A tool control method, an additive printing control method and an additive processing device

Abstract

The application discloses a tool control method, which is applied to an additive manufacturing device, the additive manufacturing device comprising a base material and a wire feeding mechanism for conveying a wire material, and the wire material is controlled to contact the base material at different positions for multiple times; when the wire material is separated from the base material, a contact signal of the wire material contacting the base material is obtained through a detection circuit to form a relative position parameter; through a plurality of relative position parameters, data for representing a surface geometric feature of the base material is generated, and the additive manufacturing device is controlled to perform a tool control operation according to the data of the surface geometric feature of the base material; the application forces the wire material to be separated from the base material after each contact, and the separation state is maintained before the relative position is changed, so that sliding friction in continuous motion is completely avoided. Through repeating the discrete event sequence, the contact parameters can be recorded at multiple positions, and the wire material itself is used as a probe.

Description

Technical Field

[0001] This invention relates to the field of additive manufacturing technology, specifically to a tool setting control method, an additive printing control method, and an additive manufacturing apparatus. Background Technology

[0002] Wire-feed additive manufacturing technology uses wire as raw material, melting and depositing it layer by layer through high-energy heat sources such as lasers, electric arcs, or electron beams to achieve rapid prototyping of metal components. Compared with traditional subtractive manufacturing and powder-feed additive manufacturing, this technology has significant advantages such as high material utilization, excellent forming efficiency, stable mechanical properties of formed parts, and relatively controllable equipment costs. It has been widely used in high-end manufacturing fields such as the manufacturing of key aerospace components, the repair and remanufacturing of mechanical equipment, mold surface modification, and the production of automotive parts.

[0003] However, laser filament additive manufacturing, as a three-dimensional layer-by-layer deposition forming technology, relies heavily on the precise control of dynamic parameters such as the filament's position, the timing of laser-filament interaction, and the molten pool morphology to ensure high-quality processing. Deviations can easily lead to defects such as filament pilling, breakage, and poor fusion, even causing processing failure. To address these process requirements, numerous studies in recent years have attempted to improve processing controllability through sensing and monitoring methods. For example, Chinese invention patent CN121945999A discloses a method and system for adjusting the transition state between the filament and substrate during laser coaxial filament melting. This method determines the transition state and adjusts the filament feeding speed by real-time acquisition of the filament feeding speed and the resistance value between the filament and the substrate. Chinese invention patent CN117445404A discloses an in-situ monitoring device for the state of filament in 3D printing extrusion forming, which uses an angle sensor to monitor the filament state.

[0004] However, a comprehensive analysis of the existing technologies reveals the following significant shortcomings in the current field of wire-feed additive manufacturing. Firstly, existing solutions often rely on indirect methods such as resistance monitoring, visual image analysis, or displacement sensors to infer the wire's state. These methods are easily affected by factors in the processing environment, such as dust, strong light from the molten pool, and electromagnetic interference. They struggle to reliably and stably obtain the true contact state between the wire and the substrate under harsh conditions, and their response speed often fails to meet the real-time control requirements of high-precision additive manufacturing. Secondly, most existing solutions only preset the initial distance between the wire feed port and the substrate through the equipment positioning system, directly using this preset value as the dry extension parameter throughout the operation without verifying or correcting the actual state of this parameter. Mechanical assembly errors, movement gaps, and thermal deformation cause the actual dry extension to deviate from the theoretical value, and this deviation cannot be detected by the system, leading to a series of process defects such as insufficient or excessive melting of the wire. Third, geometric errors on the substrate surface (such as roundness deviation of cylindrical substrates and flatness error of flat plates) directly affect the actual value of dry elongation. However, existing technologies generally regard substrate morphology detection and wire feeding control as two independent links. Morphology data cannot be fed back to the wire feeding control parameters in real time for compensation and adjustment, which makes it difficult to guarantee the dimensional accuracy of the molded parts. In particular, interlayer errors will continue to accumulate in multilayer additive manufacturing.

[0005] In summary, existing wire-feeding additive manufacturing technologies still have significant shortcomings in areas such as wire contact state detection, tool positioning accuracy, closed-loop control of wire extension, and adaptive wire feeding. There is an urgent need for a technical solution that can achieve precise wire contact detection, tool control based on measured data, and dynamically adjust wire feeding parameters accordingly, in order to improve the quality stability and automation level of wire-feeding additive manufacturing.

[0006] Therefore, this invention proposes a tool setting control method, an additive printing control method, and an additive processing apparatus. Summary of the Invention

[0007] To address the shortcomings of existing technologies, the purpose of this application is to provide a technical solution that enables precise contact detection of filaments, tool control based on measured data, and dynamic adjustment of filament feeding processing parameters, thereby improving the quality stability and automation level of filament feeding additive manufacturing.

[0008] The above-mentioned objective of this application is achieved through the following technical solution:

[0009] A tool setting control method is applied to an additive manufacturing apparatus, the additive manufacturing apparatus including a substrate and a filament feeding mechanism for conveying filaments, the additive manufacturing apparatus further having a detection circuit for detecting the contact state between the filaments and the substrate; the tool setting control method includes the following steps:

[0010] The filament is controlled to make multiple contacts with the substrate at different positions; when the filament is separated from the substrate, the contact signal obtained by the detection circuit is used to form a relative position parameter; through several of the relative position parameters, data to characterize the geometric features of the substrate surface is generated, and the additive manufacturing device is controlled to perform tool setting control operation based on the data of the geometric features of the substrate surface.

[0011] Preferably, the contact position is determined according to at least one of the following methods:

[0012] Arranged at equal or unequal intervals along a predetermined trajectory;

[0013] Alternatively, when any two adjacent relative position parameters are obtained, the deviation between the two adjacent relative position parameters and the theoretical reference value is compared to adjust the target step size of the next contact position.

[0014] Preferably, the adjustment method includes:

[0015] The target step size is adjusted by using the curvature of the relative position parameters of two adjacent contact positions as the theoretical reference value.

[0016] Alternatively, a contour curve can be fitted based on the obtained relative position parameters as the theoretical reference value to adjust the target step size.

[0017] Preferably, when no contact signal is acquired within the predetermined maximum feed stroke, the target step size is shortened so that the target step size is less than the theoretical reference value.

[0018] Preferably, the substrate is a cylindrical substrate that can rotate about an axis; the method of changing the relative spatial position between the filament and the substrate includes:

[0019] The substrate is rotated about its axis by a predetermined angle.

[0020] Alternatively, the substrate may be translated a predetermined distance along a radial straight line.

[0021] Alternatively, the wire feeding mechanism or laser head carrying the filament may be moved relative to the substrate by a predetermined step length.

[0022] An additive printing control method is applied to an additive manufacturing apparatus, the additive manufacturing apparatus including a substrate rotatable about an axis, a filament feeding mechanism, a laser, and a detection circuit. The additive printing control method includes the following steps: employing the tool setting control method described above; calculating the target dry elongation or target contact position between the filament and the substrate using data characterizing the surface geometry of the substrate; controlling the filament feeding mechanism to feed the filament by calculating the target dry elongation or target contact position at different time steps; and controlling the laser to perform a printing operation when a contact signal is obtained indicating that the filament is in contact with the substrate.

[0023] Preferably, the output power of the laser and / or the wire feeding rate of the wire feeding mechanism are adjusted according to the direction of fluctuation of the data of the geometric features of the substrate surface.

[0024] Preferably, the method for adjusting the output power of the laser and / or the wire feeding rate of the wire feeding mechanism is as follows: when the data of the geometric features of the substrate surface shows that the diameter of a certain circumferential position of the substrate is too large, the wire feeding rate or the output power at that position is reduced to reduce the thickness of the cladding layer; when the data of the geometric features of the substrate surface shows that the diameter of a certain circumferential position of the substrate is too small, the wire feeding rate or the output power at that position is increased to increase the thickness of the cladding layer.

