Automatic powder particle feeding system based on molten pool vision and spectral analysis

Abstract

The utility model discloses an automatic powder particle feeding system based on molten pool vision and spectral analysis. The automatic powder particle feeding system comprises a welding unit, a data acquisition device, an FPGA device, a computer unit and a powder feeding device. The welding unit comprises a welding spot molten pool; the data acquisition device is used for acquiring data of a welding spot molten pool; the data acquisition device transmits acquired data to the FPGA device; the computer unit is used for analyzing data transmitted into the FPGA device; the powder feeding device is used for conveying powder particles to the molten pool; data acquired by the data acquisition device is transmitted into the FPGA device, and after the computer unit analyzes the data of the FPGA device, the FPGA device controls the powder feeding rate of the powder feeding device based on the analyzed data; the data acquisition device comprises a molten pool vision unit and a spectrum acquisition unit; the molten pool visual unit is used for collecting visual images of the molten pool; the spectrum collection unit is used for collecting powder particle spectrum data of the molten pool.

Description

Technical Field

[0001] This utility model relates to the field of welding automation, specifically to an automatic powder feeding system based on visual and spectral analysis of the molten pool. Background Technology

[0002] Powder particles in titanium alloys work synergistically with other alloying elements (Cr, Mo, V, and Ti, etc.) to improve the alloy's strength, toughness, creep resistance, wear resistance, and corrosion resistance. However, adding powder particles is difficult, as it can disturb the electric arc and molten pool, causing spatter. Therefore, controlling the powder particle content in the molten pool during additive manufacturing is of great significance.

[0003] Controlling the powder particle content in the molten pool during additive manufacturing is crucial for improving weld quality and achieving a weld that matches the base material. Chinese invention patent No. 200710141482.2 discloses a cladding device based on argon arc welding, which uses a coaxial powder feeding method. The resulting cladding layer has a high dilution rate, typically around 5% to 10%, and a large area. However, when used in welding processes, the specific nitrogen content of the molten pool cannot be obtained, making precise control of the weld nitrogen content impossible. Chinese invention patent No. 201210250419.3 discloses a double-layer gas flow protection TIG welding method for high-nitrogen steel. It employs double-layer nitrogen protection, but due to gas evaporation during welding, the surface pressure of the molten pool increases, preventing nitrogen from effectively entering the pool. Therefore, its nitrogen enrichment effect is poor, and automatic powder feeding cannot be achieved. Utility Model Content

[0004] The technical problem to be solved by this utility model is to provide an automatic powder feeding system based on molten pool vision and spectral analysis that can automatically add powder and determine powder content, in order to address the above-mentioned technical status.

[0005] The technical solution adopted by this utility model to solve the above-mentioned technical problems is: an automatic powder feeding system based on molten pool vision and spectral analysis, characterized in that:

[0006] Includes welding units with weld pools;

[0007] Data acquisition device, used to collect data on the weld pool;

[0008] An FPGA device, the input of which is connected to the output of a data acquisition device, so that the data acquisition device transmits the acquired data to the FPGA device.

[0009] The computer unit has its input terminals connected to the output terminals of the FPGA device and is used to analyze the data input to the FPGA device.

[0010] The powder feeding device is used to transport powder particles to the molten pool; the data collected by the data acquisition device is transmitted to the FPGA device, the output of the computer unit is connected to the input of the FPGA device, the output of the FPGA device is connected to the powder feeding device, and the FPGA device controls the powder feeding rate of the powder feeding device based on the analyzed data.

[0011] The data acquisition device includes:

[0012] The molten pool vision unit is used to acquire visual images of the molten pool.

[0013] The spectral acquisition unit is used to acquire spectral data of powder particles from the molten pool.

[0014] To qualitatively analyze whether powder particles escape from the molten pool, preferably, the molten pool vision unit includes:

[0015] The distance between the CCD camera and the molten pool is 15-23cm;

[0016] The filter is located at the front end of the CCD camera;

[0017] A neutral density filter, positioned at the front end of the filter; and,

[0018] A data acquisition card, wherein the input end of the data acquisition card is connected to the output end of a CCD camera, and the output end of the data acquisition card is connected to the input end of an FPGA device to acquire and transmit data.

