33t+0.33m。
The selection that the product printing precision scoring, the scoring of product print time are scored with product printing consumables amount are as follows:
Default default print precision, the product printing precision is compared with the default print precision, according to than
Compared with as a result, product printing precision scoring is chosen for " outstanding ", " qualification " and one of them in " failing " this three；
Specifically, as fruit product printing precision be greater than default print precision, then product printing precision scoring be " outstanding "；As fruit product is beaten
It prints precision and is equal to default print precision, then the scoring of product printing precision is " qualification "；It is beaten as fruit product printing precision is less than default
Precision is printed, then the scoring of product printing precision is " failing "；Wherein default print precision is chosen from an interval value.
The default default print time, the product print time is compared with the default print time, according to than
Compared with as a result, product print time scoring is chosen for " outstanding ", " qualification " and one of them in " failing " this three；
Specifically, as the fruit product print time be greater than the default print time, then the product print time scoring be " outstanding "；As fruit product is beaten
The time is printed equal to the default print time, then the scoring of product print time is " qualification "；It is beaten as the fruit product print time is less than default
The time is printed, then the scoring of product print time is " failing "；Wherein the default print time is chosen from an interval value.
Default default print consumable quantity, the product printing consumables amount is compared with the default print consumable quantity,
According to comparison result, by product printing consumables amount scoring be chosen for " outstanding ", " qualification " in " failing " this three
One of them；Specifically, as fruit product printing consumables amount be greater than default print consumable quantity, then product printing consumables amount scoring is
" outstanding "；If fruit product printing consumables amount is equal to default print consumable quantity, then the scoring of product printing consumables amount is " qualification "；If
Product printing consumables amount is less than default print consumable quantity, then the scoring of product printing consumables amount is " failing "；Wherein default print
Consumable quantity is chosen from an interval value.
" outstanding " value is 100, and the value of " qualification " is 80, and the value of described " failing " is 60.
It step S4, will input parameter, result parameter and assessment records of values to tranining database.
3D printer can thus be recorded can filter out according to assessment numerical value to valuable in the input parameter of each run
The parameter of value, i.e. assessed value are higher, and the value for inputting parameter is higher, final to realize product printing precision highest, when product prints
Between it is most short, product printing consumables amount is minimum.The parameter of tranining database record 3D printing forms 3D printing experience, is engineer
Parameter decision help is provided, the quality and efficiency of raising 3D printing in this way；3D printer system can use training data simultaneously
Intelligent parameter selection is realized in library.
For different 3D printing equipment and printing technology, corresponding tranining database can be generated, to record
Corresponding 3D printing experience effectively increases the selected efficiency of the parameter of the 3D printing of engineer, the quality with 3D printing result.
Although specific embodiments of the present invention have been described above, those familiar with the art should be managed
Solution, we are merely exemplary described specific embodiment, rather than for the restriction to the scope of the present invention, it is familiar with this
The technical staff in field should be covered of the invention according to modification and variation equivalent made by spirit of the invention
In scope of the claimed protection.
Claims (10)
1. a kind of 3D printing training data base construction method based on artificial intelligence technology, it is characterised in that: include:
Step S1, the first setting input parameter in the system of 3D printer；
Step S2, after running 3D printer, the result parameter of printing is generated；
Step S3, expert's assessment is carried out to the result parameter, generates assessment numerical value；
Step S4, by the input parameter, result parameter and assessment records of values to tranining database.
2. a kind of 3D printing training data base construction method based on artificial intelligence technology according to claim 1, special
Sign is: the input parameter in the step S1 includes 3D printer performance parameter, 3D printing model parameter and 3D printing technique
Parameter.
3. a kind of 3D printing training data base construction method based on artificial intelligence technology according to claim 2, special
Sign is: the 3D printer parameter includes stamp with the size, equipment printing precision, thickness and print temperature.
4. a kind of 3D printing training data base construction method based on artificial intelligence technology according to claim 2, special
Sign is: the 3D printing model parameter includes fixed-point number, number of edges, face number, volume, surface area and normal error number.
5. a kind of 3D printing training data base construction method based on artificial intelligence technology according to claim 2, special
Sign is: the 3D printing technological parameter includes that packed density, lift height, wall thickness, print speed, sky walk speed, supporter
Long-pending and tilt angle.
