CN109909502B - Online monitoring method for laser additive manufacturing process based on multi-source heterogeneous data - Google Patents
Online monitoring method for laser additive manufacturing process based on multi-source heterogeneous data Download PDFInfo
- Publication number
- CN109909502B CN109909502B CN201910204707.7A CN201910204707A CN109909502B CN 109909502 B CN109909502 B CN 109909502B CN 201910204707 A CN201910204707 A CN 201910204707A CN 109909502 B CN109909502 B CN 109909502B
- Authority
- CN
- China
- Prior art keywords
- formula
- laser
- metal powder
- source heterogeneous
- heterogeneous data
- 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.)
- Active
Links
Classifications
-
- 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
- Powder Metallurgy (AREA)
Abstract
The invention discloses an online monitoring method of a laser additive manufacturing process based on multi-source heterogeneous data, which comprises the following steps: 1 establishing laser energy distribution; 2, obtaining the vibration speed of the medium; 3 obtaining the maximum surface temperature of the powder bed; 4 obtaining the width of a molten pool; 5, constructing a thermodynamic model of multi-field coupling in the laser melting process; 6, sparse coding of multi-source heterogeneous data; and 7, obtaining an online monitoring model. The invention can carry out real-time monitoring and process control in the laser additive manufacturing process, thereby automatically adjusting process parameters to eliminate defects when tiny defects appear, further improving the forming quality of parts and meeting the actual requirements of precision and reliability.
Description
Technical Field
The invention belongs to the technical field of sensors and monitoring, and particularly relates to a big data driven laser additive manufacturing online monitoring method.
Background
Additive manufacturing (3D printing/rapid prototyping) technology is gradually changing traditional lifestyle and production modes of people, and is considered to promote a third industrial revolution by the characteristics of rapidity, customization, digitization, networking and the like. The laser additive manufacturing technology is one of metal additive manufacturing technologies with the widest application range and the greatest practical significance at present, the technology utilizes a high-energy laser beam, melts or sinters powder according to a preset path, and then forms the powder after cooling and solidification, and the technology has obvious advantages in the manufacturing of parts which are difficult to process, such as high-performance complex components in the aerospace field, porous complex structure manufacturing in the biological manufacturing field, functional gradient material manufacturing and the like.
However, the laser additive manufacturing process is accompanied by complicated physical and chemical processes, the formed part has large temperature gradient and thermal stress, defects such as spheroidization, pores and cracks are easily generated in the forming process, the precision and the reliability of the formed part are influenced, the formed part is easy to deform and crack, the process control of the forming quality is lacked, and the application of the laser additive manufacturing technology is seriously hindered.
However, due to the complex dynamic characteristics of the laser additive manufacturing process, the high difficulty in mounting the sensor, the insufficient understanding of the related physical process and the like, the real-time monitoring and the process control of the laser additive manufacturing process are not easy to be performed, the monitoring model is limited, and most of the laser additive manufacturing process utilizes infrared temperature information and processing quality to perform modeling. However, the temperature online measurement error is large, the generation mechanism of defects in the powder melting or sintering process is complex, the reliability of quality measurement only by using temperature information is insufficient, and related research is still to be further promoted.
Although post-treatments such as sand blasting and heat treatment can achieve better surface finish and reduce fracture and delamination, the post-treatments can cause the size of the parts to change, and the post-treatment method is not suitable for parts with complex internal structures, key functional parts, precision parts and the like. Therefore, it is more critical to realize online monitoring and improve the forming quality in the laser additive manufacturing process.
