Method and system for identifying a transmission model dimensional variation

Abstract

The present application relates to the technical field of size identification, in particular to a transmission device model size variation degree identification method and system, comprising the following steps: acquiring a node angular response direction under load, extracting a direction consistent path sequence, screening a linear response chain segment, identifying a connection abnormal area, tracking a connection state expansion trend, demarcating a structure variation trajectory path, extracting a size offset continuous feature, and constructing a size variation degree identification set. In the present application, the node chain segment with consistent direction and continuous space is constructed through the angular response path, the connection state mutation area is extracted in combination with the direction flip and position distribution, the state change trend is tracked along the node sequence, the structure variation path is demarcated, the continuous feature of size offset is extracted in the path chain segment, the identification coverage of local disturbance and non-uniform variation situation is expanded, the pertinence of node screening and the integrity of path extension are enhanced, the response omission caused by state jump is reduced, and the coherence and spatial adaptation ability of size change extraction are improved.

Description

Technical Field

[0001] This invention relates to the field of size recognition technology, and in particular to a method and system for recognizing the size variation of a transmission device model. Background Technology

[0002] The field of dimensional recognition technology belongs to the intersection of computer vision and geometric modeling. It mainly involves measuring, recognizing, and monitoring changes in the dimensions of object structures based on images, point clouds, or 3D model data. Its core aspects include dimensional parameter extraction, 3D geometric modeling, dimensional difference calculation, and trend analysis. Dimensional recognition is usually accomplished through model data acquisition, feature point recognition, parameter fitting, and geometric comparison. It is widely used in industrial manufacturing, equipment inspection, and digital modeling. Among them, the traditional method for recognizing the dimensional variation of transmission device models refers to a class of methods that address changes in the structural dimensions of transmission structures caused by factors such as wear and tear, manufacturing deviations, or assembly deformation. By constructing a geometric mapping relationship between a standard model and a target model, key point data of sensitive parts of dimensional changes are extracted. Then, boundary difference comparison based on curve contour fitting or model mesh error analysis based on spatial distance calculation is used to determine the degree of dimensional variation and distribution in the model.

[0003] Existing technologies rely on comparing the geometric configuration of models to identify dimensional differences. The process is highly dependent on preset key points and their boundary matching rules. When the structural connection state changes locally due to load, it cannot accurately identify the coherent path of the state behavior between nodes. Especially when there are sudden changes in directional response or misalignment of connection surfaces during the operation of the transmission device, the original discrimination logic lacks tracking and organization of the state evolution process, resulting in fragmented recognition results in regional division. This leads to the inability to continuously extract dimensional offset features, resulting in insufficient result coverage or misjudgment of local responses, affecting the complete perception and accurate representation of the structural change trend. Summary of the Invention

[0004] To address the technical problems existing in the prior art, embodiments of the present invention provide a method for identifying the dimensional variation of a transmission device model, comprising the following steps: To achieve the above objectives, the present invention adopts the following technical solution: a method for identifying the dimensional variation of a transmission device model, comprising the following steps: S1: Obtain the angular response direction of the nodes in the connecting sub-face region under axial load, determine whether the directional angle between nodes is continuously flipped, extract the path sequence of nodes with consistent direction, and obtain the angular response path of the nodes; S2: Based on the node angular response path, analyze the node directional consistency and spatial connectivity, filter linear response chain segments, construct continuous segments according to the node order, and obtain response perturbation chain segments; S3: Based on the response disturbance chain segment, analyze the response of the path node connection sub-face, identify discontinuous displacement, abrupt connection change and misalignment combination, filter concentrated areas, locate abnormal node segments, and obtain the abnormal connection node block. S4: Based on the connection anomaly node block, analyze whether the state change process of the connection sub-face has a gradual expansion direction along the node sequence, determine whether the displacement direction extension behavior in the path is continuously promoted, and combine the continuous area of ​​state evolution extension behavior to obtain the structural change promotion trajectory path. S5: Based on the structural change propulsion trajectory path, compare the differences in the connection response of the path nodes, identify the dimensional offset change segments, analyze the relationship between the offset range and the node position, extract the continuous dimensional change response features, and obtain the dimensional change degree identification set.

[0005] As a further aspect of the present invention, the node angular response path includes node number, direction change state, and direction consistency continuity path sequence; the response disturbance chain segment includes linear conduction characteristic chain segment, continuous response segment, node distribution order and spatial position; the connection anomaly node block includes node combination with discontinuous displacement direction, node combination with abrupt connection state change, and node combination with morphological misalignment; the structural change propagation trajectory path includes node state evolution, sequential distribution relationship, and continuous region of extension behavior; and the size variation degree identification set includes node segment with size offset amplitude change, size offset distribution range, and positional correlation relationship.

[0006] As a further aspect of the present invention, the axial load refers to a linear external force load acting on the connecting surface along the main direction of the node arrangement, thereby stimulating the response behavior of the structure in the main axis direction; The angular response direction refers to the directional change that occurs in the normal direction of the connecting sub-face after the node is loaded.

[0007] As a further aspect of the present invention, the linear response chain segment refers to a numbered segment consisting of multiple adjacent nodes with the same displacement direction in the main direction and continuous changes in connection state, which manifests as linear propagation behavior in the local structure. The subsurface response refers to the connection state changes and directional responses exhibited by the connecting subsurface under nodal loads, characterizing the dynamic features of the connection relationship between nodes.

[0008] As a further aspect of the present invention, the specific steps of S1 are as follows: S101: Obtain the angular response direction of the nodes in the connecting surface area of ​​the transmission device under axial load, arrange the nodes in sequence according to the node number, analyze the relationship between the angular response directions of adjacent nodes, identify the rotational behavior characteristics between adjacent directions, and obtain the rotational behavior sequence between nodes. S102: Based on the sequence of directional rotation behavior between nodes, filter the node segments in continuous nodes where the directional change has a flipping feature, remove the node relationships where the directional change has not changed, and obtain the set of directional flipping trend path numbers. S103: Based on the set of path numbers for the direction reversal trend, extract the number and direction change status of the nodes in the path according to the number order, identify the node segments with direction maintenance relationship, and remove broken node segments based on path continuity to obtain the node angular response path.

[0009] As a further aspect of the present invention, the specific steps of S2 are as follows: S201: Based on the node angular response path, extract the angular response direction and coordinate information of each node, compare the angular pointing relationship and position change status between adjacent nodes in numerical order, extract node data where the direction remains continuous and the adjacent distance does not jump abruptly, and obtain a continuous feature sequence of direction and position. S202: Based on the continuous feature sequence of the directional position, identify the numbering intervals where the directional change trend is consistent and the node arrangement is uninterrupted, remove data segments that are discontinuous due to sudden changes in direction and spatial movement, extract numbering segments with continuous directional changes, and obtain a linear directional extension node set. S203: Based on the linear direction extended node set, extract the corresponding coordinate information according to the node number order, track and analyze the continuous direction of the node arrangement in the path, exclude the segments with displacement trend swing and sequence misalignment, extract the path segments with continuous advancement in direction and position between adjacent nodes, and obtain the response disturbance chain segment.

