System and method for adjusting and controlling additive manufacturing powder speed through electric field based on visual feedback

Abstract

The invention discloses a system and method for adjusting and controlling the additive manufacturing powder speed through an electric field based on visual feedback, and relates to the technical field of additive manufacturing, the system comprises a laser head, a machine vision monitoring unit and a control unit, the laser head and a powder feeding nozzle are coaxially arranged, the powder feeding nozzle is connected with a powder feeder, and a high-voltage electrode plate is installed in the powder feeder; the high-voltage electrode plates are connected with high-voltage power supplies; the machine vision monitoring unit comprises a high-speed camera and a thermal infrared imager, the high-speed camera is used for collecting powder flow information, and the thermal infrared imager is used for monitoring molten pool information; the control unit is used for processing information collected by the high-speed camera and the thermal infrared imager in real time and transmitting the processed information to the high-voltage power supply, and the powder speed is changed by adjusting the output voltage. The powder flow form and the molten pool state are captured in real time through the machine vision system, control signals are generated based on image processing to dynamically adjust the electric field intensity, intelligent closed-loop control over the powder speed is achieved, splashing is effectively restrained, and the forming quality and the powder utilization rate are improved.

Description

Technical Field

[0001] This invention relates to the field of additive manufacturing technology, and in particular to a system and method for controlling the speed of additive manufacturing powders based on visual feedback and electric field. Background Technology

[0002] In the Directed Energy Deposition (DED) process, the stability of the powder flow is crucial. Although the concept of using an electric field to control the powder velocity has been proposed, existing control strategies are mostly limited to PID regulation or open-loop control, which mainly have the following problems: Existing powder feeding systems generally employ open-loop control, which, while sufficient for manufacturing scenarios requiring general precision, has significant limitations in high-precision, high-stability additive manufacturing applications. Closed-loop control is essential for precise regulation. However, achieving high-performance closed-loop control still faces many technical challenges: First, at the sensing level, real-time monitoring of powder flow and the molten pool suffers from feedback delays, making it difficult to handle transient changes in high-speed manufacturing conditions. Second, at the decision-making level, the process exhibits complex characteristics such as strong nonlinearity and multivariate coupling, making it difficult for traditional control methods to achieve multi-objective coordination and dynamic optimization. Summary of the Invention

[0003] To address the shortcomings of existing technologies, the present invention aims to provide a system and method for controlling the speed of additive manufacturing powder using an electric field based on visual feedback. By using a machine vision system to capture the powder flow morphology and molten pool state in real time, and generating control signals based on image processing, the electric field strength is dynamically adjusted to achieve intelligent closed-loop control of powder speed, effectively suppressing splashing, improving molding quality and powder utilization.

[0004] To achieve the above objectives, the present invention is implemented through the following technical solution: In a first aspect, embodiments of the present invention provide a system for controlling the speed of additive manufacturing powder based on visual feedback using an electric field, comprising: The laser head is coaxially arranged with the powder feeding nozzle, the powder feeding nozzle is connected to the powder feeder, and a pair of high-voltage electrode plates are installed inside the powder feeder. The high-voltage electrode plates are connected to a high-voltage power supply. The machine vision monitoring unit includes a high-speed camera and an infrared thermal imager. The high-speed camera is used to collect powder flow information, and the infrared thermal imager is used to monitor molten pool information. The control unit processes information collected by the high-speed camera and infrared thermal imager in real time and transmits the processed information to the high-voltage power supply to change the powder speed by adjusting the output voltage.

[0005] As a further implementation, the high-speed camera is equipped with a backlight.

[0006] As a further implementation, the powder flow information includes the powder flow profile, concentration, and trajectory; the molten pool information includes the molten pool temperature field distribution and spatter generation.

