A method and system for reconstructing three-dimensional heat distribution and evaluating structure of tooth profile grinding

Abstract

The application discloses a tooth profile grinding three-dimensional heat distribution reconstruction and structure evaluation method and system, establishes a three-dimensional tooth surface discrete grid in combination with a grinding wheel motion track, tooth surface geometric appearance and grinding wheel macrostructure, generates a contact unit set with a time label through track mapping and path continuity, combines grinding energy input and grinding wheel abrasive particle distribution difference, constructs a weighted energy graph, forms a directed energy transmission path under a grid adjacency relation, and obtains an energy flow mode through iterative diffusion, introduces a time sequence convolution and offset extraction mechanism, generates a heat migration disturbance condition reflecting local heat diffusion abnormalities, is coupled with the three-dimensional grid, drives a non-steady-state heat conduction solver to obtain a tooth surface heat distribution field, extracts a tooth surface potential sensitive area set by using a structure evaluation operator, and realizes comprehensive evaluation of heat-force response and structure risk.

Description

Technical Field

[0001] This invention relates to the field of industrial safety technology, specifically to a method and system for reconstructing three-dimensional thermal distribution and evaluating the structure of tooth profile grinding. Background Technology

[0002] In the forming grinding process of gear teeth, there is a complex local thermo-mechanical coupling behavior between the grinding wheel and the workpiece tooth surface. Especially under the combined action of continuous feed and oscillating cutting, the distribution of grinding heat exhibits significant temporal evolution, spatial heterogeneity, and path dependence. Existing temperature monitoring methods usually rely on devices such as single-point thermocouples and infrared thermometers, which can only obtain the instantaneous temperature value at a fixed location at a certain moment, making it difficult to reflect the dynamic coupling relationship between "tool trajectory - heat diffusion path - workpiece structure and morphology" in actual grinding.

[0003] The unique characteristics of profile grinding lie in the fact that the workpiece tooth profile has three-dimensional curved surface features, and the contact area of ​​the grinding wheel continuously changes during the machining process. This causes the heat source to exhibit mobility and directionality on the workpiece surface, and local high-temperature areas migrate over time, forming a thermal peak drift phenomenon with trajectory characteristics. At the same time, the macroscopic structural parameters of the grinding wheel (such as the distribution of through holes and abrasive grain density) significantly alter the diffusion path and local accumulation mode of heat flow inside the workpiece, thus making the thermal field evolution more complex.

[0004] Existing measurement and modeling methods mostly rely on single-point detection or two-dimensional simplified models, lacking the ability to fuse and reconstruct three-dimensional coupled features of temperature, time and space. This makes it impossible to achieve feedback coupling between machining trajectory, tooth surface geometry and grinding wheel structural characteristics, thus limiting the accurate reproduction of grinding heat distribution and the accuracy of structural risk assessment.

[0005] In view of this, the present invention provides a method and system for reconstructing three-dimensional thermal distribution and evaluating the structure of tooth profile grinding, thereby solving the above problems. Summary of the Invention

[0006] The purpose of this invention is to provide a method and system for reconstructing three-dimensional thermal distribution and evaluating the structure of tooth profile grinding, in order to solve the problem of insufficient monitoring of thermal effects and structural evaluation in the existing tooth profile grinding process.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] Firstly, a method for reconstructing three-dimensional thermal distribution and evaluating the structure of tooth profiles during grinding includes:

[0009] The grinding wheel motion trajectory and tooth surface geometry during the tooth profile forming grinding process are obtained, and a three-dimensional discrete mesh of the tooth surface is generated in combination with the macroscopic structure of the grinding wheel; a set of contact units with time labels is formed on the three-dimensional discrete mesh of the tooth surface according to the contact migration path between the grinding wheel and the tooth surface.

[0010] Within the set of contact units, an energy flow pattern is constructed by utilizing the spatial difference between the grinding energy input and the distribution of abrasive grains on the grinding wheel; and during the construction process, thermal migration perturbation conditions are formed through temporal convolution superposition to reflect the local shift in thermal diffusion.

[0011] The thermal migration perturbation condition is coupled with the three-dimensional tooth surface discrete mesh to drive the unsteady heat conduction solver to generate the tooth surface thermal distribution field; and stress boundary operators are introduced in the generation process to obtain the thermo-mechanical response characteristics.

[0012] Based on the thermo-mechanical response characteristics and the energy flow pattern, a structural evaluation operator is constructed; the structural evaluation operator is used to extract the set of potential sensitive areas on the tooth surface and output the three-dimensional thermal distribution reconstruction and structural evaluation results of tooth profile grinding.

[0013] As a preferred embodiment of the first aspect of the present invention, the generation logic of the contact unit set includes:

[0014] The grinding wheel trajectory projection is located on the discrete grid of the tooth surface, and the grid patches crossed by the trajectory are identified to obtain the initial set of contact patches;

[0015] Adjacent contact surfaces are recombined according to trajectory continuity to generate a continuous contact path, and a path segment index is added.

[0016] Based on the feed direction of the trajectory and the macroscopic structure of the grinding wheel, the entry and exit relationships of each path segment are determined, and a set of directional contact paths is output.

[0017] Synchronize the directional contact path set with the timestamp to generate a time-stamped set of contact elements for use in energy modeling.

[0018] As a preferred embodiment of the first aspect of the present invention, the synchronization refinement logic of the time-stamped contact unit set includes:

[0019] Collect the actual feed time sequence of the grinding wheel trajectory and establish a one-to-one correspondence with the path segment index to form a preliminary synchronization table;

[0020] In the preliminary synchronization table, the relative duration of each path segment is calculated and coupled with the grinding wheel speed parameter to obtain a refined timing table;

[0021] The refined timing table is overlaid with the directional contact path set, and the start and end times of each unit are marked to output the set of synchronous contact units.

[0022] As a preferred embodiment of the first aspect of the present invention, the construction logic of the energy flow mode includes:

[0023] Based on the set of synchronous contact units, grinding energy is allocated to the corresponding surface patches to generate an initial energy distribution;

[0024] By combining the density differences of abrasive grains in different hole rows of the grinding wheel, the initial energy distribution is corrected to form a weighted energy map; grid adjacency relationships are established on the weighted energy map and expanded according to the abrasive grain orientation to generate a directed energy transfer path;

[0025] Iterative diffusion along the energy transfer path yields an energy flow pattern, which is used to create heating disturbance conditions.

[0026] As a preferred embodiment of the first aspect of the present invention, the formation logic of the thermal migration disturbance condition includes:

[0027] Using the energy flow pattern as input, a time convolution window is set according to the trajectory feed, convolution processing is performed on each energy node, and the convolution response map is output.

[0028] The convolutional response maps are stacked step by step along the trajectory direction to obtain the time-series cumulative energy offset table;

[0029] Extract local anomalous offset regions from the time-series cumulative energy offset table and label the offset direction and magnitude;

[0030] The abnormal offset label is merged with the time label of the contact unit to output the thermal migration disturbance condition.

[0031] As a preferred embodiment of the first aspect of the present invention, the convolution and offset extraction refinement logic for the thermal migration perturbation condition includes:

[0032] A bidirectional convolution kernel is configured for each contact unit, and convolution is performed along the grinding wheel trajectory direction and the principal curvature direction of the tooth surface, respectively, to obtain a dual response map;

[0033] By mapping the energy peak regions in the dual response diagram to the trajectory time sequence, a peak response table with consistent direction is generated.

