CN-121972977-A - Online sensing device and method for manufacturing multisource information by using arc increase and decrease materials

CN121972977ACN 121972977 ACN121972977 ACN 121972977ACN-121972977-A

Abstract

The invention discloses an on-line sensing device and method for multi-source information in arc material increasing and decreasing manufacturing, wherein the device comprises a substrate, a double mechanical arm, a material increasing and decreasing system, a multi-source signal acquisition system and a signal processing system, wherein the substrate supports a workbench, the double mechanical arm respectively executes material increasing and decreasing operation, the material increasing system realizes preliminary forming of metal parts, the material decreasing system improves precision, the multi-source signal acquisition system acquires full-flow information through a high-speed camera, a high-frequency microphone, a thermal imager and a magneto-optical imager, the full-flow information is preprocessed through a signal acquisition card and then transmitted to the signal processing system, on-line monitoring is realized through data fusion, feature extraction and machine learning modeling, and the method comprises the steps of model establishment, information acquisition, material increasing and decreasing processing and signal discrimination. The invention solves the problems that the signal acquisition is single and the whole process of material increase and decrease cannot be adapted in the prior art, realizes high-efficiency high-precision integrated forming, has high perception precision, and is suitable for the field of high-precision requirements.

Inventors

ZHANG ZHEN
FAN CHENYANG
BAI PEIKANG
ZHAO ZHANYONG
WANG LIQING
MA KAI
LI YUNLONG

Assignees

中北大学

Dates

Publication Date: 20260505
Application Date: 20251230

Claims (10)

