Intelligent optimization method for forming tooth green remanufacturing process parameters

Abstract

This invention discloses an intelligent optimization method for green remanufacturing process parameters of formed teeth, relating to the field of gear repair technology. The method includes: acquiring the 3D point cloud of the formed tooth workpiece, design tooth profile reference data, material heat treatment information, and damage area detection data; constructing a remanufacturing feasible domain and generating an allowable margin window; establishing a coupled process parameter model covering patch forming and finishing based on the remanufacturing feasible domain and the allowable margin window, and solving for the initial parameter set; performing patch forming and finishing in partitions according to the initial parameter set, and correcting local process parameters based on the comparison results of measured and predicted states; and performing feedback write-back based on the final quality judgment results and process records. This invention achieves cross-process collaborative control driven by real defect geometry, improves the consistency of final tooth profile restoration and process stability, and takes into account unit repair energy consumption, patch overflow volume, and secondary removal volume constraints.

Description

Technical Field

[0001] This invention relates to the field of gear repair technology, specifically to a method for intelligent optimization of process parameters in the green remanufacturing of formed teeth. Background Technology

[0002] Formed teeth refer to workpieces that directly bear the meshing force in transmission components such as gears, gear rings, and racks. They are typically subjected to high loads, high impacts, and long-term friction conditions, frequently resulting in failures such as wear, pitting, crack propagation, and localized tooth breakage. For tooth parts that are large in shape, have long manufacturing cycles, and require high standards for materials and heat treatment, direct scrapping and replacement are relatively expensive. Therefore, remanufacturing is used in engineering to extend their service life. Currently, common process routes include damage detection, defect removal, beveling or groove preparation, laser cladding or other additive manufacturing, machining to remove excess material, and performance analysis. For tooth surfaces with high precision requirements, forming or grinding is also necessary to obtain the final tooth profile.

[0003] Chinese patent document CN112139495A proposes a method for repairing broken gear teeth using additive remanufacturing. The document describes a complete repair process where the broken gear tooth is first ground smooth, and grooves are created on the broken tooth to provide space for subsequent material forming. These grooves are distributed along the tooth thickness and / or tooth width, and can be inverted triangles, inverted trapezoids, or U-shapes, with 1 to 10 grooves in total. The workpiece is then cleaned and degreased. Following the pretreatment, additive manufacturing with simultaneous feeding is used for the repair. The energy source can be laser, electric arc, electron beam, or plasma arc, and the feeding method can be powder feeding or wire feeding. According to the document, the process involves first filling the grooves, then stacking the remaining broken tooth, followed by additive manufacturing, and finally machining to remove the surface repair allowance, restoring the repaired shape to its original form. The bonding strength, hardness, wear resistance, and flaw detection results of the restored gear tooth are then evaluated. It is evident that this technology has a relatively complete repair process, including defect pretreatment, material filling, excess material removal, and performance evaluation.

[0004] However, the aforementioned technologies still have certain shortcomings in the remanufacturing of high-value formed teeth. On the one hand, the main process of this solution is still to complete the broken tooth repair through additive manufacturing. Although there is a process for removing excess material, the determination of process parameters still revolves around the previous material forming. For recycled parts after long-term operation, there are significant differences in the depth of defects, local curvature, remaining matrix thickness, microstructure, and heat dissipation conditions at different tooth positions. The heat input of additive manufacturing, the stability of the molten pool, and the degree of dilution will change with local geometry and material state. Therefore, the geometric distribution and metallurgical state of the repair layer are not stable. On the other hand, the final precision of the tooth surface still requires subsequent machining or grinding. During the tooth finishing process, there will be influencing factors such as fluctuations in cutting or grinding load, grinding wheel load, and heat accumulation. If the state of the previous repair layer is inconsistent with the subsequent finishing conditions, problems such as large or insufficient local allowances, difficulties in reprocessing, increased material and energy consumption, and fluctuations in the final tooth shape quality may occur, affecting the first-time repair pass rate and the economics of remanufacturing.

[0005] Therefore, the technical problem that needs to be solved in the existing technology is: how to reasonably determine the process parameters in the remanufacturing process of formed teeth by combining the differences of damaged workpieces and the coupling effects of multiple processes, while taking into account the final tooth shape restoration quality, process stability and resource and energy consumption constraints. Summary of the Invention

[0006] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides an intelligent optimization method for green remanufacturing process parameters of formed teeth. The method constructs a remanufacturing feasible domain and generates an allowable margin window. Based on the remanufacturing feasible domain and the allowable margin window, a coupled process parameter model covering both filler forming and finishing is established, and an initial parameter set is solved. Filler forming and finishing are performed in partitions according to the initial parameter set, and local process parameters are corrected based on the comparison between measured and predicted states. Feedback is written back based on the final quality judgment results and process records. This achieves cross-process collaborative control driven by real defect geometry, improves the consistency of final tooth profile recovery and process stability, and solves the technical problems described in the background art.

[0007] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: The intelligent optimization method for green remanufacturing process parameters of formed teeth includes the processing of filler forming, finishing, and inspection of the formed tooth workpiece to be remanufactured. The method comprises the following steps: Step 1: Acquire the 3D point cloud, design tooth profile reference data, material / heat treatment information, and damage area detection data of the formed tooth workpiece to construct a remanufacturing feasible region and generate an allowable margin window; Step 2: Based on the remanufacturing feasible region and allowable margin window, establish a coupled process parameter model covering filler forming and finishing, and solve for the initial parameter set; Step 3: Perform filler forming and finishing in partitions according to the initial parameter set, and correct local process parameters based on the comparison between measured and predicted states; Step 4: Perform feedback write-back based on the final quality judgment results and process records to correct the partitioning rules of Step 1 or the target weights, constraint boundaries, and initial parameter values ​​of Step 2.

[0008] Furthermore, before constructing the remanufacturing feasible domain in step one, the three-dimensional point cloud is registered with the design tooth profile reference data, and the defect depth, local curvature, remaining tooth root thickness, local reachability angle, and adjacent tooth reference relationship are extracted to form a defect geometric descriptor and partition information.

[0009] Furthermore, when generating the allowable margin window in step one, the target recovery profile of the intermediate forming layer is determined based on the defect geometry descriptor, and the upper and lower boundaries of the allowable margin window are determined based on the forming shrinkage compensation amount, the finishing removal allowance amount, and the disturbance allowance amount.

[0010] Furthermore, in step one, when there is measurement occlusion or missing point cloud, the missing area is filled in according to the reference relationship between adjacent teeth; when the remaining matrix thickness, crack level or accessibility is lower than the corresponding threshold, the corresponding position is removed from the remanufacturing feasible domain and a decision to switch to alternative processes is output.

[0011] Furthermore, in step two, the process parameter coupling model simultaneously associates the material forming parameters and the finishing parameters. The material forming parameters include laser power, scanning speed, powder feed rate, overlap rate, and layer increment, while the finishing parameters include finishing feed, finishing depth, and cooling strategy.

[0012] Furthermore, in step two, candidate process combinations are screened based on the remanufacturing feasible region and allowable margin window, and the candidate process combinations are evaluated by partition according to the geometric deviation of the intermediate forming layer, the final tooth profile deviation, the unit repair energy consumption, the filler overflow volume, and the secondary removal volume to generate an initial parameter set.

[0013] Furthermore, in step three, the material forming and finishing are performed sequentially according to the tooth root zone, tooth side zone, and tooth tip zone. During the execution, the width of the molten pool, temperature, visual profile, spindle power, and the state proxy quantity of the grinding wheel or tool are collected to form the actual measured state corresponding to the zone.

[0014] Furthermore, in step three, the measured state is compared with the predicted state corresponding to the initial parameter set. When the comparison result meets the correction conditions, the local laser power, scanning speed, and powder feeding are corrected according to the corresponding partition. Measurement and finishing feed Alternatively, the triggering conditions may be adjusted; when the comparison result meets the degradation conditions, the system may fall back to the safety parameter package.

[0015] Furthermore, in step four, a final quality judgment is generated based on the tooth profile measurement results, surface roughness, bonding quality, and flaw detection results, and the target parameter package, process execution record, and qualification mark corresponding to the final quality judgment are output. The process execution record includes partition execution record, local correction record, and downgrade record.

[0016] Furthermore, in step four, the final quality judgment is associated with the process execution record and written back to the parameter model and case library. Based on the write-back results, the partitioning rules in step one, the target weights and constraint boundaries in step two, and the initial parameter values ​​are corrected for the generation of the initial parameter set of the subsequent formed tooth workpiece.