[0010] As a further aspect of the present invention, the specific steps of S3 are as follows: S301: Based on the response disturbance chain segment, extract the connection sub-face response direction, displacement direction and connection state parameters of the nodes in the path under load, and according to the node numbering order, match the direction change, connection state change and surface profile offset between adjacent nodes to obtain the direction and state change parameter sequence. S302: Based on the sequence of direction and state change parameters, filter the number ranges of continuous node segments with direction reversal, connection state jump behavior and surface contour offset jump. According to the spatial distance distribution trend between each number segment, remove data segments with excessive spatial extension to obtain the numbered segments with dense abnormal behavior. S303: Based on the densely numbered segments of the abnormal behavior, extract the position parameters and state parameters of the nodes corresponding to the numbers, compare the direction of node position change with the continuity of connection response behavior, filter the segments where the response performance deviates continuously in the node arrangement direction, extract the distribution range where the number and position are coupled in response features, and obtain the connection abnormal node block.

[0011] As a further aspect of the present invention, the specific steps of S4 are as follows: S401: Based on the connection anomaly node block, extract the node connection sub-face state parameters in the order of node number, compare the state change direction between adjacent nodes, and combine the state change continuity direction with the number arrangement direction to obtain the connection state extension trend sequence. S402: Based on the connection state extension trend sequence, extract the node segments with continuous directional advancement, compare whether the displacement direction changes continuously along the numbered direction in the node segments, remove segments with reverse offset and sudden directional change, and obtain the displacement advancement behavior segments in the path. S403: Based on the displacement propagation behavior segments in the path, extract the corresponding node state change records and number position relationships, identify the consistency of state change amplitude, directional extension and node distribution density, extract segments with consistent state change directions and continuous node arrangement, and obtain the structural change propagation trajectory path.

[0012] As a further aspect of the present invention, the specific steps of S5 are as follows: S501: Based on the structural change propulsion trajectory path, extract the connection status parameters and spatial shape size values ​​of the nodes in the path, compare the directionality of the size difference between adjacent nodes according to the node number order, extract the numbered segments where the size change direction has a turning behavior, and obtain the size change trend distribution segment. S502: Call the size change trend distribution segment, side by side the node size value and path position corresponding to each segment number, identify the path segments in which the size value change rate of adjacent nodes shows continuous increase and compression, filter the node number group with continuous behavior, and obtain the size change extension number interval. S503: Based on the extended numbering interval of the size change, extract the size value, number position and spatial distribution relationship of the corresponding node in the path, compare the continuity and arrangement consistency of the size response difference between nodes within the numbering segment, extract the range in which the size change coherence is consistent in space, and obtain the size change degree identification set.

[0013] A transmission device model dimensional variation recognition system includes: The angular extraction module obtains the angular response direction of the nodes in the connection sub-face area of ​​the transmission device under axial load, extracts the relationship between the node number and direction change of the continuous flipping path segment of the included angle of adjacent nodes, and obtains the node angular response path. Based on the node angular response path, the disturbance construction module analyzes the consistency of node response direction and position continuity, filters path segments with consistent directional trends and continuous spatial distribution, and constructs continuous segments according to node order to obtain response disturbance chain segments. Based on the response disturbance chain segment, the connection identification module analyzes the state changes of the connection sub-face of the path node under load, identifies discontinuous displacement, abrupt connection changes and misalignment combinations, filters concentrated areas, locates abnormal node segments, and obtains the connection abnormal node block. Based on the abnormal node blocks, the change and evolution module tracks the extension trend of connection state changes along the node arrangement direction, analyzes the continuous transition mode of node state behavior in the path, and combines the continuous region of state evolution extension behavior to obtain the structural change propagation trajectory path. The variation degree recognition module, based on the structural variation propagation trajectory path, compares the differences in the connection response of the path nodes, identifies the dimensional offset change segments, analyzes the relationship between the offset range and the node position, extracts the continuous dimensional change response features, and obtains the dimensional variation degree recognition set.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, a chain of nodes with consistent direction and spatial continuity is constructed through angular response paths. By combining direction reversal and position distribution, regions of abrupt changes in connection state are extracted. The trend of state change is tracked along the node sequence to delineate the path of structural change. Continuous features of dimensional offset are extracted from the path chain, expanding the identification coverage of local disturbances and non-uniform changes. This enhances the targeting of node selection and the completeness of path extension, reduces response omissions caused by state jumps, and improves the coherence and spatial adaptability of dimensional change extraction. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention; Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0019] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent.

[0020] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0021] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0022] Please see Figure 1 This invention provides a method for identifying the dimensional variation of a transmission device model, comprising the following steps: S1: Obtain the angular response direction of the nodes in the connecting sub-face area of ​​the transmission device under axial load, determine whether the change behavior of the directional angle between adjacent nodes exhibits continuous flipping characteristics, extract the node path sequence with directional consistency and continuity, and obtain the node angular response path by corresponding node number and directional change state. S2: Based on the node angular response path, and according to the consistency of the node angular response direction and the structural spatial connectivity, analyze the trend uniformity of continuous node paths, select response chain segments with linear transmission characteristics, and construct continuous response segments according to the node distribution order and spatial position in the path to obtain response perturbation chain segments. S3: Based on the response disturbance chain segment, analyze the response performance of the connection sub-face of the path node under load, identify the continuous node combination with discontinuous displacement direction, abrupt connection state and misalignment, screen the concentrated area of ​​structure, locate the abnormal distribution segment by the correspondence between node position and state, and obtain the connection abnormal node block. S4: Based on the connection of abnormal node blocks, analyze whether the state changes of the connection sub-faces along the node sequence show a gradual expansion direction, determine whether the displacement direction extension behavior is continuously promoted in the path, and obtain the structural change promotion trajectory path based on the node state evolution and the sequential distribution relationship of the extension behavior continuous area. S5: Based on the trajectory path of structural changes, compare the differences in the connection response of nodes on the path in terms of structural morphology, identify node segments where the magnitude of dimensional offset changes, analyze the distribution range and positional relationship of dimensional offset, extract the continuously changing dimensional response features in the spatial structure, and obtain the dimensional change degree identification set.

[0023] The node angular response path includes node number, direction change state, and direction consistency continuity path sequence; the response disturbance chain segment includes linear transmission characteristic chain segment, continuous response segment, node distribution order and spatial location; the connection anomaly node block includes node combination with discontinuous displacement direction, node combination with abrupt connection state change, and node combination with morphological misalignment; the structural change propagation trajectory path includes node state evolution, sequential distribution relationship, and continuous region of extension behavior; the size variation degree identification set includes node segment with size offset amplitude change, size offset distribution range, and positional relationship.

