Intelligent welding system based on casting machining

Abstract

The invention discloses an intelligent welding system based on casting machining, relates to the technical field of casting machining, solves the problem that tiny errors exist in the welding treatment process, confirms a welding characteristic mean value and a minimum change trend principle through an optimal route selection end, intelligently calculates an optimal welding sequence, and improves the welding accuracy. The direction sudden change in the welding process is reduced, the risk of thermal stress concentration is reduced, path turning points are accurately recognized through geometric analysis, welding current is dynamically adjusted according to angle characteristics, the problem that welding is not continuous easily occurring in an acute angle area is effectively solved, the modular design supports flexible expansion of a system, the welding requirements of castings of different models can be rapidly met, and the welding efficiency is improved. And a standardized intelligent welding solution is provided for the manufacturing industry.

Description

Technical Field

[0001] This invention relates to the field of casting processing technology, specifically to an intelligent welding system based on casting processing. Background Technology

[0002] In the field of casting processing, welding is the core process for connecting individual castings. Its quality and efficiency directly determine the performance of the final product and production benefits. In the traditional casting welding process, model building often relies on manual measurement or simple scanning equipment, which makes it difficult to obtain accurate images of the casting from all angles. This results in insufficient accuracy of the generated model and blurry recognition of gaps between individual model units, which poses a hidden danger for subsequent welding positioning. Patent application CN113319458B discloses a rotary welding system, which includes a control module, a mounting bracket, a workpiece fixing mechanism, a turntable mechanism, and a welding mechanism. The workpiece fixing mechanism holds the workpiece to be welded, and the control module controls the workpiece fixing mechanism to lock the workpiece to be welded. The control module drives the turntable mechanism to rotate the workpiece fixing mechanism so that the workpiece to be welded fixed on the workpiece fixing mechanism rotates to the welding area of ​​the welding mechanism. The control module also controls the welding mechanism to weld the workpiece to be welded in the welding area. This invention fixes the workpiece to be welded through the workpiece fixing mechanism, and the turntable mechanism drives the workpiece to be welded to the welding area of ​​the welding mechanism. The intelligent control of the welding mechanism to weld the workpiece to be welded can automatically align with the area to be welded on the workpiece, realizing intelligent welding function.

[0003] During the welding process of castings, the corresponding welding mechanism is generally controlled to weld the weld according to the determined weld seam. However, in the actual welding process, the original welding trajectory is not in the optimal state. In the specific welding process, there are large trend changes between the running trajectories, which leads to small errors in the welding process and the welding accuracy is not effectively guaranteed. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an intelligent welding system based on casting processing, which solves the problem of minute errors in the welding process.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an intelligent welding system based on casting processing, comprising: On the model generation end, a relevant casting model is generated based on the all-around image of the relevant casting. The feature verification processing end locks the individual model units in the relevant casting models in turn according to the model bodies and related casting models stored in the cloud database. Based on the contact state between the corresponding individual model units, it locks the gap to be welded and then locks the same type of weldable segments in the gap to be welded in combination with the three-dimensional coordinate system. The optimal route selection process involves identifying several groups of similar weldable segments within the weld gap, confirming the welding characteristics associated with each similar weldable segment, and then selecting the optimal sorting process according to different sorting procedures to lock in the optimal route.

[0006] Preferably, the specific method by which the feature verification processing terminal locks the gap to be welded is as follows: For the generated casting models, the pre-stored model bodies are extracted from the cloud database. Different model bodies are compared and verified with the relevant casting models in turn. The model bodies that are completely consistent are locked and the corresponding model bodies are recorded as the model bodies associated with the relevant casting models. Extract individual model units with contact features from the relevant casting model, then confirm several contact points associated between pairs of individual model units, connect the several contact points confirmed by a single individual model unit to confirm the contact area, then confirm the outer edge contour of the contact area, and record the generated outer edge contour as the weld gap associated between the current pairs of model bodies.