Disclosure of Invention
Aiming at the defects and/or the improvement requirements in the prior art, the invention provides the on-line monitoring method for the laser additive manufacturing process based on the multi-source heterogeneous data, so that the real-time monitoring and the process control can be performed in the laser additive manufacturing process, the process parameters are automatically adjusted to eliminate the defects when the tiny defects occur, the part forming quality is improved, and the actual requirements on precision and reliability are met.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention relates to an online monitoring method of a laser additive manufacturing process based on multi-source heterogeneous data, which is characterized by comprising the following steps:
step 1, establishing laser energy distribution I (r) based on Gaussian distribution by using formula (1):
in the formula (1), r is the distance from a point on a laser spot to the center of the laser spot, omega is the radius of the laser spot, and phi is the central energy density of the laser spot;
step 2, obtaining the speed u of medium vibration by using the formula (2):
in the formula (2), alpha is the absorptivity of the metal powder material to laser; gamma is the adiabatic index of the gas, rho0Is the density of the metal powder material;
step 3, obtaining the maximum surface temperature T of the powder bed by using the formula (3)m：
In the formula (3), K is Boltzmann constant; k is the thermal conductivity of the metal powder material; v is the melting speed of the metal powder material; c is the hot melt ratio of the metal powder material;
step 4, obtaining the width delta of the molten pool by using the formula (4)w：
In the formula (4), Δ HlvLatent heat energy of the metal powder material from solid state to gas state; t islvIs the equilibrium temperature of the metal powder material from solid to gas; t isslIs the equilibrium temperature of the metal powder material from solid to liquid;
step 5, constructing a thermodynamic model of multi-field coupling in the laser melting process:
step 5.1, establishing a rectangular coordinate system O-XY by taking one vertex angle of the powder bed as an origin O and taking two edges connected with the origin O as an X axis and a Y axis respectively;
step 5.2, establishing a temperature field finite element model in the laser melting process by using the formula (5):
in formula (5), a is the thermal diffusivity of the metal powder material, T (t) is the instantaneous temperature ^ of the metal powder material at time t2Is Laplace operator; for a stable temperature field that does not vary with time, there is +2T (x, y, z) ═ 0; t (x, y, z) is the temperature of the metal powder material at point (x, y, z);
step 5.3, obtaining the temperature gradient along the normal direction n in the laser irradiation area by using the formula (6)
Step 6, sparse coding of multi-source heterogeneous data:
step 6.1, respectively obtaining multi-source heterogeneous data formed by temperature, sound and image signals through a sensor group, and recording the multi-source heterogeneous data as f;
step 6.2, carrying out fast Fourier transform on the multi-source heterogeneous data f to obtain transformed multi-source heterogeneous data fT；
Step 6.3, utilizing a redundant dictionary phi to carry out conversion on the multi-source heterogeneous data fTReconstructing to obtain transformed multi-source heterogeneous data f shown in formula (8)TAnd by the optimal l shown by the formula (7)0Norm constrains sparse factor f'TThe number of (2):
s.t.fT＝Φf′T,fT∈Rk (8)
in the formula (8), RkA real number field representing the k dimension; phi f'TRepresents a signal atom set obtained after sparse coding, and comprises:
representing the L-th signal atom obtained after sparse coding; l represents the number of signal atoms;
step 7, signal atom set obtained after sparse coding
Carrying out feature fusion to obtain an observation sequence V ═ V1,v2,...,vi,...,vn}，viThe observation data of the ith visual layer; 1,2, …, n; n representsDepending on the number of layers;
step 8, obtaining an online monitoring model E (V, H | theta) by using the formula (8), and realizing online monitoring of the laser additive manufacturing process by using the online monitoring model E (V, H | theta):
in formula (8), θ ═ wij,ai,bjIs a parameter on the RBM neural network model, where wijRepresents the connection weight of the ith visible layer and the jth hidden layer, aiRepresenting the bias of the ith visible layer, bjDenotes the bias of the jth hidden layer, hjRepresents the jth hidden layer; h denotes a set of hidden layers, and m denotes the number of hidden layers.