[0024] Please see Figure 2 The specific steps of S1 are as follows: S101: Obtain the angular response direction of the nodes in the connecting surface area of ​​the transmission device under axial load, arrange the nodes in sequence according to the node number, analyze the relationship between the angular response directions of adjacent nodes, identify the rotational behavior characteristics between adjacent directions, and obtain the rotational behavior sequence between nodes. First, a two-dimensional unfolding process is required for the connecting surface area. Each node is laid out sequentially on both sides of the load reference axis, and the angular response direction of each node during load application is extracted. The direction is expressed in angular numerical form in degrees. The extraction method is based on the node rotation reference point and the angular change amplitude to determine the direction. For example, the angular response direction of node number N1 is 12°, N2 is -14°, and N3 is 18°, representing the rotational offset direction of the node around its connecting surface under axial compression, respectively. After collecting the angular response directions of all numbered nodes, they are arranged in ascending order of node number to form a continuous path direction sequence. Further, the angular response relationship between adjacent nodes is compared. A pairwise comparison method is used to determine the directionality of the angular response directions of every two adjacent numbers. The determination method is to check whether the two angular directions have opposite signs and the amplitude exceeds a set value. The opposite angular direction signs refer to the two angles. The values ​​are in positive and negative ranges, and the rotation reference points are in different directions. The amplitude is determined by whether the difference in rotation amplitude is greater than the angular reversal judgment value, which is set to 30 degrees. In the example above, N1 is 12°, N2 is -14°, and the angle difference is 26°, which does not constitute a direction reversal. N2 to N3 are -14° to 18°, and the angle difference is 32°, which constitutes a reversal relationship. Continue to advance along the numbering sequence and gradually complete the direction determination operation of all adjacent node pairs. When a node pair with a reverse rotation trend is identified, the number, angle value, and direction turning relationship between the nodes are centrally processed. The angle symbol switching position and angle difference range are recorded, and the number of times they appear in the entire path is accumulated. If the same rotation trend appears between three or more consecutive groups of nodes, and the distance does not exceed the set number jump value, it is considered that there is continuous rotation behavior. The nodes of this type are numbered and summarized, and their response angle direction and corresponding offset relationship are output to obtain the sequence of directional rotation behavior between nodes.

[0025] S102: Based on the sequence of directional rotation behavior between nodes, filter the node segments in continuous nodes where the directional change has a flipping feature, remove the node relationships where the directional change has not changed, and obtain the set of directional flipping trend path numbers. First, the numbering order and corresponding angular response direction of all adjacent node pairs in the sequence are extracted. Each node pair corresponds to two numbered nodes and their angular response direction values, which are expressed in degrees. The extracted angle data must come from the node rotation offset response data under axial load. According to the numbering order, the directional relationship between each pair of nodes is judged. The judgment of directional change is based on the switching of the positive and negative signs of the angle and the jump value of the angle amplitude. If the angle value of the previous node is positive and the next node is negative, or the previous node is negative and the next node is positive, it is considered that there is a switch of direction sign. Combined with the angle jump judgment, if the difference between the direction values ​​of the two nodes is greater than the set angular turning identification value, the pair of nodes is defined as a flipped node pair. This identification value can be set to 30 degrees. By setting this value, significant node behavior with sudden rotational changes can be identified. For example, if the node numbers are N1, N2, N3, and N4, and the angular response direction values ​​are respectively 20°, -15°, 17°, -21°. Then N1 to N2 is 20° to -15°, the direction is reversed and the angle change is 35°, which meets the flip condition. N2 to N3 is -15° to 17°, which is also a direction switch with a change of 32°, which also constitutes a flip. In this way, all continuous node segments in the complete sequence are processed, and the numbered segments that meet the flip condition are marked. Next, the nodes that have not changed direction in these flipped segments are removed. That is, the node relationship where the positive or negative sign of the direction value has not changed or the angle change is less than 30 degrees is excluded. For example, if there is N5 to N6 which is 22° to 25°, with only a change of 3°, then this segment does not belong to the flipped segment. After this round of processing, the continuous numbered segments with prominent direction flip characteristics are retained to form a set of node numbered path sequences. All node pairs in this sequence have the characteristics of significant direction change and alternating direction signs. The numbered segments are output in the order of number arrangement to obtain the direction flip trend path number set.

[0026] S103: Based on the path number set of the direction reversal trend, extract the number and direction change status of the nodes in the path according to the number order, identify the node segments with direction maintenance relationship, and remove broken node segments according to the path continuity standard to obtain the node angular response path. First, the angular response direction values ​​of each node in the numbered set are retrieved and arranged in ascending order of number to ensure that the spatial node order of the path matches the numbering order. During the extraction process, the angular direction state of each node needs to be read simultaneously. This state represents the rotational offset direction of the node in the load application direction in degrees. In a practical example, if the angular values ​​corresponding to nodes numbered N1 to N6 are 20°, 22°, 25°, 27°, -30°, and -28°, then the directions of the first four nodes remain within the same positive range, while the angular values ​​of the last two nodes fall into the negative range, indicating a reverse extension state. Subsequently, the angular states in the above numbered sequence are continuously compared to determine whether the change in the direction difference between adjacent nodes remains within the same angular range. Furthermore, node segments whose angular states have not changed are extracted as path segments with a direction-maintaining relationship. If adjacent angular values ​​are within the same range and the change amplitude does not exceed the angular disturbance... The screening value, which can be set to 15 degrees, is considered to indicate that the direction is consistent. For example, if the direction value changes from node N1 to N4 within the range of 2° to 3°, it is determined to be a segment with consistent direction. Then, a path continuity verification operation is performed on all node segments. The verification is based on the difference in node numbers. It is determined whether the difference in adjacent node numbers exceeds the path breakage judgment value. This value is set to 2. That is, if the numbering interval is 1 or 2, it is considered to be a continuous numbering segment. Otherwise, it is considered that there is a jump in the path. For example, if the node numbers are N1, N2, N5, and N6, then because there is a jump in the numbering interval from N2 to N5, the numbering interval is 3 (greater than 2), which does not meet the continuous path judgment condition. This segment will be removed. Only segments such as N1 to N2 and N4 to N5 are retained. Finally, the numbered path segments with continuous numbering order and no sign change in angular response direction are output. All node segments that meet the dual continuity conditions of numbering and direction status are integrated and output to obtain the node angular response path.