[0007] Preferably, the feature verification processing terminal locks the same type of weldable sections in the following specific way: Based on the weld gaps identified between each pair of individual model units, the weld gaps are combined with a three-dimensional coordinate system. The three-dimensional spatial coordinates associated with different contact points within the weld gaps are recorded and denoted as (X... k Y k Z k ), where k represents different contact points; Next, the changing trends associated with adjacent contact points within the weld gap are confirmed sequentially, and the three-dimensional spatial coordinates of the point preceding the adjacent contact point are determined as (X... q Y q Z q The three-dimensional spatial coordinates of the next point are determined as (X... h Y h Z h ), where q∈k, h∈k, using: Confirm the changing trends of the corresponding contact segments at adjacent contact points, and then confirm the different changing trends associated with different contact segments in sequence; Based on the identified contact segments, adjacent contact segments are further identified as similar segments. These similar segments are then integrated to identify similar segments to be soldered, specifically including: The two changing characteristics associated with two adjacent contact segments are labeled as Q1 and Q2, respectively: If Q1∈[Q2-0.3×Q2,Q2+0.3×Q2], then the two contact segments are divided into segments of the same type; If Q1∉[Q2-0.3×Q2,Q2+0.3×Q2], then no division is made, and the same confirmation method is used to confirm the same type of segment for the two adjacent contact segments in turn. The confirmed same type of segments are then integrated. If the adjacent contact segments before and after are the same type of segment, and the two adjacent contact segments afterward are also the same type of segment, then they are uniformly divided into the same type of segment. The integrated same type of segment is recorded as the same type of segment to be welded. Several same type of segments to be welded that exist in the weld gap are confirmed. If there is a contact segment that does not have a same type of segment, then the corresponding single contact segment is directly recorded as the same type of segment to be welded.

[0008] Preferably, the optimal route selection terminal locks the optimal route in the following specific way: The average value of several sets of change trends associated with a single weldable segment of the same type is processed, and the confirmed average trend is used as the welding feature associated with the corresponding weldable segment of the same type. Randomly select a segment of the same type to be welded as the starting segment, and sort the welding features associated with the subsequent segments of the same type to be welded in a clockwise direction. At the same time, sort the segments of the same type that do not appear consecutively, in the same clockwise direction. After the welding features of all segments of the same type to be welded are sorted, a feature sequence is generated. The absolute value of the difference between adjacent welding features in the feature sequence is confirmed, and the confirmed sets of absolute values ​​are summed to lock the process features of the current sorting process. Then, select other similar segments to be soldered as starting segments in turn, and use the same sorting method as the above sorting process to confirm the process characteristics associated with the current sorting process. Continue in this way until all similar segments to be soldered are used as starting segments. Then, select the minimum value from the process characteristics of different sorting processes, take the sorting process associated with the minimum value as the optimal process, and take the sorting method associated with the optimal process as the optimal route.

[0009] Preferred options also include: Inflection point calibration: From the confirmed weld gaps, identify adjacent weldable sections of the same type, and mark the turning points between adjacent weldable sections of the same type as inflection points. Record the angular characteristics associated with the inflection points. The specific method is as follows: Based on the marked inflection point, the adjacent contact points associated with the inflection point before and after the inflection point are respectively recorded as the front point and the back point; Then, confirm the plane containing the previous point, the inflection point, and the subsequent point. From the direct line of sight of the plane, confirm the characteristic angle J associated with the inflection point. oHere, 'o' represents different inflection points. Connecting the previous point to the inflection point confirms the first line, and connecting the subsequent point to the inflection point confirms the second line. The minimum angle between the first and second lines is the characteristic angle J associated with the corresponding inflection point. o ; The inflection points marked in the weld seam and the corresponding characteristic angle J associated with those inflection points are identified. o Transmitted to the welding center; The welding center, based on the confirmed optimal route and the determined inflection points and angle characteristics, controls the welding of the weldment seam. The weldment used is the same as that used in commercially available automated equipment, specifically including: The welding joint is welded according to the preset operating parameters, which are the preset parameters. When the welded parts reach the designated inflection point, the welding current H is adjusted, and the adjusted welding current for different inflection points is denoted as Ho, where 'o' represents different inflection points. The formula is: Ho = H + |J o -180°|×C1, where C1 is a preset fixed coefficient factor. Welding is performed on different inflection points according to the confirmed welding current to complete the welding process of the corresponding weld seam.