[0025] An additive manufacturing apparatus, comprising:

[0026] Substrate;

[0027] Laser, used to heat filament;

[0028] The wire feeding mechanism is used to transport wire.

[0029] A detection circuit is electrically connected between the filament and the substrate to detect the contact state between the filament and the substrate and generate a contact signal.

[0030] A motion mechanism for changing the relative spatial position between the filament and the substrate;

[0031] A controller, communicatively connected to the filament feeding mechanism, the detection circuit, the motion mechanism, and the laser, is configured to execute either the tool setting control method described above or the additive printing control method described above. The filament feeding mechanism includes a drive motor for controlling filament feeding, and the drive motor is communicatively connected to the controller.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0033] I. Frictionless Multi-Point Contour Measurement: Existing tool setting methods mostly rely on single-point positioning or external ranging sensors, failing to acquire multi-point topographic data without damaging the filament and substrate. This invention forces the filament to detach from the substrate after each contact, maintaining this detached state before changing its relative position, completely avoiding sliding friction during continuous movement. By repeating this discrete event sequence, contact parameters can be recorded at multiple locations, using the filament itself as a probe to achieve low-cost scanning measurement of the substrate's surface geometry.

[0034] II. Adaptive Step-Length Sampling Strategy: Traditional equal-interval sampling methods cannot dynamically adjust according to the actual shape, resulting in data redundancy or omissions. This invention allows sampling points to be arranged according to a predetermined trajectory or at unequal intervals, and can further adjust the step length of the next contact position in real time based on the measured data: automatic densification in areas with large deviations or steep curvature, and sparse sampling in flat areas, significantly improving the targeting and efficiency of contour measurement.

[0035] III. Differentiated Wire Feeding Control Based on Measured Roundness: Traditional wire feeding control relies solely on preset parameters, failing to compensate for roundness deviations or interlayer cumulative errors in the substrate itself. This invention directly uses roundness data obtained through the tool setting method to calculate the target extension or target contact position, and accordingly employs different wire feeding speeds or laser powers at different circumferential positions: thinning the cladding layer where the diameter is larger, and thickening the cladding layer where the diameter is smaller. This closed-loop control strategy makes the outer diameter of the formed part more uniform, improving the dimensional accuracy of multilayer additive manufacturing. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the circuit required for the tool setting control method of the present invention.

[0037] Figure 2 This is a schematic diagram of Embodiment 1 of the present invention.

[0038] Figure 3 This is a timing diagram of one embodiment of the present invention.

[0039] Figure 4 This is a timing diagram of another embodiment of the present invention.

[0040] Figure 5 This is a timing diagram of another embodiment of the present invention.

[0041] Figure 6 This is a timing diagram of one embodiment of the present invention, as shown in Embodiment 2.

[0042] Figure 7 This is a timing diagram of another embodiment in Embodiment 2 of the present invention.

[0043] Figure 8This is a timing diagram of another embodiment in Embodiment 2 of the present invention.

[0044] Figure 9 This is a timing diagram of another embodiment in Embodiment 2 of the present invention.

[0045] Figure 10 This is a timing diagram of the implementation method described in Embodiment 3 of the present invention.

[0046] Figure 11 This is a timing diagram of the implementation method described in Embodiment 4 of the present invention.

[0047] Figure 12 This is a three-dimensional schematic diagram of the basic hardware devices involved in this invention.

[0048] Figure 13 This is a schematic diagram of the wire feeding mechanism in this invention from one perspective.

[0049] Figure 14 This is a schematic diagram of the wire feeding mechanism in this invention from another perspective.

[0050] Reference numerals in the attached drawings: 1. Wire material; 2. Conductive nozzle; 3. Wire feeding assembly; 4. Reducer; 5. Drive motor; 6. External encoder; 7. Mounting frame; 8. Communication interface; 9. Clamping mechanism; 10. Rotary motor; 11. Motion mechanism. Detailed Implementation

[0051] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0052] Example 1. This example provides a tool setting control method applied to a wire-feeding additive manufacturing apparatus. (See attached diagram.) Figure 1 As shown, the device includes: a filament, a filament feeding mechanism, a cylindrical substrate, an electric slip ring, a detection circuit, and a hot wire circuit. The cylindrical substrate is rotatable around its axis, and the electric slip ring is mounted on the end of the rotating shaft of the cylindrical substrate to maintain the electrical continuity between the detection circuit and the hot wire circuit during rotation. It should be noted that this embodiment focuses on the tool setting control method; the specific operation of the hot wire circuit will be detailed in Embodiment Four. Here, only the hardware configuration is described.

[0053] See attached document Figure 2The circuit diagram shown illustrates that the core of the detection loop is the TouchSense module. This module is powered by a low-voltage power supply and integrates a constant current source, sampling resistor, and signal conditioning circuitry, outputting a constant 100mA current to the detection loop. The specific path of the detection loop is as follows: the OUT terminal of the TouchSense module is connected via a wire to the conductive nozzle in the wire feeding mechanism, then through the wire, the cylindrical substrate, and the slip ring, finally returning to the GND of the power supply. When wire 1 is not in contact with the cylindrical substrate, the loop is open, and the TouchSense module outputs a low level; when wire 1 contacts the cylindrical substrate, the loop closes, the 100mA constant current flows, the TouchSense module detects the voltage change, and its OUT signal jumps to a high level, which is then sent to the PLC's I / O input. Simultaneously, the PLC communicates with a host computer or other network devices via a PoE switch for data recording and process parameter distribution. The hot wire circuit includes a hot wire drive board powered by a 40V / 20A power supply, capable of outputting a preheating current of up to 100A to the wire. This circuit also forms a loop with the cylindrical substrate via an electric slip ring. The hot wire drive board has an ENABLE pin, controlled by a PLC, used to enable or disable the preheating output as needed, thereby achieving time-sharing operation with the detection circuit and avoiding interference from high current to the detection signal. An encoder is installed inside the wire feeding mechanism for real-time detection of the wire's feed and retraction lengths. The rotation of the cylindrical substrate is driven by an external servo motor or other rotating mechanism; the specific rotation angle can be controlled by the number of pulses emitted by the rotary encoder or the PLC.

[0054] Based on the hardware configuration described above, the tool setting control method of this embodiment will be explained in detail below. (Refer to the appendix.) Figure 1 As shown, the tool setting control method of this embodiment includes the following steps. First, the controller controls the wire feeding mechanism to slowly feed the wire toward the substrate surface. During this process, the controller continuously samples the output level signal of the detection circuit in real time. When the end of the wire makes physical contact with the substrate surface, the detection circuit closes, and the controller detects the rising edge jump of the contact signal, thereby determining that the wire and substrate are in contact. The controller records the relative position parameter corresponding to the current contact position, i.e., the first relative position parameter. This relative position parameter can be the total wire feed from the initial reference position (e.g., the zero-position reference of the wire feeding mechanism) to the current contact point, or the rotation angle of the drive motor calculated by the controller based on encoder feedback. This parameter directly characterizes the spatial position information of the substrate surface at the contact point. Subsequently, the controller controls the wire to disengage from the substrate. This action can be achieved by controlling the wire feeding mechanism to retract the wire a set distance (e.g., retract 0.5mm to 2mm), so that the end of the wire completely detaches from the substrate surface.

