Intelligent Evaluation Method and System for Solidification Effect of Ductile Iron Based on Multi-parameter Fusion

Abstract

This invention relates to the field of intelligent manufacturing technology, specifically to an intelligent method and system for evaluating the solidification effect of ductile iron using multi-parameter fusion. The method includes the following steps: acquiring the crystallization interface and cooling path, identifying grain offset trends, tracking impurity offset trajectories, determining grain misalignment boundaries, analyzing pore extension directions, identifying structural path intersection relationships, and obtaining the structural interference trend manifestation during the solidification process. In this invention, the spatial offset between the grain propagation direction and the cooling path is used to identify local structural transition regions; the migration trajectory of impurities in grain abrupt change regions is correlated with boundary changes for identification; grain connection angle jumps are used to determine misalignment paths; the pore extension direction and grain arrangement relationship are used to calibrate the penetration trend; the intersection behavior between paths forms the basis for fusion judgment through spatial distribution; interference regions are extracted; and the linkage between path behaviors transforms the local structural state into a difference in the solidification process, improving the coverage accuracy of multi-parameter fusion evaluation at the structural level.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology, and in particular to an intelligent method and system for evaluating the solidification effect of ductile iron by integrating multiple parameters. Background Technology

[0002] The field of intelligent manufacturing technology involves the application of information technology, automation control technology, and artificial intelligence technology in the manufacturing process to achieve improved production efficiency, optimized resource allocation, and intelligent decision-making throughout the entire production process. Core aspects of this technology include digital modeling of the manufacturing process, acquisition and fusion of production data, status monitoring of key processes, and quality control. Its systematic nature is reflected in the real-time acquisition of multi-source data during the production process through sensors, combined with numerical models or algorithms to comprehensively judge the equipment operating status and process effects, achieving collaborative control and autonomous adjustment of the manufacturing system. Traditional methods for evaluating the solidification effect of ductile iron rely on monitoring the process using a single parameter such as temperature field or cooling rate and judging quality based on empirical charts. These methods primarily depend on fixed-point temperature measuring devices to record temperature changes during casting, estimating the solidification state based on the temperature characteristics of metal phase transformations, and manually judging the microstructure distribution and shrinkage tendency based on past production experience. Thermocouple measurements are typically used as the basis, supplemented by conclusions drawn from metallurgical process engineers' analysis as the evaluation foundation.

[0003] Existing technologies rely primarily on single-point temperature measurements, lacking spatial positioning capabilities when faced with grain orientation shifts or cooling path variations. They fail to demonstrate the behavioral correlation between impurity diffusion direction and grain orientation within the microstructure, and the directional jumps exhibited by irregular grain boundary connections are difficult to accurately reveal through temperature changes. The extension direction of pores within the boundary structure lacks an identification mechanism, and path behaviors are judged separately. Interference trends formed at the boundary of structural disturbances are difficult to classify and identify, and under the combined effect of multiple factors, the evaluation of solidification effects is prone to deviating from the judgment range. Summary of the Invention

[0004] To address the technical problems existing in the prior art, embodiments of the present invention provide an intelligent method for evaluating the solidification effect of ductile iron by integrating multiple parameters, comprising the following steps:

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an intelligent method for evaluating the solidification effect of ductile iron based on multi-parameter fusion, comprising the following steps:

[0006] S1: Obtain the initial interface of graphite sphere crystallization and the liquid phase cooling path, identify the grain propagation direction and the spatial offset trend of heat flow guidance, locate the turning point region in the grain expansion path, and obtain the grain cooling direction offset segment.

[0007] S2: Based on the grain cooling direction offset segment, extract the impurity precipitation start point, track the outward movement direction in the grain structure, identify the impurity offset performance in the grain direction abrupt change region, and obtain the impurity thermally induced offset path.

[0008] S3: Based on the impurity thermally induced migration path, analyze the trend of grain connection region direction change, determine the structural state at the point of continuous abrupt change in connection angle, identify the misalignment boundary path in grain arrangement, and obtain the grain boundary misalignment path direction.

[0009] S4: Based on the grain boundary misalignment path, track the extension direction of the pores in the misalignment region, determine the relationship between the extension characteristics of the pore path along the grain boundary and the arrangement direction, and obtain the pore connectivity offset trend structure.