[0034] In the peak response table, filter the energy distributions that deviate from the principal curvature direction and record their offset intensity and range;

[0035] The output is a set of offset features with direction and intensity labels, which serves as the perturbation condition for solving the thermal field.

[0036] As a preferred embodiment of the first aspect of the present invention, the construction logic of the structure evaluation operator includes:

[0037] Input the thermal-mechanical response characteristics and energy flow patterns of the tooth surface, establish a joint data alignment table, and generate a joint input set;

[0038] By extracting regions where both thermal gradient peaks and stress concentrations exist simultaneously from the joint input set, a set of candidate sensitive segments is formed.

[0039] A multidimensional correlation matrix is ​​established for the candidate sensitive fragment set, and the coupling strength is recorded;

[0040] The correlation matrix is ​​converted into operator weights, and a preliminary structure evaluation operator is output.

[0041] As a preferred embodiment of the first aspect of the present invention, the filtering and refinement logic of the candidate sensitive fragment set includes:

[0042] Locate anomalies in stress distribution within the peak region of the thermal gradient and generate a set of anomaly points;

[0043] The set of anomalies is compared with the path nodes of the energy flow pattern to filter out overlapping path segments;

[0044] Extract energy input mutation points from overlapping segments to generate a set of mutation segments;

[0045] The set of mutated fragments is output as the final limiting input to the set of sensitive fragments.

[0046] As a preferred embodiment of the first aspect of the present invention, the weighted and sensitive region output logic of the structure evaluation operator includes:

[0047] The set of mutant fragments is matched with a multidimensional correlation matrix to generate a weighted fragment matrix;

[0048] In the weighted segment matrix, segments with weights exceeding the average level are selected, and their spatial locations are recorded;

[0049] The spatial locations are labeled onto a discrete mesh on the tooth surface to generate a candidate set of sensitive regions;

[0050] Output a candidate set of sensitive areas as the final input for the three-dimensional thermal distribution reconstruction and structural evaluation results of tooth profile grinding.

[0051] In a second aspect, the present invention provides a three-dimensional thermal distribution reconstruction and structural evaluation system for tooth profile grinding, based on the implementation of the first aspect, including a trajectory and morphology acquisition module, an energy flow modeling module, a coupled solution module and a structural evaluation module, with each module connected by wired and / or wireless connections.

[0052] The trajectory and morphology acquisition module is used to generate a three-dimensional discrete mesh of the tooth surface based on the grinding wheel motion trajectory and tooth surface geometry during the tooth profile forming grinding process, combined with the macroscopic structure of the grinding wheel; and to form a set of contact units with time tags on the three-dimensional discrete mesh of the tooth surface according to the contact migration path between the grinding wheel and the tooth surface.

[0053] The energy flow modeling module is used to construct an energy flow pattern within the contact unit set by utilizing the spatial difference between the grinding energy input and the abrasive grain distribution of the grinding wheel, and to form thermal migration perturbation conditions through temporal convolution superposition during the construction process, so as to reflect the local offset of thermal diffusion.

[0054] The coupled solution module is used to couple the thermal migration perturbation condition with the three-dimensional tooth surface discrete mesh, drive the unsteady heat conduction solver to generate the tooth surface thermal distribution field, and introduce stress boundary operators in the generation process to obtain thermo-mechanical response characteristics.

[0055] The structural evaluation module is used to construct a structural evaluation operator based on the thermo-mechanical response characteristics and the energy flow pattern, to extract the set of potential sensitive areas on the tooth surface, and to output the three-dimensional thermal distribution reconstruction and structural evaluation results of the tooth profile grinding.

[0056] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0057] This invention introduces macroscopic structural parameters of the grinding wheel (through-hole distribution, abrasive grain arrangement direction) during tooth profile forming grinding and organically couples them with the tooth surface geometric mesh and the grinding wheel motion trajectory, thus generating a set of contact units with time labels and phase intervals. Therefore, energy input is no longer a traditional "average distribution," but rather a directed energy transfer path formed based on the differences in abrasive grain distribution and path directionality, which is then used to construct an energy flow pattern through iterative diffusion. Since the energy flow pattern is further convolved and superimposed, and local offset features are extracted, the temporal migration and spatial offset laws of local thermal diffusion during grinding can be explicitly reflected. Based on this, this invention not only ensures the accuracy of thermal field reconstruction, but also, by combining it with stress boundary operators, can simultaneously obtain thermo-mechanical coupling response characteristics, thereby achieving quantitative identification of potential sensitive areas on the tooth surface. Compared with existing schemes that rely on single-point temperature sampling or average energy assumptions, this invention avoids the defect of mismatch between energy diffusion direction and actual physical laws, significantly improving the consistency and reliability of three-dimensional thermal distribution reconstruction and structural evaluation. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0059] Figure 1 This is a schematic diagram of the three-dimensional thermal distribution reconstruction and structural evaluation process for tooth profile grinding according to the present invention;

[0060] Figure 2A system block diagram for the method of reconstructing and evaluating the three-dimensional thermal distribution of tooth profile grinding. Detailed Implementation

[0061] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art. The drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted.

[0062] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more exemplary embodiments. Numerous specific details are provided in the following description to give a full understanding of exemplary embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced with one or more specific details omitted, or other methods, components, steps, etc., can be employed. In other instances, well-known structures, methods, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0063] Example 1

[0064] like Figure 1 As shown, this embodiment provides a method for reconstructing the three-dimensional thermal distribution and evaluating the structure of tooth profile grinding, including the following steps:

[0065] The grinding wheel motion trajectory and tooth surface geometry during the tooth profile forming grinding process are obtained, and a three-dimensional discrete mesh of the tooth surface is generated in combination with the macroscopic structure of the grinding wheel; a set of contact units with time labels is formed on the three-dimensional discrete mesh of the tooth surface according to the contact migration path between the grinding wheel and the tooth surface.

[0066] It should be noted that this embodiment does not simply perform conventional geometric superposition of the grinding wheel trajectory and the tooth surface, but rather organically combines the grinding wheel motion trajectory, the three-dimensional geometric features of the tooth surface, and the macroscopic structural parameters of the grinding wheel to construct a three-dimensional discrete mesh for the tooth surface. This ensures the completeness and temporal accuracy of the contact area depiction, providing reliable input for subsequent energy flow and thermal field reconstruction.

[0067] Understandably, the establishment of a three-dimensional discrete mesh for the tooth surface is first based on the equidistant meshing of the tooth surface CAD model, which divides the tooth surface into a set of mesh patches of a controllable number. In this process, macroscopic structural parameters of the grinding wheel are introduced, including the distribution of the hole rows and the orientation of the abrasive grains. This allows the mesh to not only reflect the geometry of the tooth surface but also simultaneously bear the potential influence of the grinding wheel structure on energy transfer.

[0068] It should be understood that the formation of the contact element set is not a direct taking of the intersection points of the grinding wheel trajectory and the tooth surface. Instead, it is achieved through trajectory projection, patch mapping, and migration path tracking, transforming the dynamic interaction between the trajectory and the tooth surface mesh into spatial elements with time labels. Contact elements defined in this way not only possess geometric and temporal information but also serve as computational nodes for subsequent energy flow modeling.