[0019] Preferably, the wavelength range of the filter is 500-800nm.

[0020] To accurately analyze the particle content of the molten pool powder, preferably, the spectral acquisition unit includes an optical fiber sensor and a spectrometer, the output of which is connected to the input of the FPGA device and used to transmit data to the FPGA device.

[0021] Preferably, the welding unit includes a welding torch and a welding assembly, wherein the welding torch is used for welding and the welding assembly provides electrical power to the welding torch.

[0022] To prevent powder from sticking during delivery, preferably, the powder delivery device includes...

[0023] The powder feeding pipe has a powder outlet at the front end for discharging powder particles.

[0024] A gate, located at the front end of the powder feeding pipe, is used to control the size of the powder outlet; and,

[0025] A condenser is installed around the powder delivery pipe and is supplied with cooling water.

[0026] To prevent oxidation, preferably, the condensation device includes a first condenser tube disposed around the outer periphery of the powder feeding pipe and a second condenser tube disposed outside the first condenser tube; a cavity for containing a protective gas is formed between the first condenser tube and the second condenser tube. The protective gas prevents the powder particles from oxidizing and provides power for feeding.

[0027] To accurately control the powder feeding amount and speed, the computer unit preferably includes a visual image processing module and a spectral processing module. The visual image processing module is used to analyze the visual image of the molten pool, and the spectral processing module is used to analyze the spectral data of the powder particles in the molten pool. The visual image processing module uses MATLAB software, and the spectral processing module uses TQAnalyst software.

[0028] Compared with the prior art, the advantages of this utility model are:

[0029] ① It is equipped with a welding unit, a data acquisition device, an FPGA device, a computer unit, and a powder feeding device to realize automatic powder feeding;

[0030] ② Collect visual and spectral data of the welding process through a data acquisition device; analyze the data through an FPGA device and a computer unit, use visual images to determine whether there are powder particles escaping from the molten pool, and use spectral data of the powder particles to analyze the content of powder particles in the molten pool; control the powder feeding rate of the powder feeding device according to the working conditions to ensure the content of powder particles in the weld and thus ensure high weld quality.

[0031] ③ The FPGA device and computer unit have high computing speed. The FPGA device has high-speed encoding and decoding, and strong motion control synchronization, thus meeting the requirements for additive powder feeding synchronization. It also has programmability, which ensures the scalability of the system. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the structure of an embodiment;

[0033] Figure 2 This is a schematic diagram of the powder feeding device in an embodiment. Detailed Implementation

[0034] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0035] like Figure 1-2 The diagram shows a preferred embodiment of the present invention. An automatic powder feeding system based on molten pool vision and spectral analysis includes a welding unit, a data acquisition device, an FPGA device 9, a computer unit 10, and a powder feeding device 11.

[0036] The data acquisition device is used to acquire data of the solder pool 14; the data acquisition device transmits the acquired data to the FPGA device 9. The data acquisition device includes a solder pool vision unit and a spectrum acquisition unit. The solder pool vision unit is used to acquire visual images of the solder pool 14, and the spectrum acquisition unit is used to acquire the spectral data of the powder particles in the solder pool 14.

[0037] The fused pool vision unit includes a CCD camera 5, a filter 4, a neutral density filter 3, and a data acquisition card 6. The distance between the CCD camera 5 and the fused pool 14 is 15-23 cm. The filter 4 is located in front of the CCD camera 5, and the neutral density filter 3 is located in front of the filter 4. The input end of the data acquisition card 6 is connected to the output end of the CCD camera 5, and the output end of the data acquisition card 6 is connected to the input end of the FPGA device 9. Therefore, the data acquisition card 6 is used to acquire data from the CCD camera 5 and transmit it to the FPGA device 9. The wavelength range of the filter 4 is 500-800 nm. The spectral acquisition unit includes a fiber optic sensor 7 and a spectrometer 8 connected to the fiber optic sensor 7. The input end of the spectrometer 8 is connected to the output end of the fiber optic sensor 7, and the output end of the spectrometer 8 is connected to the input end of the FPGA device 9 and is used to transmit data to the FPGA device 9. A light-collecting device is provided in front of the fiber optic sensor 7 for light collection.

