CN114589315A - Optimal lapping step matching method for laser additive manufacturing - Google Patents

Optimal lapping step matching method for laser additive manufacturing Download PDF

Info

Publication number
CN114589315A
CN114589315A CN202210160157.5A CN202210160157A CN114589315A CN 114589315 A CN114589315 A CN 114589315A CN 202210160157 A CN202210160157 A CN 202210160157A CN 114589315 A CN114589315 A CN 114589315A
Authority
CN
China
Prior art keywords
lapping
model
optimal
channel
height
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.)
Granted
Application number
CN202210160157.5A
Other languages
Chinese (zh)
Other versions
CN114589315B (en
Inventor
谢吉兵
黄小康
徐露铭
柴泽
陈小奇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Jiaotong University
Original Assignee
Shanghai Jiaotong University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Jiaotong University filed Critical Shanghai Jiaotong University
Priority to CN202210160157.5A priority Critical patent/CN114589315B/en
Publication of CN114589315A publication Critical patent/CN114589315A/en
Application granted granted Critical
Publication of CN114589315B publication Critical patent/CN114589315B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B22—CASTING; POWDER METALLURGY
    • B22F—WORKING METALLIC POWDER; MANUFACTURE OF ARTICLES FROM METALLIC POWDER; MAKING METALLIC POWDER; APPARATUS OR DEVICES SPECIALLY ADAPTED FOR METALLIC POWDER
    • B22F10/00—Additive manufacturing of workpieces or articles from metallic powder
    • B22F10/80—Data acquisition or data processing
    • B22F10/85—Data acquisition or data processing for controlling or regulating additive manufacturing processes
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B33—ADDITIVE MANUFACTURING TECHNOLOGY
    • B33Y—ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3-D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3-D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING
    • B33Y50/00—Data acquisition or data processing for additive manufacturing
    • B33Y50/02—Data acquisition or data processing for additive manufacturing for controlling or regulating additive manufacturing processes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P10/00—Technologies related to metal processing
    • Y02P10/25—Process efficiency

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Materials Engineering (AREA)
  • Artificial Intelligence (AREA)
  • Chemical & Material Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Software Systems (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Manufacturing & Machinery (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Laser Beam Processing (AREA)
  • Finish Polishing, Edge Sharpening, And Grinding By Specific Grinding Devices (AREA)

Abstract

The invention provides an optimal lapping step matching method for laser additive manufacturing, which comprises the following steps of: s1: training to obtain a single-channel laser additive height prediction model, wherein the input of the model is a process parameter, and the output of the model is the height of a single cladding channel; s2: training to obtain a multi-channel lapping average height prediction model, wherein the input of the model is as follows: the process parameter + the initial lapping step amount, and the output of the model is the average height of the cladding layer after the multiple lapping; s3: and establishing an optimal lap joint stepping quantity matching model by using a method based on combination of forward prediction and backward iteration. The invention can quickly and directly match the optimal lapping step amount during horizontal lapping according to basic process parameters (laser power, scanning speed and powder feeding rate), does not need to measure the geometric characteristics of a single cladding channel in advance through experiments and then calculate the proper lapping step amount, or selects the optimal lapping step amount through a large number of experiments, thereby improving the efficiency and reducing the cost while ensuring the forming quality.

