DEVICE AND METHOD FOR REPAIRING COMPONENTS USING ADDITIVE MANUFACTURING

Description

[0001] The present invention relates to a device and a method for repairing components by means of additive manufacturing and in particular for repairing depressions in components. Background of the invention

[0002] To repair damaged components, the first step is usually to mill out the repair area. In a subsequent step, the milled area is scanned, and a corresponding computer model is created. Based on this data, a structural model is then created, which can be inserted into the milled area to repair the damaged area. This type of repair process makes it possible to repair a damaged area on the surface of a component, allowing the original surface to be restored. The structure to be inserted can be produced using additive manufacturing processes and incorporated into the milled area.

[0003] One repair method is known, for example, from US 2017 / 0370221 A1. In this method, the first step is to remove the defect from a component to be repaired. Using an additive manufacturing process, a repaired component can be created so that the component can once again retain its original geometry. However, known methods have the disadvantage that multiple data conversion and measurement data processing steps are necessary to create a structure from the measurement data of the component to be repaired, which can then be manufactured using the additive manufacturing process. This usually requires data processing. This results in high computational and data-intensive workloads.

[0004] In addition, CN 108 340 582 A describes an automated repair process which, by taking thermal infrared images, can identify any defects within the component to be repaired and adapt the subsequent repair process to the identified defects.

[0005] Furthermore, CN 109 676 135 A also presents a processing process based on previous imaging, in which a feedback repair system is generated using grayscale image analysis. More specifically, the aforementioned method initially detects existing defects within the component to be examined by comparing recorded image intensity data with predefined threshold values. In a subsequent process, the generated information is also passed on to an evaluation unit to enable a repair process specifically tailored to the defect.

[0006] Furthermore, CN 108 489 986 A shows a repair system based on remelting components. In this system, three-dimensional cross-sectional drawings of a component to be machined are generated by positioning a recording module consisting of two CCD cameras. These are then forwarded to an external computer for analysis. Furthermore, said computer is configured to detect existing defects within the component using implemented image analysis programs and transmit them to an additive manufacturing device including a laser device, thus generating an iterative detection and processing mechanism.

[0007] Based on the aforementioned prior art, it is an object of the present invention to solve the described problems. In particular, it is an object to provide an efficient method for repairing components using additive manufacturing, as well as a corresponding device. Summary of the invention

[0008] To achieve the above-mentioned objects of the present invention, the features of the independent claims are proposed and preferred developments are in the dependent claims.

[0009] According to the present invention, a method for the automated repair of components using an additive manufacturing system can comprise a first step in which the defective component is clamped into the manufacturing system. In a further step, a repair area can be defined that encompasses the surface of the component to be repaired. In addition, a tool path can be defined within the repair area. The tool path corresponds to the travel path that the tool of the manufacturing system travels in one cycle. In a further step, the condition of the component in the repair area can be determined (= measuring cycle). In such a measuring cycle, data values ​​can be determined along the defined tool path. The data values ​​correspond to the deviation of the surface of the component from a predetermined dimension. In particular, the determined deviation can be the depth (or the amount thereof) of the damaged area at a point (measurement point).In a further step, a selective application of filler or filling material can be carried out. The filler can be applied along the defined tool path in at least one filling cycle. In a filling cycle, part or preferably the entire tool path can be traversed and the manufacturing system can be selectively activated to apply the filler depending on the difference between the data value along the path and a predetermined threshold value. This advantageous method makes it possible to use the data resulting directly from the measuring cycle as direct input for the filling cycle. In the measuring cycle, a data array with measured values ​​for various path points (measurement points) is thus created. This data array, in turn, is the input for the filling cycle and, after a comparison with a predetermined threshold value, determines whether the manufacturing system is activated or deactivated to apply the filler.

[0010] In other words, the manufacturing system follows the defined tool path along the path points during the fill cycle. For each of the path points, a data value can be determined, which is compared with the threshold value. When a path point is reached, the tool head, e.g., a laser, of the additive manufacturing system is activated and a material can be applied. When the next path point is reached, a comparison is carried out to determine whether the tool head is deactivated or whether it remains activated. This advantageously makes it possible to create the program for the fill cycle directly during the measuring cycle, eliminating the need for further data conversion and using the data values ​​directly to activate or deactivate the tool in the manufacturing system.Because the tool path can be identical in both the measuring cycle and the filling cycle, the computational effort and data requirements can be further reduced. Furthermore, the measuring cycle and the filling cycle can be performed automatically, enabling essentially automatic repair of the component. Complex and error-prone data conversions of point clouds of measured values, which can result from measuring the damaged area of ​​the component, are no longer necessary. This particularly advantageously results in a simplified process for repairing components using additive manufacturing, which is also less susceptible to failure.