[0007] As a further implementation, the control unit includes an image processing module, an intelligent decision-making module, and a control output module. The image processing module is used to process image data and extract key feature parameters of the powder flow and the molten pool. The intelligent decision-making module generates an electric field strength control command by comparing the key characteristic parameters of the powder flow with a preset ideal process window; the control output module is used to send the electric field strength control command to the high-voltage power supply.

[0008] Secondly, embodiments of the present invention also provide a method for controlling the speed of additive manufacturing powder using an electric field based on visual feedback, employing the aforementioned system, comprising: Acquire visible light images of the powder flow and infrared thermal images of the molten pool, and extract key feature parameters of the powder flow and molten pool; By comparing key characteristic parameters with a preset ideal process window, the optimal electric field strength control command is generated. Output electric field strength control command to change the electric field strength applied to the powder.

[0009] As a further implementation, the system switches between three modes: splash suppression, flow rate compensation, and steady-state maintenance, based on the key input feature parameters.

[0010] As a further implementation, when the number of spatter particles extracted from the molten pool infrared thermal image continuously exceeds the dynamic safety threshold within a continuously set control cycle, the system switches to spatter suppression mode. Alternatively, when the aspect ratio of the molten pool suddenly increases and exceeds the threshold, switch to splash suppression mode.

[0011] As a further implementation, the control process in the splash suppression mode is divided into a first stage and a second stage, wherein the first stage is emergency braking and the second stage is proportional regulation to stabilize the flow.

[0012] As a further implementation, when the difference between the actual average velocity of the powder flow and the target velocity continues to exceed the tolerance range, the flow rate compensation mode is switched. Alternatively, when the powder flow divergence angle is greater than the maximum allowable divergence angle, switch to flow rate compensation mode; Alternatively, if the instantaneous mass flow rate estimated based on velocity and flow density is lower than the process requirement value, switch to flow rate compensation mode.

[0013] As a further implementation method, when the speed deviation is less than the set value, the flow shape is ideal, there is no splashing, and the molten pool area and temperature fluctuation are within the allowable range, the system switches to steady-state maintenance mode.

[0014] The beneficial effects of this invention are as follows: (1) The machine vision monitoring unit of the present invention includes a high-speed camera and an infrared thermal imager. The control unit can receive and process the image information collected by the high-speed camera and the infrared thermal imager in real time, and transmit the processed information to the high-voltage power supply to change the powder speed by adjusting the output voltage. It can simultaneously monitor the powder flow and the molten pool, not only control the speed, but also indirectly ensure the quality of the molten pool, and realize the synergistic optimization of multiple process parameters. The machine vision captures the powder flow morphology and molten pool state in real time, generates control signals based on image processing, and dynamically adjusts the electric field strength, thereby realizing intelligent closed-loop control of the powder speed, effectively suppressing splashing, improving molding quality and powder utilization.

[0015] (2) Based on the input characteristic parameters, the present invention switches between three modes: splash suppression, flow rate compensation and steady-state maintenance, and outputs control commands using different control strategies, forming a control scheme with multi-modal input, multi-objective optimization and multi-priority decision-making. Through intelligent state switching, safety objectives such as splash suppression are prioritized, and flow rate control is dynamically optimized, effectively solving the contradiction of multi-objective control. At the same time, the system automatically reduces control output under steady-state conditions, significantly reducing energy consumption and actuator wear while maintaining accuracy, and can achieve adaptive, high-precision and robust control of powder speed. Attached Figure Description

[0016] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0017] Figure 1 This is a schematic diagram of the system structure of the present invention according to one or more embodiments; Figure 2 This is a flowchart of a method according to one or more embodiments of the present invention.