Description

Optimal lapping step matching method for laser additive manufacturing
Technical Field
The invention relates to the technical field of 3D laser printing, in particular to a matching method for optimal lapping stepping amount in laser additive manufacturing.
Background
The additive manufacturing using laser as a heat source is an important means for precision manufacturing due to the advantages of small laser spot radius, high energy density and the like. Good forming is the most basic requirement of laser additive precision machining, and low geometric dimension error in the forming process is the basis for obtaining a prospective design model. When multiple cladding tracks are lapped, a strong coupling relation exists between basic process parameters (laser power, scanning speed and powder feeding rate) of laser material increase and horizontal lapping stepping quantity (horizontal distance of a robot moving between adjacent cladding tracks when the robot is lapped horizontally). The basic process parameters determine the formation of the single laser additive, and the horizontal lapping step determines the surface quality after lapping. After the basic technological parameters are determined, if the lapping step is too large, the problems of poor lapping, uneven surface, smaller average formed height than the single-pass height and the like can occur, and further the cladding of the next layer is influenced; if the lapping step is too small, most of the metal of the lapping channel falls on the previous channel, the deposition height is gradually accumulated, the distance between the working plane and the nozzle is smaller and smaller, the optimal working distance is deviated, and the inclined forming is not beneficial to cladding of the next layer. Until there is a lack of a reliable optimal step matching model, the optimal step can only be determined by enumerating different lap steps under the process parameters and then by a large number of preliminary experiments. When different basic process parameters are met next time, corresponding experimental operations are repeated. The mode of determining the optimal lapping stepping amount based on the trial and error method has the advantages of low efficiency, high cost and low intelligent level.
Application No.: the invention patent of CN201710669368.0 discloses an on-line monitoring method for laser additive manufacturing lap-joint rate, which realizes the rapid and reliable monitoring of the laser additive manufacturing lap-joint rate through the calibration, acquisition, preprocessing, correction, calculation and the like of images. However, the method mainly focuses on monitoring the lapping rate in the process, and does not solve the problem of determining the proper lapping step amount according to basic process parameters.
The document "arc additive manufacturing thick-wall structure welding bead spacing calculation strategy" obtains an empirical calculation formula of the lapping step amount through geometric derivation based on an equivalent area method, but the premise of the model is that the geometric characteristics of a single cladding channel need to be known, which means that the geometric characteristics of the single channel still need to be measured through pre-experiments before formal additive manufacturing lapping is carried out, and then the lapping step amount can be calculated, which is not efficient.
Disclosure of Invention
In order to solve the above problems, the present invention aims to provide an optimal lapping step matching method for laser additive manufacturing, which directly matches the optimal lapping step according to basic process parameters, thereby avoiding determining an appropriate lapping step through a large amount of time-consuming preliminary experiments, and improving the efficiency of laser additive manufacturing.
In order to achieve the purpose, the technical scheme of the invention is realized as follows:
a laser additive manufacturing optimal lapping step matching method comprises the following steps:
s1: training to obtain a single-channel laser additive height prediction model, wherein the input of the model is a process parameter, and the output of the model is the height of a single cladding channel;
s2: training to obtain a multi-channel lapping average height prediction model, wherein the input of the model is as follows: the process parameter plus the initial lapping stepping quantity, and the output of the model is the average height of the cladding layer after multiple lapping;
s3: and establishing an optimal lap joint stepping quantity matching model by using a method based on combination of forward prediction and backward iteration.
Further, the process parameters comprise laser power p, scanning speed v and powder feeding speed f.
Further, the S1 specifically includes the following steps:
s11: carrying out single-pass cladding experiments under the condition of a plurality of groups of different process parameter combinations, and then obtaining cladding pass heights corresponding to the process parameters through line structured light measurement;
s12: establishing a data set of the obtained data, and dividing the data set into a training set and a test set according to a ratio of 9: 1;
s13: the XGboost model is trained by using the training set, the accuracy of the model is evaluated by using the test set, and the optimal hyper-parameter combination of the model is obtained by adopting a gridding search mode, so that the single-channel laser additive height prediction model is obtained.
Further, the S2 specifically includes the following steps:
s21: carrying out a plurality of overlapping laser material increase experiments under the combination of a plurality of groups of different basic process parameters and overlapping steps, and then scanning the cladding layer by using line structured light to obtain the cladding layer profile under each process parameter and overlapping step;
s22: sampling the cladding layer according to a certain interval, and then averaging to obtain the average height of the cladding layer under a certain process parameter plus the lapping step;
s23: establishing a data set of the obtained data, and dividing the data set into a training set and a test set according to a ratio of 9: 1;
s24: the XGboost model is trained by using the training set, the accuracy of the model is evaluated by using the test set, the optimal hyper-parameter combination of the model is obtained by adopting a gridding search mode, and finally the multi-channel lapping average height prediction model is obtained.
Further, the S3 specifically includes the following steps:
s31: inputting technological parameters and an initial lapping step d;
s32: calling the single-channel laser additive height prediction model established in S1, inputting process parameters, and outputting to obtain the predicted height H of the first channelf；
S33: then, calling the single-layer multi-channel lapping average height prediction model established in S2, and obtaining the lapping average prediction height H by taking the process parameters and the initial lapping step amount d as inputm；
S34: then calculating the deviation e between the average lapping height and the first deposition height;
s35: judging whether the iteration number n of the current iteration exceeds the maximumLarge number of iterations Nmax(ii) a If N ═ NmaxAnd (5) exiting the loop, outputting the matched lapping step amount d, and otherwise, entering a new iteration.
Further, the S35 specifically includes the following steps:
s351: the iteration number n is n + 1;
s352: judging the current HmAnd HfWhether the absolute value of the difference is smaller than a threshold epsilon or not, if so, exiting the circulation and outputting the matched optimal lapping stepping quantity; otherwise, updating the lapping step;
s353: if H is presentm＜HfIf the step size of the lap joint is too large, the step size should be reduced during updating, namely d- Δ d, otherwise, the step size of the lap joint is too large, and the step size should be increased during updating, namely d-d + Δ d;
s354: inputting the basic process parameters and the updated lapping step quantity into a multi-channel lapping average height prediction model established in S2 to obtain a new average height HmAnd then the next iteration is started.
Has the advantages that: the invention can rapidly and directly match the optimal lapping step amount during horizontal lapping according to basic process parameters (laser power, scanning speed and powder feeding rate), does not need to measure the geometric characteristics of a single cladding channel in advance by experiments and then calculate the proper lapping step amount, or selects the optimal lapping step amount by a large number of experiments, thereby improving the efficiency and reducing the cost while ensuring the forming quality. The matching model of the invention comprises two forward reasoning models based on machine learning, and can be suitable for various different process parameter combinations as long as the coverage of training data is wide, so that the adaptability is good, and the intelligent degree is high. The iteration times, the iteration step length, the threshold value and the like of the method can be adjusted according to actual requirements, and the flexibility is good.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate an embodiment of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:
fig. 1 is a flowchart of an optimal lap step matching method for laser additive manufacturing according to an embodiment of the present invention;
fig. 2 is a schematic horizontal lapping diagram of an optimal lapping step matching method in laser additive manufacturing according to an embodiment of the present invention;
fig. 3 is a forward and reverse iteration flow chart of the laser additive manufacturing optimal lap step matching method according to the embodiment of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings.
Example 1
Referring to FIGS. 1-3: a laser additive manufacturing optimal lapping step matching method comprises the following steps:
s1: training to obtain a single-channel laser additive height prediction model, wherein the input of the model is a process parameter, and the output of the model is the height of a single cladding channel;
s2: training to obtain a multi-channel lapping average height prediction model, wherein the model has the following inputs: the process parameter + the initial lapping step amount, and the output of the model is the average height of the cladding layer after the multiple lapping;
s3: and establishing an optimal lap joint stepping quantity matching model by using a method based on combination of forward prediction and backward iteration.
According to the embodiment, the optimal lapping step amount during horizontal lapping can be rapidly and directly matched according to basic process parameters (laser power, scanning speed and powder feeding rate), the geometric characteristics of a single cladding channel are not required to be measured in advance through experiments, and then the appropriate lapping step amount is calculated, or the optimal lapping step amount is selected through a large number of experiments, so that the efficiency is improved and the cost is reduced while the forming quality is ensured.