[0011] The tool path can have a plurality of path points. During the measurement cycle, a deviation in a tool direction (e.g., a Z direction) between the surface of the component and a desired shape can be determined at each of the path points. The tool direction can, for example, be the direction orthogonal to the component surface in the repair area.

[0012] Advantageously, the data values ​​can indicate the deviation of the component surface in the tool direction, and the tool direction can be orthogonal to the workpiece surface (e.g., an XY plane) on which the tool path lies. The tool of the additive manufacturing system, such as a laser, is thus moved along the workpiece surface along the tool path. This tool path or the workpiece surface can, for example, be orthogonal to the tool direction, which determines the deviation.

[0013] In a filling cycle, the filler can be selectively applied along a path segment of the defined tool path. The filler can thus be applied during the process from one path point to the adjacent path point. Switching off the tool of the manufacturing system for applying the filler is determined at the target path point depending on the difference between the data value at the target path point and the predetermined threshold value. If, for example, the tool moves along the tool path from one path point to the next and the threshold value is always exceeded, the laser can be constantly activated so that the filler is always applied. Powder jet deposition welding, for example, can be provided as additive manufacturing. If a path point is reached during the process at which the data value is lower than the threshold value, the laser or the tool head can be switched off.Only when another path point is reached, where the corresponding data value again exceeds the threshold, is the laser activated again.

[0014] A path segment can be defined by two path points, and the path segment can preferably be straight. This particularly efficient and simple design of the path segments, which together form the tool path, can further simplify the repair process, since curved segments are avoided and only one movement in one X-direction and one Y-direction is necessary. Preferably, the data values ​​can also be averaged values ​​of path points of a segment. The tool can thus also be activated segment by segment.

[0015] The threshold value can be constant within a filling cycle. In one filling cycle, the manufacturing system's tool can thus traverse the entire repair area along the defined tool path, with each data point being compared with the threshold value of the filling cycle. After the filling cycle has been completed, a further filling cycle can be scheduled with a modified threshold value. This configuration of the filling cycles makes it possible to successively repair the defects. In each filling cycle, a plane is defined that is orthogonal to the tool direction. Each of these planes, in turn, has a different threshold value until a minimum threshold value is reached at the last plane, so that a final surface (e.g., a flat surface) of the component in the repair area can be achieved. It is particularly advantageous to apply an additional allowance once the minimum threshold value has been reached.For example, after reaching the minimum threshold value, the entire repair area can be filled with filler in one (or more) additional filling cycles.

[0016] Multiple filling cycles can be performed, and the threshold can be adjusted by one thickness value each time until a substantially uniform surface is achieved in the repair area. The original surface shape (e.g., linear or curved) of the component can thus be restored in the repair area. The thickness value can, for example, depend on the thickness of the material applied during the filling cycle.

[0017] The result of the measurement cycle can be an array of the determined data values. The threshold value for the first filling cycle can be set based on a maximum value or a minimum value (extreme value) of the array and a constant. In particular, the maximum value of the array is used to determine the deepest defect in the repair area. The threshold value can be set based on this deepest point, so that in the first filling cycle only the deepest damage areas are filled. In the subsequent filling cycle, a reduced threshold value is used so that additional areas can be filled with filler material. By successively repeating the filling cycles, the damage areas are completely filled.

[0018] The threshold value can be reduced by a thickness value in a subsequent filling cycle, wherein the thickness value corresponds to the height of the applied filler in a filling cycle, in particular at a path point or path segment. Such a height is, for example, the height of the weld seam.

[0019] The data values ​​determined during the measurement cycle can be used directly for the filling cycle. This eliminates the need for conversion, transformation, or digitization of the measurement cycle results. This provides a particularly efficient and simple repair procedure. Furthermore, the frequency of errors that can often arise due to conversion errors is avoided.

[0020] Data values ​​can be used (without data conversion) for comparison with the threshold value in the fill cycle. Thus, the unconverted data values ​​from the measurement cycle are used directly to decide, based on comparison with the threshold value at the respective path points, whether the adaptive manufacturing system tool is activated or deactivated (i.e., whether material is added).

[0021] The method can thus advantageously be a method for creating a flat and / or curved surface in a repair area. The repaired surface shape advantageously corresponds to the original surface shape of the component.