[0018] Among them, 1. Powder storage bin, 2. Powder feeder, 3. Control unit, 4. Powder feeding nozzle, 5. High-voltage electrode plate, 6. High-voltage power supply, 7. Laser head, 8. Laser, 9. Molten pool, 10. Substrate, 11. High-speed camera, 12. Backlight, 13. Infrared thermal imager. Detailed Implementation

[0019] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0020] Example 1: This embodiment provides a system for controlling the speed of additive manufacturing powder using an electric field based on visual feedback, such as... Figure 1 As shown, it mainly includes a laser head 7, a powder feeding nozzle 4, a high-voltage electrode plate 5, a high-voltage power supply 6, a machine vision monitoring unit, and a control unit 3. The powder feeding nozzle 4 is coaxially mounted on the outside of the laser head 7. The powder feeding nozzle 4 is connected to the powder feeder 2 through a powder feeding pipe. The powder feeder 2 is connected to the powder storage bin 1. The powder feeding nozzle 4 provides the powder required in the additive manufacturing process of the substrate. The laser head 7 is used to emit a laser 8 to the substrate 10 and form a molten pool 9 on the surface of the substrate.

[0021] The powder feeder 2 is equipped with a pair of high-voltage electrode plates 5, according to Figure 1 In the center view, two high-voltage electrode plates 5 are arranged vertically in sequence, i.e., along the powder feeding direction. One high-voltage electrode plate 5 is connected to the positive terminal of the high-voltage power supply 6 via a wire, and the other high-voltage electrode plate 5 is connected to the negative terminal of the high-voltage power supply 6 via a wire. The powder feeding speed is adjusted by changing the output voltage of the high-voltage power supply 6. In this embodiment, the high-voltage power supply 6 is a programmable high-voltage power supply 6, capable of generating a uniform electrostatic field with adjustable intensity; the output voltage range of the high-voltage power supply 6 is -30kV to +30kV.

[0022] In this embodiment, the high-voltage electrode plate 5 is an annular copper electrode plate, which is sleeved on the outer wall of the powder feeding pipe so that the powder flow can pass through the center of the electrode plate; the distance between the two high-voltage electrode plates 5 is 10mm.

[0023] The machine vision monitoring unit in this embodiment includes a high-speed camera 11 and an infrared thermal imager 13. The high-speed camera 11 is used to acquire powder flow information, and the infrared thermal imager 13 is used to monitor the information of the molten pool 9. The high-speed camera 11 is equipped with a backlight 12 (LED) so that it can clearly capture the contour, concentration, and trajectory of the powder flow. The information of the molten pool 9 monitored by the infrared thermal imager 13 includes the temperature field distribution of the molten pool 9 and the generation of spatter.

[0024] like Figure 1 As shown, the high-speed camera 11, the infrared thermal imager 13, and the high-voltage power supply 6 are all connected to the control unit 3. The control unit 3 includes an image processing module, an intelligent decision-making module, and a control output module. The image processing module can process the image data from the high-speed camera 11 and the infrared thermal imager 13 in real time, and extract key feature parameters of the powder flow and the molten pool 9 through algorithms such as background subtraction, threshold segmentation, and optical flow. Among them, the key feature parameters of the powder flow include the powder flow velocity distribution and divergence angle, and the key feature parameters of the molten pool 9 include the number of splashed particles and the trajectory of the splashed particles in the molten pool 9 region.

[0025] The intelligent decision-making module incorporates an expert database and optimization algorithms (such as PID controllers, fuzzy logic, or machine learning models). It can compare extracted key feature parameters with preset ideal process windows and generate optimal electric field strength control commands. Furthermore, when increased splashing is detected, a negative voltage command is output to slow down the powder; when powder flow dispersion or insufficient energy is detected, a positive voltage command is output to accelerate and focus the powder flow.

[0026] The control output module is used to send the electric field strength control command to the high voltage power supply 6 in real time, adjust its output voltage, thereby changing the electric field strength and completing one closed-loop control cycle.

[0027] This embodiment dynamically adjusts the electric field strength by acquiring image data, thereby achieving intelligent closed-loop control of powder velocity, effectively suppressing splashing, improving molding quality and powder utilization.