Compared with the prior art, the invention has the beneficial effects that:
1. the method provided by the invention has the advantages that the characteristics of sound, temperature and image signals in different processing states in the laser additive manufacturing process are explored by using methods such as signal processing, image analysis and the like, the relation between the characteristics and defects in the processing process is analyzed, and the physical mechanism of the laser additive manufacturing process is disclosed; establishing a new absorption model of the metal powder to the laser light source, performing thermodynamic simulation on the laser melting process, analyzing the influence rule of the processing parameters on the processing temperature, and providing a research basis for realizing the feedback control of the process; and the accuracy of online quality monitoring in the processing process is improved through multi-source signal characteristic fusion processing. The on-line monitoring of the laser additive manufacturing forming process is one of core technologies for researching and developing high-precision high-quality manufacturing equipment of metal parts with complex structures, so that the method is particularly suitable for application occasions of the metal additive manufacturing technology in the fields of aerospace, medical treatment, material manufacturing and the like.
2. The invention researches the fusion of multi-source sensing data and the consistency expression of multi-dimensional heterogeneous information in monitoring based on the analysis of the physical process of metal additive manufacturing forming and by combining the multi-modal characteristics of the monitoring information, sparsely encodes heterogeneous sensing information such as temperature, sound and images by using the mathematical theory of sparse decomposition, simultaneously realizes the compression, denoising and characteristic expression of consistency of the information, excavates the inherent relevance of the multi-source heterogeneous data, and solves the basic premise of deep learning and fusion of the multi-source heterogeneous information.
Detailed Description
In this embodiment, an online monitoring method for a laser additive manufacturing process based on multi-source heterogeneous data is performed according to the following steps:
in the selective laser melting process, the laser interacts with the metal powder to produce a process of melting, vaporization and reconstitution. There are a variety of signal sources in this laser ablation process, including acoustic, plasma, ultrasonic, infrared radiation, and electrical signals. The formation and the change of the signals are closely related to the formation and the change of a molten pool, and the research on the sound, the temperature, the image formation and the change mechanism in the laser melting process has important significance for disclosing the thermodynamic law of the laser melting process.
The sound, temperature, image formation and change mechanism in the laser melting process are researched, and the laser energy distribution when the laser adopts Gaussian distribution is established, so that the basis for disclosing the thermodynamic rule of the laser melting process is established.
Step 1, establishing laser energy distribution I (r) based on Gaussian distribution by using formula (1):
in the formula (1), r is the distance from a point on a laser spot to the center of the laser spot, omega is the radius of the laser spot, and phi is the central energy density of the laser spot;
step 2, obtaining the speed u of medium vibration by using the formula (2):
in the formula (2), alpha is the absorptivity of the metal powder material to laser; gamma is the adiabatic index of the gas，ρ0Is the density of the metal powder material;
since the absorption of a high power laser beam can rapidly generate a thermal effect, the surface temperature rapidly rises to the melting temperature. Maximum surface temperature TmVaries with the thermal conductivity k, the melting speed v, and the thermal melting ratio c of the material, so that:
step 3, obtaining the maximum surface temperature T of the powder bed by using the formula (3)m：
In the formula (3), K is Boltzmann constant; k is the thermal conductivity of the metal powder material; v is the melting speed of the metal powder material; c is the hot melt ratio of the metal powder material;
step 4, obtaining the width delta of the molten pool by using the formula (4)w：
In the formula (4), Δ HlvLatent heat energy of the metal powder material from solid state to gas state; t islvIs the equilibrium temperature of the metal powder material from solid to gas; t isslIs the equilibrium temperature of the metal powder material from solid to liquid;
the changes in sound, temperature, image are directly related to the width and maximum surface temperature of the melt pool, and the plasma species splattering, which in turn are related to the degree of spheroidization, porosity, residual stress of the selective laser melted part. Through the research on the generation and change mechanisms of sound, temperature and image signals of selective laser melting, key factors influencing the quality of selective laser melting parts are found out, key signals related to the quality are collected, and a reliable theoretical basis is provided for online monitoring of the quality of the parts by using multiple sensors.