[0022] The tool path can advantageously be defined taking into account a working diameter of the manufacturing system. Such a working diameter of the adaptive manufacturing system can, for example, be the diameter of the laser (in powder-jet cladding). Taking this working diameter into account, the tool path can advantageously be defined in such a way that the entire repair area can be completely covered.

[0023] The tool path can be a non-overlapping, continuous line. More preferably, the tool path can be configured in a meandering shape. The tool path can also consist of line segments, each of which runs in a straight line. Thus, a particularly error-free process can be provided with an optimized tool path. Furthermore, the tool path can be configured to be contour-parallel to the repair area.

[0024] The repair area can be defined so that the entire surface of the component to be repaired can be completely covered. This allows the process to create a flat surface (or, advantageously, a curved surface based on the original component shape) at the repair site.

[0025] In a final fill cycle, the additive manufacturing system's tool can be advantageously activated continuously while traversing the defined tool path. This approach makes it possible to compensate for any remaining unevenness. Furthermore, it can provide a stock allowance across the entire surface of the repair area.

[0026] The defined tool path advantageously represents the process path of the additive manufacturing system's tool and, at the same time, the process path of the measuring instrument for determining the deviation. Converting the process path or tool path is therefore not necessary between the measuring cycle and the filling cycle, or between all filling cycles. This further reduces the computational effort and minimizes the risk of errors.

[0027] The tool of the additive manufacturing system can be moved along the defined tool path from one path point to an adjacent path point. At each path point, a comparison of the determined data value of the respective path point can be performed with a threshold value. If the threshold value is exceeded, the tool can be activated. If the threshold value is exceeded, the tool can be deactivated (or vice versa, with the opposite sign). Advantageously, it is thus possible to directly provide a command array based on the measured values ​​at the path points, which activates or deactivates the laser or the tool of the additive manufacturing system.

[0028] Particularly advantageous is that the measuring cycle and the subsequent filling cycle are performed automatically, thus eliminating the need for manual intervention. Another advantage is that the measuring cycle and all subsequent filling cycles are performed automatically. The data values ​​are thus the same for all filling cycles. Alternatively, an additional measuring cycle can be advantageously scheduled after each filling cycle for improved accuracy.

[0029] The method may be a method for filling depressions of a component surface in a damaged area.

[0030] The method can advantageously comprise the step of moving the tool along path points of the defined tool path, wherein a laser is activated upon reaching a path point at which the data value is greater than the threshold value, and the laser is deactivated upon reaching a path point at which the data value is not greater than the threshold value. A plurality of fill cycles can be provided, wherein the threshold value can be reduced from fill cycle to fill cycle. Advantageously, the threshold value for the final fill cycle is reduced such that all data values ​​of any data points are greater than the threshold value. Particularly advantageously, the threshold value can be set such that several layers of stock are applied in the repair area.

[0031] The defined tool path may advantageously comprise parallel segments and the distance between adjacent segments may be determined depending on the working diameter of the laser.

[0032] The additive manufacturing system may comprise a laser, wherein switching the laser on and off at path points is dependent on the result of the comparison between the data value at the path point and the threshold value.

[0033] For the measurement cycle (to determine the data values), a probe or an optical sensor can be advantageously used. A capacitive and / or inductive sensor can also be used for the determination.

[0034] Advantageously, a data processing device can be provided for carrying out the above-mentioned method. Furthermore, a computer-readable storage medium can be provided, which comprises instructions that, when executed by a computer, cause the computer to carry out the above-mentioned method. An additive manufacturing system can be provided that is configured to carry out the above-mentioned method and has a corresponding computing unit for this purpose.

[0035] In the following, the invention is described by way of examples with reference to the accompanying figures. Figures

[0036] Fig. 1: shows an overview of an area to be repaired with a drawn-in tool path PF; Fig. 2: shows a detailed view of the tool path PF; Fig. 3a: shows a first view in a fill cycle; Fig. 3b: shows another view in a fill cycle; Fig. 3c: shows another view of the tool path in the fill cycle; Fig. 3d: shows a view of a well-advanced fill cycle; Fig. 3e: shows a final fill cycle; Fig. 4: shows a detailed view of the tool path in the fill cycle; Detailed description

[0037] The following features of the embodiments can be combined in whole or in part, and the present invention is in no way limited to the described embodiments. In the drawings, identical or similar features are designated by identical reference numerals.

