Electron beam equipment welding processing technology for realizing dynamic tracking of welding seam

Abstract

The invention discloses an electron beam equipment welding processing technology and equipment for realizing dynamic tracking of a welding seam. The equipment comprises a back scattering electron detector, an FPGA (Field Programmable Gate Array) high-speed parallel processing system, a photoelectric conversion circuit and the like. A deflection scanning instruction is transmitted to the high-frequency deflection system through the FPGA, an electron beam scans a to-be-welded area, back scattering electrons are collected to obtain welding seam position information, and therefore the electron beam is guided to hit the center of a welding seam in the next period, and the steps are repeated till the welding seam is repaired. The device can adapt to electron beam welding seam repair of various workpiece materials, various workpiece surface shapes and various welding seam directions, welding flux does not need to be additionally added, and meanwhile the welding seam is not oxidized.

Description

Technical Field

[0001] The invention relates to the field of welding processing, in particular to the phenomenon that a weld cannot be tracked in real time during electron beam welding, and a dynamic tracking technology of an electron beam weld is realized by collecting backscattered electrons for data analysis. Background Art

[0002] Electron beam welding technology is a fusion welding process that uses a high-speed focused electron beam to weld metal workpieces. It plays an increasingly important role in aerospace, marine ships, automobile manufacturing and other fields. With the advancement of industrialization, metal parts are becoming more and more complex during processing, and the requirements for their performance and processing accuracy are getting higher and higher. Traditional processing technology is difficult to adapt to the development trend of high-precision manufacturing technology. Many emerging technologies have emerged, among which electron beam processing technology is a processing technology with significant advantages. Electron beam processing technology uses an electron gun to emit a high-energy electron beam to bombard the surface of the workpiece to achieve processing. The power supply of the electron beam equipment can usually reach thousands of volts, and the power supply is connected to the cathode and anode inside the electron gun. Due to the huge voltage difference between the two poles, the electrons generated from the tungsten filament will rush into the magnetic field with extremely high kinetic energy under the action of the large electric field, pass through the centering coil, deflection coil, focusing coil and anti-astigmatism coil, and finally hit the surface of the workpiece to be processed. Although there are many processing advantages, the environment of electron beam processing must be a vacuum environment, and the welding process is often accompanied by welding defects such as pores and cracks, so it is impossible to observe and dynamically adjust the processing effect at close range. Therefore, proposing a new technology that can realize dynamic electron beam welding without breaking the vacuum environment has become one of the important challenges that needs to be solved urgently.

[0003] Dynamic tracking technology for welds is a new task that the electron beam welding industry urgently needs to solve. First, it is difficult to weld workpieces with complex geometric shapes. Some metal parts with complex geometric shapes in harsh outdoor environments are prone to cracks, which make them unusable. However, only local damage will greatly increase the cost if they are directly scrapped, so the engineering is more inclined to repair the welds so that they can continue to be used. However, surface welding is mainly carried out by manual welding. The most important disadvantage of manual welding is that the processing efficiency is low, and the workers cannot work continuously and at high intensity like machines; and the processing quality of manual welding is difficult to guarantee. Manual welding cannot guarantee that each welding has the same processing effect. Changing people or welding by one person multiple times will result in uneven results; manual welding is usually carried out in space, and the high temperature during the welding process will cause the weld to oxidize, and the strength and stiffness of the weld will be greatly reduced after completion.

[0004] Secondly, when using electron beam equipment for processing, the working space must first be evacuated to a vacuum through a vacuum pump set, and this process requires more than 30 minutes of preheating. At the same time, the working space cannot be opened immediately after the electron beam processing is completed. The pump set needs to be turned on to inject air and slowly restore it to atmospheric pressure before the workpiece can be taken out. Therefore, it is impossible to enter the working area to observe the workpiece and monitor the welding quality during the processing. If planning is not done before processing, processing errors will occur, thus affecting processing quality and efficiency.

[0005] Third, electron beam processing is mainly used for welds with large depth-to-width ratios. The diameter of the electron beam spot is usually only 200 microns, so it can accurately cut and weld workpieces like a precision scalpel. In the actual processing process, the welds will have different depths, and the electron beam rapid processing process simply relies on manual real-time adjustments or advance programming of the processing process based on the actual situation of the weld. These solutions are huge and difficult to implement. The processing quality is difficult to guarantee, and weld post-processing is required, which seriously affects the processing quality and has low production efficiency.

[0006] Fourth, the shape and direction of the weld are irregular. In many cases, the weld will not be formed along a straight line, but will be unpredictable, presenting an irregular curve. There will even be many branches on the basis of the main line, with different widths, depths and depths. Manual welding is difficult to complete, but machine welding is also difficult to operate. Therefore, the ability to automatically track welds is an urgent problem to be solved in the welding industry.