[0017] (III) Beneficial Effects This invention provides an intelligent optimization method for green remanufacturing process parameters of formed teeth, which has the following beneficial effects: The 3D point cloud of the formed tooth workpiece, the design tooth profile reference data, material heat treatment information, and damage area detection data are registered into a remanufacturing feasible region, and an allowable allowance window is generated. This ensures that subsequent processes such as patch forming and finishing are based on the actual defect geometry, avoiding deviations caused by inconsistencies in the basis of data between processes. Defect depth, local curvature, remaining tooth root thickness, local reachable angle, and adjacent tooth reference relationships are used as partitioning information. Unrepairable and repairable areas are distinguished before process decisions are made, and the initial parameter set is generated with object boundaries.

[0018] Using the remanufacturing feasible region and allowable margin window as the process parameter coupling model covering the supplementary forming and finishing, the intermediate forming layer and the final tooth profile are placed on the same decision chain, the serial process is changed to a collaborative process, and the supplementary forming and finishing are carried out according to the tooth root partition, tooth side partition and tooth tip partition according to the initial parameter set. The local process parameters are corrected by comparing the measured state with the predicted state, so that the local deviation is absorbed in the process and no longer accumulates to the final quality judgment stage.

[0019] By placing the melt pool width, temperature, visual profile, spindle power, and grinding wheel or tool status proxy quantities into the same partition execution chain, and cooperating with the allowable margin window and safety parameter package, the stability of the filler forming and the continuity of finishing are consistent, reflecting the collaborative constraints of multiple types of status information. Attached Figure Description

[0020] Figure 1 This is an overall architecture diagram of the closed-loop control system for remanufacturing formed teeth in an embodiment of the present invention; Figure 2 This is a flowchart of the defect boundary generation and remanufacturing feasible domain construction in an embodiment of the present invention; Figure 3 This is a schematic diagram of tooth surface partitioning and allowable margin window in an embodiment of the present invention; Figure 4 This is a flowchart of the partition parameter compilation and evaluation chain in an embodiment of the present invention; Figure 5 This is a closed-loop control diagram of online execution, local correction, and conservative degradation in an embodiment of the present invention; Figure 6 This is a closed-loop flowchart of final quality determination and rule rewriting in an embodiment of the present invention. Detailed Implementation

[0021] 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.

[0022] Please see Figures 1-6 This invention provides an intelligent optimization method for green remanufacturing process parameters of formed teeth, including: All of the following processing actions are executed by the remanufacturing control terminal, which is connected to the three-dimensional measurement device, crack detection device, clamping turntable, process database and field display terminal.

[0023] The first step is to transform the damaged formed tooth workpiece into a unified object that can be directly called by subsequent processes: bind the 3D point cloud, the design tooth profile reference data, the material heat treatment file and the damage detection data to the same tooth surface coordinate, and then output the defect geometric descriptor, partition information and allowable margin window.

[0024] Step 1: Before proceeding with the filler forming process, determine which parts of the damaged tooth workpiece will be included in subsequent processes, which parts should be excluded, and what filler boundaries should be retained for the parts included in subsequent processes.

[0025] Once a damaged formed tooth enters the remanufacturing unit, there are no shortage of means to visualize the damage on-site. The real challenge lies in transforming the observed defects into boundaries that can be directly used in subsequent processes. The tooth tip, tooth flank, and tooth root of a formed tooth differ in curvature, heat conduction path, and tool entry direction. The same defect at different locations corresponds to different repair risks. If step one outputs a regular point cloud or a regular defect map, step two cannot distinguish between the designed repair area and the actual material loss, nor can it place crack risk, remaining matrix thickness, and on-site entry channels on the same decision chain.

[0026] Therefore, step one does not end with the scanning process, but rather integrates the design tooth profile reference data, 3D point cloud, material heat treatment archive, and damage detection data into boundary results that directly drive subsequent processes. For example, the operator clamps the cleaned, damaged tooth workpiece onto a gripping turntable. The 3D measuring device scans along the tooth width direction across the tooth tip, flank, and root. The crack detection device then performs supplementary detection on the tooth root and fracture edges. Simultaneously, the on-site display terminal displays the envelope of the damaged area, the undamaged reference area, and the obscured area. The remanufacturing control terminal then overlays these results onto the design tooth profile reference data.

[0027] The design employs a collaborative principle: using tooth profile reference data to define the framework, 3D point cloud data to determine the current state, material heat treatment records to define load-bearing boundaries, and damage detection data to define risk boundaries. These four types of data are first unified to tooth surface coordinates, then compared point-by-point at the same location. This integrates three key questions—how much is missing, whether it can withstand subsequent heat input, and whether there is an execution path on-site—into a single processing chain. The material heat treatment record includes at least the steel grade, surface hardening method, carburized layer depth, tempering records, and existing repair records to simultaneously determine geometric and load-bearing boundaries at the same tooth surface location.

[0028] After this process, once any location is marked as a repairable area, an unrepairable area, or an alternative process area, the mark corresponds to the same tooth surface location, and subsequent steps can directly inherit it.

[0029] The remanufacturing control terminal first expands the design tooth profile reference data into a tooth surface coordinate skeleton, and then registers the three-dimensional point cloud with an undamaged reference band. After registration, it first identifies the defect boundary, crack boundary and remaining matrix boundary, and then superimposes the local reachability conditions to obtain the remanufacturing feasible region. Within the remanufacturing feasible region, the target recovery profile and allowable margin window are generated.

[0030] Furthermore, the first step is to address the depth of the defect, but instead of using rigid fitting of the entire tooth, an undamaged reference zone is selected first, and then the remaining areas are dragged into a unified coordinate system using this undamaged reference zone. The remanufacturing control terminal extracts the theoretical tooth flank surface, theoretical tooth tip transition line, and theoretical tooth root transition fillet line from the design tooth profile reference data and combines them into the target tooth surface skeleton; subsequently, a continuous and stable undamaged reference zone is identified in the 3D point cloud. The undamaged reference zone is preferentially selected from the tooth flank region far from the fracture edge, and secondarily from the undamaged region corresponding to adjacent teeth. It is important to note that if the entire tooth point cloud is directly registered, large volume defects will pull the registration result inwards towards the fracture, and the subsequent defect depth will be underestimated.

[0031] The design tooth profile reference data is preferably obtained from the original design model, tooth profile inspection files of new parts of the same model, or manufacturing archives. When the original design model is missing, the design tooth profile reference data is reconstructed from the tooth flank surfaces, tooth tip transition lines, and workpiece pitch circle parameters of adjacent complete teeth. The material heat treatment archive is preferably obtained from the manufacturing archives. When the manufacturing archives are missing, they are supplemented by spectral composition detection, surface hardness gradient detection, and metallographic verification results. Damage detection data is generated by at least one of magnetic particle testing, eddy current testing, and phased array ultrasonic testing, and is uniformly mapped to tooth surface coordinates after import.

[0032] The remanufacturing control terminal first employs a rigid registration method driven by an undamaged reference band to transform the original measurement point cloud into a registration point cloud. Registration is achieved by minimizing the normal distance between the undamaged reference band and the designed tooth profile reference data to obtain the rotation matrix and displacement vector. Subsequently, the registration point cloud is converted into tooth surface coordinate positions through tooth surface unfolding mapping, where the tooth height coordinates are normalized coordinates along the tooth height direction and the tooth width coordinates are normalized coordinates along the tooth width direction. This allows subsequent defect depth, crack grade, remaining matrix thickness, and reachability conditions to be invoked under the same positional semantics.

[0033] After completing the reference band registration, the control terminal is then manufactured to establish the defect depth field along the design normal: Among them, the depth field of the defect Tooth surface coordinate position Material loss along the design normal, with a value range of non-negative real numbers, used to convert surface geometric differences into depth values ​​directly used in subsequent processes; tooth surface coordinate position. : A single position identifier after unfolding along the tooth height and tooth width directions. The value range is the effective unfolded area of ​​the damaged formed tooth workpiece surface, used to ensure that subsequent parameters are called at the same position; design normal vector Design tooth profile reference data at the coordinate position of the tooth surface The unit normal at a given location, with its value range being the set of unit vectors, is used to limit the physical direction for shape complement determination; Design position vector Design tooth profile reference data at the coordinate position of the tooth surface The spatial position, with a value range of three-dimensional spatial coordinates, is used to provide a theoretical tooth surface reference; the measurement position vector : Coordinate position of 3D point cloud on tooth surface The registered spatial position, with a value range of three-dimensional spatial coordinates, is used to provide the current surface condition of the damaged formed tooth workpiece; When occlusion or point cloud defects occur, the manufacturing control terminal first determines whether the missing location belongs to the tooth tip occlusion area, the tooth root reflection area, or the fracture shadow area. If it belongs to one of these areas, it sequentially calls the two-level completion rules of adjacent tooth mirror mapping and same-tooth reference band interpolation, and then substitutes the completion result back into the measurement position vector. The missing position.