[0038] In this embodiment, the FPGA device 9 adopts the ALTERA EP1C20; the output terminal of the FPGA device 9 is connected to the powder feeding device 11, so that the FPGA device 9 controls the powder feeding rate of the powder feeding device 11 based on the analyzed data.

[0039] The computer unit 10 has its input terminals connected to the output terminals of the FPGA device 9 and is used to analyze the data input to the FPGA device 9. The computer unit 10 is equipped with a visual image processing module and a spectral processing module. The visual image processing module uses MATLAB software, and the spectral processing module uses TQ Analyst software. The MATLAB software is used to analyze the visual image of the molten pool 14; the TQ Analyst software is used to analyze the spectral data of the powder particles in the molten pool 14.

[0040] The welding unit has a weld pool 14 and includes a welding torch 2 and a welding assembly 1. The welding torch 2 is used for welding, and the welding assembly 1 provides power to the welding torch 2. The welding torch 2 is a CMT welding torch. The CCD camera 5 and the fiber optic sensor 7 are both fixedly connected to the welding torch 2 using clamps, and the fields of view centers of the CCD camera 5 and the fiber optic sensor 7 are aligned with the weld pool 14. The welding assembly 1 includes a welding power source, a control cabinet, and a shielding gas located within the control cabinet.

[0041] like Figure 2As shown, the powder feeding device 11 includes a powder feeding pipe 17, a gate 18, and a condensing device 16. The powder feeding pipe 17 has a powder outlet at its front end for discharging powder particles; the gate 18 is located at the front end of the powder feeding pipe 17 and controls the size of the powder outlet; the condensing device 16 is arranged around the powder feeding pipe 17 and is supplied with cooling water. The condensing device 16 includes a first condensing pipe located around the outer periphery of the powder feeding pipe 17 and a second condensing pipe located outside the first condensing pipe; a cavity 15 is formed between the first condensing pipe and the second condensing pipe to accommodate a protective gas.

[0042] The powder feeding device 11 is used to convey powder particles to the molten pool 14; the data collected by the data acquisition device is transmitted to the FPGA device 9. After the computer unit 10 analyzes the data of the FPGA device 9, the FPGA device 9 controls the powder feeding rate of the powder feeding device 11 based on the analyzed data. That is, the FPGA device 9 sends pulse trains to the servo motor of the powder feeding device 11 and internally realizes pulse distribution, completes system data encoding and decoding, and forms motion commands to control the movement of the servo motor, thereby controlling the opening and closing and opening size of the gate 18.

[0043] The workflow of the automatic powder feeding system based on molten pool vision and spectral analysis is as follows:

[0044] Step 1: Weld the workpiece 12. Use CCD camera 5 and data acquisition card 6 to acquire visual images of the molten pool. Use fiber optic sensor 7 to obtain the powder particle spectral data of the molten pool 14. Input the visual image acquisition results of the molten pool 14 and the molten pool 14 powder particle spectral data of the prior threshold into FPGA device 9 for preprocessing.

[0045] Step 2: Judge the processing results through computer unit 10: Perform online image processing with MATLAB on the obtained visual image to qualitatively analyze whether there is any powder particle escaping from the molten pool 14: If bubbles are generated, determine the powder particle content of the molten pool 14 online through the prior threshold TQ Analyst software spectral processing of the obtained spectral data. The preset range of powder particle content in the molten pool 14 can be adjusted as needed.