[0027] Please see Figure 3 The specific steps of S2 are as follows: S201: Based on the node angular response path, extract the angular response direction and coordinate information of each node, compare the angular pointing relationship and position change status between adjacent nodes in numerical order, extract node data where the direction remains continuous and the adjacent distance does not jump abruptly, and obtain a continuous feature sequence of direction and position. First, the angle response direction and 3D coordinate information of each node in the path are extracted. The angle response direction is obtained by the rotation direction of the node under axial load conditions. The angle value is expressed in degrees, and the rotation direction is represented by positive values ​​for clockwise and negative values ​​for counterclockwise. The coordinate information is represented by triplets to show the position of the node in 3D space. Each node has a unique number, angle value, and spatial coordinates. For example, node N1 has an angle of 20° and coordinates (45, 12, 8), node N2 has an angle of 23° and coordinates (48, 13, 8), and node N3 has an angle of 21° and coordinates (51, 14, 9). The nodes are arranged in numerical order to form a node sequence. Then, the angle pointing relationship and position change status between adjacent nodes are compared. First, it is compared whether the angle value change of adjacent nodes is in a direction-maintaining state. Direction maintenance requires that the angle change value is within the angular deviation reference range, which is set to ±10°. That is, if the angle difference between the current node and the previous node is within this range, the direction is determined. The change is a continuation state. Taking the above example, the angle difference between nodes N1 and N2 is 3°, which is a continuation direction. The angle difference between N2 and N3 is 2°, which is also a continuation direction. If the angle of node N4 is -25°, the difference between it and the previous node's 21° direction is 46°, which is a sudden change in direction. This segment is not considered a direction-maintaining segment. After confirming the direction continuation, the coordinate distance between adjacent nodes is processed. The spatial jump behavior is judged by the difference in three-dimensional spatial coordinates. The distance increment between two nodes in the X, Y, and Z coordinates is calculated to see if it exceeds the path segment spacing jump judgment value. This judgment value is set to 5 units of length. When the distance increment in any direction exceeds this value, it is considered a spatial jump behavior. For example, if nodes N1 and N2 differ by 3, 1, and 0 in the X, Y, and Z directions, respectively, then the position is continuous. If N2 and N4 differ by 9 in the X direction, then a jump has occurred and must be excluded. Finally, the node segments that simultaneously satisfy the continuity of angle direction and the continuity of three-dimensional coordinate position are extracted. The node numbers and corresponding parameters of such nodes are output sequentially to obtain the direction and position continuity feature sequence.

[0028] S202: Based on the continuous feature sequence of directional position, identify the numbering intervals with consistent directional change trends and uninterrupted node arrangement, remove data segments with abrupt directional changes and discontinuous spatial movement, extract numbering intervals with consistent directional changes, and obtain a linear directional extended node set; First, the angle and direction values ​​are compared sequentially by number to determine the continuity of directional changes between adjacent nodes. Continuity of direction means that the directional changes between adjacent nodes maintain the same trend. The standard for a consistent trend is that the angle changes are within the same sign range, and the magnitude of the change remains within the angle extension reference range, which is ±15 degrees. If the angle values ​​of nodes N1 to N4 are 25°, 28°, 31°, and 35° respectively, and the direction continues to increase positively, they are considered to have a consistent trend. If N5 is -32°, then the direction between N4 and N5 changes abruptly and cannot be classified into a continuous trend node group. Based on the above judgment, N1 to N4 are grouped together, and a segment is created between N4 and N5. Then, the coordinate data within the numbered groups with consistent trends are spatially arranged to verify whether there is a positional interruption between adjacent nodes in the three-dimensional coordinate system. The judgment criteria are as follows. To determine whether the change in coordinate values ​​between two nodes in any of the X, Y, and Z axes exceeds a spatial jump reference value, which is set to 6 units, if the coordinates of node N2 are (80, 45, 30) and N3 are (84, 46, 32), the difference is 4, 1, and 2, indicating spatial continuity. If N4 is (96, 48, 31), the difference in the X direction from the previous node is 12, exceeding the jump reference value, indicating a spatial interruption, and the segment containing that node is removed. This process is performed cyclically along the entire path in numerical order, extracting all numbered segments with a continuous trend in angle and direction and no sudden changes in spatial coordinates. Nodes with jumps in direction values, nodes with excessive spatial displacement, and nodes that are broken in the middle are removed. Node data segments with continuous numbering order, consistent direction changes, and stable coordinate extension are retained, and their corresponding numbered segment information is output to obtain a linearly extended node set.

[0029] S203: Based on the linear direction extension node set, extract the corresponding coordinate information according to the node number order, track and analyze the continuous direction of the node arrangement in the path, eliminate the displacement trend swing and sequence misalignment segments, extract the path segments with continuous advancement in direction and position between adjacent nodes, and obtain the response disturbance chain segment. First, the nodes are sorted according to their numbers to ensure that adjacent nodes have a physical connection in space. Then, the spatial coordinates between each group of adjacent numbered nodes are extracted sequentially, and their positional differences along the X, Y, and Z axes are judged to identify whether the node pair constitutes a continuous progression relationship in three-dimensional space. The judgment criterion is that there should be no directional reversal in any of the three axes. For example, for nodes numbered N1 to N4, the X coordinates are 42, 45, 49, and 53 respectively, indicating that the direction is continuous along the X axis. If the X coordinate of the 5th node changes to 47, it means that this node has a reverse swing behavior relative to the previous node on the X axis, and this segment needs to be removed from the continuous path. Then, the Y and Z axes are judged. If the Y-axis of the above nodes is 10, 12, 14, 17, and 19 respectively, and the Z-axis is 3, 3, 4, 4, and 5, it means that there is no sequential misalignment in the other directions except for the X-axis. This path segment can be retained until a swing occurs. Next, each continuous path segment is... The internal process performs spatial path extension judgment. If the distance between nodes in the path maintains the same direction within a continuous interval, that is, all increments are positive or close to zero, it is considered a directional continuous path. If a pair of nodes differs in the opposite direction from the previous pair of nodes in terms of coordinates, for example, the X-axis increment of the previous pair of nodes is +4 and the next pair is -6, it is judged as a swing segment. The path containing this node is removed from the valid path, and the remaining segments are processed. If the numbering order does not jump and the direction remains unidirectional, the path segment is considered a dual continuous advancement path in both direction and position. All path segments that meet the above conditions are merged and output to form a complete disturbance propagation clue. For example, in the path numbered N1 to N7, if there is a reverse coordinate jump behavior between N4 and N5, the path can be divided into two segments: N1 to N4 and N5 to N7. Only N1 to N4 meets the continuous advancement condition, so the data of this path segment is output to obtain the response disturbance chain segment.

[0030] Please see Figure 4 The specific steps of S3 are as follows: S301: Based on the response disturbance chain segment, extract the response direction, displacement direction and connection state parameters of the connection sub-face of the node in the path under the load. According to the node numbering order, the direction change, connection state change and surface profile offset between the corresponding paired adjacent nodes are obtained to obtain the direction and state change parameter sequence. First, extract the response direction, displacement direction, and connection status parameters of each node in the path under load. The response direction is represented by an angle value, indicating the direction of rotational offset of the node surface under external load. The displacement direction is referenced to the direction of the three-dimensional coordinate vector. The connection status parameters are qualitatively encoded to represent the degree of contact tightness or inter-surface slippage of different nodes. For example, the response direction of node N1 is 28°, the displacement direction is towards the positive X-axis, and the connection status is tight, designated as status code 1. The response direction of node N2 is 32°, the displacement direction is biased towards the XZ plane at an angle of 45°, and the connection status is slight slippage, designated as status code 2. After arranging the node numbers in the path in ascending order, each adjacent node pair is extracted and paired. During the pairing process, three sets of parameters need to be matched: node number, response direction, displacement direction, and connection status are grouped into pairs within adjacent numbers. Direction change recognition uses adjacent response directions. The system uses a dual-condition judgment based on the sign of the value and the difference in values. An angle difference limit of 25 degrees is set. If the difference in the response directions of two consecutive nodes exceeds this limit, it is considered a direction change. For example, if N1 is 28° and N2 is -15°, a sign change and a difference of 43° satisfy the direction change condition. Displacement direction comparison is based on the cosine difference of the coordinate vector directions of the two nodes. When the included angle is greater than 60 degrees, it indicates an excessively large direction difference, considered a sudden change in displacement direction. Connection state changes are judged by the difference in status code values. A difference of 1 indicates a slight change in state, while a difference of 2 or higher indicates a significant change in connection state. Combining the above three pairing judgment criteria, consistency verification is performed on all adjacent nodes in the entire disturbance chain segment. The direction change flag, displacement difference flag, and connection state change flag are recorded for each pair of paired nodes, and then uniformly organized into a set of direction and state change information corresponding to the numbered sequence, ultimately obtaining the sequence of direction and state change parameters.