1. The online sensing device for multi-source information in arc additive manufacturing is characterized by comprising a substrate (10), a mechanical arm, an arc additive manufacturing forming system, a material reduction system, a multi-source signal acquisition system and a signal processing system (19), wherein the substrate (10) is provided with a rotatable base and is used for supporting a workbench (9) to adapt to the material addition and reduction operation of the mechanical arm, the mechanical arm comprises a first mechanical arm (4) and a second mechanical arm (12) which are respectively used for executing the material reduction processing operation and the material addition forming operation, the arc additive manufacturing forming system is matched with the second mechanical arm (12) to realize the preliminary forming of a metal piece with a complex structure, the material reduction system is matched with the first mechanical arm (4) to process the metal piece which is formed preliminarily, the multi-source signal acquisition system acquires multi-source information in the whole process of the material addition and reduction manufacturing in real time and transmits the multi-source information to the signal processing system (19), and the signal processing system (19) carries out fusion processing, characteristic parameter extraction and machine learning discrimination on the multi-source information; The electric arc additive manufacturing forming system comprises a first control system, a wire feeding system, an electric welding system, an atmosphere protection system and a water tank (13), wherein the wire feeding system comprises a welding wire extrusion device (14) and a wire ring (15) which are used for cooperatively conveying wires to the electric welding system, the electric welding system comprises a water-cooling argon arc welding gun (11) arranged at the front end of a second mechanical arm (12), the atmosphere protection system comprises a suction system (26), an air purifier (25) and an argon protection device (16), the suction system (26) and the air purifier (25) are used for absorbing and purifying harmful gases generated in the processing process, the argon protection device (16) is used for releasing argon to provide a protective atmosphere, the first control system comprises a control host (18) and a teaching box (17) which are used for controlling the operation of all parts of the electric arc additive manufacturing forming system, and the water tank (13) is used for cooling all parts of the electric arc additive manufacturing forming system through water pipes; The material reduction system comprises a second control system, a driving system, a processing system and a cooling system, wherein the driving system comprises a power assembly (3) for driving a first mechanical arm (4) and the processing system to operate, the processing system comprises a clamp (5) and a milling cutter (6) which are arranged at the end part of the first mechanical arm (4), the cooling system comprises a liquid storage tank (2) and a nozzle (7), cutting fluid is contained in the liquid storage tank (2), and is conveyed to a processing area through the nozzle (7) to realize cooling and chip removal; The signal processing system (19) comprises a data fusion module, a characteristic parameter extraction module and a machine learning module, wherein the data fusion module is used for carrying out arrangement and organic fusion on received multi-source signals, the characteristic parameter extraction module is used for extracting relevant characteristic parameters from fused multi-source information, and the machine learning module is used for receiving the characteristic parameters and establishing a classification discrimination model.
2. The on-line sensing device for manufacturing multi-source information by using the arc increase/decrease material according to claim 1 is characterized in that the multi-source signal acquisition system comprises a signal acquisition card (20), a high-speed camera (21), a magneto-optical imager (22), a thermal imager (23) and a high-frequency microphone (24), wherein the high-speed camera (21) is used for acquiring image signals of a component (8) to be processed, the Gao Pinmai g of the microphone (24) is used for acquiring acoustic signals in the processing process, the thermal imager (23) is used for acquiring temperature signals in the processing process, the magneto-optical imager (22) is used for acquiring magnetic field intensity signals and current distribution signals in the processing process, and the signal acquisition card (20) receives analog signals or digital signals output by various sensors, and transmits the analog signals or digital signals to the signal processing system (19) after noise reduction, denoising and normalization processing.
3. An arc increase and decrease material manufacturing multi-source information online sensing method based on the device of claim 1, which is characterized by comprising the following steps: S1, starting all parts of an arc additive manufacturing forming system, independently starting a high-frequency microphone (24) after all the parts run stably, and collecting sound data of equipment running as a reference basis; s2, building a 3D model of the structural member, performing numerical simulation in advance, and generating a processing execution file; S3, starting a high-speed camera (21), a magneto-optical imager (22) and a thermal imager (23), and collecting multisource information of the material increasing and decreasing manufacturing process in real time; s4, based on the processing execution file, completing preliminary forming of the metal piece through an arc additive manufacturing forming system; s5, performing high-precision processing on the preliminarily formed metal piece through a material reduction system; s6, the signal acquisition card (20) receives the multi-source information, performs filtering and normalization processing, and transmits the processed multi-source information to the signal processing system (19); And S7, the signal processing system (19) fuses the processed multi-source information and extracts characteristic parameters, and establishes a classification and discrimination model through machine learning to realize online perception of the material increasing and decreasing manufacturing process.
4. The online sensing method of multi-source information for manufacturing arc increase/decrease materials according to claim 3, wherein the step S2 specifically comprises: S2.1, establishing a 3D model of the structural member by adopting three-dimensional modeling software, and outputting the model as an STL format file; S2.2, slicing, region division and path planning are carried out on the STL format file through slicing processing software, and LS format file is generated after coordinate calibration is completed; s2.3, performing numerical simulation verification on the machining track in advance by using virtual simulation software, and finally outputting the LP format file to the arc additive manufacturing and forming system.
5. The method for online sensing of multi-source information in arc material manufacturing according to claim 3, wherein step S4 specifically comprises: s4.1, carrying out coordinate calibration on the second mechanical arm (12) and the workbench (9), and carrying out actual printing track verification by referring to a numerical simulation result; S4.2, sequentially activating a water tank (13), a welding wire extrusion device (14), a metal wire ring (15), an argon protection device (16), an air purifier (25) and a suction system (26), and starting printing after each part stably works; S4.3, starting a second mechanical arm (12) and a water-cooling argon arc welding gun (11) through a teaching box (17), forming an electric arc between the water-cooling argon arc welding gun (11) and the substrate (10), and accumulating and forming molten metal wires; And S4.4, after printing a layer, the second mechanical arm (12) drives the water-cooling argon arc welding gun (11) to lift away from the substrate (10) to pause printing, and after the molten pool is cooled, the molten pool is moved to an arcing position to continue printing, and the circulation is performed until the preliminary forming is completed.
6. The method for online sensing of multi-source information in arc material manufacturing according to claim 3, wherein step S5 specifically comprises: S5.1, operating the control panel (1) to enable the milling cutter (6) to be close to a member (8) to be processed, and finishing tool setting through the tool setting function of the control panel (1) after observing the contact condition; S5.2, simulating and running a machining program in advance, and starting the material reduction machining after confirming no errors.
7. The online sensing method for manufacturing multi-source information by using the arc increase/decrease material according to claim 3, wherein the feature extraction processing of the multi-source information in the step S6 comprises the steps of extracting the contour features of welding seams and parts by adopting an edge detection algorithm on dynamic image information acquired by a high-speed camera (21), extracting the magnetic field intensity peak value, the valley value and the current density distribution features on magnetic field and current distribution data acquired by a magneto-optical imager (22), extracting the temperature field data acquired by a thermal imager (23), extracting the temperature distribution mean value, the variance, the hot spot position and the temperature change rate features, and converting acoustic data acquired by a high-frequency microphone (24) into frequency domain features through Fourier transformation to extract the amplitude and the frequency features.
8. The method for online sensing of multi-source information in arc material manufacturing according to claim 3, wherein the method is characterized in that dynamic image data collected by a high-speed camera (21) is subjected to filtering treatment, an intermediate value is taken as a new pixel value, and then the pixel value is mapped to a [0,1] interval for normalization treatment, wherein the formula is as follows: Wherein X is the original pixel value.
9. The online sensing method of multi-source information for manufacturing arc increase/decrease materials according to claim 3, wherein the characteristic parameters extracted in the step S7 include weld edge characteristic E, amplitude a, frequency F, temperature distribution mean M, variance V, hot spot position H, temperature change rate R, magnetic field intensity peak value P, magnetic field intensity valley value G, and current distribution density D.
10. The method for online sensing of multi-source information in arc material manufacturing according to claim 3, wherein the machine learning adopts a linear discriminant analysis method, and the established discriminant function is: LDA1=w 1 ·E+w 2 ·A+w 3 ·F+w 4 ·M+w 5 ·V+w 6 ·H+w 7 ·R+w 8 ·P+w 9 ·G+w 10 ·D LDA2=w 1 ·E+w 2 ·A+w 3 ·F+w 4 ·M+w 5 ·V+w 6 ·H+w 7 ·R+w 8 ·P+w 9 ·G+w 10 ·D wherein, the w 1 ,w 2 ,w 3 . Is a weight coefficient.