[0028] Example 2: This embodiment provides a method for controlling the speed of additive manufacturing powder using an electric field based on visual feedback, employing the system described in Embodiment 1, including: Acquire visible light images of the powder flow and infrared thermal images of the molten pool 9, and extract key feature parameters of the powder flow and molten pool 9; By comparing key characteristic parameters with a preset ideal process window, the optimal electric field strength control command is generated. Output electric field strength control command to change the electric field strength applied to the powder.

[0029] Specifically, such as Figure 2 As shown, it includes the following steps: Step 1: Set process parameters, including target powder velocity, splash tolerance threshold, ideal temperature range of molten pool 9, etc.

[0030] Step 2: Start the high-speed camera 11 and the infrared thermal imager 13 to simultaneously acquire visible-light images of the powder flow and infrared thermal images of the molten pool 9.

[0031] Step 3: The image processing module processes the image data in real time. Through algorithms such as background subtraction, threshold segmentation, and optical flow, it extracts key feature parameters such as powder flow velocity distribution, powder flow divergence angle, number and trajectory of splashed particles in the molten pool 9 region, and quantifies the powder flow state and the stability of the molten pool 9.

[0032] Step 4: The intelligent decision-making module switches between splash suppression mode, flow rate compensation mode and steady-state maintenance mode based on the input key characteristic parameters, and generates output instructions using different control strategies.

[0033] like Figure 2As shown, the triggering conditions for splash suppression mode, flow rate compensation mode and steady-state maintenance mode are determined sequentially.

[0034] Traditional single control modes cannot simultaneously handle different types of operating conditions. For example, suppressing splash requires rapid deceleration, while improving efficiency requires precise acceleration. These two objectives will seriously conflict in a single control loop. Therefore, this embodiment provides an intelligent mechanism with priority judgment capabilities, which dynamically switches the most suitable control strategy based on real-time process status (such as whether splash occurs, whether the speed deviates, and whether the operation is stable), thereby realizing the leap from "passive reaction" to "active decision-making".

[0035] The triggering conditions for splash suppression mode include: Primary trigger condition: Number of splash particles extracted from the Molten Pool 9 image ( If the dynamic safety threshold is exceeded continuously for M consecutive control cycles (e.g., M=3), This means switching to splash suppression mode. The continuous triggering mechanism effectively avoids misjudgments caused by noise in a single frame of image. Auxiliary triggering conditions: Switching to splash suppression mode occurs when the aspect ratio of the molten pool 9 suddenly increases and exceeds a threshold (indicating instability in the molten pool 9 and impending splashing), or when an extra-large (greater than 200µm) splash particle is detected. The triggering condition for splash suppression mode has the highest priority; once the condition is met, the system immediately interrupts other current control modes and forces entry into splash suppression mode.

[0036] The control strategy for splash suppression mode is as follows: Control objective: Suppress splash as quickly as possible, with priority over speed tracking.

[0037] Based on the control objective of the splash suppression mode, its control algorithm adopts a hybrid algorithm combining nonlinear Bang-Bang control (also known as hysteresis control or ripple regulator control) and proportional control (P). It can provide a suppression force that is precisely matched with the real-time deviation, and achieve fine adjustment of the force that can be controlled. It includes: the first stage (Bang-Bang emergency braking): once entered, the control unit 3 immediately outputs a pre-calibrated, fixed maximum deceleration voltage command. This command is sufficient to generate a strong reverse electric field force in the shortest time, so that the powder flow decelerates sharply. This stage lasts for K control cycles (e.g., K=2) and aims to quickly suppress the splashing trend.

[0038] Second stage (proportional adjustment and stabilization): After emergency braking, the controller switches to proportional control; its input is the deviation between the number of splashes and the threshold. The output is .in, This is the proportional gain, which is typically large to ensure strong suppression of splash. At this point, the controller's goal is to... It tends to zero, rather than the speed deviation.