[0069] For example, the logic for generating the contact unit set includes the following sub-steps:

[0070] Discretized mesh on tooth surface The upper positioning grinding wheel trajectory is projected, and the grid patches traversed by the trajectory are identified to obtain the initial set of contact patches. ;

[0071] in: This represents the mesh area traversed by the trajectory. This indicates the moment of first interaction, ensuring that the initial contact relationship between the trajectory and the tooth surface can be fully captured;

[0072] It should be noted that the initial set of contact facets Simply recording discrete facets cannot reflect the continuity of the grinding wheel feed trajectory; therefore, adjacent contact facets are recombined to form a continuous contact path that corresponds one-to-one with the trajectory process.

[0073] Will Adjacent contact surfaces are recombined according to trajectory continuity to generate a continuous contact path, and a path segment index is added. ;

[0074] in: Indicates the first Path index of a series of contact paths, For path Inner A sequence of facets, For path Inner The timing sequence corresponding to each patch ensures that there is a one-to-one correspondence between the discrete patches and the feeding process, thus avoiding single-frame instability.

[0075] It should be understood that, in practical implementation, the "continuity" of adjacent contact patches can be determined by sharing edges or vertices. If they are also temporally adjacent, they are assigned to the same path segment; if a temporal break occurs or the patches are not adjacent, a new path segment index is automatically generated. This ensures the continuity of the contact path set. It maintains consistency in both space and time, but only contains a set of continuous paths. This is insufficient to characterize "incident and exit" in a physical sense, so it is necessary to combine the macroscopic structural parameters of the grinding wheel to assign directional attributes to each path segment.

[0076] Based on the feed direction of the trajectory and the macroscopic structure of the grinding wheel, the entry and exit relationships of each path segment are determined, and the directional contact path set is output, as shown in the formula:

[0077] ;

[0078] in: Indicates the first Path index of a series of contact paths, For path Inner A sequence of facets, For path Inner The timing sequence corresponding to each patch This represents the direction attribute of the path segment, with a value of incident ( ) or exit ( );

[0079] Understandably, this directional attribute is determined by the trajectory tangent. tooth surface normal and the orientation of the grinding wheel hole rows Determining the vector relationship: when If the condition is met, it is considered an incident event; otherwise, it is considered an exit event.

[0080] It should be understood that through this vector-based determination method, the contact path not only possesses geometric continuity but is also endowed with a physical directionality consistent with the macroscopic structure of the grinding wheel, thereby providing the correct input direction for subsequent energy flow patterns. Furthermore, the directionally directed contact path... It remains a path-level collection and has not yet formed a modular description that can directly drive energy modeling. Therefore, it is necessary to divide the path segments into time segments and generate a set of contact units with time labels.

[0081] Synchronize the directional contact path set with the timestamp to generate a time-stamped set of contact elements for use in energy modeling.

[0082] For example, a set of contact units with time tags can be defined as:

[0083] ;

[0084] in: Indicates the first Path index of a series of contact paths, For a subset of faces within that time segment, Indicates the direction attribute of incident or exit; Indicates the start and end times of the contact unit; This indicates the corresponding grinding wheel rotation phase range.

[0085] It is understandable that the contact path segment will be based on the timestamp of the trajectory feed. It is divided into several sub-intervals, and each sub-interval corresponds to a set of face patches. and directional attributes This ultimately forms a contact unit.

[0086] It should be understood that, The introduction of this allows the unit to not only have temporal attributes, but also to map the structural period of the grinding wheel, thereby avoiding the misalignment of temporality and structure in subsequent energy flow and thermal migration disturbance modeling.

[0087] To further explain, the synchronization refinement logic of the time-stamped contact unit set includes:

[0088] Collect the actual feed time sequence of the grinding wheel trajectory and establish a one-to-one correspondence with the path segment index to form a preliminary synchronization table;

[0089] It should be noted that before unitization, the correspondence between global processing time and local path time must first be established. The initial synchronization table is defined as follows:

[0090] ;

[0091] in: The global time of the machine tool is sampled; To map to path segment Local time.

[0092] Understandably, this table establishes a connection between the machine tool's global time axis and the path segment's local time axis by mapping the cumulative arc length of the trajectory to the tool position sampling time. If there is no... If the path segment time does not correspond to the actual processing time, the accurate timing of the contact unit cannot be marked subsequently.

[0093] In the preliminary synchronization table, the relative duration of each path segment is calculated and coupled with the grinding wheel speed parameter to obtain a refined timing table;

[0094] It should be noted that global and local time mapping alone are insufficient to reflect the influence of the macroscopic structural period of the grinding wheel. Therefore, it is necessary to combine the angular velocity information of the grinding wheel spindle to generate a refined time series table containing phase intervals. The refined time series table is defined as follows:

[0095] ;

[0096] in: Path segment The start and end times; To use the spindle angular velocity signal provided by the machine tool control system The angular velocity data is obtained through calculation. It originates from feedback from the spindle encoder and is a real-time parameter that the CNC system can directly acquire.

[0097] It should be understood that the phase function Time integral of angular velocity The calculation yielded the result; in actual calculations, the discrete grinding wheel angular velocity sequence was used. With sampling interval Perform numerical integration. Thus, the phase interval is intercepted at the start and end times of the path segment:

[0098] ;

[0099] Understandably, this phase interval not only reflects the temporal duration of the path segment but also corresponds to the macroscopic structure of the grinding wheel (through-hole distribution, abrasive grain arrangement direction), ensuring that the contact unit can simultaneously bind to the "time dimension" and the "structural periodic dimension." Without introducing the phase interval, the same path segments within different rotation cycles would be incorrectly considered equivalent, leading to the inaccurate representation of the dispersion or aggregation patterns of energy flow in local space. By defining the phase interval, temporal-structural consistency in subsequent energy flow modeling can be guaranteed.

[0100] The refined timing table is overlaid with the directional contact path set, and the start and end times of each unit are marked to output a time-stamped set of synchronously contacting units.

[0101] It should be noted that the refined timing table Sync1 alone is still at the path segment level and cannot be directly used as input for energy modeling and thermal migration perturbations. Therefore, it is necessary to overlay the refined timing table with the directional contact path set to generate a set of contact elements with geometric, temporal, directional, and phase attributes. That is, the final set of synchronized contact elements is defined as follows: :

[0102] ;

[0103] in: Indicates the path segment index; This represents a subset of facets within the corresponding time interval; Indicates the incident or exit direction attribute; Indicates the start and end times of the contact unit; The corresponding grinding wheel rotation phase interval.

[0104] Understandably, the process of generating this set involves... Provided time range and Binding to directional path The path segment is then broken down into several sub-units, each of which contains geometric fragments, timestamps, directionality, and structural phase.

[0105] It should be understood that the selection method for the patch subset is as follows: within the path segment, a local patch sequence is extracted based on the index corresponding to the refined time interval, thereby ensuring consistent coupling between geometry and time, and generating a set of contact elements. This set is the sole input data object for constructing energy flow patterns. Without it, energy input cannot be accurately allocated to local space and specific temporal sequences, leading to distortions in the construction of thermal migration perturbation conditions. Through the definition of this set, each contact element is not only traceable in the temporal and spatial dimensions but also distinguishable in rotational phase and directional attributes, providing rigorous data support for subsequent energy flow modeling.