[0055] Furthermore, after the filament detaches from the substrate and remains detached, the controller changes the relative spatial position between the filament and the substrate, moving to a new contact position different from the first contact position, i.e., a second relative position. The method of changing the relative spatial position can be flexibly selected according to the actual device structure, such as driving the substrate to rotate (for cylindrical substrates), driving the substrate to translate (for flat substrates), or driving the motion platform of the filament feeding mechanism to move (for fixed substrate scenarios). After changing to the new position, the above feeding, contact detection, and detachment steps are repeated to obtain another relative position parameter for the second contact position, i.e., the second relative position parameter. Repeating this process yields relative position parameters for several discrete contact points distributed on the substrate surface. Based on the obtained first relative position parameters and multiple second relative position parameters, combined with the spatial coordinate information (e.g., rotation angle or translation distance) corresponding to each contact point, the controller generates data characterizing the geometric features of the substrate surface. These geometric feature data may include, but are not limited to: the roundness error curve of a cylindrical substrate, the flatness deviation distribution of a flat substrate, the location and magnitude of local protrusions or depressions, and pits or gaps reflected by contact loss (i.e., the wire feed reaching the preset maximum stroke without triggering a contact signal). Finally, based on the generated data characterizing the geometric features of the substrate surface, the controller performs tool setting control operations on the additive manufacturing apparatus. Taking the roundness error curve as an example, the tool setting control operations include, but are not limited to:

[0056] First, adjust the wire feeding parameters during subsequent additive manufacturing processes, such as wire feeding rate, retraction length, and / or wire feeding acceleration.

[0057] Second, adjust the laser processing parameters, such as laser output power, duty cycle, defocusing amount and / or start / stop timing;

[0058] Third, adjust the relative positional relationship between the filament and the substrate, such as correcting the zero-point offset or coordinate origin and / or generating a compensation path for subsequent additive manufacturing.

[0059] In one specific implementation, to prevent damage to the wire end or substrate surface due to sliding friction between the wire and the substrate during multi-point sampling, the controller executes the following discrete event sequence sequentially after each contact signal is received:

[0060] S1, Detachment and Holding: The moment the controller detects the contact signal, it immediately sends a retraction command to the wire feeding mechanism, causing the wire to retract a set distance, such as 0.5mm, to ensure that the end of the wire is completely detached from the substrate surface and to maintain this detached state.

[0061] S2, Relative Position Change: With the filament completely detached from the substrate and stationary, the controller drives the motion mechanism to move the substrate or filament feeding mechanism to the next preset sampling position. Since the filament has been retracted and detached, it never comes into contact with the substrate during the relative position change, thus avoiding frictional damage.

[0062] S3, Refeed: After the relative spatial position stabilizes, the controller once again controls the wire feeding mechanism to drive the wire to slowly feed towards the substrate, entering the next round of contact detection cycle.

[0063] It is worth noting that this embodiment ensures that each contact between the filament and the substrate during multi-point sampling is an independent, static, and frictionless contact event, thereby effectively protecting the geometry of the filament end and the integrity of the substrate surface. Furthermore, this discrete event sequence can be extended to substrates of various shapes. For example, for a flat substrate, the rotational action in S3 can be replaced by a translational action along the X-axis or Y-axis by a preset step length; and for scenarios where the filament feeding mechanism can move independently, the solution in S2 can be replaced by driving the filament feeding mechanism or laser head to move by a preset step length.

[0064] In this embodiment, one method for determining the second contact position is as follows: the sampling contact points on the substrate surface are distributed according to a preset trajectory. For cylindrical substrates, the preset trajectory can be evenly spaced along the circumference, for example, sampling one point every 15°, or it can be unevenly spaced, for example, performing dense sampling every 5° in the area near the estimated weld or defect, and sparse sampling in areas with relatively regular shapes and small deviations. For flat substrates, the preset trajectory can be a grid-like dot matrix on the XY plane, where the grid spacing can be uniformly set or non-uniformly arranged based on prior knowledge. Based on this preset trajectory, after measuring each contact point, the controller sequentially drives the motion mechanism to move the relative position of the wire and the substrate to the next preset coordinate position.

[0065] In this embodiment, another way to determine the second contact position is to dynamically adjust the target step size of the next contact position in real time based on the previous measurement results.

[0066] For example, after the controller performs the first round of circumferential sampling, if it finds that the relative position parameters of one sampling point and the next sampling point in the circumferential direction are small, that is, the contour deviation is gentle, it is judged that the surface shape in this area is relatively regular, and the sampling step size can be appropriately increased to reduce the number of measurements and improve efficiency. Conversely, if the relative position parameters of adjacent sampling points in a certain area are large, such as the deviation value jumps beyond the preset threshold, it indicates that there are significant protrusions, depressions or morphological changes in this area, and the controller will automatically shorten the step size and increase the sampling density in this area to obtain more refined contour details.

[0067] For example, when the controller performs wire feeding at a preset sampling position, if the wire feeds to the preset maximum stroke but does not trigger a contact signal, it is determined that there is a depression or missing material at that position. The controller immediately shortens the step size and automatically inserts a new sampling point near the current sampling position to detect the boundary and depth of the depression.

[0068] For example, the controller can be pre-set with a step size adjustment coefficient α (0 < α < 2, initial value α = 1). When the absolute difference between the relative position parameters of two adjacent sampling points and the theoretical reference value is less than the deviation tolerance, for example, 0.05 mm, the controller increases the step size coefficient α to 1.2, increasing the sampling step size; when the absolute difference exceeds the deviation tolerance, the controller decreases the step size coefficient α to 0.5, shortening the sampling step size to increase the sampling density.

[0069] This embodiment integrates discrete event sequences, multiple positional change methods, and detachment methods into a single implementation scenario, using a cylindrical substrate as an example for detailed explanation. (Refer to the appendix.) Figure 1 and Figure 3 As shown, the substrate is a cylindrical substrate that can rotate around an axis. A servo motor drive system is connected to the end of its rotating shaft, and a rotary encoder is provided for real-time feedback of the rotation angle. The wire feeding mechanism is installed in a fixed position, and the laser head is fixed relative to the wire feeding mechanism. The controller first controls the wire feeding mechanism to drive the wire towards the surface of the cylindrical substrate at a low speed. When the end of the wire contacts the surface of the cylindrical substrate, the detection circuit generates a contact signal. The controller captures this signal and immediately stops feeding, recording the current wire feeding length as the reference zero point L0. Subsequently, the controller executes a disengagement action: controlling the wire feeding mechanism 5 to reverse and retract a set distance S, so that the end of the wire 1 completely detaches from the surface of the cylindrical substrate 2. This retraction distance S should not be less than the sum of the expected surface roughness and the safety margin to ensure the reliability of the disengagement action, but it should not be too large to save reciprocating stroke time.

[0070] It is worth noting that the controller can also execute the following loop according to a preset circumferential sampling scheme, such as equidistant sampling: sampling point number N=36, one sampling point every 10°:

[0071] S1. The controller controls the cylindrical substrate 2 to rotate around its axis by a preset angle Δθ. After reaching the position, the position is confirmed by the angle feedback of the rotary encoder.

[0072] S2, Wire feeding: The controller controls the wire feeding mechanism 5 to drive the wire 1 to feed onto the surface of the cylindrical substrate 2;

[0073] S3. Contact Detection and Recording: When the end of the wire 1 touches the surface of the cylindrical substrate 2, a contact signal is triggered, and the controller records the current wire feed amount L. i and the current rotation angle θi ;

[0074] S4. Withdrawal and separation: The controller immediately controls the wire feeding mechanism 5 to withdraw a set distance S, so that the wire 1 is separated from contact.

[0075] S5. Loop judgment: If i < N, return to the previous step S1 and rotate to the next sampling angle; if i = N, exit the loop.