[0007] In response to the above pain points, there is currently no dynamic tracking technology for welds of electron beam equipment, and other welding methods will have corresponding solutions. For example, robotic arm welding will be equipped with a camera, which will capture the weld status of the area to be welded and transmit the current image to the processor. The algorithm will identify the weld and generate a robotic arm welding trajectory to achieve closed-loop dynamic tracking of welds. The present invention proposes and adopts a technical method for dynamic tracking of welds for electron beam equipment, filling the current gap in the electron beam industry. Summary of the invention

[0008] The present invention provides an electron beam welding processing technology method for realizing dynamic tracking of welds. This technology can be directly transplanted to any existing electron beam processing equipment, making the electron beam processing process intelligent, adaptive, and efficient. Electron beam processing is not picky about workpiece materials, and the equipment completely realizes the automation of the processing process. To achieve the above purpose, the present invention adopts the following technical solutions:

[0009] The first step is to install a backscattered electron detector in the vacuum chamber of the electron beam equipment processing area. The probe is equipped with two stepper motors, which can realize the Z-axis position and deflection angle of the probe externally. This backscattered electron detector is the key to the source of weld data. When the electron beam hits the surface of the workpiece, a large number of electrons will be reflected. These electrons carry a large amount of weld information. By extracting the information, the condition of the weld can be known. The detection head has a laser rod. When electrons hit the laser rod, bright light will be emitted. Combined with optical fiber and photoelectric conversion circuit, the optical signal of the backscattered electron is converted into an electrical signal. This step is an important foundation for all subsequent work.

[0010] In the second step, the weld electrical signal transmitted by backscattered electrons is used to realize signal transmission between the external circuit and the internal processor through a 12-bit AD module. The data output by the AD module will be transmitted to the PL end of the FPGA for Gaussian parallel data processing, and the image data will be filtered. The salt and pepper noise and Gaussian noise in the image are filtered out through median filtering and Gaussian filtering algorithms, and the imaging signal is filtered through a pipeline. A real-time image of the weld can be generated in the processor, so that the position that the electron beam needs to hit in the next cycle can be obtained to ensure that it can hit the middle of the weld.

[0011] The third step is to transmit the backscattered electron signal processed by FPGA on the PL side to the PS side. The PS side is a processor system that processes the data and performs floating-point operations to optimize the weld details, thereby improving the welding quality and efficiency from a system perspective. After completing a cycle of data processing on the PS side, a new cycle of scanning is about to begin, and the PS side sends an enable signal to reset the counter. After the counter is reset, the electron beam spot is pulled back to the starting point of the scan, and the next cycle of scanning is performed at the starting point of the scan. The PS side controls each module through the GP interface and transmits the final generated weld position information back to the PL side. The overall control of data transmission is completed on the PS side, and the data transmission process is simulated online.

[0012] The fourth step is to transmit the data from the PS end to the PL end. At this time, the position information of the weld has been fully analyzed. Next, it is necessary to guide the electron beam welding work based on this information. The PL end generates corresponding signals to control the high-frequency scanning drive module, dynamically adjust the electron beam energy distribution and scanning path through the high-frequency deflection system, track the weld position in real time, automatically generate trajectories along irregular curves and adjust the electron beam parameters to ensure that the electron beam accurately hits the center of the weld and repairs the cracks.

[0013] Step 5: Repeat steps 3 and 4 until the entire weld repair process is finally completed.

[0014] The electron beam welding dynamic tracking technology works through the above modules and steps. The invention can effectively solve the problems currently faced by the welding industry. The beneficial effects of the invention are as follows:

[0015] Compared with the backscattered electron detectors currently used in the electron beam processing process, the scintillation crystal detector has advantages such as good signal-to-noise ratio, temperature resistance and suitability for high-voltage environments. It has better applicability and good sensitivity in the complex environment of the electron beam welding area. In addition, through optical fiber transmission, the photoelectric conversion amplifier circuit is placed away from the workpiece area, which effectively reduces the impact of space charge on the circuit system, thereby extending the service life of the circuit system.

[0016] The ZYNQ chip is mainly composed of a dual-core ARM processor and a high-speed interface AXI bus that interacts with the system at the programmable logic unit PL end. Compared with traditional FPGAs, it can use ARM's integrated NEON technology to efficiently process floating-point numbers and optimize the design according to specific data processing requirements; compared with traditional ARM processor systems, it has stronger parallel data processing capabilities, and can also directly program hardware circuits as needed to reduce software execution operations and achieve high-speed data processing.