[0106] Within the set of contact units, an energy flow pattern is constructed by utilizing the spatial difference between the grinding energy input and the distribution of abrasive grains on the grinding wheel; and during the construction process, thermal migration perturbation conditions are formed through temporal convolution superposition to reflect the local shift in thermal diffusion.

[0107] It should be noted that the contact element set only provides a spatial-temporal distribution framework for energy input. However, if energy is directly distributed evenly among the elements, it cannot reflect the influence of the grinding wheel's macroscopic structure and abrasive grain arrangement on the energy transfer path. Therefore, this step introduces the difference between the grinding energy input and the spatial distribution of grinding wheel abrasive grains to construct an energy flow pattern, and further uses temporal convolution superposition to form thermal migration perturbation conditions.

[0108] Understandably, the energy flow model is not merely an energy distribution map, but an energy transfer network containing directional expansion relationships and local weighting mechanisms, capable of accurately describing the propagation direction and local accumulation trend of energy on the tooth surface. Without this model, subsequent thermal migration perturbations can only be calculated based on the independent temperature rise of each element, failing to reflect the spatial offset and temporal migration characteristics of heat diffusion in actual grinding, leading to a significant deviation between simulation results and the real process.

[0109] For example, the construction logic of the energy flow pattern includes:

[0110] Based on the set of synchronous contact units, grinding energy is allocated to the corresponding surface patches to generate an initial energy distribution;

[0111] It should be noted that the initial energy distribution is based on the synchronous contact unit set. Based on this, the total grinding energy input is... The units are allocated to the corresponding facets according to their contact area and time interval.

[0112] For example, it is defined as: ;

[0113] in: Represents a piece of dough The contact area; Represents path segment All the pieces inside area Accumulate; This represents the energy input during the corresponding time segment; Indicates allocation to dough pieces The initial energy value.

[0114] Understandably, this distribution method ensures that the energy input is distributed proportionally to the contact area in space, rather than simply evenly, thus conforming to the physical meaning of energy density. However, the initial energy distribution only reflects the geometric contact characteristics and does not take into account the uneven distribution of grinding wheel abrasive grains, so it still needs to be corrected.

[0115] By combining the density differences of abrasive grains in different hole rows of the grinding wheel, the initial energy distribution is corrected to form a weighted energy map;

[0116] It should be noted that the grinding wheel is not an ideal homogeneous medium; the difference in abrasive grain density between the rows of holes directly affects the actual energy distribution on the tooth surface. This is determined by the abrasive grain distribution function. Weighted corrections are applied to the initial energy distribution:

[0117] ;

[0118] in: Represents a piece of dough Spatial coordinates; Indicates the abrasive grains of the grinding wheel on the surface. The density function of the spatial coordinates can be obtained from the microstructure of the grinding wheel or statistical modeling; This represents the weighted energy value.

[0119] Understandably, this weighting not only reflects the proportional relationship of the contact area but also the spatial differences in the arrangement of abrasive grains in the grinding wheel, thus forming a more realistic energy map. This provides input for establishing a directed energy transfer path; omitting this step would cause the energy flow model to ignore the influence of the grinding wheel's microstructure, reducing the model's accuracy.

[0120] Establish grid adjacency relationships on the weighted energy map and expand it according to the abrasive grain orientation to generate a directed energy transfer path;

[0121] It should be noted that the energy diagram has been weighted. It remains a static distribution, and the direction of energy propagation and adjacency relationships on the tooth surface are not yet apparent. Therefore, it is necessary to establish a directed energy transfer path by combining the topology of the tooth surface mesh and the abrasive grain arrangement direction, as shown in the formula:

[0122] ;

[0123] in: The starting facet representing energy transfer; Indicates the starting facet Adjacent target patches; Indicates the starting facet The set of adjacent faces; The weights are determined by the abrasive grain orientation vector. The angle between the vector of the line connecting the center of the facet and the vector of the facet.

[0124] Understandably, if the direction of the line connecting the facets is consistent with the orientation of the abrasive grains, the transfer weight is greater; if it deviates from the orientation direction, the weight is reduced, ensuring that the energy propagation has physical guidance. Through the construction of this directional path, energy is no longer limited to local accumulation, but can be extended along a physically reasonable direction, avoiding the simplistic assumption of "equidistant diffusion".

[0125] Iterative diffusion along the energy transfer path yields an energy flow pattern, which is used to create heating disturbance conditions.

[0126] It should be noted that a directed energy path not only provides the direction of propagation, but also requires an iterative diffusion process to transfer energy layer by layer in order to form a complete energy flow pattern, as shown in the formula:

[0127] ;

[0128] in: Indicates the first Patch during round iteration Energy value; Represents the energy weight transferred from adjacent faces; cumulative term This indicates that energy is transferred from the neighbor to the target facet along the path.

[0129] It is understandable that the iterative process will propagate energy layer by layer in a local area, so that the energy distribution originally concentrated in the contact unit gradually expands into a global energy map, forming a dynamically evolving energy flow pattern. This pattern not only includes the energy value of each unit, but also implies the propagation direction, speed and accumulation relationship. Therefore, it can truly reflect the local heat accumulation and diffusion law that occurs on the tooth surface during the grinding process.

[0130] Furthermore, the energy flow model, as an output, will be directly used for the subsequent calculation of thermal migration perturbation conditions. Without this model, the heat conduction solution will lack a correct energy input structure, leading to deviations in the heat distribution simulation and failing to meet the accuracy requirements of subsequent 3D reconstruction and structural evaluation.

[0131] Furthermore, the energy flow model only describes the iterative transfer of energy on the spatial grid, which is insufficient to reflect the dynamic migration characteristics of local thermal diffusion. Therefore, this embodiment, based on the energy flow model, further forms thermal migration perturbation conditions through temporal convolution and migration extraction, so that the perturbation conditions can simultaneously constrain energy evolution in both the temporal and structural dimensions.

[0132] It is understandable that the introduction of convolution operation can identify the temporal correlation of the trajectory feed in the energy flow mode, while offset extraction is used to capture the region where the local energy distribution deviates from the main transfer direction, thereby generating perturbation conditions that are more consistent with the actual grinding heat diffusion behavior. If convolution and offset extraction are not performed and only static energy distribution is relied upon, the abnormal energy accumulation or dispersion phenomenon will not be effectively characterized, which will lead to the distortion of subsequent unsteady heat conduction solutions.

[0133] To further explain, the formation logic of the aforementioned thermal migration disturbance conditions includes:

[0134] Using the energy flow pattern as input, a time convolution window is set according to the trajectory feed, and convolution processing is performed on each energy node to output a convolution response map; wherein, the convolution operation can be represented as:

[0135] ;

[0136] in: Represents a piece of dough The convolution response value at time t; For energy flow patterns in time Energy value; The temporal convolution kernel function; For The central time window In the time window Any point in time within the range is the independent variable of the convolution operation.