[0031] S302: Based on the sequence of direction and state change parameters, filter the number range of continuous node segments with direction reversal, connection state jump behavior and surface contour offset jump. According to the spatial distance distribution trend between each number segment, remove data segments with excessive spatial extension to obtain the numbered segments with dense abnormal behavior. First, traverse all node pairs in the sequence according to their numbers, extracting the direction difference, connection status indicator change, and surface contour coordinate offset parameters between each pair of nodes. The direction difference is obtained by subtracting the angular response values ​​of the two consecutive nodes. The condition for direction reversal is set as an angle difference greater than 30 degrees. If the angle of node N1 is 22° and N2 is -18°, then the direction difference between them is 40°, which is judged as a reversal behavior. The connection status indicator consists of the contact status level of each node, which is represented by numbers. For example, a status value of 1 indicates close contact, 2 indicates slight slippage, 3 indicates significant slippage, and 4 indicates separation. If the difference between the status codes of the two consecutive nodes is greater than 1, it is considered a jump. For example, if the status changes from 1 to 3, the difference is 2, which meets the jump condition. The surface contour offset is based on the change in the Z-axis direction of the node's three-dimensional coordinates. If the Z-axis increment changes by more than 10 units between two adjacent nodes, it is judged as a jump behavior. In the example, if N1 is Z=15 and N2 is Z=28, then the offset is 13, constituting a jump. After obtaining the above three parameters, based on the node number, all node pairs that meet any one of the abnormal behaviors are labeled. After labeling, adjacent abnormal node pairs are continuously numbered and extracted to form preliminary abnormal segments. Then, spatial density detection is performed on the coordinate information of all nodes in each abnormal number interval. The absolute distance range of all nodes in the X, Y, and Z directions in three-dimensional space is extracted. The sum of the differences between the maximum and minimum coordinate values ​​in the numbered segment is calculated as the spatial extension index of the segment. If the value exceeds 60 coordinate units, the segment is judged to be an overly scattered data segment and needs to be removed. For example, the spatial range span of the numbered N5 to N12 is 78, which exceeds the removal baseline of 60, so the segment is not retained. Conversely, if the spatial range of the numbered N13 to N18 is 42, it is included in the valid numbered segment. The entire path is traversed and screened according to this standard. Finally, the continuous numbered segments that meet the conditions of direction flip, state jump, or contour jump and whose spatial distribution is compact are output, resulting in the densely numbered segments of abnormal behavior.

[0032] S303: Based on the densely numbered segments of abnormal behavior, extract the position parameters and state parameters of the nodes corresponding to the numbers, compare the direction of node position change with the continuity of connection response behavior, filter the segments where the response performance deviates continuously in the node arrangement direction, extract the distribution range where the number and position are coupled in response features, and obtain the connection abnormal node block. First, extract the node coordinate information and connection status parameters corresponding to each number. The node coordinate information includes X, Y, and Z axis coordinate values, used to describe the node's position in space. The connection status parameters use numerical values ​​to indicate different degrees of contact behavior, for example, status code 1 represents contact, 2 represents slight misalignment, 3 represents slippage, and 4 represents disconnection. After extraction, arrange them according to the number order. Compare the direction of position change between each node and its adjacent nodes with the change in connection status pairwise. First, determine whether the direction of position change is continuous. That is, if the increment of the X-axis direction between nodes N1 and N2 is +4, and the increment between nodes N2 and N3 is +3, then the direction is considered to be continuous and consistent. If the direction increment between N3 and N4 is -5, the direction is reversed, which is judged as a swing behavior. This section is interrupted and the number of this section is removed. Then, judge the connection response status behavior. The judgment standard is whether the change of status parameters is continuous. If the status codes are 1, 2, and 3 in sequence, it is considered as behavior extension. If there is a jump from 1 to 3 or a reversal from 3 to 1, it is considered as non-connection. Continuing from the previous step, combining the two judgment criteria mentioned above, each group of numbered segments in the path is screened sequentially, retaining the numbered segments that simultaneously satisfy both spatial direction continuity and state behavior continuity. Then, the retained segments are evaluated for response offset synchronicity, judging whether the nodes within the numbered interval continuously deviate from the state in the arrangement direction. For example, if the state code continuously increases from 1 to 3 from N5 to N8, it indicates that the segment has a trend of gradual connection instability. In addition, combined with the change in position direction, the coordinate increment direction between nodes is statistically analyzed from the X-axis, Y-axis, and Z-axis directions to see if it is consistent with the numbering sorting direction. If all are positive or in the same direction, it is considered that there is a coupling between the arrangement direction and the response state. If multiple retreating nodes appear in the middle, that is, the position change direction is reversed or the state change falls back, then the segment does not meet the response consistency screening condition and is excluded. Finally, all segments that show synchronous response in numbering sorting, spatial advancement, and continuous change in connection state are extracted, and their corresponding node number intervals and spatial coordinate ranges are output to obtain the connection abnormal node block.

[0033] Please see Figure 5 The specific steps of S4 are as follows: S401: Based on the connection anomaly node block, extract the node connection sub-face state parameters in the order of node number, compare the state change direction between adjacent nodes, and combine the state change continuity direction with the number arrangement direction to obtain the connection state extension trend sequence. First, the connection sub-surface state parameters corresponding to each node number are extracted. These parameters numerically mark the degree of difference in node surface contact behavior; for example, a state value of 1 indicates stable contact, 2 indicates slight surface separation, 3 indicates significant misalignment, and 4 indicates loss of contact. After extraction, the parameters are sorted by node number from smallest to largest, ensuring the path direction matches the structural arrangement. After sorting, the state values ​​of any two adjacent nodes are compared, and their numerical difference and direction of change are extracted. If the state value of the later node is higher than that of the earlier node, it indicates that the connection is trending towards separation; conversely, it indicates that the state is trending towards recovery or approximation. Then, the entire numbering sequence is traversed segment by segment in the same manner to generate a state change direction marker between each group of nodes. A positive direction indicates a separation trend, a negative direction indicates a fitting trend, and a zero value indicates no change in state. A continuous change recognition condition is set during the judgment process: when three or more consecutive nodes... If the state changes in the same direction and the state values ​​increase or decrease continuously, it is considered a segment of the state direction extension trend. For example, if the state values ​​of nodes N1 to N5 are 1, 2, 2, 3, and 4 respectively, it can be classified as an extension trend segment. If there is a jump between numbers or the state is reversed, such as N6 being 2, then it will no longer be classified as a trend extension from that point onwards. Next, the trend direction and the numbering direction need to be matched to determine whether the numbering increase direction and the state value change direction are in the same direction. If the numbering increases and the state value also shows an increasing trend, it can be considered that the path extension and state evolution direction are consistent. If the numbering order is inconsistent with the state change direction, such as the numbering increasing while the state value decreases, then the trend segment does not meet the consistency requirement and is excluded. Finally, the numbering segments that connect the continuous state changes and are consistent with the numbering increase direction in the node arrangement direction are selected. The corresponding directional state trend information is output according to the numbering sequence to obtain the connection state extension trend sequence.