[0034] For tooth surface coordinates where occlusion or point cloud defects occur, the remanufacturing control terminal first reads the reference position vectors of adjacent complete teeth at the same unfolded position, and then reads the mirror reference position vectors about the tooth groove symmetry plane. The coordinates are then completed using a combination of confidence coefficients obtained by normalizing the adjacent tooth integrity score and the symmetry tooth integrity score. If only one is available, that coefficient is directly used as the completion result. After completion, the obtained position vectors are then substituted back to the missing positions of the measured position vectors.

[0035] For example, in a scenario involving chipped tooth roots of a carburized and quenched gear, voids appear on the inner side of the fracture surface after the first scan. The remanufacturing control terminal first calls the contour of adjacent teeth at the same height to restore the outer boundary, and then uses the reference band of this tooth to advance towards the edge of the fracture surface, ultimately forming a continuous depth color band on the on-site display terminal. This separates the design modification from the actual material loss and provides a depth basis with a consistent normal direction for subsequent boundary screening.

[0036] After establishing the defect depth field, not all depth defect locations proceed to subsequent processes. After chipping or breaking teeth, damaged formed tooth workpieces often have near-surface cracks, localized structural deterioration, or insufficient load-bearing thickness near the tooth root. If these factors are ignored, although subsequent patch forming covers the surface defects, the substrate still cannot withstand the heat input and finishing load.

[0037] Therefore, step one continues to construct the remanufacturing feasibility domain around the remaining matrix thickness, crack grade, and local accessibility conditions. Here, local accessibility conditions are not a single nozzle angle, but rather a comprehensive result of the combined effects of the filler forming head entering the channel, the detection probe coverage direction, and the subsequent finishing tool entering the corridor.

[0038] The remanufacturing control terminal generates a remanufacturing feasible domain based on this: Among them, remanufacturing is feasible. : The set of tooth surface coordinate positions that are allowed to proceed to subsequent processes, with values ​​ranging from the tooth surface coordinate positions. A subset of, used to define the boundaries for continued remanufacturing; remaining matrix thickness Tooth surface coordinate position The effective load-bearing thickness is measured along the tooth root inwards, and its value range is a non-negative real number. It is used to prevent further heat input from being applied to a weak substrate. Thickness threshold : Minimum remaining matrix thickness required to enter the remanufacturing feasible region, with a value range of positive real numbers, used to give the lower limit of load-bearing capacity; Crack grade Damage detection data at the coordinate position of the tooth surface The crack risk level given is a closed interval. A higher value indicates a higher risk, and is used to include surface and near-surface defects in the assessment; level threshold. The highest crack risk level that allows for continued remanufacturing, with a value range that is a closed interval. Local reachability coefficient Tooth surface coordinate position The combined entry conditions for the material forming execution head and the subsequent finishing tool are within a closed interval. A larger value indicates a more complete channel, which is used to exclude locations with insufficient channels. Reachable threshold The minimum local reachability coefficient that allows for continued remanufacturing, with a value range that is a closed interval. Used to specify the execution boundary; upper limit of depth The maximum defect depth that is still allowed to enter the remanufacturing feasible region under the current process route. The value range is a non-negative real number, used to exclude ultra-deep defects. The local reachability coefficient is the smaller of the nozzle reachability score and the tool reachability score, and is determined by the incident angle of the forming head relative to the local normal at the current position, the minimum clearance between the nozzle or tool profile and adjacent teeth, and the continuous free travel required to complete the current machining. The result is trimmed to the range of zero to one. The thickness threshold, grade threshold, reachability threshold, and depth limit are not manually specified, but are retrieved from the process database according to the three-dimensional index of material heat treatment category-damage type-equipment capability; when no complete match is found in the process database, the linear interpolation result of two adjacent parameter levels is used.

[0039] In terms of microscopic implementation, the remaining matrix thickness Crack grade determined jointly by 3D point cloud and material heat treatment archive. Obtained by any one of the following methods: magnetic particle testing, eddy current testing, or phased array ultrasonic testing; local reachability coefficient. The calculation is based on the incident angle of the material forming head, the nozzle outline and the gap between adjacent teeth, and the entry trajectory of the finishing tool.

[0040] For example, if a tooth root location on the on-site display terminal simultaneously exhibits deep defects, high crack levels, and obstruction of the tool entry corridor, the remanufacturing control terminal will directly mark it as an unsuitable repair area; if a tooth side location has spalling but the remaining matrix is ​​continuous, the crack level is low, and the execution channel is intact, then that location is retained within the remanufacturing feasible domain.

[0041] Furthermore, whether it is worth repairing, whether it can be repaired, and whether it can bear the load after repair are combined into a single boundary judgment, so that the partition information output in step one directly has process significance.

[0042] After the remanufacturing feasibility domain is determined, if the original design tooth profile reference data is used as the endpoint of the filler forming, dimensional omissions are likely to occur after subsequent finishing. If the filler forming height is increased indiscriminately, it will lead to more ineffective removal. Therefore, the output of step one is an allowable allowance window with clear upper and lower boundaries, so that as long as the subsequent filler forming surface falls within this window, it can be smoothly connected with the subsequent finishing.

[0043] The upper and lower boundaries of the allowable margin window are written as: Among them, the lower boundary margin Tooth surface coordinate position The minimum allowance required to ensure sufficient material for subsequent finishing is defined here, and its value range is a non-negative real number; upper boundary allowance. Tooth surface coordinate position The maximum allowable reserve amount is a non-negative real number that is not less than the lower boundary margin. Used to prevent excessive local shaping; molding shrinkage compensation amount Tooth surface coordinate position The allowance for additional material replenishment due to thermal shrinkage and matrix constraints is a non-negative real number, used to offset the geometric shrinkage after the filler is formed; the allowance is removed during finishing. Tooth surface coordinate position This is the amount of material retained for the subsequent finishing tool to cut stably. Its value range is a non-negative real number, used to ensure that there is a continuous load in the subsequent finishing process. Disturbance Reserve Tooth surface coordinate position This is an additional allowance for clamping attitude fluctuations, path offsets, and local reachability fluctuations. The value range is a non-negative real number, used to leave a buffer boundary for on-site execution. The forming shrinkage compensation amount is a coarse compensation amount in step one. It is retrieved from the process database based on the material heat treatment type, defect depth field, and pre-selected process route, and is used to give the initial boundary of the allowable allowance window before step two. After the process combination is determined in step two, the coarse compensation amount is refined and corrected based on the predicted amount of the intermediate forming layer. The finishing removal allowance is jointly determined by the minimum stable cutting depth of the finishing tool and the target surface formation requirements; the disturbance allowance is jointly determined by the upper bound of the clamping positioning error, the upper bound of the path following error, and the upper bound of the local reachable fluctuation.

[0044] In the specific generation process, the remanufacturing control terminal generates a candidate band of the target recovery profile along the normal direction of the designed tooth profile reference data, and then the lower boundary allowance is... and upper boundary margin It is superimposed on the candidate band, and the abrupt change position is flattened according to the continuity of local curvature and the continuity of tooth root transition fillet.

[0045] For example, in the tooth flank spalling pit scenario, a double-layer boundary will appear on the on-site display terminal, enclosing the spalling pit. The inner boundary corresponds to the lower boundary allowance, and the outer boundary corresponds to the upper boundary allowance. In the tooth root chipping scenario, the remanufacturing control terminal prioritizes ensuring the continuity of the tooth root transition fillet contour before generating the allowable allowance window for the tooth flank area. This changes the definition of "acceptable" from a single height value to a continuous window, ensuring that subsequent material forming neither results in insufficient material placement nor excessive material height.

[0046] Preferably, the three-dimensional measurement device uses a blue light structured light scanning head or a line laser scanning head, the crack detection device uses a magnetic particle detection device and a phased array ultrasonic detection device arranged in series, the clamping turntable uses a CNC turntable with angle locking function, and the remanufacturing control terminal uses an industrial computer. The line laser scanning head samples circle by circle along the tooth width direction, and the CNC turntable locks after each rotation of a single tooth angle distance to ensure repeated coverage of the coordinate positions of each tooth surface.