[0038] The present invention relates to an optimized method for the automated repair of components that, for example, exhibit damage. Such damage can be bulges, scratches, score lines, etc. In order to repair the indentation in the component, a method is proposed in which the computational effort can be reduced to a minimum. Data processing can thus be reduced to a minimum, so that the conversion errors that commonly occur can be avoided. Additional processing of the measurement data in order to create a model (e.g., a CAD model) that can be used for the repair is not necessary, since according to the present invention, the data from the measurement cycle is used directly as input for a filling cycle. In such a filling cycle, material is applied to the component layer by layer using additive manufacturing.The present invention can be used particularly advantageously for additive manufacturing processes that operate on a laser basis.

[0039] In Fig. 1 A first view of a repair area O is shown. The repair area O completely covers the surface of the metallic component to be repaired, whereby the edge of the defined repair area O should have a minimum distance to the surface to be repaired. As in Fig. 1 As shown, the damage D to be repaired on the component is a depression. This depression is completely filled using the method according to the invention, so that the component has a flat surface at the end of the process.

[0040] In Fig. 1 An example tool path PF is shown. The tool path PF defines the path that the tool of the manufacturing system travels in a cycle. However, the tool path PF is not limited to the Fig. 1The tool path PF is not limited to the straight-line configuration shown, but could also include curved sections. However, the tool path PF advantageously includes only straight sections, so that the computational effort can be further reduced. The tool path PF can also advantageously be defined as a function of the working diameter (WorkDia). As shown in Fig. 1 As shown, the diameter of the manufacturing system is, for example, the diameter of the laser. The distance between adjacent segments of the tool path PF is determined based on the diameter of this laser. It is particularly advantageous if the distance between two adjacent line segments of the tool path PF corresponds to at least half the working diameter plus 5%. This particularly advantageous distance has resulted in very reliable repair results.

[0041] The Fig. 1The tool path shown, for example, comprises a rectangular outer tool path PF and a meandering inner tool path PF, in which the start point and the end point are arranged opposite each other. The tool path can be a continuous tool path or can advantageously be divided into several spaced tool paths. Advantageously, no line segment of the tool path PF overlaps. The tool path PF is defined in such a way that, taking into account the working diameter, the entire area in the repair area O is covered. In both the measuring cycle and the filling cycle, the tool path PF remains unchanged, which further reduces the computational effort and data expenditure. Fig. 1The view shown shows an example of the XY plane, which in this case is a flat plane. The deviations of the component surface from a specified dimension, which simultaneously form the data values ​​DW, are measured in the tool direction (e.g., a Z direction). The tool direction is, for example, the direction in which the tool of the manufacturing system applies the filler.

[0042] Particularly advantageously, the tool path PF can also be an uninterrupted line path that runs continuously from a starting point to an end point and (always) comprises straight line segments that preferably do not overlap. In such a case, the array generated in the measuring cycle can also be used directly for the fill cycle, corresponding to the path points. In a further development, curved line segments can be used. For example, a polynomial curve can be used as a line segment.

[0043] In Fig. 2 A detailed view of the repair area O is shown. The tool path PF is formed by path points PFp. These path points are arranged in a row, with the path segments PFs between them. These path segments PFs are, in this example, straight line sections. The distance between adjacent path points PFp can vary along the tool path, so that optimal path points PFp are provided. In particular, however, the path points PFp can be distributed evenly over the entire tool path PF, with at least one path point PFp being provided in each corner of the tool path PF, as is also the case in Fig. 2 is shown. In such a corner point, for example, a change in the direction of the tool path PF may occur.

[0044] In Fig. 2Arrows are also shown which indicate the direction along which the tool of the manufacturing system, in particular the laser, is moved along the tool path PF. The tool path PF, which is directly above the damaged area, is preferably a single, uninterrupted tool path which is traversed once in one direction in a cycle. For example, a tool path can be interrupted if the tool has to be set down or retracted (e.g. along the tool direction), so that application of filler is not possible. In an advantageous further development, the tool path is designed such that in a filling cycle only segments are approached in which the filler is to be applied. In such a development, the tool path in the filling cycle can differ from the tool path in the measuring cycle. In addition, the tool path can differ from filling cycle to filling cycle.

[0045] As in Fig. 2 shown, data values ​​DW are provided, which are determined for each of the path points PFp.

[0046] The process for repairing components using additive manufacturing involves, in a first step, clamping the component to be repaired into the manufacturing system. Once the component is in the manufacturing system, a repair area can be defined that covers the entire (or only a portion) of the component to be repaired. The tool path (PF) can then be defined within this repair area. All of these steps can be considered an initialization process.