[0046] Step 3: Based on the judgment result of computer unit 10 in step 2, if the powder particle content in the molten pool 14 is lower than the set threshold data, computer unit 10 sends data to FPGA device 9. FPGA device 9 controls powder feeding device 11, and controls the powder feeding rate of powder feeding device 11 through gate 18, adding powder particles to the molten pool 14. Through arc electromagnetic stirring and welding torch 2 oscillation, the powder particles are fully mixed and melted with the molten base metal, ultimately forming weld 13 that meets the target powder particle content, thereby increasing the powder particle content of weld 13. When the powder particle content in the molten pool 14 reaches the preset extreme value, powder feeding is stopped, ultimately forming weld 13 that meets the target powder particle content composition.

Claims

1. A powder particle automatic feeding system based on molten pool vision and spectrum analysis, characterized in that: it comprises a welding unit; a welding spot molten pool (14); a data acquisition device for acquiring data of the welding spot molten pool (14); an FPGA device (9), an input end of the FPGA device (9) being connected to an output end of the data acquisition device, so that the data acquisition device transmits the acquired data into the FPGA device (9); a computer unit (10), an input end of the computer unit (10) being connected to an output end of the FPGA device (9) and being used for analyzing the data transmitted into the FPGA device (9); and a powder feeding device (11) for feeding powder particles to the molten pool (14); the data acquired by the data acquisition device is transmitted into the FPGA device (9), an output end of the computer unit (10) is connected to an input end of the FPGA device (9), and an output end of the FPGA device (9) is connected to an input end of the powder feeding device (11), so that the FPGA device (9) controls a powder feeding rate of the powder feeding device (11) based on the analyzed data; the data acquisition device comprises: a molten pool vision unit for acquiring a visual image of the molten pool (14), and a spectrum acquisition unit for acquiring powder particle spectrum data of the molten pool (14); the molten pool vision unit comprises: a CCD camera (5) with a distance of 15-23 cm from the molten pool (14); a filter (4) arranged at a front end of the CCD camera (5); a light reduction sheet (3) arranged at a front end of the filter (4); and a data acquisition card (6) with an input end connected to an output end of the CCD camera (5) and an output end connected to an input end of the FPGA device (9) to acquire and transmit data; a wave band of the filter (4) is 500-800 nm; the spectrum acquisition unit comprises a fiber sensor (7) and a spectrometer (8), an input end of the spectrometer (8) being connected to an output end of the fiber sensor (7), and an output end of the spectrometer (8) being connected to an input end of the FPGA device (9) and being used for transmitting data to the FPGA device (9); the welding unit comprises a welding torch (2) for welding and a welding assembly (1) for providing electric energy for the welding torch (2); the powder feeding device (11) comprises: a powder feeding pipe (17) with a powder outlet at a front end for feeding powder particles; a gate (18) arranged at the front end of the powder feeding pipe (17) and capable of controlling a size of the powder outlet; and a condensing device (16) arranged around the powder feeding pipe (17) and connected to cooling water; the condensing device (16) comprises a first condensing pipe arranged around the powder feeding pipe (17) and a second condensing pipe arranged outside the first condensing pipe; a cavity (15) for accommodating protective gas is formed between the first condensing pipe and the second condensing pipe; the computer unit (10) is provided with a visual image processing module and a spectrum processing module; the visual image processing module is used for analyzing the visual image of the molten pool (14); and the spectrum processing module is used for analyzing the powder particle spectrum data of the molten pool (14). ​ ​ ​ ​ ​ ​ ​ ​ 2. The powder particle automatic feeding system based on molten pool vision and spectral analysis according to claim 1, characterized in that: ​ ​ ​ ​ ​ 3. The powder particle automatic powder feeding system based on molten pool vision and spectral analysis according to claim 2, characterized in that: ​ 4. The powder particle automatic powder feeding system based on molten pool vision and spectral analysis according to claim 1, characterized in that: ​ 5. The powder particle automatic powder feeding system based on molten pool vision and spectral analysis according to claim 1, characterized in that: ​ 6. The powder particle automatic feeding system based on molten pool vision and spectral analysis according to claim 1, characterized in that: ​ ​ ​ ​ 7. The powder particle automatic feeding system based on molten pool vision and spectral analysis according to claim 6, characterized in that: ​ 8. The powder particle automatic feeding system based on molten pool vision and spectral analysis according to claim 1, characterized in that: ​

Images