[0047] In terms of software workflow, the remanufacturing control terminal first runs the tooth surface coordinate unfolding program, followed by the registration program, the defect depth field generation program, the boundary screening program, and the allowable margin window generation program. If any program detects missing input, control is returned to the previous program to complete the input. As a parallel extension, the 3D measurement device can also use a combination of a contact tooth surface measurement probe and a line laser scanning head. When the material heat treatment file is insufficient, it can be replaced by metallographic verification records or hardness gradient detection records, with local reachability coefficients... It can also be deduced from the simulation results of the preset path. For double helical teeth, internal gear rings or large gear rings, only the corresponding tooth surface coordinate expansion method needs to be changed, and the logic of establishing the defect depth field, constructing the remanufacturing feasible region and generating the allowable margin window remains unchanged.

[0048] When in use, the coordinate position of each tooth surface is mapped to a unique partition state. On-site operators can directly check whether the fracture edge, crack edge and allowable allowance window coincide on the on-site display terminal, avoiding misalignment and shape compensation caused by position semantic drift in subsequent processes.

[0049] Step 2: Based on the boundary and allowable margin window formed in Step 1, generate the initial parameter set for material forming and finishing.

[0050] After the damaged formed tooth workpiece enters the remanufacturing unit, Step 1 has already identified which locations can proceed to subsequent processes and what kind of reserve layer needs to be retained at these locations. The root, flank, and tip regions of the formed tooth differ in local curvature, heat dissipation channels, nozzle entry direction, and tool contact method; the intermediate forming layer generated by the filler material directly determines the cutting continuity, depth of cut load, and final tooth profile formation process of the subsequent finishing process. If the filler material forming parameters and finishing parameters are only looked up separately in tables, situations often arise where the previous filler is too high and the subsequent finishing process is too heavy, or the previous filler is too thin and the subsequent process cannot form continuous contact. Therefore, Step 2 uses the remanufacturing feasibility domain output from Step 1. Starting with the allowable margin window, the candidate range of process combinations is narrowed down first. Then, in the same evaluation chain, the intermediate forming layer, the final tooth profile, the unit repair energy consumption, the filler overflow volume, and the secondary removal volume are examined simultaneously. Finally, the partition execution order, switching boundary, and parameter package are determined together.

[0051] Any process combination must first be accepted within the remanufacturing feasibility domain given in step one. The selection process includes filtering for allowance windows; establishing process combinations for tooth root, tooth flank, and tooth tip zones, but sharing the same material heat treatment file, patch material file, and equipment capability boundary; the reserved layer left in the patch forming stage is judged not by whether it can be fully filled, but by whether it can be continuously accepted by subsequent finishing processes. The remanufacturing control terminal sequentially calls the candidate process domain shrinkage program, the zone evaluation program, and the parameter compilation program. The output of the previous program is directly used as the input of the next program; therefore, the defect depth field output in step one... Remanufacturing Feasibility Domain Lower boundary margin and upper boundary margin In step two, the coordinates are always transferred along the same tooth surface, without starting a second set of position semantics.

[0052] In the following text, the reserved layer refers to the intermediate forming layer after the patch material is formed; the process combination refers to a single candidate solution in the candidate solution set; and the initial parameter set refers to the set of selected solutions compiled after zoning evaluation. The term "green" refers only to the joint constraints of three measurable indicators: unit repair energy consumption, patch material overflow volume, and secondary removal volume, and does not replace specific process indicators with abstract low-carbon expressions.

[0053] The remanufacturing control terminal first reads the partition information and then divides the remanufacturing feasible domain. The process is divided into tooth root zone, tooth flank zone, and tooth tip zone. Then, the material heat treatment file, filler material file, nozzle file, tool file, and cooling file are loaded. Candidate process domains are first established for each zone, and then the intermediate forming layer prediction results and finishing cut prediction results are generated. After the prediction is completed, the manufacturing control terminal performs a unified evaluation on each zone and outputs the initial parameter set, zone execution order, and zone switching boundary.

[0054] The remanufacturing control terminal establishes process combinations within each zone. This includes at least laser power. Scanning speed Powder delivery volume Overlap rate Layer increment Finishing feed Fine cutting depth and cooling strategy Among them, the root zone prioritizes limiting heat input and layer increment, the flank zone prioritizes limiting overlap rate and finishing feed, and the tip zone prioritizes limiting edge augmentation and final cut depth. The reason for this division is that the root zone is more sensitive to heat accumulation and transition fillet continuity, the flank zone is more sensitive to meshing profile continuity, and the tip zone is more sensitive to edge protrusions.

[0055] Subsequently, the remanufacturing control terminal calculates the predicted amount of the intermediate forming layer at each tooth surface coordinate position. And based on this, determine the process combination. Whether to retain it within the candidate process domain. Where: Among them, the predicted amount of intermediate forming layer Process combination At the coordinate position of the tooth surface The thickness of the reserved layer formed at the location is a non-negative real number, used to determine whether the local thickness after the filler material is formed is within the allowable allowance window; process combination : Determined by laser power Scanning speed Powder delivery volume Overlap rate Layer increment Finishing feed Fine cutting depth and cooling strategy The partitioned process objects, whose value range is a set of combinations allowed by the process database, are used to carry the unified configuration of the material forming stage and the finishing stage in step two; tooth surface coordinate position Maintaining the same mapping as step one, the value range is the remanufacturing feasible region. Effective set of positions within; laser power The energy output of the material forming equipment is within the range of allowable power registered in the equipment capacity database. Scan speed The path forward speed of the material forming actuator head, with a value range defined by the permissible speed range registered in the equipment capacity database; powder feeding rate. The feed rate is the amount of material supplied to the molten pool per unit time, and the range of values ​​is the executable supply segment of the powder feeding device; overlap rate. The coverage ratio of adjacent scan channels, with values ​​within a closed interval. Layer increment This represents the normal lifting amount between adjacent forming layers, with a range of positive real numbers, used to constrain the lifting height of a single layer. Local deposition mapping coefficient The coordinate position of the tooth surface The energy-powder conversion factor at the point of application, with a range of positive real numbers, is used to absorb the effects of local curvature, incident attitude, and material thermal conductivity. Speed ​​compensation coefficient The coordinate position of the tooth surface The low-speed translation amount at the location is a non-negative real number, used to avoid abrupt changes in conversion in the low-speed region; the overlap thickness coefficient. The coordinate position of the tooth surface The overlap thickening coefficient at the location, taking values ​​in the range of non-negative real numbers, is used to reflect the impact of overlapping adjacent melt channels on the thickening of the reserved layer; the layer increment suppression coefficient The coordinate position of the tooth surface The layer increment suppression coefficient at the location is a non-negative real number in the range of values. It is used to prevent sacrificing interlayer stability by trading excessive layer increment for surface transmittance. The local deposition mapping coefficient, velocity compensation coefficient, overlap thickening coefficient, and layer increment suppression coefficient are not set freely, but are obtained by looking up tables in the calibration database according to the material heat treatment category, local curvature, incident angle, and zone label. When the current position is between two adjacent levels, linear interpolation is used to determine the intermediate forming layer prediction amount to ensure that the prediction amount has a definite input source.

[0056] Remanufacturing control terminals only retain those that meet the requirements. and process combination Proceed to the next level of judgment. For example, after the operator selects a carburized and quenched damaged tooth workpiece with chipped tooth roots and flaking tooth sides on the on-site display terminal, the remanufacturing control terminal first retrieves the corresponding filler material file and nozzle file, then removes the high-heat input combination from the tooth root section and removes the excess material exceeding the upper boundary from the tooth side section. High-compensation, high-compensation combinations were eliminated, ultimately retaining three different cooling strategies. Candidate process combinations .

[0057] Therefore, step two converts the allowable margin window given in step one into a process combination. The entry boundary is determined by the schemes that avoid obvious boundary violations from the outset, allowing them to be included in subsequent comparisons.

[0058] After the candidate process domains have been narrowed down, step two continues to address the question of how to select among multiple process combinations that satisfy the boundary conditions. The core here is not to find the lowest value of a single indicator, but to maintain a continuous evaluation chain between supplementary forming and finishing. If only the predicted value of the intermediate forming layer is pursued... If the material falls within the allowable margin window, the subsequent finishing process may result in intermittent contact. If only the local contour after finishing is close to the restored contour, the process may result in excessive filler overflow and excessive secondary removal volume.