[0010] S5: Based on the aforementioned pore connectivity offset trend structure, analyze the intersection of grain misalignment path, impurity offset trajectory and pore extension direction, identify the segments of path superposition and direction convergence region within the structure, and obtain the structural interference trend performance during solidification.

[0011] As a further aspect of the present invention, the grain cooling direction offset segment includes the grain propulsion direction vector, the spatial distribution relationship of heat flow guidance, the boundary orientation deviation trend, and the structural turning section of the propulsion path. The impurity thermally induced offset path includes the precipitation initiation distribution, migration path turning characteristics, the motion trajectory of the abrupt arrangement region, and the non-principal axis diffusion region. The grain boundary misalignment path includes the arrangement connection angle variation point, multi-directional boundary intersection structure, grain angle abrupt change mode, and the continuous jump region of the connection direction. The pore connectivity offset trend structure includes the pore structure extension trajectory, the expansion trend along the boundary direction, the grain region connection mode, and the convergence of pore and arrangement directions. The solidification process structural interference trend includes the grain jump and impurity turning overlap region, the pore extension and connection abrupt change boundary zone, and the state of the interference behavior co-aggregation region.

[0012] As a further aspect of the present invention, the grain expansion path refers to extracting the continuous directional line segment of the grain growth path in space during the solidification process of ductile iron, and analyzing the growth trend and direction offset characteristics under the influence of the main heat flow axis.

[0013] The impurity precipitation start point refers to the spatial location point in the grain cooling direction offset region where the impurity first precipitates from the melt. Based on the relationship between the temperature gradient direction at the location and the main axis of grain arrangement, it is determined whether the impurity is located in the position segment where the grain arrangement trend changes.

[0014] As a further aspect of the present invention, the misaligned boundary path refers to, based on the grain connection angle dataset, detecting continuously jumping angle change segments in the grain connection section, tracking the position sequence of boundary direction abrupt changes, and identifying the set of spatial line segments in the grain arrangement structure that exhibit discontinuous direction and multi-directional connection phenomena.

[0015] The impurity migration trajectory refers to tracing the migration path direction in the grain structure, screening continuous migration segments with consistent orientation, and identifying the spatial migration path morphology of impurities under thermal induction in the non-principal axis region.

[0016] As a further aspect of the present invention, the specific steps of S1 are as follows:

[0017] S101: Obtain the grain boundary trajectory and liquid phase region cooling path in the initial interface of graphite sphere crystallization, extract the spatial extension line between the continuous grain propagation direction and the cooling guide, analyze the angle distribution between the grain propagation direction and the heat flow guide, and obtain the grain propagation direction extension relationship;

[0018] S102: Based on the grain propulsion direction extension relationship, compare the direction of the main axis of heat flow guidance, identify whether the grain propulsion direction deviates from the trend of the main axis of heat flow in a local section, analyze the length change of the angle deflection section and the heat flow direction jump node in the propulsion path, and obtain the grain propulsion direction deviation segment.

[0019] S103: Based on the grain propulsion direction deviation segment, screen the spatial angle change region within the grain propulsion path, locate the position region where the grain propulsion direction turns from the dominant heat flow direction in the segment, identify the grain boundary direction switching segment, and obtain the grain cooling direction offset segment.

[0020] As a further aspect of the present invention, the specific steps of S2 are as follows:

[0021] S201: Based on the grain cooling direction offset segment, identify the spatial starting point of impurity precipitation, detect the angular difference between the temperature gradient direction of the starting point region and the main axis direction of grain arrangement, determine whether the impurity release position is distributed in the region where the grain arrangement trend changes, and obtain the distribution of impurity precipitation starting position.

[0022] S202: Based on the distribution of the starting position of the impurity precipitation, track the movement trajectory of the impurity in the grain structure, extract the angle value between the trajectory direction vector and the grain principal axis direction, calculate the change range of the angle difference in the differentiated segment, and locate the continuous segment of the direction change to obtain the direction of the impurity migration offset trajectory.