6. a kind of 3D printing training data base construction method based on artificial intelligence technology according to claim 1, special
Sign is: the result parameter in the step S2 includes product printing precision, product print time and printed material consumption.
7. a kind of construction method of the tranining database of artificial intelligence 3D printing according to claim 6, it is characterised in that:
Expert's assessment is carried out to the result parameter in the step S3 specifically: using assessment formula:
P=Ad+Bt+Cm；
Wherein, P is that expert assesses numerical value, and A, B and C are weight parameter, A ∈ [0,1], B ∈ [0,1], C ∈ [0,1], and A+B+
C=1, d are the scoring of product printing precision, and t is the scoring of product print time, and m is the scoring of product printing consumables amount.
8. a kind of 3D printing training data base construction method based on artificial intelligence technology according to claim 7, special
Sign is: the assessment formula specifically:
P=0.34d+0.33t+0.33m.
9. a kind of 3D printing training data base construction method based on artificial intelligence technology according to claim 7, special
Sign is: the selection that the product printing precision scoring, the scoring of product print time are scored with product printing consumables amount are as follows:
Default default print precision, the product printing precision is compared with the default print precision, is tied according to comparing
Product printing precision scoring is chosen for " outstanding ", " qualification " and one of them in " failing " this three by fruit；
The default default print time, the product print time is compared with the default print time, is tied according to comparing
Product print time scoring is chosen for " outstanding ", " qualification " and one of them in " failing " this three by fruit；
Default default print consumable quantity, the product printing consumables amount is compared with the default print consumable quantity, according to
Product printing precision scoring is chosen for " outstanding ", wherein one in " qualification " and " failing " this three by comparison result
It is a.
10. a kind of 3D printing training data base construction method based on artificial intelligence technology according to claim 9, special
Sign is: " outstanding " value is 100, and the value of " qualification " is 80, and the value of described " failing " is 60.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910085587.3A CN109814817A (en) | 2019-01-29 | 2019-01-29 | A kind of 3D printing training data base construction method based on artificial intelligence technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910085587.3A CN109814817A (en) | 2019-01-29 | 2019-01-29 | A kind of 3D printing training data base construction method based on artificial intelligence technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109814817A true CN109814817A (en) | 2019-05-28 |
Family
ID=66605629
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910085587.3A Pending CN109814817A (en) | 2019-01-29 | 2019-01-29 | A kind of 3D printing training data base construction method based on artificial intelligence technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109814817A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022161000A1 (en) * | 2021-01-29 | 2022-08-04 | 苏州奇流信息科技有限公司 | Training method and training apparatus for machine learning model, and evaluation system |
| CN117103692A (en) * | 2023-04-13 | 2023-11-24 | 上海轮廓科技有限公司 | 3D printing methods, devices, storage media and computer program products |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050026121A1 (en) * | 2003-07-30 | 2005-02-03 | Leonhard Christoph H. | Multimedia social skills training |
| US20100271655A1 (en) * | 2009-04-23 | 2010-10-28 | Xerox Corporation | Method and system for replenishing consumables in a printing configuration based on confidence intervals |
| WO2016011163A2 (en) * | 2014-07-15 | 2016-01-21 | Macdonald-Korth Emily M | Standard system and method for assigning ratings to art materials and works of art based on the projected stability of the constituents |
| CN106202779A (en) * | 2016-07-20 | 2016-12-07 | 重庆电子工程职业学院 | A kind of method of Rapid Prototype Design |
| CN106182765A (en) * | 2016-07-05 | 2016-12-07 | 上海交通大学 | 3D printer model scale error Forecasting Methodology based on support vector machine |
| CN107591074A (en) * | 2017-09-05 | 2018-01-16 | 杭州市第三人民医院 | A kind of Flexible ureteroscope simulation learning method made based on 3D printing |