[0076] After the above loop, the controller obtains N groups of data pairs. If the cylindrical substrate 2 is an ideal regular cylinder and is installed concentrically, then all L i should be equal; if there are actual roundness deviations, eccentricities or surface defects, then L i will change with the angle θ i . In other words, the controller can compare the feed length L i at each sampling point with the theoretical reference value L_ref (which can be obtained from the standard position measured during initial tool setting or the theoretical value of the CAD model), and calculate the radial deviation ΔR i = L i - L ref . Plotting the change of ΔR i with θ i as a curve can obtain the roundness error distribution curve of the cylindrical substrate 2, visually showing the convex and concave regions on its circumference. The controller can also calculate roundness error statistics based on the distribution of ΔR i , such as the maximum deviation, minimum deviation, roundness tolerance value, etc. According to the obtained roundness error curve, the controller performs one or a combination of the following tool setting control operations:

[0077] First, if the average value of the feed length at each sampling point deviates from the theoretical reference value, indicating that there is a systematic deviation in the zero position of the wire end relative to the substrate surface, the controller automatically corrects the zero coordinates;

[0078] Second, during subsequent processing, dynamically adjust the radial distance between the wire feeding port and the cylindrical substrate according to the roundness error curve, so that the actual dry elongation remains constant at different angles;

[0079] Third, the controller superimposes the roundness error curve onto the motion control instructions for subsequent multi-layer additive manufacturing, so that the cladding deposition of each layer adaptively compensates for the initial shape error of the substrate.

[0080] It should be noted that during the entire sampling process, since the wire is withdrawn and separated after each contact, the wire never contacts the substrate during the rotation of the substrate, thus avoiding sliding friction. The controller generates roundness error data for the circumferential surface of the substrate based on the obtained multiple relative position parameters, and accordingly performs tool setting control operations, such as correcting the zero offset during subsequent processing or adjusting the motion trajectory.

[0081] It should be further noted that when the controller completes the contact measurement of the first sampling point and obtains the relative position parameter L1, and based on L1 and the theoretical reference value L... ref After calculating the next rotation step size Δθ based on the deviation, the following comparison strategy can be further executed: If the absolute value of the deviation is less than a preset threshold, it indicates that the contour of the region is regular, and the controller sets Δθ to a larger value to reduce the number of sampling points and improve efficiency; if the absolute value of the deviation is greater than the threshold, it indicates that there are significant protrusions or depressions in the region, and the controller reduces Δθ to a smaller value to densify sampling and obtain more refined contour data. After completing the measurement of the second sampling point and obtaining L2, the controller readjusts the next rotation step size based on the rate of change of L2 and L1 or the deviation from the theoretical reference value. This process is repeated until the entire circumference is covered, and its overall timing is as follows: Figure 4 As shown. This adaptive step-size sampling strategy allows for sparse sampling in regions with gentle contours and dense sampling in regions with dramatic contour changes, ensuring both the accuracy of roundness measurement and saving tool setting time.

[0082] In another embodiment, this embodiment will further describe the position change method and the disengagement action method. These methods can be combined to adapt to different equipment configurations and processing scenarios.

[0083] First, when the substrate is cylindrical and can rotate around its axis, the controller drives the substrate to rotate around its axis by a preset angle. This method is suitable for additive manufacturing of rotating parts, such as cladding repair of shafts and rollers.

[0084] Second, when the substrate is a flat plate or cuboid and mounted on an XY linear motion platform, the controller drives the substrate to translate a preset distance along a radial linear direction. This method is suitable for grid point sampling on flat substrates, such as substrate flatness detection before large-area cladding.

[0085] Third, when the substrate is fixed in place, and the wire feeding mechanism or the laser head is integrated onto the motion mechanism, the controller drives the wire feeding mechanism or laser head to move relative to the substrate by a preset step length. This method is suitable for scenarios where the substrate is large, heavy, and difficult to move, such as the partial repair of large molds.

[0086] Fourth, the wire feeding mechanism is controlled to reverse and retract the wire by a set distance. Upon detecting a contact signal, the controller immediately sends a reverse pulse to the drive motor of the wire feeding mechanism, causing the wire feeding wheel to reverse and retract the wire by, for example, 0.5mm to 2mm. The retraction distance can be precisely controlled by an encoder. The advantages of this method are fast response, short stroke, and no dependence on the substrate's motion mechanism.

[0087] Fifth, control the substrate away from the filament. When the filament feeding mechanism lacks a retraction function or its retraction accuracy is insufficient, the controller can drive the motion platform carrying the substrate to move a set distance away from the filament, thereby detaching the filament from the substrate. This method is suitable for scenarios where the filament feeding mechanism has a simple structure and can only feed filament in one direction, but it requires the substrate motion platform to have a rapid response capability.

[0088] In practice, one of the above methods can be selected based on the equipment hardware configuration, or both methods can be used simultaneously to accelerate the detachment speed. Taking a device with a cylindrical substrate, a fixed wire feeding mechanism, and a rotatable and radially micro-movable substrate as an example, the controller can use substrate rotation to change its circumferential position and move the substrate away from the wire to achieve the detachment action, while simultaneously using method one as a backup detachment method. Please refer to [link to relevant documentation]. Figure 5 The specific workflow for this scenario is as follows:

[0089] S1. The wire is fed to the surface of the substrate, and the detection circuit triggers a contact signal.

[0090] S2. The controller records the current rotation angle and wire feed length.

[0091] S3. The controller drives the substrate to move radially away from the filament by a set distance, for example, 0.3 mm, so that the filament separates from the substrate;

[0092] S4. The controller drives the substrate to rotate around the axis by a preset angle, such as 10°;

[0093] S5. The controller drives the substrate to move radially closer to the filament, restoring it to its initial radial position;

[0094] S6. Repeat the above steps to perform the next round of feed and contact detection.

[0095] This combination fully leverages the multi-degree-of-freedom advantage of the substrate motion platform, avoiding frequent wire retraction, and is especially suitable for situations where the wire is relatively soft and repeated retraction can easily lead to bending and deformation.

[0096] Example 2. Based on the tool control method described in Example 1, this example further details the specific implementation of dynamically adjusting the target step size. This example presents four dynamic adjustment methods that can be used individually or in combination, respectively adjusting the sampling step size in real time based on the magnitude of the deviation, curvature change, fitting error, and contact loss response. The technical principles and implementation steps of each method are described below.

[0097] In one implementation, such as Figure 6As shown, the controller dynamically adjusts the target step size for the next contact position based on the absolute value of the deviation between the obtained relative position parameters and the theoretical reference value. Specifically, a theoretical reference value needs to be determined first. This theoretical reference value can be the standard feed length obtained through a single contact measurement at a reference position on the substrate surface at the start of the tool setting process, or it can be a theoretical radial dimension preset based on the substrate's CAD model or process documents. The controller denotes this theoretical reference value as L. ref During multi-point sampling, after each sampling point is measured, the relative position parameter L of that point is obtained. i Then, the controller calculates the deviation value Δ i =|L i -L ref |. Then, according to Δ i The size is compared with a preset threshold to determine the next sampling step size (e.g., the rotation angle Δ of the cylindrical substrate). θ , or the translational distance Δ of the flat substrate d ):

[0098] For example, when Δ i When the value is greater than the first threshold (e.g., 0.10 mm), it indicates that there is a significant protrusion or depression at that location, and the geometry of the substrate surface deviates significantly from the theoretical value. The controller automatically shortens the sampling step size, for example, by reducing the step size to 50% of the original step size, in order to increase the density of measurements in subsequent sampling and to capture the contour changes of the area more precisely.

[0099] For example, when Δ i When the value is less than the second threshold (e.g., 0.02 mm), it indicates that the position is very close to the theoretical value and the surface is relatively regular. The controller automatically increases the sampling step size, for example, by increasing the step size to 150% of the original step size, in order to reduce the number of sampling points and improve tool setting efficiency.

[0100] For example, when Δ i When the value is between the first threshold and the second threshold, the controller maintains the current step size.

[0101] It should be noted that the first and second thresholds can be flexibly set according to factors such as the processing accuracy requirements of the substrate, the wire diameter, and the expected contour error. For high-precision additive manufacturing scenarios, the thresholds can be set smaller, and the step size adjustment range can be smoother; for roughing or rapid tool setting scenarios, the thresholds can be appropriately relaxed. The controller can set the maximum and minimum limits for the step size to prevent excessively large step sizes from causing details to be missed, or excessively small step sizes from resulting in too many sampling points and excessively long sampling time.