[0010] This invention provides an intelligent welding system based on casting processing. Compared with existing technologies, it has the following advantages: The innovative "same type of weldable segment" locking mechanism divides the welding path into segments with consistent direction by analyzing the trend of changes in the spatial coordinates of the contact points, laying the foundation for subsequent optimization. By confirming the average value and minimum variation trend of welding characteristics through the optimal route selection end, the optimal welding sequence is intelligently calculated to reduce directional abrupt changes during the welding process and reduce the risk of thermal stress concentration. The inflection point calibration end accurately identifies the path inflection point through geometric analysis and dynamically adjusts the welding current according to the angle characteristics, effectively solving the welding discontinuity problem that is prone to occur in acute angle areas. The modular design allows for flexible system expansion, enabling rapid adaptation to the welding needs of different casting models and providing standardized intelligent welding solutions for the manufacturing industry. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the principle framework of the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] As an example of this application, please refer to Figure 1 This application provides an intelligent welding system based on casting processing, including a model generation end, a feature verification and processing end, a cloud database, an inflection point calibration end, an optimal route selection end, and a welding center; Among them, the model generation end and the cloud database are electrically connected to the input node of the feature verification processing end, and the feature verification processing end is electrically connected to the input nodes of the optimal route selection end and the inflection point calibration end, respectively, and the inflection point calibration end and the optimal route selection end are electrically connected to the input node of the welding center. In the model generation stage, machine vision equipment is used to acquire omnidirectional images of the relevant castings. Based on these images, a model of the casting belonging to the current casting is generated. After model generation, multiple individual models are placed in a complete state, with corresponding gaps between them. This allows for effective differentiation of individual models. Subsequent analysis confirms the corresponding gaps. Specifically, the method of generating a model body from a target image is quite common in existing technologies, so it will not be elaborated on here. It is disclosed in patent publications such as CN120219530A, CN116229201A, CN120047564A, and CN120031090A. The associated generation steps generally include: Align the camera / sensor intrinsic parameters with the multi-device coordinate system, plan the scan path, and set the overlap rate to 50%-80%; Point cloud and RGB images are acquired by scanning from multiple angles, and the stitching effect is previewed in real time to ensure that there are no obvious gaps. Point cloud noise and outliers are removed by filtering, and image denoising and edge contrast are enhanced. Feature matching is performed using coarse registration, and then global textured point clouds are generated using fine registration. By calculating the point cloud normal vectors, an implicit surface is fitted, and a triangular mesh is generated; Then, UV unwrapping of the parametric mesh is used to map RGB pixels, and interpolation is used to fill the texture. Multi-view maps are fused, and the accuracy is verified by comparing with CAD drawings to generate a 3D model belonging to the current image.

[0014] In the feature verification processing end, based on the model body and related casting models pre-stored in the cloud database, the individual model units existing in the relevant casting models are locked in sequence. Based on the contact state between the corresponding individual model units, the gap to be welded is locked. In combination with the three-dimensional coordinate system, the similar welding segments existing in the gap to be welded are locked. Specifically, the so-called similar welding segments are welding segments with relatively consistent trends. After the similar welding segments are confirmed, it is convenient to determine the specific route in the subsequent process, so as to achieve the best welding effect.

[0015] In the optimal route selection process, based on several groups of similar weldable segments identified within the weld gap, the welding characteristics associated with each similar weldable segment are confirmed. Then, according to different sorting processes, the optimal sorting process is selected to lock in the optimal route. The optimal route is the route with the least change in the corresponding trend during the welding process, which facilitates the subsequent welding process.