[0017] After colliding with the scintillating crystal, the backscattered electrons will appear in the form of fluorescence with different brightness. Designing a weak light signal conversion circuit with high sensitivity and gain can better capture and retain the detailed information of the weld surface morphology. Avalanche diode (APD) is a photoelectric conversion device based on the avalanche effect. It has been widely used due to its advantages such as fast response speed and high sensitivity.

[0018] The design and simulation of the hardware driver module in the imaging system are realized on the PL side, including the high-frequency scanning driver module, the weld data acquisition module, and the weld data processing and cache module. Finally, the overall control of data transmission is completed on the PS side, and the data transmission process is simulated online.

[0019] The electron beam energy distribution and scanning path are dynamically adjusted through the high-frequency deflection system, and the trajectory is generated along the irregular curve and the electron beam parameters are adjusted to ensure that the electron beam hits the center of the weld accurately and repairs the cracks.

[0020] The invention adopts a closed-loop control architecture, which aims to improve welding quality and efficiency through real-time monitoring, analysis and adjustment. The architecture integrates the three functions of monitoring, analysis and execution throughout the entire welding process to form a closed loop to achieve dynamic tracking of the weld. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a schematic diagram of an electron beam welding process for realizing dynamic tracking of weld seams:

[0022] Numbers in the figure: 1. electron gun; 2. deflection coil; 3. electron beam weld scanning area; 4. backscattered electron detector; 5. optical fiber; 6. photoelectric signal conversion circuit; 7. signal threshold adjustment circuit; 8. AD acquisition data conversion; 9. ZYNQ main control system; 10. DA output data conversion; 11. high-frequency scanning drive circuit; DETAILED DESCRIPTION

[0023] Now the invention is further described in detail with reference to the accompanying drawings. The above drawings are simplified schematic diagrams, which only illustrate the basic structure of the invention in a schematic manner, and therefore only show the components related to the invention.

[0024] The present invention provides a welding process design idea for electron beam equipment for realizing dynamic tracking of welds, comprising an electron gun 1, a deflection coil 2, an electron beam weld scanning area 3, a backscattered electron detector 4, an optical fiber 5, a photoelectric signal conversion circuit 6, a signal threshold adjustment circuit 7, an AD acquisition data conversion 8, a ZYNQ main control system 9, a DA output data conversion 10, and a high-frequency scanning drive circuit 11.

[0025] First, an electron beam is emitted from the electron gun 1. When the electron beam passes through the deflection coil 2, the deflection coil 2 generates a uniform magnetic field in the area because the circuit is connected. The magnetic field deflects the electron beam, and the direction of the red electron beam changes as shown in the figure. When the electron beam hits the surface of the weld, the secondary electrons and backscattered electrons on the surface of the weld will escape from the surface of the workpiece due to external bombardment, and these electrons will be captured by the nearby backscattered electron detector 4. The backscattered electrons hit the scintillation crystal on the backscattered electron detector, which will generate a weak light signal. These light signals are transmitted to the photoelectric signal conversion circuit 6 along the optical fiber 5. The photoelectric conversion circuit 6 first amplifies the signal through the avalanche diode APD, converts the weak light signal into a current signal, and then converts the current signal into a voltage signal through the operational amplifier circuit. However, the voltage signal at this time is very weak and has a large amount of ripples, so a decoupling circuit is used to eliminate noise, and a dual-cross group amplifier is used to amplify the signal, and the voltage signal is output to the signal threshold adjustment circuit 7. The signal threshold adjustment circuit 7 adjusts the brightness and contrast of the weld information. The pre-processed data will be converted into digital signals through AD acquisition data conversion 8, and the digital signals will be transmitted to the ZYNQ main control system 9 for high-speed parallel algorithm processing. The ZYNQ main control system 9 is divided into PL and PS ends. The digital signal is input to the weld data processing module of the PL end for data processing, and then the processed data is transmitted to the DDR3 memory of the PS end through the AXI interconnection module and the S_AXI_HP high-speed interface by DMA for storage. After completing the acquisition of a cycle of data, the PS end normalizes the frame data in DDR3 to pixel grayscale and transmits it to the AXI4-Stream to Video Out IP core through VDMA. The AXI4-Stream to Video Out IP core converts the data in the AXI4-Stream format into weld position information vector data to guide the welding of the electron beam in the next cycle. The processed digital signal is then converted into a voltage signal through DA output data conversion 10. The high-frequency scanning drive circuit 11 responds to the voltage signal quickly and follows it well, transmits the current signal to the deflection coil 2, uses the analyzed data to guide the electron beam to hit the middle of the weld in the next cycle, and scans the electron beam welding scanning area 3, continues to obtain the weld position, and repeats the above operations until the weld is completed.