[0102] In another implementation, such as Figure 7As shown, the controller dynamically adjusts the target step size for the next contact position based on the curvature or rate of change reflected by the relative position parameters of two or more adjacent contact positions. This method is particularly suitable for substrate surfaces with drastic curvature changes, such as steps, edges, and local pits. Specifically, the controller obtains the relative position parameters L of three consecutive sampling points... i-1 、L_i、L i+1 After obtaining the corresponding spatial coordinates, the curvature or rate of change of the region's contour can be approximately calculated. A simplified method is to calculate the slope change between two adjacent points: Let k1 = (L... i -L i-1 ) / (θ i -θ i-1 ), k2=(L i+1 -L i ) / (θ i+1 -θ i If ), then the curvature representation can be defined as |k2-k1| / (θ) i+1 -θ i-1 When the curvature representation exceeds a preset gradient threshold, it indicates a sharp change in the contour in that region, such as convexity, depression, or edge. The controller then automatically shortens the sampling step size for the next step to increase sampling density in that region and obtain more refined contour information. Conversely, when the curvature representation is less than the gradient threshold, it indicates a gentle contour change, and the controller can increase the step size or maintain the original step size. The advantage of this method is that it does not rely on theoretical benchmark values ​​but rather adaptively adjusts based on the local variation characteristics of the measured data itself. Therefore, even if the substrate has overall eccentricity or systematic offset, this method can still effectively identify local morphological features.

[0103] In another implementation, such as Figure 8 As shown, the controller fits the contour curve of the substrate surface based on the obtained relative position parameters and evaluates the fitting error in local areas of the fitted curve. When the fitting error exceeds the preset accuracy requirement, the sampling point density in that local area is increased (i.e., the step size is shortened). The principle of this method is that during the measurement process, as the number of sampling points increases, the controller uses interpolation or spline fitting algorithms in real time to generate an approximate substrate contour curve. For the sampled area, the error between the fitted curve and the actual measurement points can be calculated. If the fitting error in a certain local area is large, it indicates that the contour change in that area has not been fully represented by the existing sampling points, and new sampling points need to be added in that area. The controller can automatically backtrack to that area and insert additional sampling points with a smaller step size until the local fitting error meets the accuracy requirement. The typical implementation process of this method is as follows:

[0104] S1. The controller initially adopts a large step size, for example, sampling one point every 30° for the circumference of the cylindrical substrate, quickly completing the initial sampling of the entire circumference and obtaining a sparse set of contour points.

[0105] S2. The controller performs spline interpolation on the sparse point set to generate a continuous fitted contour curve.

[0106] S3. The controller calculates the fitting error at each original sampling point, that is, the difference between the measured value and the fitted value, and finds the local area with the largest fitting error.

[0107] S4. If the maximum fitting error exceeds the preset accuracy threshold, the controller inserts a new sampling point between two adjacent points in the region with the maximum error, that is, shortens the step size of the segment, and drives the substrate or wire feeding mechanism to move to the position for supplementary measurement.

[0108] S5. Add the supplementary measurement points to the point set, refit the contour curve, and re-evaluate the fitting error.

[0109] S6. Repeat S4 to S5 until the maximum error of the entire fitted contour is less than the accuracy threshold, or the preset maximum number of sampling points is reached.

[0110] In another implementation, such as Figure 9 As shown, in response to the failure to obtain a contact signal within the predetermined maximum feed stroke, the controller automatically shortens the target step length to detect the boundary or depth of the depression. Specifically, when the filament is fed to the substrate surface at a preset feed speed, if the detection circuit is not triggered even after feeding to the preset maximum allowable stroke (e.g., 1.5 times the theoretical dry elongation), the controller determines that a depression, gap, or material loss exists at the sampling location. In this case, simply recording "no contact" does not provide sufficient morphological information; further detection of the boundary and depth of the depression area is required. The controller then performs the following operations:

[0111] S1. Immediately stop the current feed and pull the wire back to a safe position.

[0112] S2. Reduce the sampling step size to 1 / 3 or less of the original step size, and then perform encrypted sampling on both sides of the current missing position, that is, in the front and back directions, to determine the starting and ending boundaries of the depression.

[0113] S3. Regarding the depth of the depression, the controller can attempt to feed at multiple positions within the depression area, recording the feed length from the start of feeding to the trigger contact signal, i.e., the feed length reaching the bottom of the depression, thereby obtaining the depression depth value at that location. If a position is fed to its maximum stroke but still fails to make contact, it is marked as "through-hole missing" (e.g., through hole).

[0114] S4. The detected depression boundary points and depth information are incorporated into the substrate surface geometric feature dataset for subsequent tool control operations, such as filling and compensating the depression area during additive manufacturing.

[0115] Based on the above implementation methods, the controller can combine multiple dynamic adjustment methods to achieve the best sampling effect. For example, contour sampling can be completed first using a method based on fitting error. For local areas where the fitting error still does not meet the requirements, if the area contains contact gaps, the contact gap-based method can be used for detection first. Furthermore, the controller can assign a priority to each adjustment method; for example, contact gaps have the highest priority, followed by curvature changes, and finally, the magnitude of the deviation.

[0116] Example 3. Based on Examples 1 and 2, this example further introduces a hot wire circuit and a CCD industrial camera as auxiliary detection and process enhancement means for the tool setting control method. This allows the tool setting process to not only obtain the geometric features of the substrate surface, but also to improve the cladding performance of the wire by using hot wire preheating, and to verify contact reliability and monitor preheating uniformity by using visual and thermal imaging information.

[0117] Combined with appendix Figure 1 and attached Figure 2 The hardware architecture shown includes a hot filament drive board, a 40V / 20A power supply, and slip rings in the hot filament circuit. The hot filament drive board is controlled by the PLC's ENABLE signal and can apply a preheating current of up to 100A to the filament when needed, ensuring the filament reaches a predetermined temperature (e.g., 200℃~600℃) before entering the molten pool. A CCD camera (not shown separately in the attached diagram, but can communicate with the PLC or host computer via a PoE switch) is mounted near the laser head, with its optical axis aligned with the contact area between the filament end and the substrate, for real-time image acquisition of the contact point. Additionally, an infrared thermal imager or a CCD camera with an infrared filter can be optionally added to detect the temperature distribution of the filament after preheating.

[0118] It is important to note that since the detection circuit uses a weak constant current signal of 100mA, while the hot wire circuit can draw up to 100A during operation, if both are simultaneously connected to the wire-substrate circuit, the large current will severely interfere with the small signal and may even damage the detection circuit. Therefore, this embodiment employs a time-sharing interlock strategy in its control logic: when performing tool setting detection, the PLC forcibly disables the hot wire drive board (ENABLE=0), at which point only the detection circuit operates; after the tool setting detection is completed and the preheating or cladding stage begins, the hot wire circuit is re-enabled, and the output of the detection circuit can be ignored or physically disconnected. The following detailed description, in conjunction with specific implementation methods, further illustrates this.

[0119] In one implementation, when the filament contacts the substrate surface and triggers a contact signal, the controller simultaneously triggers the CCD camera to capture an image. Image processing helps confirm that the filament end is indeed in contact with the substrate surface, rather than a false contact caused by scale, oil, etc. If the CCD image shows a significant gap between the filament end and the substrate, but the detection circuit outputs a contact signal, it is determined that the detection circuit may have a short circuit or interference, and the controller issues an alarm and pauses the tool setting. Simultaneously, the CCD image can also be used to determine the morphological condition of the filament end. If abnormalities such as pilling or hooking are found at the filament end, the controller can automatically perform a retraction and re-trimming of the filament to ensure the accuracy of subsequent measurements. The controller records the pixel coordinates of the filament end in the image at each contact. If the actual distance between the pixel coordinates of two consecutive contact points is inconsistent with the arc length corresponding to the theoretical rotation step, it indicates that the substrate may have ellipticity or installation misalignment.