[0023] S203: Based on the migration and offset trajectory of the impurities, analyze the path direction of the impurities at the outer edge of the grain structure, determine whether they continue to expand along the non-principal axis region, identify the continuity of movement and similar characteristics of the direction during the outward migration process, and obtain the thermally induced offset path of the impurities.

[0024] As a further aspect of the present invention, the specific steps of S3 are as follows:

[0025] S301: Based on the impurity thermally induced offset path, identify the angular change position of the grain arrangement connection region, obtain the grain principal axis direction vectors on both sides of the connection point, calculate the angle between adjacent principal axis vectors, and obtain the connection direction angle dataset.

[0026] S302: Based on the connection direction angle dataset, analyze the change of the included angle value in the grain connection path, filter the angle segments where the direction changes, determine whether the direction trend is discontinuous with the adjacent area, and obtain the boundary connection direction jump interval.

[0027] S303: Based on the boundary connection direction jump interval, detect the grain arrangement path, extract the connection line segments corresponding to the continuous direction change, track the corresponding direction trend in the structural connection path, and obtain the grain boundary misalignment path direction.

[0028] As a further aspect of the present invention, the specific steps of S4 are as follows:

[0029] S401: Based on the grain boundary misalignment path, monitor the directional trend of adjacent pores in the area associated with the path, identify the connection trend of the two ends of the pores in space, determine the angle distribution between the extension direction and the grain misalignment path, and obtain the pore extension direction vector set.

[0030] S402: Based on the set of pore extension direction vectors, according to the angle sequence between the pore extension line segment and the adjacent grain arrangement direction, filter the connecting segments with continuously changing directional trends and consistent orientation, separate the spatial path regions with consistent orientation, and obtain the directionally consistent distribution region.

[0031] S403: Based on the directionally consistent distribution region, analyze the lateral continuity of the pore path in the grain arrangement space, identify the continuous extended structural segments without path breaks, divide the spatially connected pore path set, and obtain the pore connectivity offset trend structure.

[0032] As a further aspect of the present invention, the specific steps of S5 are as follows:

[0033] S501: Based on the pore connectivity offset trend structure, analyze the correspondence between the grain misalignment path and the impurity offset trajectory in space, identify the segments in the grain arrangement direction jump segment that intersect with the impurity offset trajectory, and obtain the grain-impurity overlap region.

[0034] S502: Based on the grain impurity overlap region, compare the spatial intersection position of the pore extension path and the grain connection abrupt segment, extract the position segment where the connection direction turns and intersects with the pore path, and obtain the path intersection structure segment.

[0035] S503: Based on the path intersection structure segment, identify the structural positions where grain jumps, impurity turning and pore extension interact in space, locate the intersecting path states, and obtain the structural interference trend of the solidification process.

[0036] A multi-parameter integrated intelligent evaluation system for the solidification effect of ductile iron includes:

[0037] The grain offset recognition module obtains the initial interface of graphite sphere crystallization and the liquid phase cooling path, extracts the grain propulsion direction vector, compares it with the spatial relationship of heat flow guidance, detects whether the grain boundary in the propulsion path has a directional offset, filters the position segments of structural extension and turning, and obtains the grain cooling direction offset segment.

[0038] The impurity migration extraction module extracts the impurity precipitation start point based on the grain cooling direction offset segment, tracks the direction change in the grain path, identifies the offset trend of impurities in the abrupt arrangement region, locates the position segment that causes offset along the grain boundary, and obtains the impurity thermally induced offset path.

[0039] Based on the impurity thermally induced offset path, the structural misalignment identification module extracts the position of the angle change in the grain connection region, determines whether there is a direction switch in the grain boundary connection, detects the segments of extension and change in the connection path, identifies the structural segments of grain misalignment, and obtains the direction of the grain boundary misalignment path.

[0040] The pore extension identification module extracts the extension direction of adjacent pores based on the grain boundary misalignment path, determines whether the pores are connected along the misalignment boundary region, compares the overlap between the path and the grain arrangement direction, extracts the directional features of the pores extending inside the structure, and obtains the pore connectivity offset trend structure.

[0041] The interference trend assessment module, based on the pore connectivity offset trend structure, compares the spatial positions of grain misalignment, impurity offset and pore penetration path, identifies the intersection relationship between behavioral paths, extracts the regional performance of paths superimposed and adjacently extended in the structure, and obtains the structural interference trend performance of the solidification process.