[0034] S402: Based on the connection state extension trend sequence, extract the node segments with continuous directional advancement, compare whether the displacement direction changes continuously along the numbered direction in the node segments, remove segments with reverse offset and sudden directional change, and obtain the displacement advancement behavior segments in the path. First, extract the node information corresponding to each consecutive numbered segment in the sequence, including the node number, three-dimensional spatial coordinates, and connection sub-face status values. Arrange the node data within the consecutive numbered segments according to the numbering order, and extract the coordinate change values ​​between every two adjacent nodes to determine the extension behavior in the displacement direction. The displacement direction determination is performed using a three-axis incremental comparison method, that is, calculating the difference between the coordinate values ​​of the subsequent node and the preceding node in the X, Y, and Z directions respectively. If a negative increment appears in any of the three directions, it is judged as a reverse offset. For example, if the coordinates of node N1 are (45, 22, 12), N2 is (47, 23, 13), and N3 is (46, 24, 14), then the increment in the N1 to N2 direction is (2, 1, 1), continuing the forward movement; the increment in the N2 to N3 direction is (-1, 1, 1), showing a reverse behavior in the X direction, and this segment is judged to have an offset swingback. After the judgment is completed, the segments with such swing characteristics are removed. At the same time, the direction increment vector of each segment node is judged for a sudden change in direction. The judgment condition is a sudden change in the direction of the angle between two adjacent displacement vectors. In specific execution, the directionality of the two direction increments is extracted. If the previous segment is a fully positive direction increment and the subsequent segment is a mixed positive and negative direction increment, then this segment is also regarded as a direction change segment and needs to be interrupted. In the example, if there is a node N6 in the segment N4 to N8 with the same direction as N5 but the direction of N7 is offset in the opposite direction, then the segment between N6 and N7 is marked as a sudden change point. After processing, only the segment from N4 to N6 is retained as a continuous advancement path. The subsequent number groups are traversed until the entire trend sequence is traversed. Finally, all node segments that satisfy the condition that the displacement direction has not reversed or changed in direction on the three axes are extracted to form a set of numbered segments with continuous extension of response in the path, and the displacement advancement behavior segment in the path is obtained.

[0035] S403: Based on the displacement propagation behavior segments in the path, extract the corresponding node state change records and number position relationship, identify the consistency of state change amplitude, directional extension and node distribution density, extract segments with consistent state change direction and continuous node arrangement, and obtain the structural change propagation trajectory path. First, the node numbers and corresponding state change records within the path segment are extracted. These records contain the node's response behavior along the normal and tangential directions during contact at the connection surface, displacement direction transformation information, and the update sequence of node state values. During data extraction, nodes are read sequentially in ascending order of their numbers, and the correspondence between the corresponding numbers and state change values ​​is recorded. These state values ​​are then converted into discrete change amplitudes, meaning the difference between the state values ​​of adjacent nodes reflects the actual magnitude of their state updates. This allows for a preliminary judgment of the change amplitude. If the absolute value of this difference is greater than a set change reference value of 1.5, the location is marked as a node with a significant response update. Simultaneously, vector comparisons are performed on the state change directions of adjacent nodes segment by segment according to their numbering order. If the angle between the state change directions of three adjacent nodes is less than 15 degrees, the state change direction in that segment is determined to be stable. To ensure consistency, for example, the state directions of numbers N5, N6, and N7 are (1, 2, 0), (2, 4, 0), and (3, 6, 0) respectively, which can be considered as consistent directions. Further, it is determined whether the nodes within this numbered segment are densely arranged. The criteria for determining node density is that the three-dimensional coordinate distance between adjacent numbered nodes does not exceed a set reference value of 3. If the coordinate spacing between all nodes in numbered segments N5 to N9 is within 1, then this numbered segment meets the requirement of dense distribution. Based on the above conditions, small segments that simultaneously satisfy the requirements of continuous state amplitude, consistent directional extension, and uniform node spacing are selected from the original path segments. These small segments are numbered, recorded, and organized to form a set of continuous numbered segments. Within the set, points of directional change and jumps are removed, while segments with strong directional extension are retained. Finally, a set of segments with ordered numbers, consistent state change directions, and continuous node arrangement is obtained, resulting in the structural change propagation trajectory path.

[0036] Please see Figure 6 The specific steps of S5 are as follows: S501: Based on the structural change propulsion trajectory path, extract the connection status parameters and spatial morphological size values ​​of the nodes in the path, compare the directionality of the size difference between adjacent nodes according to the node number order, extract the numbered segments where the size change direction has a turning behavior, and obtain the size change trend distribution segment. First, the connection status parameters and spatial dimensions of each numbered node in the path are read one by one. The connection status parameters are quantified by the contact area, deformation response, or contact distance changes between adjacent connecting surfaces of the nodes. The dimensions refer to the boundary measurement values ​​or maximum projection range of the nodes in the three spatial axes. The node numbers are arranged in ascending order and a number index is constructed. Then, the dimension difference between adjacent numbered node pairs is read according to the number index, the dimension difference vector is calculated, and the difference direction is obtained by projecting it according to the coordinate axis direction. Subsequently, a sequential comparison operation is performed on the dimension difference directions of multiple consecutive nodes to determine whether the difference direction continues to extend in the same direction. If it is found that the dimension difference direction from a certain node to the direction of the previous node pair has an angle exceeding 45 degrees, then the node number is taken as the starting point of the turning point, and it is combined with the starting number of the previous segment to form a numbered segment with a sudden change in the dimension change direction. For example, between the numbers N8 and N14, if N8 to The direction of N11 is offset towards the positive X direction, but the direction after N12 is offset towards the positive Z direction. Therefore, it can be determined that a directional change occurs at the N12 location. Further, the difference in the magnitude of the directional change before and after the change is compared in continuous segments. Cases where the directional change angle does not exceed the preset reference angle are eliminated. For example, if the included angle is less than 20 degrees, it is considered that there is no obvious change, thus avoiding misjudgment. During the execution process, the length of each segment judgment range is no less than 3 sets of numbered pairs to ensure that the directional changes are comparable. For the acquisition of the size difference between nodes, the maximum projection size of the three axes in the node surface measurement data is extracted. For example, if the dimensions of nodes N10 and N11 are 5.3mm and 6.0mm respectively, the difference is 0.7mm. The directionality can be determined by the projection of the adjacent coordinate difference to find the main axis of change. Further, the number range of the directional change of each segment is stored, and finally, a set of numbered segments with significant turning behavior in the directional changes of each size is formed, resulting in the distribution segment of the size change trend.