[0120] In practical applications, such as Figure 10 As shown, the controller corrects the actual angle of the next rotation based on visual feedback, ensuring the wire end aligns with the expected sampling position and preventing the wire from missing the preset sampling point due to eccentricity. After completing the tool setting sampling for the entire circumference, if preheating of the substrate surface or trial operation before actual additive manufacturing is required, the hot wire circuit can be activated. The controller enables the hot wire drive board and sets the preheating current (e.g., 80A), heating the wire to a bright red during the wire feeding process. At this time, an infrared thermal imager or CCD with a filter is aimed at the area where the wire and substrate will contact, acquiring a real-time thermal distribution image. The controller analyzes whether the wire surface temperature is uniform: if the temperature gradient along the wire axis is normal, but the wire temperature is significantly lower at certain circumferential angle positions, it indicates that the contact thermal resistance between the wire and substrate is too high at that location, or that the actual dry elongation at that location is too large, resulting in a decrease in the preheating effect. The controller can mark these locations as defect areas and correspondingly increase the laser power or extend the preheating time to ensure the quality of subsequent cladding.

[0121] Another implementation combines position-changing and disengagement actions with a CCD camera, suitable for continuous rotational machining of cylindrical substrates. The scheme described in Example 1 requires continuous substrate rotation and continuous wire feeding. However, during the tool setting stage, to accurately measure roundness, a discrete touch, retraction, rotation, and re-touch pattern is required. In this implementation, after discrete tool setting sampling, the controller calculates the theoretical dry elongation correction value for each angular position based on the obtained roundness error curve. Then, the controller switches to continuous rotation mode and simultaneously enables the hot wire circuit for actual cladding deposition. During continuous rotation, the CCD camera monitors the molten pool morphology and the preheating status of the wire end in real time. If the molten pool in a certain angular area is found to be too small or the wire is not completely melted, the controller can fine-tune the wire feeding speed or laser power in that area in real time based on the previous roundness error curve.

[0122] Similarly, when implementing the technical solution described in Embodiment 1, especially when retracting the filament or removing the substrate, a CCD camera can be used to verify the effectiveness of the detachment action. Specifically, after the controller issues a retraction command, a certain delay is made, and the CCD camera captures an infrared image. Image analysis confirms that a clear gap has been formed between the end of the filament and the substrate. If the image shows that the filament has not yet detached, the controller performs retraction again or increases the retraction distance until visual confirmation of successful detachment. This closed-loop verification mechanism significantly improves the reliability of discrete sampling and avoids friction or collision during the next feed due to incomplete detachment. When the substrate rotates, the contact resistance may fluctuate because the hot filament circuit is powered by an electric slip ring, resulting in uneven circumferential preheating temperature of the filament. This embodiment uses an infrared thermal imager or a CCD camera equipped with an infrared filter to capture the temperature distribution of the filament at each rotation angle. If the filament temperature is found to be significantly lower than the average value in a certain angle range, the controller can determine that the electric slip ring has poor contact or the substrate surface is dissipating heat too quickly, and take one of the following measures:

[0123] First, temporarily increase the current in the hot wire to compensate for the temperature.

[0124] Second, reduce the substrate rotation speed within this angle range to extend the preheating time;

[0125] Third, mark the area and increase the laser power during subsequent cladding.

[0126] It should be noted that these adjustments can all be pre-compensated based on the roundness error data obtained during the tool setting stage, thus achieving multi-parameter coordinated control.

[0127] For example, consider the roundness detection and preheating wire feeding of a cylindrical substrate:

[0128] S1. Initial tool setting: Disable the hot wire circuit, execute the sampling scheme described in Example 1, obtain the relative position parameters of each point on the circumference, and generate a roundness error curve.

[0129] S2. Visual calibration: Use a CCD camera to capture images of each contact point to verify the authenticity of the wire contact and record the morphology of the wire end.

[0130] S3. Roundness Compensation Calculation: The controller calculates the target dry elongation correction value at each angular position based on the roundness error curve.

[0131] S4. Preheat Enable: Enables the hot wire circuit, sets the preheat current, and starts wire feeding.

[0132] S5. Preheating uniformity test: The substrate rotates at low speed for one revolution, and the infrared thermal imager collects the heat distribution of the filament at different angle positions. If local temperature abnormalities are found, it is determined whether they are related to geometric deviations based on the roundness error curve, and the hot wire current or substrate rotation speed is adjusted accordingly.

[0133] S6. Cladding process: Start the laser and perform circumferential cladding according to the corrected dry elongation and wire feeding parameters. The CCD camera continuously monitors the state of the molten pool.

[0134] S7. Interlayer alignment: After completing one layer of cladding, repeat the above steps, but this time use the surface of the newly formed layer as the substrate, remeasure and compensate for the roundness of the current layer.

[0135] Example 4. This example further provides a structural scheme for a wire feeding mechanism that can specifically implement the above method, as well as its electrical connection and coordinated control method with the controller, detection circuit, and hot wire circuit.

[0136] like Figure 11 The hardware foundation shown includes a clamping mechanism 9 for clamping the roller. A rotary motor 10 drives the clamping mechanism 9 to perform rotational movements. The motion mechanism 11 is a multi-axial linear motion device, such as an XYZ axis motion device, with a wire feeding assembly 3 mounted on its end effector.

[0137] like Figures 12-13As shown, the wire feeding mechanism described in this embodiment includes: wire 1, conductive nozzle 2, wire feeding assembly 3, reducer 4, drive motor 5, external encoder 6, mounting frame 7, and communication interface 8. The mounting frame 7 serves as the supporting foundation for the entire wire feeding mechanism and is fixedly installed on the motion platform of the additive manufacturing device or near the laser head. The wire 1 is clamped and fed by the wire feeding assembly 3, guided by the conductive nozzle 2, and extends outwards, pointing towards the substrate surface. The drive motor 5 is connected to the wire feeding assembly 3 via the reducer 4, providing power for wire feeding and retraction. The external encoder 6 is installed on the driven wheel side of the wire feeding assembly 3 or at the output end of the drive motor 5, used to detect the actual feed length and movement speed of the wire 1 in real time. The communication interface 8 is located on the mounting frame 7, used to realize the electrical connection and signal interaction between the wire feeding mechanism and the controller, detection circuit, and hot wire circuit.

[0138] Specifically, both the drive motor 5 and the rotary motor 10 are preferably servo motors. The drive motor 5 is connected to the wire feeding assembly 3 via a reducer 4. The controller communicates with the drive motor 5 and the rotary motor 10 through a communication interface 8, and is used to send wire feeding commands and retraction commands to the drive motor 5, and to send roller angle adjustment commands to the rotary motor 10. When the controller needs to perform the wire feeding action, it sends a forward rotation command to the drive motor 5. The drive motor 5 drives the wire feeding assembly 3 to rotate forward via the reducer 4, conveying the wire 1 towards the substrate surface. The feeding speed is set by the controller according to the process requirements, preferably a low-speed feed to ensure the detection accuracy at the moment of contact. After the detection circuit feeds back a contact signal to the controller through the communication interface 8, the controller immediately sends a stop command and a reverse retraction command to the drive motor 5 through the communication interface 8. The drive motor 5 reverses, and drives the wire feeding assembly 3 to rotate in the opposite direction via the reducer 4, retracting the wire 1 by a set distance, so that the end of the wire 1 is completely detached from the substrate surface.

[0139] Furthermore, in the additive printing control stage, the controller calculates the target wire elongation or target contact position varying with angle based on the geometric feature data of the substrate surface measured during the tool setting stage, and sends corresponding wire feed rate adjustment commands to the drive motor 5 via the communication interface 8. For example, when the diameter of a certain circumferential position of the substrate is too large, the controller reduces the wire feed rate command value corresponding to that position; when the diameter is too small, the controller increases the wire feed rate command value. The drive motor 5 responds to this command and adjusts its speed in real time to achieve differentiated wire feed control. The drive motor 5 uses servo control technology in conjunction with the reducer 4 to ensure that the wire feed assembly 3 remains stable even when running at low speeds, avoiding defects such as filament pilling or breakage caused by speed fluctuations.