[0059] Therefore, the remanufacturing control terminal establishes a unified evaluation expression for each partition, compressing the position of the intermediate forming layer deviating from the center of the allowable margin window, the risk of final tooth formation, the unit repair energy consumption, and the burden of material removal volume into the same calculation chain. Specifically: Among them, the zoning evaluation value Process combination The comprehensive evaluation result in the current partition, with values ​​ranging from non-negative real numbers, serves as a unified comparison benchmark for initial parameter set selection; number of discrete locations. : The total number of local locations after the current partition is divided, with values ​​ranging from positive integers; Local prediction value. Process combination In the The intermediate forming layer prediction value, formed at discrete locations, takes the value of a non-negative real number and is used to continue the aforementioned intermediate forming layer prediction value. Discrete representation; Lower boundary margin : No. The lower boundary allowance corresponding to each discrete position, taking values ​​in the range of non-negative real numbers, is used to specify the minimum allowance requirement for material forming; the upper boundary allowance... : No. The upper boundary margin corresponding to each discrete position takes values ​​that are non-negative real numbers and are not less than the lower boundary margin. Geometric weights Final form weights Energy consumption weight and volume weight All are the first The non-negative real numbers at discrete locations respectively adjust the centering degree of the intermediate forming layer, the final tooth shape formation, the unit repair energy consumption, and the attention intensity of the volume burden in the current partition. Final shape deviation estimation term : No. A discrete location in the current process combination The remaining local contour values ​​that may still be retained after finishing are non-negative real numbers, used to incorporate the continuity of finishing into the evaluation in advance; energy consumption estimation item. : No. A discrete location in the current process combination The unit repair energy consumption corresponding to the completion of patch forming and finishing is given, with a value range of non-negative real numbers, used to uniformly convert the resource input of the patch forming stage and the finishing stage; volume estimation item. : No. A discrete location in the current process combination The combined volume of the overflow volume of the supplementary material and the volume of the secondary removal is a non-negative real number. In practice, the remanufacturing control terminal first determines the shape of the finishing tool and the finishing depth of cut. A local contact zone is generated, and then a final shape deviation estimation term is formed based on the breakage situation between the local contact zone and the restored contour. Then, based on the energy input per unit length of the filler material forming, the zoned dwell time for cooling, and the cutting time of the finishing tool, an energy consumption estimate is generated. Finally, a volume estimation term is formed based on the amount of material accumulated outside the allowable margin window and the fine-tuning removal path. .

[0060] Among them, the final shape deviation estimation term is the absolute value of the normal difference between the local contour after virtual finishing and the target restored contour under the current process combination. Virtual finishing is obtained by geometric intersection of the finishing tool envelope and the predicted contour of the intermediate forming layer. The energy consumption estimation term consists of the energy consumption of material forming and finishing. The former is calculated based on the energy input per unit length of the path, and the latter is calculated based on the spindle load baseline, finishing feed and finishing depth of cut. The volume estimation term consists of the sum of the overflow volume of material exceeding the upper boundary allowance and the volume removed by virtual finishing. The geometric weight, final shape weight, energy consumption weight and volume weight are read from the process database according to the partition type, material heat treatment category and defect level, and normalized to make the sum of each weight equal to one.

[0061] Preferably, geometric weights Increase the final shape weight in the meshing zone on the tooth flank. Increase energy consumption weight in the tooth root transition zone. Increasing volume weight in large-area shape-filling regions Improvement is achieved in edge-protrusion-sensitive areas. For example, if a certain process combination... The tooth flank partitioning brings the predicted amount of the intermediate forming layer closer to the center of the allowable margin window, but the tooth root partitioning results in a larger volume estimation term. Then the process combination It will not be incorporated into the initial parameter set; conversely, if another process combination This process combination balances the contact continuity of the tooth root zone with the central positioning of the reserved layer in the tooth flank zone. It will be retained.

[0062] Therefore, step two uses the same partition evaluation value. The two processes converge around the same recovery contour. After completing the zonal evaluation, the remanufacturing control terminal compiles the selected process combinations for each zonal into an initial parameter set that can be directly called in step three.

[0063] Preferably, continuous variables are solved using sequential quadratic programming, while cooling strategies are used for discrete variables. Branch filtering is used for path type and tool type; when the process database is stored in form, the remanufacturing control terminal first performs cubic spline interpolation on the form nodes, and then extracts candidate points of continuous variables from the interpolation surface to reduce jumps caused by excessively coarse process table granularity. The compiled initial parameter set not only includes laser power Scanning speed Powder delivery volume Overlap rate Layer increment Finishing feed and fine cutting depth It also includes cooling strategies. Partition execution order, partition switching conditions, and exception boundaries.

[0064] In the preferred implementation, the remanufacturing control terminal first compiles the initial parameter set of the tooth root section, then compiles the initial parameter set of the tooth flank section, and finally compiles the initial parameter set of the tooth tip section. The reason for this arrangement is that the tooth root section relies more on the stable transition fillet boundary, the tooth flank section relies more on the continuously unfolded recovery contour, and the tooth tip section undertakes the end sealing task.

[0065] For example, after the compilation is completed, the on-site display terminal displays the tooth root partition parameter package, tooth flank partition parameter package, and tooth tip partition parameter package respectively, and displays whether a dwell cooling was performed before the zone change and when abnormal retreat was triggered. Extending further, for damaged formed tooth workpieces of internal gear rings, only the nozzle entry direction, tool entry direction, and local accessibility conditions need to be rewritten to the inner contour version; for large gear rings repaired on-site, only the equipment capability database needs to be switched to the capability tables corresponding to mobile patch forming equipment and portable finishing equipment, while the process combination... , Zone Evaluation Value The partitioning and reorganization logic remains unchanged.

[0066] Therefore, the initial parameter set output in step two can be directly called in step three, and the same terminology and boundary system as in step one can be maintained when switching regions, cooling down, or rolling back abnormally.

[0067] When using it, rely on the partition evaluation value By incorporating intermediate forming layer deviation, final tooth shape formation risk, unit repair energy consumption, filler overflow volume, and secondary removal volume into the same evaluation chain, process trade-offs are no longer driven by single indicators. Finally, step two writes the partition execution order, partition switching conditions, and abnormal boundaries into the initial parameter set.

[0068] All of the following actions are executed by the remanufacturing control terminal. The remanufacturing control terminal is connected to the material forming equipment, finishing equipment, molten pool observation device, temperature acquisition device, contour acquisition device, spindle power acquisition device, grinding wheel load acquisition device, and process database, and receives the initial parameter set, partition execution order, and abnormal trigger boundary output from step two.

[0069] Step 3: Implement the initial parameter set formed in Step 2 into on-site actions for tooth root partitioning, tooth flank partitioning, and tooth tip partitioning. When the local state deviates from the predicted state, correct the material forming parameters and finishing parameters point by point according to the tooth surface coordinates, so that the intermediate forming layer and finishing recycling process always proceed around the allowable margin window.

[0070] The process combination obtained in step two Only the partition evaluation value must be met before execution. The parameter results are as follows. After the damaged formed tooth workpiece enters the field, the nozzle incident angle, local heat dissipation, clamping posture, and grinding wheel contact state will all change. If it is still pushed directly into the lumped area according to the initial parameters, the filler forming layer will deviate from the lower boundary allowance. or upper boundary margin Finishing loads can also accumulate in the tooth root transition zone or tooth flank corner zone.

[0071] Therefore, step three transforms the parameter results from step two into a continuous action chain of path execution, status acquisition, position comparison, local correction, and conservative degradation, ensuring that the boundaries established in the first two steps do not fail in the field.

[0072] The remanufacturing control terminal first generates a path queue according to the execution order of the partitions, and then maps each material feeding path and each finishing path to the same tooth surface coordinates. After each path is completed, the on-site status is compared with the predicted status in step two at the same location; if the comparison result shows that the current path is approaching the allowable margin window boundary or the finishing load boundary, the laser power is immediately corrected within the current partition. Scanning speed Powder delivery volume Finishing feed Fine cutting depth Or cooling strategy This process avoids waiting for the tooth-forming process to finish. With this method, the filler forming stage and the finishing stage are no longer isolated, sequential actions, but rather continuous closures along the same boundary.

[0073] The predicted values ​​for molten pool width, temperature, profile, spindle power, grinding wheel load, and residual profile are all generated by path discretization simulation of the process combination selected in step two. The remanufacturing control terminal discretizes the material filling path and finishing path into position sequences with fixed step sizes. The fixed step size is the larger value between the tooth surface coordinate grid step size and the minimum control step size of the equipment, to ensure that the predicted state and the actual state are compared at the same discrete position.

[0074] The remanufacturing control terminal first reads the initial parameter set and then sends parameter messages to the patch forming equipment or finishing equipment; the parameter messages contain partition identifiers, path identifiers, and laser power. Scanning speed Powder delivery volume Overlap rate Layer increment Finishing feed Fine cutting depth Cooling strategy Lower boundary margin and upper boundary margin After executing the current path, the equipment returns a feedback message; the feedback message contains the same partition identifier, the same path identifier, and the corresponding field status in the melt pool width, temperature, profile, spindle power, and grinding wheel load. The remanufacturing control terminal decides whether to continue execution, make partial corrections, or trigger a conservative degradation based on the feedback message.