[0037] S502: Call the size change trend distribution segment, side by side the node size value and path position corresponding to each segment number, identify the path segments in which the size value change rate of adjacent nodes shows continuous increase and compression, filter the node number group with continuous behavior, and obtain the size change extension number interval. First, the identified dimensional direction turning point number ranges from the previous stage are read segment by segment. Within each segment, nodes are arranged in numerical order, and their dimensional values ​​and corresponding path coordinates in the three spatial axes are extracted one by one. The dimensional difference between each pair of adjacent numbered nodes is calculated sequentially, and combined with their path coordinate differences to form a combined dataset. For each node combination, the trend of the difference magnitude is recorded sequentially. It is determined whether the change between the current dimensional difference and the previous set of differences shows a continuous increase or decrease in value. For example, if the dimensional difference between nodes N50 and N51 is 0.6mm, N51 to N52 is 0.9mm, and N52 to N53 is 1.3mm, it is considered a continuous increase. If it is 0.9mm→0.6mm→0.3mm, it is considered a continuous compression. Furthermore, the number of groups with continuous changes is counted. If the number of nodes with continuous behavior is greater than or equal to 3... If the path segment is determined to have a continuous rate of change in size, in actual execution, it is necessary to eliminate small fluctuations caused by individual measurement errors. Therefore, the minimum change step size threshold is set to 0.3mm. Data with a change step size smaller than this threshold are not included in the continuity judgment. Within a size trend segment, there may be multiple size increase and compression sequences at the same time. Each sequence is numbered and its start and end numbers are recorded. During the screening process, if the size change trend of adjacent nodes does not maintain a consistent change behavior in more than three directions, the segment number will not participate in the formation of the extended number interval. For example, if the numbers N100 to N102 are continuous but the change trend is irregular, they will not be adopted. After traversing all segments, the node number groups with continuous change behavior are merged and their number interval information is output to obtain the size change extended number interval.

[0038] S503: Based on the extended numbering interval of size change, extract the size value, numbering position and spatial distribution relationship of the corresponding node in the path, compare the continuity of size response difference and arrangement consistency between nodes within the numbering segment, extract the range in which the size change coherence is consistent in space, and obtain the size change degree recognition set. First, extract the three-axis spatial coordinates and dimensions of the corresponding nodes within each numbered segment. Establish a correspondence between the dimension value of each node and its number position, forming a two-dimensional data pair set. Then, combine the node's arrangement order in the path to indicate its spatial position order. Select adjacent node pairs in each numbered segment for difference calculation, recording the magnitude and direction of dimension changes. Perform a continuity judgment on the dimension difference results for all node pairs. If the direction of the dimension difference changes between consecutive nodes is consistent, and the magnitude of the dimension change between adjacent nodes shows a monotonically linear correlation with the numbering interval between adjacent nodes, then the numbered segment is considered to have continuous change characteristics. For example, if the dimension difference between nodes N101 and N102 is 0.5mm, between N102 and N103 it is 0.6mm, and between N103 and N104 it is 0.7mm, and the numbering interval between each pair of nodes is 1, then the change in direction and stride is continuous. The spatial distribution relationship is determined in one step. After extracting the three-axis coordinates of each group of nodes, the spatial distance change of each pair of nodes is calculated sequentially and compared with the numbering sequence. If the numbering interval is constant and the coordinate position changes linearly and smoothly without sudden jumps or reversals, the numbering segment is judged to have consistent arrangement. If three or more consecutive groups of nodes in a numbering interval have the characteristics of consistent size change direction, linear size step change, and progressive spatial position along the path direction, the numbering interval is classified as a continuous size change segment. Conversely, if there are behaviors such as reversal of size change direction between node pairs, drastic step fluctuation, or sudden coordinate deviation, the numbering interval is removed and not included in the processing result. After performing the above determination and filtering on each numbering segment in the entire extended numbering set, the set of all numbering segments that meet the continuous change condition is output, resulting in the size variation degree identification set.

[0039] Please see Figure 7 A transmission device model dimensional variation recognition system, including: The angular extraction module obtains the angular response direction of the nodes in the connection sub-face area of ​​the transmission device under axial load, extracts the relationship between the node number and direction change of the continuous flipping path segment of the included angle of adjacent nodes, and obtains the node angular response path. The perturbation construction module analyzes the consistency of node response direction and positional continuity based on the node angular response path, filters path segments with consistent directional trends and continuous spatial distribution, and constructs continuous segments according to the node order to obtain response perturbation chain segments. The connection identification module analyzes the state changes of the connection sub-face of the path node under load based on the response disturbance chain segment, identifies the combination of discontinuous displacement, abrupt connection change and misalignment, filters concentrated areas, locates abnormal node segments, and obtains the connection abnormal node block. The change and evolution module is based on the connection anomaly node block. It tracks the extension trend of connection state changes along the node arrangement direction, analyzes the continuous transition mode of node state behavior in the path, and combines the continuous region of state evolution extension behavior to obtain the structural change propagation trajectory path. The variation degree identification module identifies the dimensional offset change segments by comparing the differences in the connection response of path nodes based on the trajectory path of structural change, analyzes the relationship between the offset range and the node position, extracts the continuous change dimensional response features, and obtains the dimensional variation degree identification set.

[0040] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for identifying dimensional variation of a transmission device model, characterized in that, Includes the following steps: S1: Obtain the angular response direction of the nodes in the connecting sub-face region under axial load, determine whether the directional angle between nodes is continuously flipped, extract the path sequence of nodes with consistent direction, and obtain the angular response path of the nodes; S2: Based on the node angular response path, analyze the node directional consistency and spatial connectivity, filter linear response chain segments, construct continuous segments according to the node order, and obtain response perturbation chain segments; S3: Based on the response disturbance chain segment, analyze the response of the path node connection sub-face, identify discontinuous displacement, abrupt connection change and misalignment combination, filter concentrated areas, locate abnormal node segments, and obtain the abnormal connection node block. S4: Based on the connection anomaly node block, analyze whether the state change process of the connection sub-face has a gradual expansion direction along the node sequence, determine whether the displacement direction extension behavior in the path is continuously promoted, and combine the continuous area of ​​state evolution extension behavior to obtain the structural change promotion trajectory path. S5: Based on the structural change propulsion trajectory path, compare the differences in the connection response of the path nodes, identify the dimensional offset change segments, analyze the relationship between the offset range and the node position, extract the continuous dimensional change response features, and obtain the dimensional change degree identification set.