[0137] This can be understood as: filtering energy on the time axis: for a certain moment It not only considers the energy at that moment, but also the energy at neighboring moments. The energy is superimposed according to the convolution kernel weights, resulting in the final result. It reflects the cumulative trend of energy within a local time range, providing input data for the subsequent "thermal migration perturbation conditions" and ensuring that the energy response reflects not only the instantaneous energy value, but also its cumulative effect during the trajectory feeding process.

[0138] The convolutional response maps are stacked step by step along the trajectory direction to obtain the time-series cumulative energy offset table;

[0139] It should be noted that the generated convolutional response map still represents the independent outputs of each unit, lacking an overall trend of trajectory evolution. Therefore, the convolutional responses are progressively superimposed along the trajectory feed direction to obtain a time-series cumulative energy offset table.

[0140] It is understandable that this overlay operation essentially uses the trajectory direction as the main sorting dimension, merging the energy contributions of different units under the convolution response step by step to form a continuous temporal offset sequence. The temporal cumulative energy offset table can intuitively present the migration trajectory of the energy peak in the trajectory direction, providing a data basis for the subsequent identification of abnormal offset regions.

[0141] Extract local anomalous offset regions from the time-series cumulative energy offset table and label the offset direction and magnitude;

[0142] It should be noted that energy does not migrate strictly along the trajectory during grinding; it often shifts due to local changes in abrasive grain density or abrupt changes in tooth surface curvature. Therefore, these localized abnormal regions need to be extracted in the time-series cumulative energy offset table.

[0143] Understandably, the method for identifying anomalous offset regions includes: setting a main direction offset threshold; when the local response deviates from the main direction of the trajectory by more than the threshold, it is marked as an anomalous offset region; simultaneously, the offset amplitude and direction are recorded. Extracting the offset region is not only about numerical labeling, but also about providing a "local anomaly driving source" for thermal disturbance conditions. If this step is omitted, the thermal field simulation results will lack the actual local thermal peak drift phenomenon. The core objective of local anomaly offset extraction is to identify regions in the energy distribution that abruptly change or deviate from the main transmission direction, and to label their direction and amplitude as offset vectors.

[0144] ;

[0145] in: Indicates the location of the detected abnormal energy peak. This represents the projection of the peak point onto the main transmission direction of the trajectory; Represents a piece of dough Local energy shift;

[0146] It should be understood that, The introduction of this feature ensures that each outlier not only has an absolute position but also an offset vector relative to the main propagation direction, thus accurately characterizing its offset direction and magnitude. Furthermore, to avoid single-point errors, this embodiment further calculates the neighborhood average energy offset:

[0147] ;

[0148] in, Represents a piece of dough The neighborhood set, Indicates a patch within a neighborhood The local energy shift is achieved by introducing The offset determination result is extended from a single point to a regional level, effectively filtering out false peaks caused by random abrasive grain disturbances.

[0149] For example, in actual calculations, when the local energy change... When the preset threshold is exceeded, the system automatically marks the area as an abnormal offset zone and records its offset direction and magnitude. The threshold can be set through statistical analysis of historical grinding test data or obtained through empirical optimization.

[0150] The abnormal offset label is merged with the time label of the contact unit to output the thermal migration disturbance condition;

[0151] Furthermore, after extracting local anomaly offsets, these offset features need to be fused with the time labels of the contact unit set to form complete thermal migration perturbation conditions. The output logic is defined as follows: ;

[0152] in, This represents the corresponding set of contact elements. Represents the offset vector (including direction and magnitude labels). Indicates a time label.

[0153] It is understandable that the final thermal migration perturbation conditions will be... This not only includes spatial location and local energy characteristics but also incorporates temporal information, ensuring that the perturbation conditions remain consistent with actual grinding heat diffusion behavior at both temporal and spatial levels. Omitting this step and relying solely on energy flow patterns for heat conduction solutions would fail to capture anomalous energy drift in local areas of the tooth surface, leading to deviations in the reconstructed heat distribution. By introducing heat migration perturbation conditions, the heterogeneity of local energy diffusion can be effectively compensated, making the unsteady-state heat conduction solution closer to the actual grinding process.

[0154] More specifically, the convolution and offset extraction refinement logic for the thermal migration perturbation condition includes:

[0155] A bidirectional convolution kernel is configured for each contact unit, and convolution is performed along the grinding wheel trajectory direction and the principal curvature direction of the tooth surface, respectively, to obtain a dual response map;

[0156] It should be noted that each contact unit is configured with both a trajectory-direction convolution kernel and a curvature-direction convolution kernel. The trajectory-direction convolution kernel is used to capture the energy accumulation relationship along the grinding wheel feed direction, while the curvature-direction convolution kernel is used to reflect the lateral diffusion effect caused by the tooth surface curvature. The bidirectional convolution ensures that abnormal energy distribution is neither ignored due to coverage along the trajectory direction nor distorted due to differences in lateral curvature. The convolution response result is as follows:

[0157] ;

[0158] ;

[0159] in, The convolutional response is in the trajectory direction. The response is the convolutional response in the curvature direction. This is the corresponding convolution kernel function.

[0160] By mapping the energy peak regions in the dual response diagram to the trajectory time sequence, a peak response table with consistent direction is generated.

[0161] It should be noted that although the convolution response map can reflect the trend of energy change, it has not yet been mapped to the actual trajectory feed time. The energy peak region is extracted from the dual response map and matched one by one with the trajectory feed time to obtain a peak response table with the same direction. The generation of the peak response table ensures that the convolution result is bound to the machine tool time axis, so that the offset can be tracked to the specific machining time.

[0162] In other words, the peak response table is defined as:

[0163] ;

[0164] in, Indicates the time anchor point of the trajectory. This indicates the peak value of the bidirectional convolution at that moment.

[0165] In the peak response table, filter the energy distributions that deviate from the principal curvature direction and record their offset intensity and range;

[0166] It should be noted that in the peak response table, to identify energy distributions deviating from the principal curvature direction, this step uses a threshold determination for filtering. If the response amplitude in the curvature direction at a certain moment... Response to trajectory direction If the ratio exceeds a preset threshold, the region is identified as an abnormal offset region. The range of the preset threshold can be set according to the results of typical grinding experiments to ensure that the noise peak is not misjudged as an offset feature.

[0167] The output is a set of offset features with direction and intensity labels, which serves as the perturbation condition for solving the thermal field.

[0168] It should be noted that after the offset region is filtered, the result needs to be output as a perturbation feature set containing direction and intensity labels. The direction label is used to distinguish the incident or exit characteristics of the offset vector, and the intensity label is used to quantitatively describe the offset amplitude. Without the dual labeling of direction and intensity, the perturbation conditions cannot provide accurate constraints in the solution of unsteady heat conduction.

[0169] In other words, the offset feature set is defined as: ;

[0170] in, Represents the offset vector. Indicates direction label, This represents the intensity of the migration. The resulting migration feature set serves as the specific input for the thermal migration perturbation conditions, providing data support for subsequent unsteady-state heat conduction solutions.

[0171] The thermal migration perturbation condition is coupled with the three-dimensional tooth surface discrete mesh to drive the unsteady heat conduction solver to generate the tooth surface thermal distribution field; and stress boundary operators are introduced in the generation process to obtain the thermo-mechanical response characteristics.