[0039] The exit condition for splash suppression mode is: when (Compare The control unit 3 can exit only after the aspect ratio of the molten pool 9 returns to the normal range (e.g., 5 control cycles) and the speed returns to a lower safety recovery threshold. After exiting, the control unit 3 does not immediately return to the original target speed, but enters the "speed ramp recovery" process, gradually increasing the target speed from the current lower value to the original set value at a preset, slower slope, so as to avoid sudden speed changes from impacting the molten pool 9 again.

[0040] For the flow rate compensation mode, the triggering condition is: Condition B1 (Insufficient velocity): Actual average velocity of powder flow ( ) and target speed ( The difference between the two values ​​consistently exceeds the tolerance range; Condition B2 (Stream divergence): The divergence angle of the powder stream ( () greater than the maximum allowable divergence angle () This indicates poor powder focusing and reduced utilization. Condition B3 (Insufficient flow rate estimation): The instantaneous mass flow rate estimated based on velocity and flow density is lower than the process requirement value.

[0041] The control strategy for the flow rate compensation mode is as follows: Target control: Ensure that the actual velocity accurately tracks the target velocity and maintains the focused stream.

[0042] Control algorithm: The system adopts parameter self-tuning PID control, which has parameter self-tuning function and meets the requirements of accurate and stable tracking of target speed in flow rate compensation mode.

[0043] For condition B1: the input is the speed deviation. The output is .

[0044] For condition B2: the input is the divergence angle deviation. The output is Increasing the voltage can usually both accelerate and focus the signal, so the two outputs are usually positively correlated.

[0045] Synthesized Output: The final voltage command is a weighted sum of the two: , where w is the weighting coefficient.

[0046] PID parameters ( The flow rate is not fixed and is adjusted online based on the real-time quality flow rate. The higher the flow rate, the greater the system inertia, therefore the flow rate is automatically increased. (Differential gain) is adjusted to suppress overshoot. (Proportional gain) to avoid oscillation.

[0047] For the steady-state maintenance mode, both the triggering and maintenance conditions must be met simultaneously: (Speed ​​deviation is extremely small); This indicates a speed deviation, typically 0.15. (Ideal stream shape); This indicates the stream divergence angle, typically 0.5°. (Almost no splashing); The area and temperature fluctuation of the molten pool 9 are within the allowable range.

[0048] The control strategy for maintaining steady-state mode is as follows: Control objectives: Maintain the current excellent state while reducing control actions, saving energy and extending the life of the high-voltage power supply.

[0049] Control algorithm: Integral separation PID control with dead zone is adopted.

[0050] Dead Zone: A dead zone is set near steady state. When the speed deviation is within this dead zone, the controller output is zero, avoiding unnecessary and minute voltage adjustments, thereby eliminating high-frequency oscillations. Integral Separation: When entering steady-state mode, the integral term (I term) is cleared. This prevents overshoot caused by integral values ​​that may have accumulated before entering steady state. At this time, the controller actually only operates within a very small proportional band (P), maintaining smooth and stable system operation.

[0051] The goal of steady-state maintenance mode is to maintain stability to the maximum extent and save energy and reduce consumption when the system has reached its optimal operating condition. It adopts PID with dead zone integral separation to achieve the goals of energy saving, consumption reduction and extending equipment life by actively reducing intervention.

[0052] Step 5: Cyclic execution: Repeat steps 2 to 4 to achieve real-time, adaptive closed-loop control of the entire manufacturing process.

[0053] This embodiment, by simultaneously monitoring the powder flow and the molten pool 9, can not only control the speed but also indirectly ensure the quality of the molten pool 9, achieving synergistic optimization of multiple process parameters. Furthermore, image-based processing and decision-making are completed within milliseconds, which can promptly suppress splashing, compensate for flow rate fluctuations, and significantly improve the surface quality and internal density of the deposited layer.

[0054] This embodiment forms an intelligent control strategy with multimodal input, multi-objective optimization, and multi-priority decision-making, realizing adaptive, high-precision, and robust control of powder speed.