2. The method for identifying the dimensional variation of a transmission device model according to claim 1, characterized in that, The node angular response path includes node number, direction change state, and direction consistency continuity path sequence. The response disturbance chain segment includes linear conduction characteristic chain segment, continuous response segment, node distribution order and spatial position. The connection anomaly node block includes node combination with discontinuous displacement direction, node combination with abrupt connection state change, and node combination with morphological misalignment. The structural change propagation trajectory path includes node state evolution, sequential distribution relationship, and continuous region of extension behavior. The size change degree identification set includes node segment with size offset amplitude change, size offset distribution range, and positional correlation relationship.

3. The method for identifying the dimensional variation of a transmission device model according to claim 1, characterized in that, The axial load refers to the linear external force load acting on the connecting surface along the main direction of the node arrangement, which excites the response behavior of the structure in the main axis direction; The angular response direction refers to the directional change that occurs in the normal direction of the connecting sub-face after the node is loaded.

4. The method for identifying the dimensional variation of a transmission device model according to claim 1, characterized in that, The linear response chain segment refers to a numbered segment consisting of multiple adjacent nodes with the same displacement direction in the main direction and continuous changes in connection state, which manifests as linear propagation behavior in the local area of ​​the structure; The subsurface response refers to the connection state changes and directional responses exhibited by the connecting subsurface under nodal loads, characterizing the dynamic features of the connection relationship between nodes.

5. The method for identifying the dimensional variation of a transmission device model according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Obtain the angular response direction of the nodes in the connecting surface area of ​​the transmission device under axial load, arrange the nodes in sequence according to the node number, analyze the relationship between the angular response directions of adjacent nodes, identify the rotational behavior characteristics between adjacent directions, and obtain the rotational behavior sequence between nodes. S102: Based on the sequence of directional rotation behavior between nodes, filter the node segments in continuous nodes where the directional change has a flipping feature, remove the node relationships where the directional change has not changed, and obtain the set of directional flipping trend path numbers. S103: Based on the set of path numbers for the direction reversal trend, extract the number and direction change status of the nodes in the path according to the number order, identify the node segments with direction maintenance relationship, and remove broken node segments based on path continuity to obtain the node angular response path.

6. The method for identifying the dimensional variation of a transmission device model according to claim 1, characterized in that, The specific steps of S2 are as follows: S201: Based on the node angular response path, extract the angular response direction and coordinate information of each node, compare the angular pointing relationship and position change status between adjacent nodes in numerical order, extract node data where the direction remains continuous and the adjacent distance does not jump abruptly, and obtain a continuous feature sequence of direction and position. S202: Based on the continuous feature sequence of the directional position, identify the numbering intervals where the directional change trend is consistent and the node arrangement is uninterrupted, remove data segments that are discontinuous due to sudden changes in direction and spatial movement, extract numbering segments with continuous directional changes, and obtain a linear directional extension node set. S203: Based on the linear direction extended node set, extract the corresponding coordinate information according to the node number order, track and analyze the continuous direction of the node arrangement in the path, exclude the segments with displacement trend swing and sequence misalignment, extract the path segments with continuous advancement in direction and position between adjacent nodes, and obtain the response disturbance chain segment.

7. The method for identifying the dimensional variation of a transmission device model according to claim 1, characterized in that, The specific steps for S3 are as follows: S301: Based on the response disturbance chain segment, extract the connection sub-face response direction, displacement direction and connection state parameters of the nodes in the path under load, and according to the node numbering order, match the direction change, connection state change and surface profile offset between adjacent nodes to obtain the direction and state change parameter sequence. S302: Based on the sequence of direction and state change parameters, filter the number ranges of continuous node segments with direction reversal, connection state jump behavior and surface contour offset jump. According to the spatial distance distribution trend between each number segment, remove data segments with excessive spatial extension to obtain the numbered segments with dense abnormal behavior. S303: Based on the densely numbered segments of the abnormal behavior, extract the position parameters and state parameters of the nodes corresponding to the numbers, compare the direction of node position change with the continuity of connection response behavior, filter the segments where the response performance deviates continuously in the node arrangement direction, extract the distribution range where the number and position are coupled in response features, and obtain the connection abnormal node block.

8. The method for identifying the dimensional variation of a transmission device model according to claim 1, characterized in that, The specific steps of S4 are as follows: S401: Based on the connection anomaly node block, extract the node connection sub-face state parameters in the order of node number, compare the state change direction between adjacent nodes, and combine the state change continuity direction with the number arrangement direction to obtain the connection state extension trend sequence. S402: Based on the connection state extension trend sequence, extract the node segments with continuous directional advancement, compare whether the displacement direction changes continuously along the numbered direction in the node segments, remove segments with reverse offset and sudden directional change, and obtain the displacement advancement behavior segments in the path. S403: Based on the displacement propagation behavior segments in the path, extract the corresponding node state change records and number position relationships, identify the consistency of state change amplitude, directional extension and node distribution density, extract segments with consistent state change directions and continuous node arrangement, and obtain the structural change propagation trajectory path.

9. The method for identifying the dimensional variation of a transmission device model according to claim 1, characterized in that, The specific steps of S5 are as follows: S501: Based on the structural change propulsion trajectory path, extract the connection status parameters and spatial shape size values ​​of the nodes in the path, compare the directionality of the size difference between adjacent nodes according to the node number order, extract the numbered segments where the size change direction has a turning behavior, and obtain the size change trend distribution segment. S502: Call the size change trend distribution segment, side by side the node size value and path position corresponding to each segment number, identify the path segments in which the size value change rate of adjacent nodes shows continuous increase and compression, filter the node number group with continuous behavior, and obtain the size change extension number interval. S503: Based on the extended numbering interval of the size change, extract the size value, number position and spatial distribution relationship of the corresponding node in the path, compare the continuity and arrangement consistency of the size response difference between nodes within the numbering segment, extract the range in which the size change coherence is consistent in space, and obtain the size change degree identification set.

10. A system for recognizing dimensional variations in a transmission device model, characterized in that, The system is used to implement the method for identifying the dimensional variation of a transmission device model as described in any one of claims 1-9, and the system includes: The angular extraction module obtains the angular response direction of the nodes in the connection sub-face area of ​​the transmission device under axial load, extracts the relationship between the node number and direction change of the continuous flipping path segment of the included angle of adjacent nodes, and obtains the node angular response path. Based on the node angular response path, the disturbance construction module analyzes the consistency of node response direction and position continuity, filters path segments with consistent directional trends and continuous spatial distribution, and constructs continuous segments according to node order to obtain response disturbance chain segments. Based on the response disturbance chain segment, the connection identification module analyzes the state changes of the connection sub-face of the path node under load, identifies discontinuous displacement, abrupt connection changes and misalignment combinations, filters concentrated areas, locates abnormal node segments, and obtains the connection abnormal node block. Based on the abnormal node blocks, the change and evolution module tracks the extension trend of connection state changes along the node arrangement direction, analyzes the continuous transition mode of node state behavior in the path, and combines the continuous region of state evolution extension behavior to obtain the structural change propagation trajectory path. The variation degree recognition module, based on the structural variation propagation trajectory path, compares the differences in the connection response of the path nodes, identifies the dimensional offset change segments, analyzes the relationship between the offset range and the node position, extracts the continuous dimensional change response features, and obtains the dimensional variation degree recognition set.

Images