Step 5, constructing a thermodynamic model of multi-field coupling in the laser melting process:
a thermodynamics and dynamics analysis model based on laser solidification multi-field coupling is established, various phenomena in the powder evolution process, such as hole formation, molten pool flowing and the like, are disclosed, a new thought is provided, a numerical model based on the physical process and a monitoring system based on multi-source data are mutually verified, and more accurate and reliable online monitoring is achieved. The laser scans over the bed along a fractal path, and the heat transfer process, including radiation, convection between the powder layer and the environment and heat conduction between the powder layer and the submount, latent heat of fusion, is very large. Both the change in powder state and the corresponding change in thermal properties complicate the heat transfer process.
Step 5.1, establishing a rectangular coordinate system O-XY by taking one vertex angle of the powder bed as an origin O and taking two edges connected with the origin O as an X axis and a Y axis respectively;
step 5.2, establishing a temperature field finite element model in the laser melting process by using the formula (5):
in the formula (5), a is the thermal diffusivity of the metal powder material, T (t) is the instantaneous temperature of the metal powder material at time t,
is Laplace operator; for a stable temperature field that does not vary with time, there is
T (x, y, z) is the temperature of the metal powder material at point (x, y, z);
step 5.3, obtaining the temperature gradient along the normal direction n in the laser irradiation area by using the formula (6)
According to the change of the enthalpy value and the boundary condition under the actual condition, the influence of the change of the temperature field of the molten pool on the uniformity and the thermal stress of the laser melting part is researched. Therefore, the reliability of monitoring the laser melting process can be enhanced by mutually verifying the thermodynamic modeling of the laser melting process and the monitored sound, temperature and image signals, and theoretical basis is provided for realizing feedback control and melting compensation of the selective laser melting process in the future.
Step 6, sparse coding of multi-source heterogeneous data:
the research of sparse coding obtains the consistency representation of multi-source heterogeneous monitoring information, meanwhile, data compression is achieved, and the learning speed and the recognition accuracy of a deep network are improved. According to the traditional sampling theorem, high-frequency characteristic information common in monitoring, such as acoustic emission signals, ultrasonic signals and other high-harmonic components, needs to be acquired at the sampling frequency of several megahertz, which brings difficulty to the real-time performance of the monitoring system. In addition, the high-dimensional data such as temperature images are large in calculation amount and slow in speed, and are not beneficial to training of the deep DBN model.
According to a compression sensing theory proposed by Candes and Donoho in 2006, signal compression and sampling are combined, and sparse representation (measured value) of a signal is obtained through learning of a redundant dictionary, so that the measured data volume of the signal is far smaller than that required by a traditional sampling method, and the real-time processing speed is improved. Since the monitoring signal is almost non-sparse in the time domain, the theory described above cannot be directly applied to compressed sampling of the monitoring signal. But we can reconstruct the sparse signal by sparse representation through fast fourier transform and then solve an optimal norm problem through an over-fitting equation (redundant dictionary).
Step 6.1, respectively obtaining multi-source heterogeneous data formed by temperature, sound and image signals through a sensor group, and recording the multi-source heterogeneous data as f;
step 6.2, carrying out fast Fourier transform on the multi-source heterogeneous data f to obtain transformed multi-source heterogeneous data fT；
Step 6.3, utilizing redundant dictionary phi to change oppositelyTransformed multi-source heterogeneous data fTReconstructing to obtain transformed multi-source heterogeneous data f shown in formula (8)TAnd by the optimal l shown by the formula (7)0Norm constrains sparse factor f'TThe number of (2):
s.t.fT＝Φf′T,fT∈Rk (8)
in the formula (8), RkA real number field representing the k dimension; phi f'TRepresents a signal atom set obtained after sparse coding, and comprises:
representing the L-th signal atom obtained after sparse coding; l represents the number of signal atoms; the higher the sparsity of the signal, the less measurement data will be needed to recover the original signal using the above equation. After the multi-source heterogeneous sensing information passes through the redundant dictionary, sparse representation of consistency is obtained. In addition, for a monitoring signal with low signal-to-noise ratio, a dictionary learning method can be used based on sparse representation, atoms of pure noise and a signal containing noise are considered to be learned respectively, and the noise is separated by utilizing the correlation of the atoms.