Specifically, the process parameters include laser power p, scanning speed v, and powder feeding rate f.
Note that, referring to fig. 2: basic process parameters (laser power p, scanning speed v, powder feeding rate) of the present examplef) After determination, the form size of the first pass can be determined. The ideal lap formation should be the average height H after lapmHeight H near first pass formingf. If the lapping step d is too large, the adjacent cladding channels will not contact or contact too little, and the lapping formation will have gullies, then the average height HmWill be less than the first deposition height Hf(ii) a Otherwise, HmWill be greater than Hf. Therefore, an optimal lap step matching model can be established using this fundamental relationship.
In a specific example, the S1 specifically includes the following steps:
s11: carrying out single-pass cladding experiments under the condition of a plurality of groups of different process parameter combinations, and then obtaining cladding pass heights corresponding to the process parameters through line structured light measurement;
s12: establishing a data set for the obtained data, and dividing the data set into a training set and a test set according to the ratio of 9: 1;
s13: the XGboost model is trained by using the training set, the accuracy of the model is evaluated by using the test set, and the optimal hyper-parameter combination of the model is obtained by adopting a gridding search mode, so that the single-channel laser additive height prediction model is obtained.
In a specific example, the S2 specifically includes the following steps:
s21: carrying out a plurality of overlapping laser material increase experiments under the combination of a plurality of groups of different basic process parameters and overlapping steps, and then scanning the cladding layer by using line structured light to obtain the cladding layer profile under each process parameter and overlapping step;
s22: sampling the cladding layer at a certain interval, and then averaging to obtain the average height of the cladding layer under a certain process parameter plus the lapping stepping amount;
s23: establishing a data set of the obtained data, and dividing the data set into a training set and a test set according to a ratio of 9: 1;
s24: the XGboost model is trained by using the training set, the accuracy of the model is evaluated by using the test set, the optimal hyper-parameter combination of the model is obtained by adopting a gridding search mode, and finally the multi-channel lapping average height prediction model is obtained.
The matching model of the embodiment comprises two forward reasoning models (a single-channel laser additive height prediction model and a multi-channel lapping average height prediction model) based on machine learning, and can be suitable for different process parameter combinations as long as the coverage of training data is wide, so that the adaptability is good, and the intelligent degree is high.
In a specific example, the S3 specifically includes the following steps:
s31: inputting technological parameters and an initial lapping step d;
s32: calling the single-channel laser additive height prediction model established in the S1, inputting process parameters, and outputting to obtain the first-channel predicted height Hf;
s33: then, calling the single-layer multi-channel lapping average height prediction model established in S2, and obtaining the lapping average prediction height H by taking the process parameters and the initial lapping step amount d as inputm；
S34: then calculating the deviation e between the average lapping height and the first deposition height;
s35: judging whether the iteration number N of the current iteration exceeds the maximum iteration number N or notmax(ii) a If N ═ NmaxAnd (5) exiting the loop, outputting the matched lapping step amount d, and otherwise, entering a new iteration.
In this embodiment, the overlap step amount d is continuously updated by the above method of forward inference and reverse iteration, so as to gradually reduce HmAnd HfUntil the absolute value of the deviation e is smaller than the set threshold value, and then the optimal horizontal lapping stepping amount d is output.
In a specific example, the S35 specifically includes the following steps:
s351: the iteration number n is n + 1;
s352: judging the current HmAnd HfWhether the absolute value of the difference is smaller than a threshold epsilon or not, if so, exiting the circulation and outputting the matched optimal lapping stepping quantity; otherwise, updating the lapping step;
s353: if H is presentm＜HfThe overlap step is too large at this time, and should be updatedIf the step size is too large, the step size should be increased, i.e. d + Δ d;
s354: inputting the basic process parameters and the updated lapping step amount into the multi-channel lapping average height prediction model established in the step S2 to obtain a new average height Hm, and then starting the next iteration.
It should be noted that, in the specific implementation, the user may customize the threshold epsilon and the number of iterations N as neededmaxAnd the step size Δ d of each update of the lap step amount d. When the method is used, only basic process parameters (laser power, scanning speed and powder feeding rate) are required to be input into the established optimal lapping step matching model, and then the model automatically calculates to obtain the optimal lapping step.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and should not be taken as limiting the scope of the present invention, which is intended to cover any modifications, equivalents, improvements, etc. within the spirit and scope of the present invention.