[0075] Before the remanufacturing control terminal enters a certain partition, it first corrects the posture of the damaged formed tooth workpiece to make it consistent with the tooth surface coordinates in step one; after the posture is consistent, it only sends the parameter message corresponding to the current path and does not cross to the next path.

[0076] Preferably, the root section is executed first, followed by the flank section, and finally the tip section. The root section expands towards the flank along the transition fillet, the flank section advances layer by layer along the meshing direction, and the tip section completes the top closing after sealing the edges on both sides. This arrangement is because the root section is more sensitive to the continuity of the transition fillet, the flank section is more sensitive to the continuity of the contour, and the tip section handles edge sealing. For example, on a damaged formed tooth workpiece with chipped root corners and peeling flanks, the remanufacturing control terminal first sends parameter messages for the first pass of the first layer of the root section; after the material forming equipment completes this pass, the melt pool observation device returns an image of the melt pool width, the temperature acquisition device returns the temperature at the end of the path, and the contour acquisition device returns the local contour. The remanufacturing control terminal places these data onto discrete positions on the current path, establishing a positional basis for subsequent corrections. Step three immediately converts the execution results of each path into a comparable state, so subsequent corrections no longer rely on human experience.

[0077] After the material forming equipment completes a certain path, the remanufacturing control terminal simultaneously reads the molten pool width, path temperature, and local contour. If the molten pool width expands, the path temperature drops more slowly, and the local contour approaches the upper boundary allowance... This indicates that the original laser power should continue. Scanning speed and powder delivery volume Execution will cause local height increases; if the molten pool width shrinks, the path temperature decreases, and the local contour approaches the lower boundary margin. This indicates that subsequent finishing will lack continuously available material. Therefore, the remanufacturing control terminal calculates the material offset at each discrete location. .

[0078] Among them, the deviation of the filler material This represents the combined deviation between the actual and predicted conditions of the material forming process, expressed as a non-negative real number, used to determine whether correction is necessary; the measured value of the molten pool width. With predicted pool width These represent the measured and predicted molten pool widths at the current location, respectively, both being non-negative real numbers; the measured temperature value... Compared with temperature prediction These represent the measured path temperature and the predicted path temperature at the current location, respectively. Both values ​​are non-negative real numbers used to characterize local heat accumulation; the measured profile value... Compared with the predicted profile value These represent the measured profile height and the predicted profile height at the current position, respectively. Both values ​​are non-negative real numbers and are used to characterize whether the intermediate forming layer is close to the boundary of the allowable margin window. Weight , , The participation weights for the corresponding state variables are all non-negative real numbers; the translation amount , , The normalized translation is represented by positive real numbers to maintain the stability of the denominator; the tooth surface coordinate position... Maintain the same mapping as in step one, and take the value as the remanufacturing feasible region. The effective position within is used to ensure that the correction corresponds to the boundary. When the material deviation During the elevation adjustment, the remanufacturing control terminal adjusts the laser power based on whether the current profile is above or below the center line of the allowable margin window. Scanning speed and powder delivery volume : Among them, the corrected laser power Corrected scanning speed And the corrected powder delivery amount These represent the local execution parameters at the current position, with their value ranges constrained by the capacity boundaries of the material forming equipment and the powder feeding device, respectively; correction coefficients. , , This is the correction factor for the parameter corresponding to the current position; all values ​​are non-negative real numbers. (Sign function) The format is: input greater than 0 takes +1, equal to 0 takes 0, and less than 0 takes -1, used to specify the correction direction; where laser power... Scanning speed and powder delivery volume The initial parameters for the current position given in step two; lower boundary margin. With upper boundary margin This represents the allowable margin window boundary output in step one; The weights of each component in the material deviation are retrieved from the process database and normalized according to the partition label; each translation is taken as the lower limit of the minimum effective range of the corresponding sensor quantity in the calibration database; each correction coefficient is retrieved from the local sensitivity table according to the material heat treatment category, partition label, and local curvature. The material correction threshold, material degradation threshold, and continuous material out-of-bounds count are all retrieved from the process database according to the partition type, material heat treatment category, and equipment capability; when the material deviation exceeds the material correction threshold, local correction is performed; when multiple consecutive discrete locations exceed the material degradation threshold, the current path is stopped and the safety parameter package is invoked.

[0079] For example, the measured profile value of the first half of the third section of the tooth flank. Near the upper boundary margin And the actual temperature value The descent is slow, and the remanufacturing control terminal reduces the laser power at that location. and powder delivery volume Increase scanning speed And insert cooling strategies The preset dwell interval is used; after executing the next step, the local contour returned by the contour acquisition device falls back into the allowable margin window. Step 3 combines the melt pool width, path temperature, and local contour into the same correction basis, and fixes the correction direction to the center line of the allowable margin window.

[0080] After the filler material is formed and passes intermediate inspection, the remanufacturing control terminal calls the finishing path queue. When the finishing equipment executes the current path, the spindle power acquisition device acquires the spindle power, the grinding wheel load acquisition device acquires the grinding wheel load, and the contour acquisition device acquires the local residual contour.

[0081] If the spindle power increases, the grinding wheel load accumulates, and the residual profile does not decrease, it indicates that the original finishing feed... and fine cutting depth The current path is no longer suitable; continuing would introduce localized high loads into subsequent paths. Therefore, the remanufacturing control terminal calculates the finishing deviation at the current location. .

[0082] Among them, the finishing deviation This represents the combined deviation between the actual finishing condition and the predicted condition, and is a non-negative real number used to determine whether the finishing path needs adjustment; the measured value of the spindle power. Compared with the predicted spindle power These represent the measured spindle power and predicted spindle power at the current position, respectively. Both values ​​are non-negative real numbers used to characterize the input load; measured value of grinding wheel load. Compared with the predicted value of grinding wheel load These represent the measured and predicted grinding wheel loads at the current position, respectively, both being non-negative real numbers used to characterize the clogging trend on the grinding wheel surface; the measured value of the residual profile... Compared with the predicted residual profile These represent the measured residual contour and the predicted residual contour at the current location, respectively. Both values ​​are non-negative real numbers and are used to characterize whether the reserved layer is continuously recycled. Weight , , The participation weights for the corresponding state variables are all non-negative real numbers; the translation amount , , The normalized translation is represented by all positive real numbers; the tooth surface coordinate position... Maintaining the same mapping as in step one, this is used to locate the finishing load and residual profile at the same discrete location; When the precision deviation During the ascent, the remanufacturing control terminal first performs a local correction: reducing the finishing feed. Reduce finishing depth and in cooling strategy Triggering the trimming interval; if the trimming deviation at multiple consecutive positions is... If the abnormal trigger boundary is exceeded, or if the high-temperature position in the material forming stage overlaps with the high-load position in the finishing stage, then the current path is stopped, and the process reverts to the safety parameter package compiled in step two. The safety parameter package must contain at least the reduced finishing feed. Reduced finishing depth and extended cooling strategy For example, during tooth root section finishing, the measured value of spindle power... Measured value of grinding wheel load Simultaneously raised along the transition fillet, while the measured value of the residual profile... If the descent is not synchronized, the remanufacturing control terminal will first trigger a trimming process, and then reduce the finishing feed. and fine cutting depth If the final deviation after the two paths If the boundary is still exceeded, then the safety parameter package will be used instead.

[0083] The weights of each component in the finishing deviation are retrieved from the process database and normalized according to the partition label, and each translation is taken as the lower limit of the minimum effective range of the corresponding sensor quantity; the finishing correction coefficient is retrieved from the local sensitivity table. When the finishing deviation exceeds the finishing correction threshold, the remanufacturing control terminal reduces the finishing feed, decreases the finishing depth, and triggers the dressing interval; when multiple consecutive discrete positions exceed the finishing degradation threshold, or when the deviation does not fall back below the finishing correction threshold after dressing, the safety parameter package is invoked. Optionally, when acoustic emission monitoring is used, the measured value and its predicted value of acoustic emission energy are used as parallel or alternative inputs for the grinding wheel load state.

[0084] Step three divides the finishing stage into two levels of actions: local correction and conservative downgrading. Minor deviations will not spread, and serious deviations will no longer be forcibly pushed forward.

[0085] Preferably, the patch forming equipment uses a coaxial powder-feeding laser cladding head, and the finishing equipment uses a forming grinding spindle or a finishing spindle with equivalent function; for damaged forming gear workpieces with internal gear rings, only the path direction and nozzle incident direction are rewritten to the inner contour version, and the patch deviation is... Finishing deviation The logic for calling the safety parameter package remains unchanged, thus ensuring that the same terminology system can cover external teeth, internal teeth, and large gear rings in the field.