The information collected by multiple sensors for selective laser melting processes is complex and diverse. The selective laser melting quality of the part has a direct relationship with the state of the melting process, and corresponding state changes such as spheroidization, porosity and cracks are contained in temperature, sound and image information. By utilizing the latest development result of artificial intelligence, a deep belief neural network (DBN) with associative memory and self-adaptive processing capability is established, so that comprehensive and complete information can be obtained, and the quality of the part subjected to selective laser melting can be rapidly and accurately monitored in an online state.
Step 7, after sparse codingThe resulting set of signal atoms
Carrying out feature fusion to obtain an observation sequence V ═ V1,v2,...,vi,...,vn}，viThe observation data of the ith visual layer; 1,2, …, n; n represents the number of visible layers;
step 8, obtaining an online monitoring model E (V, H | theta) by using the formula (8), and realizing online monitoring of the laser additive manufacturing process by using the online monitoring model E (V, H | theta):
in formula (8), θ ═ wij,ai,bjIs a parameter on the RBM neural network model, where wijRepresents the connection weight of the ith visible layer and the jth hidden layer, aiRepresenting the bias of the ith visible layer, bjDenotes the bias of the jth hidden layer, hjRepresents the jth hidden layer; h denotes a set of hidden layers, and m denotes the number of hidden layers.
Claims (1)
1. An on-line monitoring method of a laser additive manufacturing process based on multi-source heterogeneous data is characterized by comprising the following steps:
step 1, establishing laser energy distribution I (r) based on Gaussian distribution by using formula (1):
in the formula (1), r is the distance from a point on a laser spot to the center of the laser spot, omega is the radius of the laser spot, and phi is the central energy density of the laser spot;
step 2, obtaining the speed u of medium vibration by using the formula (2):
in the formula (2), alpha is the absorptivity of the metal powder material to laser; gamma is the adiabatic index of the gas, rho0Is the density of the metal powder material;
step 3, obtaining the maximum surface temperature T of the powder bed by using the formula (3)m：
In the formula (3), K is Boltzmann constant; k is the thermal conductivity of the metal powder material; v is the melting speed of the metal powder material; c is the hot melt ratio of the metal powder material;
step 4, obtaining the width delta of the molten pool by using the formula (4)w：
In the formula (4), Δ HlvLatent heat energy of the metal powder material from solid state to gas state; t islvIs the equilibrium temperature of the metal powder material from solid to gas; t isslIs the equilibrium temperature of the metal powder material from solid to liquid;
step 5, constructing a thermodynamic model of multi-field coupling in the laser melting process:
step 5.1, establishing a rectangular coordinate system O-XY by taking one vertex angle of the powder bed as an origin O and taking two edges connected with the origin O as an X axis and a Y axis respectively;
step 5.2, establishing a temperature field finite element model in the laser melting process by using the formula (5):
in the formula (5), a is the thermal diffusivity of the metal powder material, and T (t) is the metal powder at time tThe instantaneous temperature of the material(s) is,
is Laplace operator; for a stable temperature field that does not vary with time, there is
T (x, y, z) is the temperature of the metal powder material at point (x, y, z);
step 5.3, obtaining the temperature gradient along the normal direction n in the laser irradiation area by using the formula (6)
Step 6, sparse coding of multi-source heterogeneous data:
step 6.1, respectively obtaining multi-source heterogeneous data formed by temperature, sound and image signals through a sensor group, and recording the multi-source heterogeneous data as f;
step 6.2, carrying out fast Fourier transform on the multi-source heterogeneous data f to obtain transformed multi-source heterogeneous data fT；
Step 6.3, utilizing a redundant dictionary phi to carry out conversion on the multi-source heterogeneous data fTReconstructing to obtain transformed multi-source heterogeneous data f shown in formula (8)TAnd by the optimal l shown by the formula (7)0Norm constrains sparse factor f'TThe number of (2):
s.t.fT＝Φf′T,fT∈Rk (8)
in the formula (8), RkA real number field representing the k dimension; phi f'TAfter representing sparse codingThe resulting set of signal atoms incorporates:
representing the L-th signal atom obtained after sparse coding; l represents the number of signal atoms;
step 7, signal atom set obtained after sparse coding
Carrying out feature fusion to obtain an observation sequence V ═ V1,v2,...,vi,...,vn}，viThe observation data of the ith visual layer; 1,2, …, n; n represents the number of visible layers;
step 8, obtaining an online monitoring model E (V, H | theta) by using the formula (9), and realizing online monitoring of the laser additive manufacturing process by using the online monitoring model E (V, H | theta):
in formula (9), θ ═ wij,ai,bjIs a parameter on the RBM neural network model, where wijRepresents the connection weight of the ith visible layer and the jth hidden layer, aiRepresenting the bias of the ith visible layer, bjDenotes the bias of the jth hidden layer, hjRepresents the jth hidden layer; h denotes a set of hidden layers, and m denotes the number of hidden layers.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910204707.7A CN109909502B (en) | 2019-03-18 | 2019-03-18 | Online monitoring method for laser additive manufacturing process based on multi-source heterogeneous data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910204707.7A CN109909502B (en) | 2019-03-18 | 2019-03-18 | Online monitoring method for laser additive manufacturing process based on multi-source heterogeneous data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109909502A CN109909502A (en) | 2019-06-21 |
| CN109909502B true CN109909502B (en) | 2021-04-27 |
Family
ID=66965365
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910204707.7A Active CN109909502B (en) | 2019-03-18 | 2019-03-18 | Online monitoring method for laser additive manufacturing process based on multi-source heterogeneous data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109909502B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110490866B (en) * | 2019-08-22 | 2020-05-15 | 四川大学 | Metal additive forming size real-time prediction method based on depth feature fusion |
| CN110605393B (en) * | 2019-09-25 | 2021-06-08 | 中国兵器装备集团自动化研究所 | Laser three-dimensional forming process detection method and system and application |
| CN110942150A (en) * | 2019-11-09 | 2020-03-31 | 上海航天设备制造总厂有限公司 | Intelligent diagnosis method for forming failure of parts of selective laser melting equipment |
| CN111283192B (en) * | 2020-01-22 | 2021-08-31 | 南京理工大学 | Laser powder bed melting additive manufacturing molten pool monitoring and pore control method |
| CN111504980B (en) * | 2020-04-26 | 2021-04-16 | 长春工业大学 | LIBS online monitoring device and method for metal additive manufacturing process |
| CN112122741A (en) * | 2020-08-14 | 2020-12-25 | 江苏理工学院 | Welding bead forming control system in electric arc additive manufacturing process and parameter optimization method thereof |
| CN113172242B (en) * | 2021-04-27 | 2021-12-28 | 武汉科技大学 | Selective laser melting forming real-time monitoring device and implementation method |
| SE2250597A1 (en) * | 2022-05-19 | 2023-11-20 | Cellink Bioprinting Ab | Multi-sensor evaluation of a printing process |