[0016] As a further embodiment of Embodiment 1, this mainly focuses on the related processing steps of the feature verification processing end. The specific confirmation process of the feature verification processing end for the weld gap includes: For the generated casting models, pre-stored model bodies are extracted from the cloud database. Each model body is associated with a single casting. The model bodies are compared and verified with the relevant casting models in turn. Model bodies that are completely consistent are identified, and the corresponding model areas are recorded as the model units associated with the corresponding model bodies. In other words, they are compared one by one. For example, for a base, there is a corresponding base model for comparison. Based on the overall contour features of the corresponding model, a completely consistent model unit can be found. The corresponding model unit is the corresponding model body. During the comparison, the center points of the two models are aligned, and then the rotation comparison verification method is used to confirm whether the two model bodies are completely consistent. If they are completely consistent, they are the same model bodies. If they are not completely consistent, other model bodies are verified. Since the component models involved in the relevant casting models are all fixed, it is easy to identify the model associated with the corresponding component. In other words, it is to verify the different model bodies contained in the relevant casting models one by one. Based on the different model units associated with different related casting models, we identify pairs of model units with contact points. These contact points are points that are adjacent in spatial features. We then confirm several contact points associated with each pair of model units (i.e., the gap area between the two model bodies; a contact point is a three-dimensional spatial feature that exists during the construction of the corresponding model body, which can be understood as three-dimensional spatial coordinates. Based on these three-dimensional spatial coordinates, we can identify two adjacent points between two corresponding model bodies, indicating a contact point). We connect the several contact points identified for each individual model unit to confirm the contact area. Then, we confirm the outer edge contour of the contact area and record the generated edge contour as the weld gap associated with each pair of model units. Each pair of model units represents a single model unit. Since a single model unit cannot generate weld seams, the weld seam is confirmed based on the contact points on the model unit during the subsequent welding process. This effectively confirms the accuracy of the corresponding seam. In the preset cloud database, different model units have pre-defined splicing states. In the corresponding splicing state, the corresponding weld seams between two model units can be determined. The so-called outer edge contour refers to two surfaces A and B. When A and B come into contact, the contact area between the two surfaces is the inner area. The outer contour of the inner area is the contact seam that can be seen by the naked eye. The corresponding contact seam is the specific seam that needs to be welded. In actual operation, relevant personnel will also assemble the corresponding two sets of model units according to the pre-defined splicing states to facilitate subsequent welding.

[0017] Specifically, the feature verification processing end uses the following method to lock similar sections to be welded: Based on the weld gaps identified between each pair of model bodies, the weld gaps are combined with a three-dimensional coordinate system. The three-dimensional spatial coordinates associated with different contact points within the weld gaps are recorded and denoted as (X... k Y k Z k ), where k represents different contact points; Next, the variation characteristics associated with adjacent contact points within the weld gap are confirmed sequentially, and the three-dimensional spatial coordinates of the point preceding the adjacent contact point are determined as (X... q Y q Z q The three-dimensional spatial coordinates of the next point are determined as (X... h Y h Z h ), where q∈k, h∈k, using: Identify the changing characteristics of the corresponding contact segments at adjacent contact points, and then identify the different changing characteristics associated with different contact segments in sequence. Based on the identified contact segments, adjacent contact segments are further identified as being of the same type: The two changing characteristics associated with two adjacent contact segments are labeled as Q1 and Q2, respectively: If Q1∈[Q2-0.3×Q2,Q2+0.3×Q2], then the two contact segments are divided into segments of the same type; If Q1∉[Q2-0.3×Q2,Q2+0.3×Q2], then no division is made, and the same confirmation method is used to confirm the same type of segment for the two adjacent contact segments in turn (that is, the confirmation process of the running trend of the corresponding weld trajectory. When the running trends are basically the same, then they belong to the corresponding same type of segment. When the corresponding running trends are significantly different, then they do not belong to the corresponding same type of segment). The confirmed same type of segments are integrated. If the adjacent contact segments before and after are the same type of segment, and the two adjacent contact segments afterward are also the same type of segment, then they are uniformly divided into the same type of segment. The integrated same type of segment is recorded as the same type of segment to be welded. Several same type of segments to be welded within the weld gap are confirmed. If there is a contact segment that does not have a same type of segment (which is unlikely), then the corresponding single contact segment is directly recorded as the same type of segment to be welded. Specifically, a weld seam may have several bending areas. In order to effectively identify and divide the trend of the corresponding bending areas, it is necessary to identify the change characteristics between adjacent contact points. From the specific change characteristics, identify similar segments with relatively consistent trends to facilitate the welding process of subsequent welding components and to effectively select the trajectory. During welding, the welding components should avoid the areas with the greatest change characteristics to effectively ensure accuracy.