[0086] When using it, utilize the material deviation amount The melt pool width, temperature, and profile are unified under the same correction criteria to ensure that the intermediate forming layer does not deviate from the allowable allowance window. This is achieved by utilizing the finishing deviation. By merging spindle power, grinding wheel load, and residual profile into a single treatment basis, and through a two-level path control of local correction and conservative degradation, the finishing load is prevented from accumulating locally.

[0087] All of the following actions are performed by the remanufacturing control terminal. The remanufacturing control terminal is connected to the tooth profile measuring device, surface roughness measuring device, bonding quality inspection device, flaw detection device, process database, parameter model and case library, and receives the record from step three.

[0088] Step 4: Match the final quality measurement results with the partition execution records formed in Step 3 point by point, output the qualification mark and execution record, and write the final quality deviation back to the parameter model and case library to form rules.

[0089] If the final quality assessment only retains a "pass" or "fail" conclusion, the boundary information, path information, and local correction information accumulated in steps one through three cannot be converted into experience that can be inherited by the next damaged formed tooth workpiece. The final quality of the formed tooth is not a single dimensional issue, but a composite state determined by the final tooth profile, surface roughness, bonding quality, and flaw detection results; and these states are causally related to the path position, correction direction, and conservative degradation actions in step three. Therefore, step four does not simply archive the measured values, but reprojects the measured values ​​to the same tooth surface coordinates as in step one, recouples the final quality deviation with the process record of step three, and then writes the results back to the partitioned evaluation chain of step two.

[0090] First, the tooth profile measuring device, surface roughness measuring device, bonding quality inspection device, and flaw detection device output the final tooth profile, surface integrity, repair layer bonding status, and internal defect status, respectively. Second, the remanufacturing control terminal projects all these measurement results onto the same tooth surface coordinate system and correlates them with the path markers, zone markers, and correction records from step three. Finally, based on the spatial distribution of the final quality deviation, the remanufacturing control terminal reverse-engineers the zone evaluation values ​​from step two. The corresponding weights and boundaries are determined, and the correction rules are written back to the parameter model and the case library.

[0091] The remanufacturing control terminal first reads the workpiece number, partition identifier, path identifier, position index, and corrected laser power from the process record. Corrected scanning speed Corrected powder delivery amount The system records finishing feed, finishing depth of cut, cooling strategy execution, and safety parameter package calls. Then, it calls the tooth profile measuring device, surface roughness measuring device, and combines the quality inspection device and flaw detection device to complete the final measurement. After the final measurement is completed, the manufacturing control terminal generates the final quality deviation field, performs the qualification judgment, outputs the final quality result package, and writes the final quality deviation field back to the parameter model and case library after coupling it with the process record.

[0092] The process involves several steps: a tooth profile measurement device performs tooth profile and tooth direction scanning on the repaired damaged tooth workpiece; a surface roughness measurement device collects surface roughness data along the tooth root, tooth flank, and tooth tip zones; a quality inspection device inspects the interface between the repair layer and the substrate; and a flaw detection device verifies internal incomplete fusion, crack propagation, and porosity. Subsequently, the remanufacturing control terminal reprojects the final tooth profile measurement results, surface roughness measurement results, combined with the quality inspection results and flaw detection results, back onto the tooth surface coordinates used in step one. The key here is ensuring that each final measurement position returns to the same tooth surface coordinate position. This causes the material to deviate from the required amount in step three. Finishing deviation It directly corresponds to the final quality result.

[0093] For example, the operator transfers the finished damaged tooth workpiece to the final measurement station. The tooth profile measuring device first scans the tooth side profile, then moves to the tooth root transition fillet. The surface roughness measuring device samples again along the path that was just scanned. Combined with the quality inspection device, the bottom of the repair layer is continuously scanned. The flaw detection device then conducts supplementary inspections on the tooth root zone and the fracture reconstruction zone. After the final measurement is completed, the on-site display terminal displays the final shape residual zone of the tooth side zone, the bonding state zone of the tooth root zone, and the flaw detection state zone of the tooth tip zone.

[0094] After completing the in-situ projection, the remanufacturing control terminal generates the final mass deviation field. : Among them, the final mass deviation field Tooth surface coordinate position The final quality comprehensive deviation is a non-negative real number, used to provide a unified basis for acceptance judgment and write-back; the final measured value of the tooth profile. Compared with the final tooth profile target value These represent the final shape measurement result and the target restored contour result at the current position, respectively, both of which are non-negative real numbers; the measured surface roughness value. With the target surface roughness value These represent the measured surface roughness result and the target result at the current location, respectively, and are non-negative real numbers. Combined with measured values ​​of the state Combined with the target value of the state These represent the measured and target results of the bonding quality at the current location, respectively, with non-negative real numbers, used to characterize the bonding state between the repair layer and the substrate; the measured value of the flaw detection state. With the target value of the flaw detection status Representing the measured result and target result of flaw detection at the current location, respectively, with non-negative real numbers, used to characterize the internal defect state; weight , , , The weight of each final test item at its current position is a non-negative real number, used to distinguish the degree of attention given to different final test items by the three partitions; translation amount. , , , This is a normalized translation, with all values ​​being positive real numbers, used to avoid fluctuations caused by excessively small denominators; lower boundary margin. With upper boundary margin The allowable margin window boundary output from step one is used to connect the final tooth profile deviation with the boundary of the previous reserved layer. The final tooth profile target value is taken from the target recovery profile determined in step one and confirmed by the process chains in steps two and three; the surface roughness target value is read from the process database according to the zoning function requirements; the bonding state target value and the flaw detection state target value are read according to the bonding level standard and the non-destructive testing level standard, respectively. Before entering the final quality deviation field calculation, the original bonding quality inspection quantity and the original flaw detection quantity are first normalized and mapped to the bonding state measured value and the flaw detection state measured value, respectively, so as to be compared with the final tooth profile measured value and the surface roughness measured value in the same evaluation chain.

[0095] Furthermore, it compresses the four types of final test results—final tooth profile, surface roughness, mating state, and flaw detection state—into the same final quality deviation field. In this process, the tooth surface coordinate semantics are maintained in accordance with those in steps one and three, thus laying the positional foundation for subsequent attribution and write-back. This is done in the formation of the final quality deviation field. Subsequently, the remanufacturing control terminal does not use the average value of the whole tooth for judgment, but first performs position-level judgment within the partition, then partition-level judgment, and finally workpiece-level judgment.

[0096] The root and flank sections of the formed tooth serve different functions. If only a single average value is used to summarize them, severe local deviations will be masked by the good condition of other areas. The remanufacturing control terminal first determines the final quality deviation field at each tooth surface coordinate position. If the location threshold is exceeded, then determine whether the location is within the remanufacturing feasible region. The edge position, whether it is accompanied by a safety parameter package call record, and whether it is located at the end of the path repeatedly corrected in step three; then, the position-level judgment results are merged into the zone-level status, and a workpiece-level qualified mark is formed based on the zone-level status.

[0097] For example, when the final mass deviation field is located at most positions of the tooth flank partition If the position threshold is kept within the range, but a continuous boundary zone appears in the tooth root partition fracture reconstruction area, the remanufacturing control terminal will not directly output the whole part qualified mark. Instead, it will mark the workpiece status as final quality pending verification, and at the same time mark the specific path and location index of the tooth root partition fracture reconstruction area in the result package.

[0098] Preferably, the final quality result package includes at least the workpiece number, partition identifier, path identifier, location index, and final quality deviation field. The system includes: workpiece-level qualification indicator, zone-level status, location band requiring verification, target parameter package version number, and process record version number. The remanufacturing control terminal simultaneously sends the final quality result package to the field display terminal and the case library storage interface; the field display terminal is used to show the specific location to the operator, and the case library storage interface is used to save the final quality result of this damaged formed tooth workpiece.

[0099] As a parallel extension, if the final mass deviation field If a small number of isolated locations deviate from the boundary, and these locations do not involve the bonding or flaw detection states, the remanufacturing control terminal will output a local rework indicator; if the final quality deviation field... If the continuous path area is out of bounds and accompanied by the joint state or flaw detection state being out of bounds, a switching alternative process identifier will be output.

[0100] When the workpiece-level acceptance indicator is partial rework, the remanufacturing control terminal extracts the rework location set and resends this set along with the latest safety parameter package to step three. Only the rework location set is subject to material replenishment or finishing rework. When the workpiece-level acceptance indicator is switching to an alternative process, the current process route is terminated, and the workpiece is marked as an alternative process part. Furthermore, acceptance criteria are bound to specific locations, paths, and final test items, eliminating the situation where only the overall result is seen without understanding the underlying reasons.