[0055] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A system for controlling the speed of additive manufacturing powder using an electric field based on visual feedback, characterized in that, include: The laser head is coaxially arranged with the powder feeding nozzle, the powder feeding nozzle is connected to the powder feeder, and a pair of high-voltage electrode plates are installed inside the powder feeder. The high-voltage electrode plates are connected to a high-voltage power supply. The machine vision monitoring unit includes a high-speed camera and an infrared thermal imager. The high-speed camera is used to collect powder flow information, and the infrared thermal imager is used to monitor molten pool information. The control unit processes information collected by the high-speed camera and infrared thermal imager in real time and transmits the processed information to the high-voltage power supply to change the powder speed by adjusting the output voltage.

2. The system for controlling the speed of additive manufacturing powder based on visual feedback using an electric field, as described in claim 1, is characterized in that... The high-speed camera is equipped with a backlight.

3. The system for controlling the speed of additive manufacturing powder based on visual feedback using an electric field, as described in claim 1, is characterized in that... The powder flow information includes the powder flow profile, concentration, and trajectory; the molten pool information includes the molten pool temperature field distribution and spatter generation.

4. The system for controlling the speed of additive manufacturing powder based on visual feedback using an electric field, as described in claim 1, is characterized in that... The control unit includes an image processing module, an intelligent decision-making module, and a control output module. The image processing module is used to process image data and extract key characteristic parameters of powder flow and molten pool. The intelligent decision-making module generates an electric field strength control command by comparing the key characteristic parameters of the powder flow with a preset ideal process window; the control output module is used to send the electric field strength control command to the high-voltage power supply.

5. A method for controlling the speed of additive manufacturing powder using an electric field based on visual feedback, characterized in that, The system described in any one of claims 1-4 includes: Acquire visible light images of the powder flow and infrared thermal images of the molten pool, and extract key feature parameters of the powder flow and molten pool; By comparing key characteristic parameters with a preset ideal process window, the optimal electric field strength control command is generated. Output electric field strength control command to change the electric field strength applied to the powder.

6. The method for controlling the powder speed in additive manufacturing based on visual feedback using an electric field, as described in claim 5, is characterized in that... Based on the input key feature parameters, it switches between three modes: splash suppression, flow rate compensation, and steady-state maintenance.

7. The method for controlling the powder speed in additive manufacturing based on visual feedback using an electric field, as described in claim 6, is characterized in that... When the number of spatter particles extracted from the infrared thermal image of the molten pool continuously exceeds the dynamic safety threshold within a continuously set control cycle, switch to spatter suppression mode. Alternatively, when the aspect ratio of the molten pool suddenly increases and exceeds the threshold, switch to splash suppression mode.

8. The method for controlling the powder speed of additive manufacturing based on visual feedback using an electric field according to claim 7, characterized in that, The control process in the splash suppression mode is divided into a first stage and a second stage, wherein the first stage is emergency braking and the second stage is proportional regulation to stabilize the flow.

9. The method for controlling the powder speed of additive manufacturing based on visual feedback using an electric field, as described in claim 6, is characterized in that... When the difference between the actual average velocity of the powder flow and the target velocity continues to exceed the tolerance range, switch to flow rate compensation mode. Alternatively, when the powder flow divergence angle is greater than the maximum allowable divergence angle, switch to flow rate compensation mode; Alternatively, if the instantaneous mass flow rate estimated based on velocity and flow density is lower than the process requirement value, switch to flow rate compensation mode.

10. The method for controlling the powder speed of additive manufacturing based on visual feedback using an electric field, as described in claim 6, is characterized in that... When the following conditions are met simultaneously: the velocity deviation is less than the set value, the flow pattern is ideal, there is no splashing, and the molten pool area and temperature fluctuation are within the allowable range, switch to steady-state maintenance mode.

Images