[0042] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0043] In this invention, the spatial offset between the grain propagation direction and the cooling path is used to identify local structural transition regions. The migration trajectory of impurities in the grain abrupt change region is associated with the boundary changes for identification. The grain connection angle jump is used to determine the misalignment path. The relationship between the pore extension direction and the grain arrangement is used to calibrate the penetration trend. The intersection behavior between paths forms the basis for fusion judgment through spatial distribution. Interference regions are extracted. The linkage between path behaviors transforms the local structural state into the difference in the solidification process, thereby improving the coverage accuracy of multi-parameter fusion evaluation at the structural level. Attached Figure Description

[0044] 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.

[0045] Figure 1 This is a schematic diagram of the steps of the present invention;

[0046] Figure 2 This is a detailed schematic diagram of S1 of the present invention;

[0047] Figure 3 This is a detailed schematic diagram of S2 of the present invention;

[0048] Figure 4 This is a detailed schematic diagram of S3 of the present invention;

[0049] Figure 5 This is a detailed schematic diagram of S4 of the present invention;

[0050] Figure 6 This is a detailed schematic diagram of S5 of the present invention;

[0051] Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0052] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0053] 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.

[0054] 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, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0055] 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.

[0056] 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.

[0057] Please see Figure 1 This invention provides an intelligent method for evaluating the solidification effect of ductile iron by integrating multiple parameters, comprising the following steps:

[0058] S1: Obtain the initial interface of graphite sphere crystallization and the liquid phase cooling conduction path, extract the grain propulsion direction vector, compare it with the extension relationship of heat flow guidance in spatial distribution, detect whether the continuous direction of grain boundary deviates from the heat flow direction during the propulsion process, screen the position segment where the structural direction changes in the grain propulsion path, and obtain the grain cooling direction offset segment.

[0059] S2: Based on the grain cooling direction offset segment, extract the impurity precipitation space starting point, track the adjustment trajectory of migration direction in the grain path, identify whether the movement direction changes in the grain arrangement abrupt region, monitor whether the impurities diffuse outward along the non-main axis region, and locate the position segment where the outward movement occurs to obtain the impurity thermally induced offset path.

[0060] S3: Based on the impurity thermally induced migration path, extract the position of the angle change in the grain arrangement connection region, analyze whether the grain boundary forms a multi-directional intersection structure, detect the extension angle jump behavior between grains, identify the path region of continuous abrupt change in the boundary connection direction, determine whether structural connection misalignment occurs, and obtain the grain boundary misalignment path direction.

[0061] S4: Based on the grain boundary misalignment path, identify the extension direction of adjacent pore structures, determine whether they extend along the grain misalignment boundary, track the connection mode of pore paths in the grain arrangement region, analyze whether there is overlap and convergence between the pore extension direction and the grain arrangement direction, and obtain the pore connectivity offset trend structure.

[0062] S5: Based on the pore connectivity shift trend structure, the spatial distribution of grain misalignment, impurity outward movement and pore penetration path is analyzed simultaneously. The overlap range between the grain arrangement jump position and the impurity migration direction is identified. The spatial intersection relationship between the pore extension region and the grain connection abrupt segment is compared. The regional state of the intersection of interference behaviors in the structure is extracted to obtain the structural interference trend performance during the solidification process.

[0063] The grain cooling direction offset segment includes the grain propagation direction vector, the spatial distribution relationship of heat flow guidance, the boundary deviation trend, and the structural turning section of the propagation path. The impurity thermally induced offset path includes the distribution of precipitation initiation points, migration path turning characteristics, the movement trajectory of the abrupt arrangement region, and the non-principal axis diffusion location. The grain boundary misalignment path includes the arrangement connection angle variation point, multi-directional boundary intersection structure, grain angle abrupt change mode, and the continuous jump region of the connection direction. The pore connectivity offset trend structure includes the pore structure extension trajectory, the expansion trend along the boundary direction, the grain region connection mode, and the convergence of pore and arrangement directions. The solidification process structural interference trend includes the overlapping area of ​​grain jump and impurity turning, the boundary zone of pore extension and connection abrupt change, and the state of the interference behavior co-aggregation region.