[0018] As a further embodiment of Embodiment 1, the focus is on the processing procedures related to the optimal route selection end. The optimal route selection end's process for locking the optimal route includes: The average value of several sets of change trends associated with a single weldable segment of the same type is processed, and the confirmed average trend is used as the welding feature associated with the corresponding weldable segment of the same type. A random segment of the same type to be welded is selected as the starting segment. The welding features associated with subsequent segments of the same type to be welded are sorted in a clockwise direction. Simultaneously, segments of the same type that do not appear consecutively are sorted in the same clockwise direction. This process continues until the welding features of all segments of the same type to be welded are sorted, generating a feature sequence. (If the gap to be welded is a closed loop, it is not necessary to sort segments of the same type that do not appear consecutively. If it is not a closed loop, and the selected segment of the same type to be welded is located in the middle of the gap, then after sorting, the last group of segments of the same type to be welded is not connected to the first group of segments of the same type to be welded. In this case, after the original sorting sequence, the first group of segments of the same type to be welded is sorted again in a clockwise direction. The sorted numerical sequence is placed after the original sorting sequence to form a complete sequence.) The absolute value of the difference between adjacent welding features in the feature sequence is confirmed, and the confirmed absolute values ​​are summed to lock the process feature (that is, the summed total value) of the current sorting process. Then, select other similar segments to be soldered as starting segments in turn, and use the same sorting method as the above sorting process to confirm the process characteristics associated with the current sorting process. Continue in this way until all similar segments to be soldered are selected as starting segments. Then, select the minimum value from the process characteristics of different sorting processes, and take the sorting process associated with the minimum value as the optimal process. Take the sorting method associated with the optimal process as the optimal route (there are starting segments and ending segments during sorting, so the route can be determined directly). Example: Given a weld gap containing five similar weldable segments, A, B, C, D, and E, with confirmed welding characteristics of 1, 1.3, 3.1, 2.4, and 2.5 respectively, and segment A as the starting point, the confirmed characteristic sequence is {1, 1.3, 3.1, 2.4, 2.5}, and the resulting process characteristic is: 0.3 + 1.8 + 0.7 + 0.1 = 2.9. With B as the starting segment, the corresponding feature sequence is {1.3, 3.1, 2.4, 2.5, 1}, and the resulting process feature is: 1.8 + 0.7 + 0.1 + 1.5 = 4.1; With C as the starting segment, the characteristic sequence is {3.1, 2.4, 2.5, 1, 1.3}, and its process characteristic is: 0.7 + 0.1 + 1.5 + 0.3 = 2.6; With D as the starting segment, the characteristic sequence is {2.4, 2.5, 1, 1.3, 3.1}, and its process characteristic is: 0.1+1.5+0.3+1.8=3.7; With E as the starting segment, the characteristic sequence is {2.5, 1, 1.3, 3.1, 2.4}, and its process characteristic is: 1.5 + 0.3 + 1.8 + 0.7 = 4.3; Therefore, if C is taken as the starting segment and the optimal solution is found, then the optimal route is CDEAB. This method of determining the optimal route is most applicable when the gap to be welded is in a closed loop state. Of course, it is also applicable when it is not in a closed loop state. It can effectively reduce the change trajectory and change parameters between the corresponding welding components, and thus effectively control the error. If the change state is large during welding, the probability of the error will also be greater.

[0019] As a second embodiment of this application, the first embodiment pertains to the process of determining the optimal route during welding. In order to achieve a better determination method, it is also necessary to confirm the inflection points in the route. In order to achieve a better fixing effect, the welding current at the inflection points during welding is adjusted to achieve optimal stability. In the second embodiment, the main execution terminals are the inflection point calibration terminal and the welding center; Among them, the inflection point calibration end identifies adjacent similar weldable sections from the confirmed weld gaps, marks the turning points associated with adjacent similar weldable sections as inflection points, and records the angular characteristics associated with the inflection points. The welding center, based on the confirmed optimal route and the determined inflection points and angle characteristics, controls the welding of the welded parts to perform welding treatment on the weld gaps. The welded parts are the same as those used in automated equipment on the market. As a further embodiment of Example 2, the specific confirmation method for the angular feature confirmation process of the inflection point calibration end includes: Based on the marked inflection point, the adjacent contact points associated with the inflection point before and after the inflection point are respectively recorded as the front point and the back point; And confirm the plane containing the previous point, the inflection point, and the subsequent point (because it is an inflection point, that is, a point of change, the adjacent points before and after it are not on the same straight line as this inflection point, so only one plane can be confirmed, not multiple planes). From the direct line of sight of the plane, confirm the characteristic angle J associated with the inflection point. o Here, 'o' represents different inflection points. Connecting the previous point to the inflection point confirms the first line, and connecting the subsequent point to the inflection point confirms the second line. The minimum angle between the first and second lines is the characteristic angle J associated with the corresponding inflection point. o ; The inflection points marked in the weld seam and the corresponding characteristic angle J associated with those inflection points are identified. o It is then transferred to the welding center.