[0140] It is worth noting that the external encoder 6 is installed on the driven wheel side of the wire feeding assembly 3 or at the output end of the drive motor 5, and is connected to the controller via the communication interface 8. The external encoder 6 is used to detect the rotation angle and rotation speed of the wire feeding assembly 3 in real time, and then convert them into the actual feed length and actual feed speed of the wire 1, and transmit the feedback signal to the controller via the communication interface 8. In the tool setting control method described in Embodiment 1, when the end of the wire 1 contacts the surface of the substrate and triggers a contact signal, the controller needs to record the relative position parameter corresponding to the current contact position. The acquisition of this relative position parameter is achieved by reading the cumulative number of pulses of the external encoder 6 at the current moment. The controller converts the number of pulses fed back by the external encoder 6 into the total wire feed from the initial reference position to the current contact point, as the relative position parameter L of the contact point. i Since the external encoder 6 directly detects the rotation of the driven wheel of the wire feeding assembly 3 (which is in direct contact with the wire 1), it measures the actual displacement of the wire 1, rather than the theoretical rotation of the drive motor 5. Therefore, it effectively eliminates errors caused by factors such as wire feeding wheel slippage and wire elastic deformation, ensuring the accuracy of the tool setting measurement. Furthermore, after the controller sends a reverse retraction command to the drive motor 5, the external encoder 6 can also provide real-time feedback on the actual retraction displacement of the wire 1. The controller compares this feedback value with a preset retraction distance: if the actual retraction displacement reaches the preset value, the disengagement action is confirmed to be complete; if the actual retraction displacement is insufficient, the controller continues to send retraction commands until the preset value is reached, or automatically compensates for the deviation.

[0141] It should be further noted that the communication interface 8 is located on the mounting frame 7 and includes at least three functional channels: a detection signal interface, a hot wire power interface, and a control signal interface, as well as their corresponding wiring channels.

[0142] First, the detection signal interface is used to electrically connect the detection circuit and the filament 1. As described in Embodiment 1, the detection circuit includes a TouchSense module (powered by a low-voltage power supply, outputting a constant 100mA small current), whose OUT terminal is connected to the conductive nozzle 2 via a wire, and then flows back to the power supply GND via the filament 1, the substrate, and the slip ring. Specifically, in the filament feeding mechanism of this embodiment, the connection path of the detection circuit is as follows: the OUT terminal of the TouchSense module is connected to the inside of the filament feeding mechanism via the detection signal interface of the communication interface 8, and is electrically connected to the conductive nozzle 2; the filament 1 maintains electrical contact with the conductive nozzle 2 under the clamping of the filament feeding assembly 3; when the end of the filament 1 contacts the surface of the substrate, the detection circuit closes, and the 100mA constant current forms a complete circuit through the filament 1, the substrate, and the slip ring. The TouchSense module detects the voltage change and outputs a contact signal. This contact signal is transmitted to the controller via the communication interface 8. Since the conductive nozzle 2 and the filament 1 have a sliding electrical contact, the detection signal interface of the communication interface 8 adopts a low-impedance design to ensure that the weak current signal is not interfered with during transmission.

[0143] Second, the hot wire power interface is used to electrically connect the hot wire circuit and the conductive nozzle 2. As described in Embodiment 3, the hot wire circuit includes a hot wire drive board and a 40V / 20A power supply, which can output a preheating current of up to 100A to the wire 1. The output end of the hot wire drive board is connected to the inside of the wire feeding mechanism via the hot wire power interface of the communication interface 8, and is electrically connected to the conductive nozzle 2; the wire 1 is in electrical contact with the conductive nozzle 2 under the clamping of the wire feeding assembly 3, and the preheating current is conducted to the wire 1 through the conductive nozzle 2 to preheat the wire 1. The hot wire circuit also forms a return path through the substrate and the slip ring. It should be noted that the detection signal interface and the hot wire power interface of communication interface 8 are physically isolated from each other, and are electrically designed with time-sharing interlock: when the controller performs tool setting detection, it sends a disable signal (ENABLE=0) to the hot wire drive board through the control signal interface of communication interface 8 to forcibly disconnect the hot wire circuit; when the tool setting detection is completed and the preheating or cladding stage is entered, the controller re-enables the hot wire circuit (ENABLE=1), and the output signal of the detection circuit can be ignored or physically disconnected by the controller.

[0144] Third, the control signal interface is used to realize the communication connection between the controller and the drive motor 5 and the external encoder 6. Specifically, the control signal interface includes at least a drive command channel and an encoder feedback channel. One end of the drive command channel is connected to the controller (e.g., the pulse output port or analog output port of the PLC), and the other end is connected to the driver input terminal of the drive motor 5, used to transmit wire feeding speed commands, rotation direction commands, and start / stop commands. One end of the encoder feedback channel is connected to the signal output terminal of the external encoder 6, and the other end is connected to the controller, used to transmit the pulse signal of the external encoder 6, which is converted by the controller into the actual feed length and actual feed speed of the wire 1.

[0145] In one implementation, after obtaining the roundness error curve, the controller calculates the target extension length or target contact position between the wire and the substrate based on the curve. The extension length refers to the effective length of the wire extending from the wire feed nozzle to the surface of the substrate or the molten pool. Ideally, without considering factors such as thermal deformation, the extension length should equal the feed length measured during tool setting. However, due to roundness deviations in the substrate, using a fixed extension length parameter will result in inconsistent actual spacing between the wire and the substrate at different angular positions, thus affecting the cladding quality. Therefore, the controller sets a reference extension length and then superimposes the roundness deviation onto the reference extension length to obtain the target extension length that varies with the angle. In other words, at positions with a larger diameter, the target extension length should be increased accordingly to avoid excessive wire compression; at positions with a smaller diameter, the target extension length should be decreased accordingly to ensure effective contact between the wire and the molten pool.

[0146] Another approach is to directly determine the target contact position based on the roundness error. The controller uses the theoretical profile of the substrate surface as a reference, converting the deviation between the actual profile and the theoretical profile into a correction amount for the relative distance between the wire feed nozzle and the substrate. When the substrate rotates to a certain angle, the controller drives the wire feeding mechanism or the radial motion mechanism of the substrate to ensure that the wire end is always aligned with the theoretical contact point at the current angle, rather than a fixed absolute position. After determining the target wire elongation or target contact position, the controller controls the wire feeding mechanism to feed the wire according to the calculated target value, while simultaneously activating the laser for additive manufacturing. During processing, the substrate rotates at a uniform speed, the wire is continuously fed in, and the laser beam is focused on the contact area between the wire end and the substrate surface, forming a molten pool and depositing it layer by layer.

[0147] During the wire feeding process, although the target wire elongation has been preset based on the roundness error, various disturbances may still occur in actual processing, such as slight changes in the substrate contour caused by thermal deformation, fluctuations in wire feed speed, and surface tension effects caused by changes in the shape of the molten pool. These disturbances will cause the actual contact state between the wire and the substrate or the formed layer to deviate from the expectation, manifesting as an unstable contact signal output by the detection circuit—possibly exhibiting "jumping" phenomena such as signal discontinuity, frequent jumps, or abnormal amplitude. This embodiment utilizes this feature to continuously acquire the contact signal of the detection circuit during the wire feeding process after the laser is started, and adjusts the output power of the laser and / or the wire feeding rate of the wire feeding mechanism in real time according to the stability of the contact signal (i.e., the degree of jump).