[0047] After initialization, a measuring cycle and several filling cycles can preferably follow fully automatically. In a measuring cycle, the manufacturing system can automatically determine the depth values ​​in the Z direction (or generally in the tool direction) at all path points PFp of the tool path PF. This is Fig. 2represented by the designation Z_result[n]. The data values ​​are thus, for example, the measured values ​​in the Z direction (or generally in the tool direction) at the path points PFp. In a particularly advantageous embodiment, these measured values ​​are stored sequentially in a data array. The data array can thus be viewed as an ordered arrangement of measured depth values. The depth values ​​can be measured in relation to a predefined zero plane, which, for example, corresponds to an ideal surface depth of the component.

[0048] After the data values ​​have been determined and a data array has been created, the first filling cycle can follow directly, in which an additive laser selectively fills the damaged area of ​​the component in the repair area O. Such filling cycles are shown in the Figures 3a to 3e . In an exemplary first filling cycle, as shown in Fig. 3aAs shown, the tool of the additive manufacturing system travels the entire tool path PF. At each path point PFp reached, a comparison of the respective data value with the threshold value is performed. The threshold value is set depending on the determined depth of the damage. In the first filling cycle, as shown in Fig. 3a As shown, the maximum value in the data array is determined and the threshold value is assumed to be the maximum value (or preferably slightly lower, by around 1%). This means that in the first filling cycle, only the damage area with the maximum depth is filled. All areas where the damage is less deep are not filled with filler in the first filling cycle. This is also shown in Fig. 3awhere the activated laser A is shown as a solid line along the tool path PF. In areas of the tool path where the threshold is not reached or exceeded, the laser is deactivated B, which is Fig. 3a is shown by the dashed line. In Fig. 3a Thus, only the innermost area of ​​the damaged area is filled with filling material.

[0049] In the subsequent filling cycle, as described in Fig. 3b As shown, the threshold value has been modified. In particular, the threshold value has been reduced, for example, by the fill level (e.g., calculated weld thickness) as it was applied in the first filling cycle. Thus, by further reducing the threshold value, Fig. 3The filling cycle shown is used to apply filler to a larger area of ​​the damaged area. The laser is activated A both in the area of ​​the laser processing area determined in the first filling cycle and in the additional area beyond it, where the data value of the path points exceeds the threshold.

[0050] Analog is in Fig. 3c and 3d the filling area is increased and the laser is activated in increasingly larger areas.

[0051] In Fig. 3e A final filling cycle is shown as an example, in which the laser is essentially continuously activated A. Only in the horizontal sections between two adjacent line segments of the meandering tool path PF is the laser deactivated B in order to further increase the surface quality.

[0052] In Fig. 4a detailed view of a filling cycle is shown. The laser is activated A between two adjacent path points PFp when a path point PFp is reached in the direction of travel of the laser that exceeds the threshold value. Only when a path point PFp is reached at which the threshold value is again undershot is the laser deactivated B. For all path points where the data value is less than the threshold value, the laser remains deactivated or is deactivated if it was previously activated. In the present exemplary embodiment, the activation and deactivation of the laser is described as dependent on the measured data value of a threshold value being exceeded. However, it is also possible to do this with the opposite sign. In other words, the measured values ​​in the tool direction can be assumed to be negative values. In such a configuration, the laser is then activated when the threshold is undershot.However, the above description, in which the laser is activated when the data value exceeds a threshold, concerns, as an example, the amount of the measured depth in the repair area.

[0053] The component surface to be repaired can also be curved in its original form (including three-dimensionally curved). The method according to the invention can therefore restore the originally curved surface in the repair area. The measuring direction preferably corresponds to the tool direction and can preferably be orthogonal to the surface of the component in the measuring cycle so that a desired depth of the damaged area can be determined. The tool path can preferably also be in a surface which is parallel to the original (i.e. undamaged) surface of the component. The absolute direction of the measurement can therefore change from measuring point to measuring point, for example if measurements are taken along a curved surface. This depth can be filled up accordingly in the filling cycle. The application layers of the tool in the filling cycle are preferably applied in the tool direction.