| CN115115578B (en) * | 2022-05-23 | 2023-02-10 | 厦门宇电自动化科技有限公司 | Defect detection method and system in additive manufacturing process |
| CN115138867B (en) * | 2022-07-22 | 2024-04-26 | 南京航空航天大学 | Device and method for real-time monitoring feedback and optimization of molding quality of gradient material in laser additive manufacturing |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202576563U (en) * | 2012-05-30 | 2012-12-05 | 上海海事大学 | Laser cladding real-time monitoring system based on zigbee wireless sensing technology |
| CN105855743A (en) * | 2016-05-04 | 2016-08-17 | 华中科技大学 | On-line monitoring system and method for weld pool dynamic process |
| CN107328391A (en) * | 2017-06-28 | 2017-11-07 | 武汉理工大学 | A kind of multi-sensor cooperation monitoring device and method for broadband laser cladding |
| CN107992904A (en) * | 2017-12-22 | 2018-05-04 | 重庆邮电大学 | Forest Ecology man-machine interaction method based on Multi-source Information Fusion |
| CN108038266A (en) * | 2017-11-17 | 2018-05-15 | 西安铂力特增材技术股份有限公司 | A kind of method for numerical simulation of selective laser repair process |
| CN108062432A (en) * | 2017-11-17 | 2018-05-22 | 西安铂力特增材技术股份有限公司 | A kind of method for numerical simulation of selective laser fusion process |
| WO2018183003A1 (en) * | 2017-03-31 | 2018-10-04 | Konica Minolta Laboratory U.S.A., Inc. | 3d imaging by multiple sensors during 3d printing |
| CN109034391A (en) * | 2018-08-17 | 2018-12-18 | 王玲 | The multi-source heterogeneous information RBM network integration framework and fusion method of automatic Pilot |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102519387B (en) * | 2011-10-27 | 2014-03-05 | 哈尔滨工业大学 | Visual inspection method of electron beam welding pool shape parameter |
| CN103506756B (en) * | 2013-09-11 | 2015-07-08 | 上海交通大学 | Laser lap welding gap detecting system and laser lap welding gap detecting method based on molten pool image visual sensing |
| US11117194B2 (en) * | 2017-03-15 | 2021-09-14 | Applied Materials, Inc. | Additive manufacturing having energy beam and lamp array |
| JP2018184623A (en) * | 2017-04-24 | 2018-11-22 | 株式会社東芝 | Inherent strain calculation device for laminated molding, inherent strain calculation method therefor, inherent strain calculation program therefor, analysis device therefor, analysis method therefor, and lamination molding device |
| US20190004497A1 (en) * | 2017-06-30 | 2019-01-03 | Norsk Titanium As | Technologies of controlling additive manufacturing systems |
| CN109447939B (en) * | 2018-07-10 | 2022-01-28 | 南京理工大学 | Weld width prediction method before molten pool forming based on mixed spectrum vision |
-
2019
- 2019-03-18 CN CN201910204707.7A patent/CN109909502B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202576563U (en) * | 2012-05-30 | 2012-12-05 | 上海海事大学 | Laser cladding real-time monitoring system based on zigbee wireless sensing technology |
| CN105855743A (en) * | 2016-05-04 | 2016-08-17 | 华中科技大学 | On-line monitoring system and method for weld pool dynamic process |
| WO2018183003A1 (en) * | 2017-03-31 | 2018-10-04 | Konica Minolta Laboratory U.S.A., Inc. | 3d imaging by multiple sensors during 3d printing |
| CN107328391A (en) * | 2017-06-28 | 2017-11-07 | 武汉理工大学 | A kind of multi-sensor cooperation monitoring device and method for broadband laser cladding |
| CN108038266A (en) * | 2017-11-17 | 2018-05-15 | 西安铂力特增材技术股份有限公司 | A kind of method for numerical simulation of selective laser repair process |
| CN108062432A (en) * | 2017-11-17 | 2018-05-22 | 西安铂力特增材技术股份有限公司 | A kind of method for numerical simulation of selective laser fusion process |
| CN107992904A (en) * | 2017-12-22 | 2018-05-04 | 重庆邮电大学 | Forest Ecology man-machine interaction method based on Multi-source Information Fusion |
| CN109034391A (en) * | 2018-08-17 | 2018-12-18 | 王玲 | The multi-source heterogeneous information RBM network integration framework and fusion method of automatic Pilot |