[0172] It should be noted that the thermal migration perturbation condition, as a correction input for energy distribution, only reflects the local shift trend in the spatial and temporal dimensions, but does not translate into the overall temperature field distribution of the tooth surface. This step couples the perturbation condition with the three-dimensional discrete mesh of the tooth surface to drive the unsteady heat conduction solver to obtain the complete thermal distribution field of the tooth surface.

[0173] Understandably, the three-dimensional discrete mesh of the tooth surface serves as the physical computation domain, providing geometric constraints and boundary conditions, while the thermal migration perturbation condition is input to the mesh elements as an energy source term for the time-space distribution. The solver iteratively calculates the evolution of the temperature field over time using numerical methods (such as the finite element method or the finite difference method).

[0174] It should be understood that without introducing thermal migration perturbation conditions, the temperature distribution can only reflect the homogeneous diffusion process and cannot capture the thermal peak drift effect in actual grinding, leading to deviations in the simulation results.

[0175] Specifically, the unsteady-state heat conduction equation is expressed as:

[0176] ;

[0177] in: Density of the tooth surface material For specific heat capacity, This is the partial derivative of temperature with time; Thermal conductivity, For gradient operators, The temperature gradient represents the temperature in spatial coordinates. The direction and magnitude of the change. The thermal conductivity term represents the temperature diffusion effect caused by the internal thermal conduction of the material; Represents spatial coordinates With time Below is the heat source term generated by grinding contact friction on the tooth surface. Its value can be calculated based on the contact area, grinding power, and energy distribution coefficient, which is a conventional energy input term. This represents the thermal migration perturbation conditions extracted by local offset.

[0178] Furthermore, this embodiment introduces a stress boundary operator while generating the thermal distribution field on the tooth surface, enabling synchronous coupling between the thermal and mechanical fields. The stress boundary operator calculates the thermal strain and thermal stress of each mesh node based on the material's thermal expansion coefficient and boundary constraints, thereby outputting the thermo-mechanical response characteristics.

[0179] It should be understood that the thermo-mechanical response characteristics include: nodal temperature sequence, nodal stress tensor, and deformation caused by local thermal expansion; these results serve as inputs to the structural evaluation operator in the next step, ensuring that the tooth surface safety analysis is based not only on temperature distribution but also on thermo-mechanical effects.

[0180] Based on the thermo-mechanical response characteristics and the energy flow pattern, a structural evaluation operator is constructed; the structural evaluation operator is used to extract the set of potential sensitive areas on the tooth surface and output the three-dimensional thermal distribution reconstruction and structural evaluation results of tooth profile grinding.

[0181] It should be noted that although the complete three-dimensional unsteady temperature field and corresponding thermo-mechanical response characteristics of the tooth surface have been obtained in step S103, these results still belong to the original field distribution and do not directly point to potential failure risk points in the tooth surface structure. To this end, this embodiment further proposes the construction logic of the structural evaluation operator, which is used to couple the thermal field, force field and energy flow mode, extract the set of potential sensitive areas of the tooth surface, and finally output the three-dimensional thermal distribution reconstruction and structural evaluation results of tooth profile grinding.

[0182] It is understandable that the so-called "structural evaluation operator" is not a simple threshold determination, but rather a multi-level modeling process that gradually converges to reliable sensitive region identification through a joint input set, candidate fragment set, correlation matrix, weight operator, and sensitive region candidate set.

[0183] It should be understood that the key innovation of this structural evaluation operator lies in:

[0184] A joint alignment table for time and space is introduced to ensure that the thermal field and force field are analyzed under the same reference frame;

[0185] Integrate abrupt changes in energy flow patterns into the decision logic to ensure a response to real processing disturbances;

[0186] By using a multidimensional correlation matrix and weight mapping, we can achieve quantitative and repeatable screening of sensitive areas, rather than relying on empirical speculation.

[0187] Specifically, the construction logic of the structure evaluation operator includes:

[0188] Input the thermal-mechanical response characteristics and energy flow patterns of the tooth surface, establish a joint data alignment table, and generate a joint input set;

[0189] It should be noted that this joint input set enables the analysis of temperature, stress, and energy evolution within the same coordinate frame by synchronizing node indices and timestamps. Without establishing a joint input set, the thermal field, force field, and energy flow patterns would be separate, making effective coupled analysis impossible.

[0190] Specifically, the joint input set is defined as:

[0191] ;

[0192] in: Indicates node temperature. Indicates nodal stress. This indicates the intensity of the energy flow pattern at the node. It is a set of grid nodes; the joint input set enables the three elements of heat, force and energy to be calculated under a unified framework, avoiding data misalignment or loss.

[0193] By extracting regions where both thermal gradient peaks and stress concentrations exist simultaneously from the joint input set, a set of candidate sensitive segments is formed.

[0194] It should be noted that within the joint input set, if a region simultaneously exhibits both a thermal gradient peak and a stress concentration, then that region is determined to be a candidate sensitive segment, denoted as the set:

[0195]

[0196] in: , These are the thermal gradient and stress threshold, respectively. The candidate sensitive segment set is only a preliminary screening result to avoid mistakenly identifying single-point random disturbances as sensitive areas. Only when the thermal gradient and stress concentration occur simultaneously does it mean that there is a potential risk of local failure on the tooth surface. The candidate sensitive segment set is a higher-level result for subsequent screening, not the final sensitive area.

[0197] A multidimensional correlation matrix is ​​established for the candidate sensitive fragment set, and the coupling strength is recorded;

[0198] It should be noted that, in order to quantify the coupling relationship between thermal field, force field and energy flow mode in candidate sensitive fragments, this step establishes a multidimensional correlation matrix and records the coupling strength. Each dimension of the correlation matrix corresponds to a physical quantity (such as temperature gradient, stress amplitude, energy mutation rate), and the matrix elements represent the coupling strength between different physical quantities. This matrix provides a numerical basis for subsequent weighted calculations.

[0199] This embodiment constructs a multidimensional correlation matrix:

[0200] ;

[0201] in, Representing the correlation function, the matrix elements are not only used to describe the statistical correlation between variables, but also serve as inputs for subsequent operator weight calculations.

[0202] To further explain, the filtering and refinement logic of the candidate sensitive segment set includes:

[0203] Anomalies in stress distribution are located within the peak region of the thermal gradient, and a set of anomalies is generated, which reflects the initial intersection region of temperature and stress.

[0204] The set of outliers is compared with the path nodes of the energy flow pattern, and overlapping path segments are filtered out to ensure that the identification results have a real energy input background and that the candidate results are coupled with the energy flow pattern, rather than isolated numerical anomalies.

[0205] Extract energy input mutation points from overlapping segments to generate a set of mutation segments. Mutation points usually correspond to high-risk areas caused by local abrasive grain distribution differences in the grinding wheel or fluctuations in feed parameters.

[0206] The set of mutant fragments is output as the final limiting input to the sensitive fragment set, providing precise boundaries for the computation of the structure evaluation operator.

[0207] The correlation matrix is ​​converted into operator weights, and a preliminary structure evaluation operator is output.