Claims

1. Method for the automated repair of components using an additive manufacturing system, comprising the following steps: - a) defining a repair region (0) which comprises the surface of a component to be repaired and defining a tool path (PF) within the repair region (0); - b) determining the nature of the component in the repair region (0) in a measuring cycle by means of a measuring instrument of the additive manufacturing system by determining data values (DW) along the defined tool path (PF) for the deviation of the surface of the component from a predefined dimension; and - c) selectively applying filler along the defined tool path (PF) by means of a tool of the additive manufacturing system in at least one filling cycle, wherein, in a filling cycle, part or all of the tool path (PF) is traversed and, in the process, the manufacturing system is selectively activated to apply the filler, depending on the difference between the data value (DW) and a predeterminable threshold value, wherein the tool path (PF) corresponds to the travel path which the tool of the manufacturing system traverses in at least one filling cycle and at the same time corresponds to the travel path which the measuring instrument traverses for the determination of the deviation.

2. Method according to one of the preceding claims, wherein the tool path (PF) has a multiplicity of path points (PFp) and, in step b), a deviation is determined at each of the path points (PFp) in a tool direction between the surface of the component and a desired shape.

3. Method according to one of the preceding claims, wherein the data values (DW) specify the deviation of the component surface in the tool direction and the tool direction is orthogonal to a selected surface in which the tool path (PF) lies.

4. Method according to one of the preceding claims, wherein, in a filling cycle, the filler is selectively applied along a path segment (PFs) of the defined tool path (PF) if the determined data value (DW) for the path segment (PFs) or a path point (PFp) of the defined tool path (PF) is greater than or equal to a threshold value.

5. Method according to the preceding claim, wherein a path segment (PFs) is delimited by two path points (PFp) and the path segment (PFs) preferably runs in a straight line; and / or wherein the threshold value is constant in a filling cycle; and / or wherein a plurality of filling cycles are carried out and the threshold value is in each case adapted by a thickness value until a final surface which corresponds to a desired surface is present in the repair region.

6. Method according to one of the preceding claims, wherein the result of step b) is an array consisting of the determined data values (DW) and the threshold value for the first filling cycle is defined on the basis of a maximum value or minimum value of the array and / or a constant.

7. Method according to one of the preceding claims, wherein the threshold value is adapted by a thickness value in a further filling cycle and wherein the thickness value corresponds to the height of the applied filler in a filling cycle.

8. Method according to one of the preceding claims, wherein the determined data values (DW) of the measuring cycle are used directly for the filling cycle; and / or wherein the determined data values (DW) are used without data conversion for comparison with the threshold value in the filling cycle.

9. Method according to one of the preceding claims, wherein the tool path (PF) is defined taking into account a working diameter (WorkDia) of the manufacturing method, in particular a diameter of a laser, and covers the entire repair region (O); and / or wherein the tool path (PF) is a nonoverlapping, continuous linear profile which is preferably of meandering configuration; and / or wherein the defined tool path (PF) is the travel path of the tool of the additive manufacturing system and the travel path of a measuring instrument for the determination of the deviation.

10. Method according to one of the preceding claims, wherein the tool of the additive manufacturing system is moved along the defined tool path (PF) from one path point (PFp) to an adjacent path point (PFp) and, at each path point (PFp), a comparison of the determined data value (DW) of the respective path point (PFp) and the threshold value is carried out and, if the threshold value is exceeded, the tool is activated; and / or wherein at least steps b) and c) can be carried out fully automatically.

11. Method according to one of the preceding claims, wherein step c) comprises: moving the tool along path points (PFp) of the defined tool path (PF), wherein, when a path point (PFp) is reached at which the data value (DW) is greater than the threshold value, the laser is activated (A) and, when a path point (PFp) is reached at which the data value (DW) is not greater than the threshold value, the laser is deactivated (B), and wherein, after a filling cycle, a further filling cycle follows in which the threshold value is reduced.

12. Method according to one of the preceding claims, wherein the defined tool path (PF) comprises parallel segments and the distance between adjacent segments is defined depending on the working diameter of the laser; and / or wherein the additive manufacturing system comprises a laser and step c) comprises: switching the laser on and off at path points (PFp) depending on the result of the comparison between the data value (DW) at the path point (PFp) and threshold value; and / or wherein a measuring probe, an optical sensor, a capacitive sensor or an inductive sensor is used to determine the deviation.

13. Additive manufacturing system having a device for data processing, comprising means for carrying out the method according to Claim 1.

14. Computer-readable storage medium, comprising instructions which, when executed by a computer, cause the latter to carry out the method according to Claim 1.

15. Use of an additive manufacturing system according to Claim 13 for the repair of components by means of powder-jet deposition welding.