Claims (6)

1. The optimal lapping step matching method for laser additive manufacturing is characterized by comprising the following steps of:
s1: training to obtain a single-channel laser additive height prediction model, wherein the input of the model is a process parameter, and the output of the model is the height of a single cladding channel;
s2: training to obtain a multi-channel lapping average height prediction model, wherein the input of the model is as follows: the process parameter + the initial lapping step amount, and the output of the model is the average height of the cladding layer after the multiple lapping;
s3: and establishing an optimal lap joint stepping quantity matching model by using a method based on combination of forward prediction and backward iteration.
2. The laser additive manufacturing optimal lap step matching method according to claim 1, wherein said process parameters include laser power p, scanning speed v, powder feeding rate f.
3. The laser additive manufacturing optimal lap step matching method according to claim 1, wherein the S1 specifically includes the steps of:
s11: carrying out single-pass cladding experiments under the condition of a plurality of groups of different process parameter combinations, and then obtaining cladding pass heights corresponding to the process parameters through line structured light measurement;
s12: establishing a data set of the obtained data, and dividing the data set into a training set and a test set according to a ratio of 9: 1;
s13: the XGboost model is trained by using the training set, the accuracy of the model is evaluated by using the test set, and the optimal hyper-parameter combination of the model is obtained by adopting a gridding search mode, so that the single-channel laser additive height prediction model is obtained.
4. The laser additive manufacturing optimal lap step matching method according to claim 1, wherein the S2 specifically includes the steps of:
s21: carrying out a plurality of overlapping laser material increase experiments under the combination of a plurality of groups of different basic process parameters and overlapping steps, and then scanning the cladding layer by using line structured light to obtain the cladding layer profile under each process parameter and overlapping step;
s22: sampling the cladding layer according to a certain interval, and then averaging to obtain the average height of the cladding layer under a certain process parameter plus the lapping step;
s23: establishing a data set of the obtained data, and dividing the data set into a training set and a test set according to a ratio of 9: 1;
s24: the XGboost model is trained by using the training set, the accuracy of the model is evaluated by using the test set, the optimal hyper-parameter combination of the model is obtained by adopting a gridding search mode, and finally the multi-channel lapping average height prediction model is obtained.
5. The laser additive manufacturing optimal lap step matching method according to claim 1, wherein the S3 specifically includes the steps of:
s31: inputting technological parameters and an initial lapping step d;
s32: call middle building of S1Inputting technological parameters and outputting to obtain the predicted height H of the first laser additivef；
S33: then, calling the single-layer multi-channel lapping average height prediction model established in S2, and obtaining the lapping average prediction height H by taking the process parameters and the initial lapping step amount d as inputm；
S34: then calculating the deviation e between the average lapping height and the first deposition height;
s35: judging whether the iteration number N of the current iteration exceeds the maximum iteration number N or notmax(ii) a If N ═ NmaxAnd (5) exiting the loop, outputting the matched lapping step amount d, and otherwise, entering a new iteration.
6. The laser additive manufacturing optimal lap step matching method according to claim 5, wherein the S35 specifically includes the steps of:
s351: the iteration number n is n + 1;
s352: judging the current HmAnd HfWhether the absolute value of the difference is smaller than a threshold epsilon or not, if so, exiting the circulation and outputting the matched optimal lapping stepping quantity; otherwise, updating the lapping step;
s353: if H is presentm＜HfIf the step size of the lap joint is too large, the step size of the lap joint is reduced during updating, namely d- Δ d, otherwise, the step size of the lap joint is too large, and the step size of the lap joint is increased during updating, namely d + Δ d;
s354: inputting the basic process parameters and the updated lapping step quantity into a multi-channel lapping average height prediction model established in S2 to obtain a new average height HmAnd then the next iteration is started.
CN202210160157.5A 2022-02-22 2022-02-22 Optimal lapping step matching method for laser additive manufacturing Active CN114589315B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210160157.5A CN114589315B (en) 2022-02-22 2022-02-22 Optimal lapping step matching method for laser additive manufacturing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210160157.5A CN114589315B (en) 2022-02-22 2022-02-22 Optimal lapping step matching method for laser additive manufacturing

Publications (2)

Publication Number Publication Date
CN114589315A true CN114589315A (en) 2022-06-07
CN114589315B CN114589315B (en) 2022-12-16

Family

ID=81806371

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210160157.5A Active CN114589315B (en) 2022-02-22 2022-02-22 Optimal lapping step matching method for laser additive manufacturing

Country Status (1)

Country Link
CN (1) CN114589315B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115859746A (en) * 2023-02-13 2023-03-28 无锡祝融航空航天科技有限公司 Copper material additive manufacturing forming precision control method based on deep learning

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160236416A1 (en) * 2015-02-16 2016-08-18 Arevo Inc. Method and a system to optimize printing parameters in additive manufacturing process
CN108559995A (en) * 2018-02-28 2018-09-21 东北大学 A kind of method of laser cladding technological parameter optimization in plane
CN110110862A (en) * 2019-05-10 2019-08-09 电子科技大学 A kind of hyperparameter optimization method based on adaptability model
US20200247063A1 (en) * 2018-04-02 2020-08-06 Nanotronics Imaging, Inc. Systems, methods, and media for artificial intelligence process control in additive manufacturing
CN111688192A (en) * 2020-06-24 2020-09-22 西安文理学院 Selective laser melting main process parameter matching optimization method
CN112214864A (en) * 2020-08-04 2021-01-12 沈阳工业大学 Method for predicting size of multi-channel multi-layer laser cladding layer
CN113379536A (en) * 2021-06-29 2021-09-10 百维金科(上海)信息科技有限公司 Default probability prediction method for optimizing recurrent neural network based on gravity search algorithm
US20210405613A1 (en) * 2020-06-30 2021-12-30 Atos Spain Sociedad Anonima Predicting system in additive manufacturing process by machine learning algorithms
US20220042924A1 (en) * 2020-08-07 2022-02-10 Sigma Labs, Inc. Defect identification using machine learning in an additive manufacturing system