[0064] Please see Figure 2 The specific steps of S1 are as follows:

[0065] S101: Obtain the grain boundary trajectory and liquid phase region cooling path in the initial interface of graphite sphere crystallization, extract the spatial extension line between the continuous grain propagation direction and the cooling guide, analyze the angle distribution between the grain propagation direction and the heat flow guide, and obtain the grain propagation direction extension relationship;

[0066] First, the grain contour and heat transfer lines in the liquid phase are extracted from the cross-sectional image of the casting during the early solidification stage. Based on the gray-scale abrupt change lines in the grain edge region of the metallographic image, the spatial location covered by the grain elongation path is located. Simultaneously, cooling conduction lines are drawn in the liquid phase region according to the temperature gradient change direction. For the spatial arrangement of the grain elongation path and heat transfer lines in the same region, the grain growth propagation vector set and the liquid phase cooling path vector set are extracted. For each set of grain propagation vectors and corresponding cooling vectors, the spatial angle value sequence between them is obtained. Further, it is detected whether these angle values ​​show a concentrated change in a continuous spatial region. Based on the trend of grain growth, and combined with examples of five consecutive grain propagation path angles of 14 degrees, 17 degrees, 19 degrees, 20 degrees, and 18 degrees, this region can be considered to have the grain propagation direction and the overall extension of the heat flow direction consistent. If the angle value changes to 45 degrees after entering the next propagation segment, it is considered that the grain growth direction deviates from the cooling direction, indicating a tendency of spatial turning. Such abnormal angle change regions have a staged distribution characteristic in the grain path. The range of segments in which the grain propagation direction extends and changes can be delineated by matching the turning point position in the grain propagation path with the intersection position of the liquid phase path, and finally the grain propagation direction extension relationship can be obtained.

[0067] S102: Based on the extension relationship of the grain propulsion direction, compare the direction of the main axis of heat flow guidance, identify whether the grain propulsion direction deviates from the trend of the main axis of heat flow in a local section, analyze the length change of the angle deflection section and the jump node of the heat flow direction in the propulsion path, and obtain the grain propulsion direction deviation segment.

[0068] First, the propulsion vector groups extracted from the image are spatially compared with the heat flow segments. An angular correspondence is established between each grain propulsion vector group and the main heat flow axis on a two-dimensional plane. Continuous propulsion segments are selected from multiple propulsion path sequences, and the changing behavior of the propulsion trend is measured using angle data. If the angle between the grain propulsion direction and the main heat flow axis changes continuously within the same region, there may be a deviation in the propulsion direction in that region. Further, segments with more frequent changes in the angle sequence are extracted, and the angular fluctuations of three to five adjacent grain propulsion path segments are recorded. For example, if the angle values ​​of a continuous segment of a grain path are 17 degrees, 19 degrees, 21 degrees, 35 degrees, and 33 degrees, it can be preliminarily determined that a deviation occurs at 35 degrees. Next, it is determined whether the total length of the line segment in the angled transition region exceeds the average length of the grain growth segment. When the total length of the angled deflection segment is found to be 4.8 mm, while the average path segment in other regions is 3 mm, it can be determined that it belongs to an abnormal deviation segment. At the same time, based on whether there are nodes of turning direction or tortuous connection in this local space according to the direction of the main heat flow axis, it is compared whether the continuous path segment guided by the heat flow produces fluctuation segments in space. If the guiding path of the main heat flow axis in this region changes by more than 20 degrees, it is determined that the region has both grain propulsion direction deviation and heat flow direction jump phenomenon, thus locating the grain path segment as a deviation segment region, and finally obtaining the grain propulsion direction deviation segment.

[0069] S103: Based on the grain propulsion direction deviation segment, screen the spatial angle change region within the grain propulsion path, locate the position region where the grain propulsion direction turns from the dominant heat flow direction in the segment, identify the grain boundary direction switching segment, and obtain the grain cooling direction offset segment.