Non-Patent Citations (4)
| Title |
|---|
| GH4169合金粉末选区激光熔化成形数值模拟及试验研究;成雅徽;《中国优秀硕士学位论文全文数据库工程科技I辑》;20160815;第B023-32页 * |
| Multisensor Data Fusion for Additive Manufacturing Process Control;Vandone, Ambra; Baraldo, Stefano; Valente, Anna;《IEEE ROBOTICS AND AUTOMATION LETTERS》;20181031;第3卷(第4期);第3279-3284页 * |
| Multisensor fusion for on line monitoring of the quality of spot welding in automotive industry;Cullen, J. D.等;《MEASUREMENT》;20080531;第41卷(第4期);第412-423页 * |
| 盘形激光焊接状态多传感信息融合在线检测;孙燕;《中国优秀硕士学位论文全文数据库工程科技I辑》;20151015;第B022-43页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109909502A (en) | 2019-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109909502B (en) | Online monitoring method for laser additive manufacturing process based on multi-source heterogeneous data | |
| Carpenter et al. | On residual stress development, prevention, and compensation in metal additive manufacturing | |
| Khanzadeh et al. | In-situ monitoring of melt pool images for porosity prediction in directed energy deposition processes | |
| Ning et al. | Analytical modeling of part porosity in metal additive manufacturing | |
| CN111283192B (en) | Laser powder bed melting additive manufacturing molten pool monitoring and pore control method | |
| Xiong et al. | Determination of surface roughness in wire and arc additive manufacturing based on laser vision sensing | |
| CN108362733B (en) | Semitransparent material photo-thermal characteristic distribution measuring method based on combination of phase-locked thermal wave and optical chromatography | |
| Long et al. | Numerical simulation of thermal behavior during laser metal deposition shaping | |
| Olleak et al. | Part-scale finite element modeling of the selective laser melting process with layer-wise adaptive remeshing for thermal history and porosity prediction | |
| Staiano et al. | Experimental study on hydrodynamic performances of naval propellers to adopt new additive manufacturing processes | |
| Xia et al. | Intelligent process monitoring of laser-induced graphene production with deep transfer learning | |
| Xiaohui et al. | Effect of supporting structure design on residual stresses in selective laser melting of AlSi10Mg | |
| Zhang et al. | Compression of pulsed infrared thermography data with unsupervised learning for nondestructive evaluation of additively manufactured metals | |
| Kumar et al. | Numerical modeling of selective laser melting: influence of process parameters on the melt pool geometry | |
| Fotovvati et al. | Experimental, computational, and data-driven study of the effects of selective laser melting (SLM) process parameters on single-layer surface characteristics | |
| Criales et al. | A sensitivity analysis study on the material properties and process parameters for selective laser melting of Inconel 625 | |
| Khanzadeh et al. | Fast prediction of thermal data stream for direct laser deposition processes using network-based tensor regression | |
| Dai et al. | A prediction model of milling force for aviation 7050 aluminum alloy based on improved RBF neural network | |
| Bontha | The effect of process variables on microstructure in laser-deposited materials | |
| Li et al. | Study on parameter optimization of laser cladding Fe60 based on GA-BP neural network | |
| Wischeropp et al. | Simulation of the effect of different laser beam intensity profiles on heat distribution in selective laser melting | |
| Li et al. | Research and prospect of on-line monitoring technology for laser additive manufacturing | |
| CN107506510B (en) | Method and device for producing an object with improved surface properties | |
| Choi et al. | Simulation-based uncertainty quantification for additively manufactured cellular structures | |
| Wang et al. | An initial residual stress inference method by incorporating monitoring data and mechanism model |
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 |