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160236416A1 (en) * 2015-02-16 2016-08-18 Arevo Inc. Method and a system to optimize printing parameters in additive manufacturing process
CN108559995A (en) * 2018-02-28 2018-09-21 东北大学 A kind of method of laser cladding technological parameter optimization in plane
US20200247063A1 (en) * 2018-04-02 2020-08-06 Nanotronics Imaging, Inc. Systems, methods, and media for artificial intelligence process control in additive manufacturing
CN110110862A (en) * 2019-05-10 2019-08-09 电子科技大学 A kind of hyperparameter optimization method based on adaptability model
CN111688192A (en) * 2020-06-24 2020-09-22 西安文理学院 Selective laser melting main process parameter matching optimization method
US20210405613A1 (en) * 2020-06-30 2021-12-30 Atos Spain Sociedad Anonima Predicting system in additive manufacturing process by machine learning algorithms
CN112214864A (en) * 2020-08-04 2021-01-12 沈阳工业大学 Method for predicting size of multi-channel multi-layer laser cladding layer
US20220042924A1 (en) * 2020-08-07 2022-02-10 Sigma Labs, Inc. Defect identification using machine learning in an additive manufacturing system
CN113379536A (en) * 2021-06-29 2021-09-10 百维金科(上海)信息科技有限公司 Default probability prediction method for optimizing recurrent neural network based on gravity search algorithm

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
叶宏等: "H13钢激光熔覆Co基涂层组织及热疲劳性能", 《强激光与粒子束》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115859746A (en) * 2023-02-13 2023-03-28 无锡祝融航空航天科技有限公司 Copper material additive manufacturing forming precision control method based on deep learning
CN115859746B (en) * 2023-02-13 2023-04-25 无锡祝融航空航天科技有限公司 Deep learning-based copper material additive manufacturing forming precision control method

Also Published As

Publication number Publication date
CN114589315B (en) 2022-12-16

Similar Documents

Publication Publication Date Title
CN114589315B (en) Optimal lapping step matching method for laser additive manufacturing
WO2021129671A1 (en) Weld bead modeling method, device and system for wire-arc additive manufacture
JP3258331B2 (en) Method and apparatus for treating any 3D shaped surface with a laser, especially for polishing (polishing) and polishing (texture) a workpiece and for treating the sealing surface of a die
US5107093A (en) Method and apparatus for automatic multi-run welding
CN106513676B (en) A kind of hot spot cooperates with controllable laser metal increasing material manufacturing method with amyloid plaque automatically
CN105345237A (en) Device and process method for automatically controlling welding seam shape in longitudinal submerged arc welding
JP2002531275A (en) Method and system for determining weld bead quality
CN105759725A (en) Speed-sensitive section constant-speed curve interpolation speed planning method
JP2002532253A (en) Method and system for controlling topographical features of a weld to improve fatigue performance of a manufactured structure
CN113290302A (en) Quantitative prediction method for surplus height of electric arc welding additive manufacturing
CN110647107B (en) Porous free-form surface continuous scanning measurement trajectory planning method and system
CN109878075B (en) Method for scanning and processing by adopting continuously variable light spots in 3D printing
CN111932539A (en) Molten pool image and depth residual error network-based height and penetration collaborative prediction method
CN114564880B (en) Method for constructing digital twin module in additive manufacturing process
CN114769988A (en) Welding control method and system, welding equipment and storage medium
CN110126266A (en) A kind of three-dimension object manufacturing method
CN105809736A (en) Three-dimensional reconstruction method and device of pipeline
CN103212849B (en) The method of cut SMT screen plate
CN110977174B (en) Pulse laser high-speed same-point interval multiple processing system and processing method
CN1174246A (en) Method and apparatus for controlling galvanization coating weight
CN1802677A (en) Distributed signal control system
CN116664508A (en) Weld surface quality detection method and computer readable storage medium
JP7382912B2 (en) Additive manufacturing equipment and additive manufacturing method
CN115319287A (en) Lap joint laser scanning welding method based on linear energy density regulation
CN115255074A (en) Molding control method and system for nuclear-grade alloy steel elbow

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
GR01 Patent grant
GR01 Patent grant