[0020] The welding process for weld seams at the welding center specifically includes: According to the preset operating parameters, the welding parts are welded to the weld seam along the confirmed optimal route. The operating parameters are preset parameters, which are determined by the operator based on experience, including welding current, welding speed and specific contact force, etc. When the welded parts reach the designated inflection point, the welding current H is adjusted, and the adjusted welding current for different inflection points is denoted as Ho, where 'o' represents different inflection points. The formula is: Ho = H + |J o-180°|×C1, where C1 is a preset fixed coefficient factor. Its specific value is determined by the operator based on experience. All data are dimensionless and numerically calculated. Welding is performed on different inflection points based on the confirmed welding current to complete the welding process of the corresponding weld seam. Specifically, the sharper the corresponding inflection point angle, the greater the deviation from the 180° angle. In the actual welding process, this makes it easier for welding errors to occur (i.e., there are still tiny gaps after welding). In order to effectively reduce such welding errors, it is necessary to specifically adjust the welding current associated with the specified inflection point to achieve the optimal welding process and fully ensure the stability associated with the welding process.

[0021] Some of the data in the above formulas are numerical calculations with dimensions removed, and the contents not described in detail in this specification are all prior art known to those skilled in the art.

[0022] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. An intelligent welding system based on casting processing, characterized in that, include: On the model generation end, a relevant casting model is generated based on the all-around image of the relevant casting. The feature verification processing end locks the individual model units in the relevant casting models in turn according to the model bodies and related casting models stored in the cloud database. Based on the contact state between the corresponding individual model units, it locks the gap to be welded and then locks the same type of weldable segments in the gap to be welded in combination with the three-dimensional coordinate system. The optimal route selection process involves identifying several groups of similar weldable segments within the weld gap, confirming the welding characteristics associated with each similar weldable segment, and then selecting the optimal sorting process according to different sorting procedures to lock in the optimal route.

2. The intelligent welding system based on casting processing according to claim 1, characterized in that, The specific method by which the feature verification processing terminal locks the weld gap is as follows: For the generated casting models, the pre-stored model bodies are extracted from the cloud database. Different model bodies are compared and verified with the relevant casting models in turn. The model bodies that are completely consistent are locked and the corresponding model bodies are recorded as the model bodies associated with the relevant casting models. Extract individual model units with contact features from the relevant casting model, then confirm several contact points associated between pairs of individual model units, connect the several contact points confirmed by a single individual model unit to confirm the contact area, then confirm the outer edge contour of the contact area, and record the generated outer edge contour as the weld gap associated between the current pairs of model bodies.

3. The intelligent welding system based on casting processing according to claim 2, characterized in that, The specific method by which the feature verification processing terminal locks similar weldable sections is as follows: Based on the weld gaps identified between each pair of individual model units, the weld gaps are combined with a three-dimensional coordinate system. The three-dimensional spatial coordinates associated with different contact points within the weld gaps are recorded and denoted as (X... k Y k Z k ), where k represents different contact points; Next, the changing trends associated with adjacent contact points within the weld gap are confirmed sequentially, and the three-dimensional spatial coordinates of the point preceding the adjacent contact point are determined as (X... q Y q Z q The proposed three-dimensional spatial coordinates of the next point are (X... h Y h Z h ), where q∈k, h∈k, using: Confirm the changing trends of the corresponding contact segments at adjacent contact points, and then confirm the different changing trends associated with different contact segments in sequence; Based on the identified contact segments, adjacent contact segments are identified as similar segments, and the identified similar segments are integrated to identify similar segments to be welded.