[0208] It should be noted that a weighted fragment matrix is ​​generated based on the mutant fragment set S and the multidimensional correlation matrix M:

[0209] ;

[0210] in, Indicates at grid nodes The weighted segment values ​​calculated above represent the degree of sensitivity; The larger the value, the more likely the node is to simultaneously satisfy the coupled characteristics of thermal gradient anomaly, stress concentration, and energy mutation, and thus the higher the probability that it belongs to a potentially sensitive region.

[0211] This represents the spatial gradient of nodal temperature, i.e., the rate of change of the temperature field in a local region. The larger the value, the more drastic the temperature change in the region, which often corresponds to thermal stress concentration.

[0212] This represents the mechanical stress value at the node. The equivalent stress value after normation is the stress value. Stress concentration areas are often closely related to potential cracks or fatigue sources.

[0213] This indicates the local distribution value of the energy flow pattern at this node. Abnormal changes or shifts in energy values ​​often correspond to abnormal heat sources caused by uneven distribution of abrasive grains on the grinding wheel or fluctuations in feed parameters during the actual grinding process.

[0214] This indicates the correlation between temperature gradient and stress. If stress concentration also exists in the peak area of ​​the temperature gradient, the higher the positive correlation between the two, the easier it is to form thermomechanical damage.

[0215] This indicates the correlation between stress distribution and energy input. If the energy mutation region overlaps with the stress anomaly region, it means that the energy input directly drives the abnormal response of the mechanical effect.

[0216] This indicates the correlation between temperature gradient and energy input. If a sudden change in energy in a certain region directly leads to an increase in the peak temperature gradient, it indicates a strong causal coupling between the thermal field and the energy source.

[0217] , , These are empirical weights used to adjust the contribution of the three types of correlations in the structural evaluation operator. They are set through experimental calibration or numerical simulation verification. The weight ratios of the three can be adjusted under different gear materials or grinding parameters. In some implementation schemes, the three can satisfy... The normalization conditions are set to ensure that the operator outputs are comparable.

[0218] To further explain, the weighted and sensitive region output logic of the structure evaluation operator includes:

[0219] The set of mutant fragments is matched with a multidimensional correlation matrix to generate a weighted fragment matrix;

[0220] It should be noted that by matching the aforementioned set of mutant fragments with the multidimensional correlation matrix one by one to obtain the weighted value corresponding to each mutant fragment, a weighted fragment matrix is ​​formed. This process essentially assigns the coupling relationship of the three types of features—thermal gradient, stress, and energy—to the mutant fragments in order to quantify the sensitivity of the fragments. If a fragment exhibits high coupling in all three types of correlation, its weighted value is significantly higher than that of ordinary fragments, and such fragments will be given priority in subsequent sensitive area determination.

[0221] In the weighted segment matrix, segments with weights exceeding the average level are selected, and their spatial locations are recorded;

[0222] It should be noted that in the weighted segment matrix, segments with weight values ​​exceeding the average level or a preset threshold are selected and their spatial locations are recorded to form a highly sensitive candidate set. This selection step ensures that the identification of sensitive areas is selective, rather than including all segments. The average level or threshold can be obtained through experimental calibration. Its purpose is to eliminate false high values ​​caused by accidental fluctuations, thereby improving the reliability of sensitive area determination.

[0223] The spatial locations are labeled onto a discrete mesh on the tooth surface to generate a candidate set of sensitive regions;

[0224] It should be noted that the spatial locations of the previously selected fragments are marked on the discrete mesh of the tooth surface to generate a candidate set of sensitive areas. This process transforms the numerical calculation results into spatial distribution results, so that the sensitive areas can be intuitively located on the three-dimensional tooth surface structure. The candidate set not only contains the geometric coordinates of the sensitive areas, but also binds timestamps and weighted values. Therefore, it can be used for both spatial visualization and subsequent security assessment.

[0225] Output a candidate set of sensitive areas as the final input for the three-dimensional thermal distribution reconstruction and structural evaluation results of tooth profile grinding.

[0226] It should be noted that the candidate set of sensitive areas is used as the final output and as the direct input to the three-dimensional thermal distribution reconstruction and structural evaluation results of tooth profile grinding. The output candidate set of sensitive areas can be used to determine the potential damage areas of the tooth surface during the actual grinding process, and guide the optimization of grinding parameters and gear material selection. The output result is not only a set, but can also include spatial thermodynamic indicators (such as temperature rise rate, stress amplitude, and energy mutation rate), providing quantifiable decision-making basis for engineering applications.

[0227] Example 2

[0228] like Figure 2 As shown, the parts not detailed in this embodiment are as described in Embodiment 1. This embodiment provides a three-dimensional thermal distribution reconstruction and structural evaluation system for tooth profile grinding, including a trajectory and morphology acquisition module, an energy flow modeling module, a coupled solution module and a structural evaluation module. The modules are connected by wired and / or wireless means.

[0229] The trajectory and morphology acquisition module is used to generate a three-dimensional discrete mesh of the tooth surface based on the grinding wheel motion trajectory and tooth surface geometry during the tooth profile forming grinding process, combined with the macroscopic structure of the grinding wheel; and to form a set of contact units with time tags on the three-dimensional discrete mesh of the tooth surface according to the contact migration path between the grinding wheel and the tooth surface.

[0230] The energy flow modeling module is used to construct an energy flow pattern within the contact unit set by utilizing the spatial difference between the grinding energy input and the abrasive grain distribution of the grinding wheel, and to form thermal migration perturbation conditions through temporal convolution superposition during the construction process, so as to reflect the local offset of thermal diffusion.

[0231] The coupled solution module is used to couple the thermal migration perturbation condition with the three-dimensional tooth surface discrete mesh, drive the unsteady heat conduction solver to generate the tooth surface thermal distribution field, and introduce stress boundary operators in the generation process to obtain thermo-mechanical response characteristics.

[0232] The structural evaluation module is used to construct a structural evaluation operator based on the thermo-mechanical response characteristics and the energy flow pattern, to extract the set of potential sensitive areas on the tooth surface, and to output the three-dimensional thermal distribution reconstruction and structural evaluation results of the tooth profile grinding.

[0233] The three-dimensional thermal distribution reconstruction and structural evaluation system for tooth profile grinding provided in this embodiment is used to execute the three-dimensional thermal distribution reconstruction and structural evaluation method for tooth profile grinding disclosed in the above embodiments of the present invention. The system consists of multiple functional modules, each corresponding to a functional unit in the method steps, and achieves complete operation of the method flow through data interaction. Specifically, the trajectory acquisition module, mesh generation module, energy modeling module, and thermo-mechanical solution module in the system are used to acquire the motion trajectory, construct the three-dimensional tooth surface mesh, generate the energy flow mode, and reconstruct the thermal distribution field and structural response during tooth profile grinding, respectively. Since the system's module design corresponds one-to-one with the method steps, the relevant methods and processes can be found in the detailed description of the foregoing embodiments, and will not be repeated here.