| CN108127913A (en) * | 2017-12-22 | 2018-06-08 | 珠海天威飞马打印耗材有限公司 | Intelligent 3D printing system and its Method of printing |
| CN108790177A (en) * | 2018-06-13 | 2018-11-13 | 东莞市原力无限打印科技有限公司 | A kind of test method of test 3D printing object fineness |
-
2019
- 2019-01-29 CN CN201910085587.3A patent/CN109814817A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050026121A1 (en) * | 2003-07-30 | 2005-02-03 | Leonhard Christoph H. | Multimedia social skills training |
| US20100271655A1 (en) * | 2009-04-23 | 2010-10-28 | Xerox Corporation | Method and system for replenishing consumables in a printing configuration based on confidence intervals |
| WO2016011163A2 (en) * | 2014-07-15 | 2016-01-21 | Macdonald-Korth Emily M | Standard system and method for assigning ratings to art materials and works of art based on the projected stability of the constituents |
| CN106182765A (en) * | 2016-07-05 | 2016-12-07 | 上海交通大学 | 3D printer model scale error Forecasting Methodology based on support vector machine |
| CN106202779A (en) * | 2016-07-20 | 2016-12-07 | 重庆电子工程职业学院 | A kind of method of Rapid Prototype Design |
| CN107591074A (en) * | 2017-09-05 | 2018-01-16 | 杭州市第三人民医院 | A kind of Flexible ureteroscope simulation learning method made based on 3D printing |
| CN108127913A (en) * | 2017-12-22 | 2018-06-08 | 珠海天威飞马打印耗材有限公司 | Intelligent 3D printing system and its Method of printing |
| CN108790177A (en) * | 2018-06-13 | 2018-11-13 | 东莞市原力无限打印科技有限公司 | A kind of test method of test 3D printing object fineness |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022161000A1 (en) * | 2021-01-29 | 2022-08-04 | 苏州奇流信息科技有限公司 | Training method and training apparatus for machine learning model, and evaluation system |
| CN117103692A (en) * | 2023-04-13 | 2023-11-24 | 上海轮廓科技有限公司 | 3D printing methods, devices, storage media and computer program products |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7305367B1 (en) | Instantaneous price quotation system for custom manufactured parts | |
| EP2664443B1 (en) | Solid freeform fabrication using a plurality of modeling materials | |
| Beaman et al. | Solid freeform fabrication: a new direction in manufacturing | |
| US7406361B2 (en) | Rapid prototyping method and apparatus using V-CAD data | |
| CN109800533A (en) | A kind of 3D printing process parameter optimizing method based on artificial intelligence technology | |
| JP5408221B2 (en) | Solid object shaping apparatus and control program | |
| US20040006405A1 (en) | Support volume calculation for a CAD model | |
| EP3170649A1 (en) | Three dimensional modeling apparatus, three dimensional modeling method, program, and storage medium | |
| CA2233202A1 (en) | Method and apparatus for data manipulation and system control in a selective deposition modeling system | |
| CN102950770A (en) | Three-dimensional object molding apparatus and control program | |
| CN107530973B (en) | Processing object component data for three-dimensional objects | |
| US20180194070A1 (en) | 3d printing using preformed reuseable support structure | |
| CN109814817A (en) | A kind of 3D printing training data base construction method based on artificial intelligence technology | |
| CN105599300A (en) | Method for making rock bedding structure model based on 3D printing technology | |
| TWI584941B (en) | System of rapid prototyping and method thereof | |
| CN104441641B (en) | The implementation method and device of 3D printing based on photocureable rapid shaping | |
| JP2002292748A (en) | Colored three-dimensional forming system and method, data processing device for colored three-dimensional forming and method, data processing program for colored three-dimensional forming, and recording medium having data processing program recorded thereon | |
| CN115366412B (en) | A composite printing forming method and system for multi-material components | |
| JP2017019111A (en) | Laminate molding device and laminate molding method | |
| CN114556437A (en) | Method of generating a grid of parts, method of using a grid of parts, computer program and computer readable medium | |
| JP2019081935A (en) | Method for manufacturing metallic component | |
| CN112004654B (en) | Packing three-dimensional building bed | |
| Chartier et al. | Handbook of Advanced Ceramics: Chapter 6.5. Rapid Prototyping of Ceramics | |
| CN205009601U (en) | Three-dimensional full-color composite printing device | |
| WO2025029283A1 (en) | Utilizing transferred global context information to predict part deformation during metal sintering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190528 |