4. The intelligent welding system based on casting processing according to claim 3, characterized in that, The specific methods for integrating similar segments include: The two changing characteristics associated with two adjacent contact segments are labeled as Q1 and Q2, respectively: If Q1∈[Q2-0.3×Q2,Q2+0.3×Q2], then the two contact segments are divided into segments of the same type; If Q1∉[Q2-0.3×Q2,Q2+0.3×Q2], then no division is made, and the same confirmation method is used to confirm the same type of segment for the two adjacent contact segments in turn. The confirmed same type of segments are then integrated. If the adjacent contact segments before and after are the same type of segment, and the two adjacent contact segments afterward are also the same type of segment, then they are uniformly divided into the same type of segment. The integrated same type of segment is recorded as the same type of segment to be welded. Several same type of segments to be welded that exist in the weld gap are confirmed. If there is a contact segment that does not have a same type of segment, then the corresponding single contact segment is directly recorded as the same type of segment to be welded.

5. The intelligent welding system based on casting processing according to claim 1, characterized in that, The specific method for locking the optimal route at the optimal route selection endpoint is as follows: The average value of several sets of change trends associated with a single weldable segment of the same type is processed, and the confirmed average trend is used as the welding feature associated with the corresponding weldable segment of the same type. Randomly select a segment of the same type to be welded as the starting segment, and sort the welding features associated with the subsequent segments of the same type to be welded in a clockwise direction. At the same time, sort the segments of the same type that do not appear consecutively, in the same clockwise direction. After the welding features of all segments of the same type to be welded are sorted, a feature sequence is generated. The absolute value of the difference between adjacent welding features in the feature sequence is confirmed, and the confirmed sets of absolute values ​​are summed to lock the process features of the current sorting process. Then, select other similar segments to be soldered as starting segments in turn, and use the same sorting method as the above sorting process to confirm the process characteristics associated with the current sorting process. Continue in this way until all similar segments to be soldered are used as starting segments. Then, select the minimum value from the process characteristics of different sorting processes, take the sorting process associated with the minimum value as the optimal process, and take the sorting method associated with the optimal process as the optimal route.

6. The intelligent welding system based on casting processing according to claim 1, characterized in that, Also includes: Inflection point calibration end: From the confirmed weld gap, identify adjacent weld segments of the same type, mark the turning point associated between adjacent weld segments of the same type as the inflection point, and record the angular characteristics associated with the inflection point. The welding center, based on the confirmed optimal route and the determined inflection points and angle characteristics, controls the welding of the welded parts to perform welding treatment on the weld gaps. The welded parts are the same as those used in commercially available automated equipment.

7. The intelligent welding system based on casting processing according to claim 6, characterized in that, The specific method for confirming the inflection point angle characteristics at the inflection point calibration end is as follows: Based on the marked inflection point, the adjacent contact points associated with the inflection point before and after the inflection point are respectively recorded as the front point and the back point; Then, confirm the plane containing the previous point, the inflection point, and the subsequent point. From the direct line of sight of the plane, confirm the characteristic angle J associated with the inflection point. o Here, 'o' represents different inflection points. Connecting the previous point to the inflection point confirms the first line, and connecting the subsequent point to the inflection point confirms the second line. The minimum angle between the first and second lines is the characteristic angle J associated with the corresponding inflection point. o ; The inflection points marked in the weld seam and the corresponding characteristic angle J associated with those inflection points are identified. o It is then transferred to the welding center.

8. The intelligent welding system based on casting processing according to claim 6, characterized in that, The welding process of the welding center for the weld seam specifically includes: The welding joint is welded according to the preset operating parameters, which are the preset parameters. When the welded parts reach the specified inflection point, the welding current H is adjusted, and the adjusted welding current for different inflection points is denoted as Ho, where 'o' represents different inflection points. The formula is: Ho = H + |J o -180°|×C1, where C1 is a preset fixed coefficient factor. Welding is performed on different inflection points according to the confirmed welding current to complete the welding process of the corresponding weld seam.

Images