[0234] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for reconstructing three-dimensional thermal distribution and evaluating the structure of a tooth profile during grinding, characterized in that, include: The grinding wheel motion trajectory and tooth surface geometry during the tooth profile forming grinding process are obtained, and a three-dimensional discrete mesh of the tooth surface is generated by combining the macroscopic structure of the grinding wheel. Based on the contact migration path between the grinding wheel and the tooth surface, a set of contact units with time labels is formed on the three-dimensional discrete mesh of the tooth surface. Within the contact unit assembly, an energy flow pattern is constructed by utilizing the spatial difference between the grinding energy input and the abrasive grain distribution of the grinding wheel; During the construction process, thermal migration perturbation conditions are formed through temporal convolution superposition to reflect the local shift in thermal diffusion; wherein: The construction logic of the energy flow pattern includes: allocating grinding energy input to the corresponding facets based on the set of synchronous contact units to generate an initial energy distribution; modifying the initial energy distribution by combining the density differences of abrasive grains in different hole rows of the grinding wheel to form a weighted energy map; establishing grid adjacency relationships on the weighted energy map and expanding it according to the abrasive grain orientation to generate a directed energy transfer path; iteratively diffusing along the energy transfer path to obtain the energy flow pattern, which is used to form the heating disturbance conditions. The formation logic of the thermal migration disturbance condition includes: taking the energy flow mode as input, setting a time convolution window that follows the trajectory feed, performing convolution processing on each energy node, and outputting a convolution response map; progressively superimposing the convolution response maps along the trajectory direction to obtain a time-series accumulated energy offset table; extracting local abnormal offset regions from the time-series accumulated energy offset table and marking the offset direction and magnitude; merging the abnormal offset markings with the time labels of the contact units to output the thermal migration disturbance condition. The thermal migration perturbation condition is coupled with the three-dimensional tooth surface discrete mesh to drive the unsteady heat conduction solver to generate the tooth surface thermal distribution field; and stress boundary operators are introduced in the generation process to obtain the thermo-mechanical response characteristics. Based on the thermo-mechanical response characteristics and the energy flow pattern, a structural evaluation operator is constructed; the structural evaluation operator is used to extract the set of potential sensitive areas on the tooth surface and output the three-dimensional thermal distribution reconstruction and structural evaluation results of tooth profile grinding.

2. The method for reconstructing three-dimensional thermal distribution and evaluating the structure of tooth profile grinding according to claim 1, characterized in that, The logic for generating the contact unit set includes: The grinding wheel trajectory projection is located on the discrete grid of the tooth surface, and the grid patches crossed by the trajectory are identified to obtain the initial set of contact patches; Adjacent contact surfaces are recombined according to trajectory continuity to generate a continuous contact path, and a path segment index is added. Based on the feed direction of the trajectory and the macroscopic structure of the grinding wheel, the entry and exit relationships of each path segment are determined, and a set of directional contact paths is output. Synchronize the directional contact path set with the timestamp to generate a time-stamped set of contact elements for use in energy modeling.

3. The method for reconstructing three-dimensional thermal distribution and evaluating the structure of tooth profile grinding according to claim 2, characterized in that, The synchronization refinement logic of the time-stamped contact unit set includes: Collect the actual feed time sequence of the grinding wheel trajectory and establish a one-to-one correspondence with the path segment index to form a preliminary synchronization table; In the preliminary synchronization table, the relative duration of each path segment is calculated and coupled with the grinding wheel speed parameter to obtain a refined timing table; The refined timing table is overlaid with the directional contact path set, and the start and end times of each unit are marked to output the set of synchronous contact units.

4. The method for reconstructing three-dimensional thermal distribution and evaluating the structure of tooth profile grinding according to claim 3, characterized in that, The convolution and offset extraction refinement logic for the thermal migration perturbation conditions includes: A bidirectional convolution kernel is configured for each contact unit, and convolution is performed along the grinding wheel trajectory direction and the principal curvature direction of the tooth surface, respectively, to obtain a dual response map; By mapping the energy peak regions in the dual response diagram to the trajectory time sequence, a peak response table with consistent direction is generated. In the peak response table, filter the energy distributions that deviate from the principal curvature direction and record their offset intensity and range; The output is a set of offset features with direction and intensity labels, which serves as the perturbation condition for solving the thermal field.

5. The method for reconstructing three-dimensional thermal distribution and evaluating the structure of tooth profile grinding according to claim 1, characterized in that, The construction logic of the structure evaluation operator includes: Input the thermal-mechanical response characteristics and energy flow patterns of the tooth surface, establish a joint data alignment table, and generate a joint input set; By extracting regions where both thermal gradient peaks and stress concentrations exist simultaneously from the joint input set, a set of candidate sensitive segments is formed. A multidimensional correlation matrix is ​​established for the candidate sensitive fragment set, and the coupling strength is recorded; The correlation matrix is ​​converted into operator weights, and a preliminary structure evaluation operator is output.

6. The method for reconstructing three-dimensional thermal distribution and evaluating the structure of tooth profile grinding according to claim 5, characterized in that, The filtering and refinement logic of the candidate sensitive segment set includes: Locate anomalies in stress distribution within the peak region of the thermal gradient and generate a set of anomaly points; The set of anomalies is compared with the path nodes of the energy flow pattern to filter out overlapping path segments; Extract energy input mutation points from overlapping segments to generate a set of mutation segments; The set of mutated fragments is output as the final limiting input to the set of sensitive fragments.

7. The method for reconstructing three-dimensional thermal distribution and evaluating the structure of tooth profile grinding according to claim 6, characterized in that, The weighted and sensitive region output logic of the structural evaluation operator includes: The set of mutant fragments is matched with a multidimensional correlation matrix to generate a weighted fragment matrix; In the weighted segment matrix, segments with weights exceeding the average level are selected, and their spatial locations are recorded; The spatial locations are labeled onto a discrete mesh on the tooth surface to generate a candidate set of sensitive regions; Output a candidate set of sensitive areas as the final input for the three-dimensional thermal distribution reconstruction and structural evaluation results of tooth profile grinding.

8. A three-dimensional heat distribution reconstruction and structural evaluation system for tooth profile grinding, based on the implementation of any one of the three-dimensional heat distribution reconstruction and structural evaluation methods for tooth profile grinding according to claims 1-7, characterized in that, It includes a trajectory and topography acquisition module, an energy flow modeling module, a coupled solution module, and a structure evaluation module, with each module connected via wired and / or wireless means; The trajectory and morphology acquisition module is used to generate a three-dimensional discrete mesh of the tooth surface based on the grinding wheel motion trajectory and tooth surface geometry during the tooth profile forming grinding process, combined with the macroscopic structure of the grinding wheel. And on the three-dimensional discrete mesh of the tooth surface, a set of contact units with time tags is formed according to the contact migration path between the grinding wheel and the tooth surface; The energy flow modeling module is used to construct an energy flow pattern within the contact unit set by utilizing the spatial difference between the grinding energy input and the abrasive grain distribution of the grinding wheel, and to form thermal migration perturbation conditions through temporal convolution superposition during the construction process, so as to reflect the local offset of thermal diffusion. The coupled solution module is used to couple the thermal migration perturbation condition with the three-dimensional tooth surface discrete mesh, and drive the unsteady heat conduction solver to generate the tooth surface thermal distribution field. Stress boundary operators are introduced during the generation process to obtain thermo-mechanical response characteristics; The structural evaluation module is used to construct a structural evaluation operator based on the thermo-mechanical response characteristics and the energy flow pattern, to extract the set of potential sensitive areas on the tooth surface, and to output the three-dimensional thermal distribution reconstruction and structural evaluation results of the tooth profile grinding.

Images