[0026] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and inventive concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An electron beam welding process and equipment for realizing dynamic tracking of welds, characterized in that: include: ZYNQ main control system, AD conversion module, DA conversion module, high-frequency scanning drive circuit, backscattered electron detector, photoelectric conversion circuit and signal conditioning circuit. The backscattered electron detector is located inside the vacuum chamber. It is responsible for converting the backscattered electron signal generated by the interaction between the electron beam and the sample during the deflection scanning process into an optical signal, and transmitting the optical signal to the area away from the workpiece through the optical fiber to the signal processing system outside the vacuum chamber. The signal conditioning circuit amplifies and filters the analog signal for collection by the AD conversion module. After the data is processed by ZYNQ in high-speed parallel, the position information of the weld is recorded, and the deflection scanning signal is output at the same time, the position information of the weld is continuously updated, and the working position of the electron beam in the next cycle is guided. After the data information processed by ZYNQ is converted by DA digital-to-analog conversion, the deflection coil is driven by the high-frequency scanning drive circuit to guide the electron beam to deflect so as to hit the middle of the weld to achieve welding. At the same time, ZYNQ continuously sends scanning signals to guide the electron beam to scan the weld position in the area to be welded. Thanks to the high-frequency deflection circuit, the above process can be completed within milliseconds. The frequency of this process is much greater than the refresh rate of the human eye, so it seems to be carried out simultaneously, but in fact the electron beam is constantly repeating the same process to slowly complete the repair of the weld.

2. The electron beam welding process and equipment for dynamic tracking of weld seams as claimed in claim 1, characterized in that: The ZYNQ master control system is an integrated embedded system developed by Xilinx. It integrates the ARM Cortex-A series processors and field programmable gate arrays (FPGAs) to achieve high-performance computing and programmability of hardware logic on a single chip.

3. The electron beam welding process and equipment for dynamic tracking of weld seams as claimed in claim 1, characterized in that: The backscattered electron detector is mainly composed of two modules, backscattered electron collection and optical signal transmission, and a component base. The backscattered electron collection module includes a YAG scintillation crystal, a reflective lens, and a focusing cup; the optical signal transmission module mainly includes a focusing lens, a light guide lens, and an optical fiber.

4. The electron beam welding process and equipment for dynamic tracking of weld seams as claimed in claim 1, characterized in that: The high-frequency scanning drive circuit, after the scanning signal sent by the ZYNQ main control system is converted by the DA module, needs to be amplified by the high-frequency scanning drive circuit to drive the deflection scanning coil. By applying current to the scanning coil, an electromagnetic lens is formed to realize the deflection control of the electron beam. The high-frequency scanning drive circuit is a key module for realizing high-speed scanning of the electron beam. The circuit can realize dynamic following of high-frequency signals, and can follow frequency signals up to 60KHz without overshoot.

5. The electron beam welding process and equipment for dynamic tracking of weld seams as claimed in claim 1, characterized in that: The AD analog-to-digital conversion module and the DA digital-to-analog conversion module are important data conversion modules for the electron gun welding system to communicate with the ZYNQ main control system. The output of the DAC is updated according to the rising edge of the clock. The DAC of each channel has an independent 14-bit data input port, WRT write enable and CLK clock signal line. The AD conversion module is controlled by the PL end clock. When the rising edge of the clock arrives, the AN9238 starts to convert the collected signal. The output digital signal is read in parallel in the channel through the PL end weld data acquisition module.

6. The electron beam welding process and equipment for dynamic tracking of weld seams as claimed in claim 1, characterized in that: The photoelectric conversion circuit first converts the weak light signal of the backscattered electrons hitting the laser rod into a current signal, and then directly connects the output electrical signal to the operational amplifier circuit to convert the current signal into a voltage signal. At this time, the voltage signal is small and has a lot of interference, so a decoupling circuit is used to eliminate noise, and the signal is amplified through a dual-span amplifier to facilitate further signal processing by the FPGA.

7. The electron beam welding process and equipment for dynamic tracking of weld seams as claimed in claim 1, characterized in that: The signal conditioning circuit adjusts the weld signal in a timely manner. Different capacitor and resistor combinations can achieve a maximum gain amplification of 21 times. The bandwidth gain provided by the operational amplifier is much larger than the gain-bandwidth product in the circuit, so that the detailed features of the weld surface morphology can be retained, which is more helpful for analyzing the weld position. The circuit gain is determined by the ratio of an adjustable resistor to a fixed resistor. By adjusting the resistance of the adjustable resistor connected to the circuit, the contrast of the signal can be adjusted to improve the welding quality.

Images