[0148] For example, the controller samples the contact signal at high frequency and calculates the statistical characteristics of the signal within a time window (e.g., 50ms), including the contact signal duty cycle, signal transition frequency, or signal amplitude. Based on these statistical characteristics, the controller calculates a contact stability coefficient S, ranging from 0 to 100%, with a higher S indicating more stable contact. When S falls below a first threshold, the controller determines the contact state is abnormal and executes one or a combination of the following adjustment measures:

[0149] First, adjust the laser power: If the contact signal shows that the wire and substrate are intermittently in contact and then detaching, with the typical symptom being balling at the wire end and the ball rolling on the substrate surface, then appropriately reduce the laser power to decrease the amount of melting at the wire end and encourage the wire to re-form a stable liquid bridge contact. Conversely, if the contact signal shows that the wire has not been in contact for a long time, but the molten pool still exists, then the laser power can be appropriately increased to enhance the traction and melting ability of the wire.

[0150] Second, adjust the wire feeding rate: If the contact signal frequently changes and is accompanied by increased wire feeding resistance, it indicates that the wire may be pushed back by the molten pool. In this case, the wire feeding rate should be reduced. If the contact signal remains at a low level, but the encoder shows that the wire has been fed out, it may be that the wire is bent due to insufficient wire extension. In this case, the wire feeding rate should be increased while instructing the wire feeding mechanism to quickly retract a small distance before refeeding.

[0151] In a preferred embodiment, the controller simultaneously employs laser power adjustment and wire feed rate adjustment, with the two operating in coordination according to a certain ratio. For example, a coordination coefficient K is set such that when S decreases, the laser power is reduced by 0.6 times the deviation, and the wire feed rate is reduced by 0.4 times the deviation.

[0152] It is worth noting that after obtaining the roundness error curve, the outer circle of each newly formed layer should be as close as possible to the theoretical cylindrical surface. To this end, the geometric deviation of the lower layer can be compensated by adjusting the amount of material deposited at different angles in each layer. The amount of material deposited mainly depends on the volume of wire entering the molten pool per unit time, and this volume is proportional to the wire feeding speed. Simultaneously, the laser power affects the melting efficiency of the wire, and also influences the thickness of the cladding layer to some extent. Therefore, the strategy provided in this embodiment is as follows:

[0153] First, when the roundness error indicates that the diameter is too large at a certain circumferential position, the amount of material deposited at that position should be reduced to thin the new cladding layer there, thereby gradually correcting the overall diameter. Specific measures include reducing the wire feed speed at that position and / or reducing the laser power.

[0154] Second, if the roundness error indicates that the diameter is too small at a certain circumferential position, the amount of material deposited at that position should be increased to thicken the new cladding layer there. Specific measures include increasing the wire feed speed at that position and / or increasing the laser power.

[0155] In actual processing, the substrate rotates uniformly at an angular velocity ω. The controller discretizes the roundness error curve into N angular intervals (e.g., N=360, one interval per 1°), and pre-calculates a wire feed speed correction factor α(θ) and a laser power correction factor β(θ) for each interval. The reference wire feed speed v0 and reference laser power P0 are set according to process requirements. The timing of this process can be found in [reference needed]. Figure 11 The actual control quantity executed is: v(θ) = v0 * (1 - k) v *ΔR(θ) / T ref P(θ) = P0 * (1 - k) p *ΔR(θ) / T ref ). Among them, T ref k is the theoretical single-layer thickness. v k p These are empirical coefficients used to adjust the compensation intensity. To prevent overcompensation or undercompensation, the controller can set upper and lower limits for wire feed speed and laser power. Furthermore, considering thermal inertia, changes in wire feed speed and laser power should be smooth to avoid abrupt changes that could cause instability in the molten pool. The controller can employ first-order hysteresis filtering or linear interpolation to ensure that v(θ) and P(θ) change continuously with the angle.

[0156] It should be further noted that after the first layer of differential compensation, the roundness error of the forming layer should be significantly reduced. If there is still residual error, the same strategy can be continued in the second and third layers until the overall roundness meets the requirements, thus achieving layer-by-layer adaptive compensation in multi-layer additive manufacturing.

[0157] All the above embodiments merely illustrate implementation methods for relevant practical applications of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A tool setting control method applied to an additive manufacturing apparatus, the additive manufacturing apparatus comprising a substrate and a filament feeding mechanism for conveying filaments, characterized in that, The additive manufacturing apparatus also has a detection circuit for detecting the contact state between the filament and the substrate; The tool setting control method includes the following steps: The filament is controlled to make multiple contacts with the substrate at different positions; when the filament separates from the substrate, the contact signal of the filament in contact with the substrate obtained by the detection circuit forms a relative position parameter. Data characterizing the geometric features of the substrate surface is generated using several relative position parameters. Based on the data of the geometric features of the substrate surface, the additive manufacturing apparatus is controlled to perform tool setting control operations.

2. The tool setting control method according to claim 1, characterized in that, The contact position is determined according to at least one of the following methods: Arranged at equal or unequal intervals along a predetermined trajectory; Alternatively, when any two adjacent relative position parameters are obtained, the deviation between the two adjacent relative position parameters and the theoretical reference value is compared to adjust the target step size of the next contact position.

3. The tool setting control method according to claim 2, characterized in that, The adjustment method includes: The target step size is adjusted by using the curvature of the relative position parameters of two adjacent contact positions as the theoretical reference value. Alternatively, a contour curve can be fitted based on the obtained relative position parameters as the theoretical reference value to adjust the target step size.

4. The tool setting control method according to claim 3, characterized in that, When no contact signal is acquired within the predetermined maximum feed stroke, the target step size is shortened so that the target step size is less than the theoretical reference value.

5. The tool setting control method according to any one of claims 1 to 3, characterized in that, The substrate is a cylindrical substrate that can rotate about an axis; The methods for changing the relative spatial position between the filament and the substrate include: The substrate is rotated about its axis by a predetermined angle. Alternatively, the substrate may be translated a predetermined distance along a radial straight line. Alternatively, the wire feeding mechanism or laser head carrying the filament may be moved relative to the substrate by a predetermined step length.

6. An additive printing control method, applied to an additive manufacturing apparatus, the additive manufacturing apparatus comprising a substrate rotatable about an axis, a filament feeding mechanism, a laser, and a detection circuit, characterized in that, The additive printing control method includes the following steps: The tool setting control method according to any one of claims 1 to 5 is adopted; Using data that characterizes the geometric features of the substrate surface, the target dry elongation or target contact position between the filament and the substrate is calculated; by calculating the target dry elongation or target contact position at different time steps, the filament feeding mechanism is controlled to transport the filament. When a contact signal is obtained indicating that the filament is in contact with the substrate, the laser is controlled to perform a printing operation.

7. The additive printing control method according to claim 6, characterized in that, The output power of the laser and / or the wire feeding rate of the wire feeding mechanism are adjusted according to the direction of fluctuation of the data of the geometric features of the substrate surface.

8. The additive printing control method according to claim 7, characterized in that, The method for adjusting the output power of the laser and / or the wire feeding rate of the wire feeding mechanism is as follows: When the data of the geometric features of the substrate surface show that the diameter of a certain circumferential position of the substrate is too large, the wire feeding speed or the output power at that position is reduced to reduce the thickness of the cladding layer. When the data of the geometric features of the substrate surface show that the diameter of a certain circumferential position of the substrate is too small, the wire feeding speed or the output power at that position is increased to increase the thickness of the cladding layer.

9. An additive manufacturing apparatus, characterized in that, include: Substrate; Laser, used to heat filament; The wire feeding mechanism is used to transport wire. A detection circuit is electrically connected between the filament and the substrate to detect the contact state between the filament and the substrate and generate a contact signal. A motion mechanism for changing the relative spatial position between the filament and the substrate; The controller is communicatively connected to the filament feeding mechanism, the detection circuit, the motion mechanism, and the laser. The controller is configured to perform the tool setting control method according to any one of claims 1 to 5, or the additive printing control method according to any one of claims 6 to 8.

10. The additive manufacturing apparatus according to claim 9, characterized in that, The wire feeding mechanism includes a drive motor for controlling the wire feeding, and the drive motor is communicatively connected to the controller.

Images