Description

Online sensing device and method for manufacturing multisource information by using arc increase and decrease materials Technical Field The invention relates to the technical field of arc increasing and decreasing materials, in particular to an on-line sensing device and method for manufacturing multisource information by using arc increasing and decreasing materials. Background As a novel material processing and forming process, the additive manufacturing technology has the advantages of high processing efficiency, low processing cost and the like compared with the traditional forming processes such as casting, forging and the like, and particularly has unique advantages in the aspect of manufacturing parts with complex structures. Arc additive manufacturing techniques, which are more efficient in forming, are often used to manufacture complex structures. However, with the intensive research of researchers on additive manufacturing technology, the disadvantages of the additive manufacturing technology are gradually highlighted, that is, the performance of parts manufactured by the additive manufacturing technology is often poor, and the precision, the surface finish and the like of the parts cannot meet the high-requirement standard. This has led to great limitations in the application of additive manufacturing technology in some fields where high demands are placed on part accuracy, such as aerospace, military manufacturing, etc. Compared with additive manufacturing, the additive manufacturing process has the defects of low production efficiency, low material utilization rate, long processing period and the like, but parts processed by the traditional additive manufacturing process often have higher precision and performance. Therefore, if additive manufacturing and subtractive manufacturing can be combined, the advantages of the additive manufacturing and the subtractive manufacturing are exerted, and the forming efficiency is high, the cost is low, and the precision and the performance are high. However, there is currently a lack of multi-source information on-line sensing systems that are simultaneously suitable for arc additive and subtractive manufacturing. Patent CN116559196a discloses a system and a method for monitoring defects in arc additive manufacturing, which can detect multi-source information on the surface and internal defects of a component manufactured by arc additive through a magneto-optical imager, an infrared camera, a high-speed camera and an acoustic sensor, but cannot be suitable for material reduction manufacturing, and cannot monitor the forming process on line. Patent CN116604218a discloses a device for controlling arc additive forming quality by acoustic reflection feedback, which can realize on-line monitoring of arc additive manufacturing components through an acoustic emission acquisition system and a feedback control system, discover defects in time and adjust and remedy at any time, but cannot acquire multi-source information, and is not suitable for a material reduction process. Disclosure of Invention In order to overcome the defects, the device and the method for online sensing multi-source information in manufacturing of the arc additive are provided, firstly, an arc additive forming system is utilized to preliminarily form a metal piece with a complex structure, four high-precision sensors are utilized to collect multi-source information and transmit the multi-source information to a signal analysis system for further analysis and discrimination, then a material reduction processing system starts to operate to further process the metal piece, precision and performance are improved, a multi-source signal acquisition and processing system operates again in the process to collect information and analyze, a signal processing system extracts characteristic parameters and establishes a defect model by utilizing a machine learning method to discriminate multi-source signals, thus realizing online sensing of multi-source information, comprehensively online monitoring of components and solving the problem of the current online sensing device. The technical scheme includes that the online sensing device for multi-source information in arc additive manufacturing comprises a substrate, a mechanical arm, an arc additive manufacturing forming system, a material reduction system, a multi-source signal acquisition system and a signal processing system, wherein the substrate is provided with a rotatable base for supporting a workbench to adapt to additive and subtractive operations of the mechanical arm, the mechanical arm comprises a first mechanical arm and a second mechanical arm which are respectively used for executing the material reduction processing operations and the material addition forming operations, the arc additive manufacturing forming system is matched with the second mechanical arm to realize preliminary forming of metal parts with complex structures, the material r