[0101] Furthermore, if the final quality result package is merely an archive, the partition evaluation value in step two... The correction direction in step three and the trigger boundary of the security parameter package will not undergo any targeted changes.

[0102] Therefore, the remanufacturing control terminal will ultimately control the quality deviation field. Deviation from the filler material in step three Finishing deviation The security parameter package call record and partition execution order are coupled together, forming a write-back coupling quantity. Then, based on the write-back coupling amount The parametric model and case library were modified. Specifically: Among them, write-back coupling amount Tooth surface coordinate position The combined write-back amount of the final quality deviation and the execution process deviation, taken as a non-negative real number, is used to determine the direction and magnitude of rule correction; the final quality deviation field Deviation of supplementary materials Finishing deviation The corresponding results inherited from steps four and three respectively, all taking non-negative real numbers; Abnormal record count Tooth surface coordinate position Whether a security parameter package call, continuous correction, or path interruption occurred at this point, the value is a non-negative real number, used to introduce abnormal execution actions into the write-back; weight , , , The write-back coupling weights are all non-negative real numbers, used to distinguish between corrections caused by final test deviations and corrections caused by execution anomalies. The abnormal recording quantity is statistically obtained from the process records in step three according to the tooth surface coordinate position. When the corresponding path triggers the safety parameter package, continuous correction occurs, path stops, or trimming actions occur, the abnormal recording quantity is recorded and normalized according to the preset upper limit. To avoid confusion with the volume weight in step two, the write-back coupling weight in step four corresponds to the final measurement deviation weight, material deviation weight, finishing deviation weight, and abnormal recording weight, respectively. During write-back, if the dominant source of the write-back coupling quantity is the material deviation, the initial settings of the laser power, scanning speed, and powder feed amount corresponding to that position are adjusted; if the dominant source is the finishing deviation, the finishing feed, finishing depth, and trimming trigger boundary are adjusted; if the dominant source is the final quality deviation and is concentrated in the main contact area of ​​the tooth side, the final shape weight is increased and the upper boundary allowance is tightened.

[0103] In practice, the remanufacturing control terminal is based on the write-back coupling amount. Three types of content were corrected respectively. First, if the write-back coupling was... If the tooth flank section continues to increase, then increase the position weight corresponding to the final shape weight in step two, and tighten the upper boundary margin. The permissible range for this location. Secondly, regarding the write-back coupling amount. The main factor affecting the tooth root zone is the deviation of the filler material. The drive then sets the initial laser power of the tooth root section. Scanning speed and powder delivery volume Adjust to the new starting gear. Thirdly, if the write-back coupling amount... The tooth tip section is mainly affected by finishing deviation. and abnormal record volume The driver then writes the finishing feed record and finishing depth record of the tooth tip partition into a new safety parameter package to trigger the boundary.

[0104] For example, on a damaged shaped tooth workpiece that has been repaired, the final mass deviation field The indicator shows repeated boundary violations near the transition fillet of the tooth root section, and the safety parameter packet is triggered multiple times in this area during step three; the remanufacturing control terminal writes back the corresponding coupling amount in this area. Mark it as a high value, increase the attention weight of the tooth root partition to the engagement state and flaw detection state in the parametric model, and fix the tooth root partition execution order of this type of workpiece to the priority position.

[0105] This ensures that the final quality result is not only the conclusion of this workpiece, but also a clear input before the next workpiece enters step two.

[0106] Preferably, the parameter model adopts a hierarchical storage method categorized by workpiece type, material heat treatment type, and defect type, while the case library adopts a method categorized by workpiece number and defect depth field. Remanufacturing Feasibility Domain Final quality result package and write-back coupling The associated storage method is as follows: the tooth profile measuring device adopts a contact gear measuring center or a non-contact profile scanning device, the surface roughness measuring device adopts a stylus measuring device, the combined quality inspection device adopts an ultrasonic testing device, and the flaw detection device adopts a penetrant testing device or a magnetic particle testing device. Devices with equivalent functions are also acceptable.

[0107] In use, a final quality result package is generated through two levels of judgment: zone-level and workpiece-level, ensuring that the pass / fail indicator, verification location, and alternative process trigger location have clear coordinates. This is achieved through write-back coupling. The final quality results are sent back to the parameter model and case library so that the next remanufacturing process will have correction rules for this type of defect and this type of material before entering step two.

[0108] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0109] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0110] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0111] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0112] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for intelligent optimization of green remanufacturing process parameters for formed teeth, comprising material forming, finishing, and inspection of the formed tooth workpiece to be remanufactured, characterized in that: include, Step 1: Obtain the 3D point cloud of the formed tooth workpiece, design tooth profile reference data, material / heat treatment information and damage area detection data, construct the remanufacturing feasible domain and generate the allowable allowance window; Step 2: Based on the remanufacturing feasible region and allowable margin window, establish a coupled process parameter model covering the forming and finishing of the filler material and solve the initial parameter set; Step 3: Perform material forming and finishing in the partitioned areas according to the initial parameter set, and correct the local process parameters based on the comparison results between the measured state and the predicted state; Step 4: Based on the final quality judgment results and process records, perform feedback writing back to correct the partitioning rules in Step 1 or the target weights, constraint boundaries, and initial parameter values ​​in Step 2.

2. The intelligent optimization method for green remanufacturing process parameters of formed teeth according to claim 1, characterized in that: Before constructing the remanufacturing feasible domain in step one, the three-dimensional point cloud is registered with the design tooth profile reference data, and the defect depth, local curvature, remaining tooth root thickness, local reachability angle and adjacent tooth reference relationship are extracted to form the defect geometric descriptor and partition information.

3. The intelligent optimization method for green remanufacturing process parameters of formed teeth according to claim 2, characterized in that: When generating the allowable margin window in step one, the target recovery profile of the intermediate forming layer is determined based on the defect geometry descriptor, and the upper and lower boundaries of the allowable margin window are determined based on the forming shrinkage compensation amount, finishing removal allowance amount and disturbance allowance amount.

4. The intelligent optimization method for green remanufacturing process parameters of formed teeth according to claim 3, characterized in that: In step one, when there is measurement occlusion or missing point cloud, the missing area is filled in according to the reference relationship between adjacent teeth; when the remaining matrix thickness, crack level or accessibility is lower than the corresponding threshold, the corresponding position is removed from the remanufacturing feasible domain and the decision to switch to alternative process is output.

5. The intelligent optimization method for green remanufacturing process parameters of formed teeth according to claim 4, characterized in that: In step two, the process parameter coupling model simultaneously associates the material forming parameters and the finishing parameters. The material forming parameters include laser power, scanning speed, powder feed rate, overlap rate, and layer increment, while the finishing parameters include finishing feed, finishing depth, and cooling strategy.

6. The intelligent optimization method for green remanufacturing process parameters of formed teeth according to claim 5, characterized in that: In step two, candidate process combinations are screened based on the remanufacturing feasible region and allowable margin window, and the candidate process combinations are evaluated by partition according to the geometric deviation of the intermediate forming layer, the final tooth profile deviation, the unit repair energy consumption, the filler overflow volume, and the secondary removal volume to generate an initial parameter set.

7. The intelligent optimization method for green remanufacturing process parameters of formed teeth according to claim 6, characterized in that: In step three, the material forming and finishing are performed sequentially according to the tooth root zone, tooth side zone, and tooth tip zone. During the execution, the width of the molten pool, temperature, visual profile, spindle power, and the state proxy quantity of the grinding wheel or tool are collected to form the actual measured state corresponding to the zone.

8. The intelligent optimization method for green remanufacturing process parameters of formed teeth according to claim 7, characterized in that: In step three, the measured state is compared with the predicted state corresponding to the initial parameter set. When the comparison result meets the correction conditions, the local laser power, scanning speed, powder feeding amount, finishing feed or trimming trigger conditions are corrected according to the corresponding partition. When the comparison result meets the degradation conditions, it is rolled back to the safety parameter package.

9. The intelligent optimization method for green remanufacturing process parameters of formed teeth according to claim 8, characterized in that: In step four, a final quality judgment is generated based on the tooth profile measurement results, surface roughness, bonding quality, and flaw detection results. The target parameter package, process execution record, and qualification mark corresponding to the final quality judgment are output. The process execution record includes partition execution record, local correction record, and degradation record.

10. The intelligent optimization method for green remanufacturing process parameters of formed teeth according to claim 9, characterized in that: In step four, the final quality judgment is associated with the process execution record and written back to the parameter model and case library. Based on the write-back results, the partitioning rules in step one, the target weights and constraint boundaries in step two, and the initial parameter values ​​are corrected for the generation of the initial parameter set of the subsequent formed tooth workpiece.

Images