[0070] First, based on the distribution of the propulsion path segments reflected in the two-dimensional image, the turning points with abrupt angle changes are identified one by one. Each grain propulsion trajectory is marked according to whether the angle between the preceding and following propulsion segments changes rapidly. For example, if a grain path has a 28-degree angle in the preceding segment and a 67-degree angle in the following segment, it is considered a spatial angle abrupt change location. Then, these abrupt change segments are overlaid and matched in the heat flow-dominant direction layer. By comparing the intersections or overlapping areas of the two paths, the intersection segments where the grain propulsion direction shifts and the heat flow-oriented path changes direction are further located. For example, in an image patch, the node where the grain propulsion line segment shifts is located at pixel 42 on the X-axis of the image. At this location, which is close to the 37-pixel region on the X-axis of the heat flow guiding turning path, it is determined to be a grain propulsion turning node formed under the influence of heat flow. Subsequently, the direction line of the grain boundary in this region is extracted, and it is identified whether the connection angle between the front and rear boundaries changes direction. For example, if the inclination angle of the front boundary is 21 degrees and the rear boundary suddenly changes to 60 degrees, it is considered that there is a change in the grain boundary direction in this region. Such switching segments are accumulated in multiple paths. By identifying their extension trend in different directions and combining it with the trend of the main heat flow axis, the region segment that simultaneously satisfies the change in grain propulsion direction and the deflection of grain boundary direction is selected, and finally the grain cooling direction offset segment is obtained.

[0071] Please see Figure 3 The specific steps of S2 are as follows:

[0072] S201: Based on the grain cooling direction offset segment, identify the spatial starting point of impurity precipitation, detect the angular difference between the temperature gradient direction of the starting point region and the main axis direction of grain arrangement, determine whether the impurity release position is distributed in the region where the grain arrangement trend changes, and obtain the distribution of impurity precipitation starting position.

[0073] First, the trajectory of the advancement direction jump is segmented. The locations where the tangential angle of the grain boundary changes are selected for point-by-point verification to confirm the temperature gradient change range corresponding to the temperature field distribution layer within the region. The normal direction of the temperature contour line at each point is extracted as the gradient direction at that location. Then, the gradient direction vectors at these points are compared with the principal axis vectors of the grain arrangement at the same location. If, at a certain location, such as position 122 on the X-axis of the image, the temperature gradient vector points 22 degrees east of north, while the grain principal axis vector points due north, then the angle difference at that point is 22 degrees. This is recorded as the location where there is an angular offset between the gradient direction and the principal axis direction. Then, based on the thermal... The field evolution image sequence identifies points showing signs of brightness contraction in regions of rapid temperature decrease. For example, if a continuous decrease in grayscale occurs in the Y-axis segment from 45 to 52, it is determined to be the region where impurities begin to be released. Simultaneously, the grain arrangement change path in this region is superimposed to determine whether these impurity release points are located within the range where the grain arrangement trajectory changes direction. For example, at X-axis 115, the grain trajectory angle changes from 27 degrees in the first segment to 64 degrees in the second segment, and the impurity precipitation signal is formed in the corresponding turning segment, which is determined to be related to grain turning. Finally, it is confirmed whether the impurity precipitation initiation point is distributed in the region where the grain arrangement trend changes direction, thus obtaining the distribution of impurity precipitation initiation positions.

[0074] S202: Based on the distribution of the starting position of impurity precipitation, track the movement trajectory of impurities in the grain structure, extract the angle value between the trajectory direction vector and the grain principal axis direction, calculate the change range of angle difference in the differentiated segment, and locate the continuous segment of direction change to obtain the direction of impurity migration offset trajectory.

[0075] The specific formula for calculating the magnitude of change in angle differences within differentiated paragraphs is as follows:

[0076] ;

[0077] in, Representing the The angle between the direction of the impurity movement trajectory in the segment and the direction of the grain principal axis. This represents the principal axis reference direction angle of the current grain region. Representing the Impurity migration stability factor of the segment Representing the The range of variation in the grain arrangement direction within the segment The strain compatibility parameter representing the microregions of the grain, This represents the total number of paragraphs included in the analysis;

[0078] Segment 1 has an included angle of 15°, a reference direction of 12°, an included angle difference of 3°, a stability factor of 0.85, a strain angle of 5°, and a harmonicity of 0.90. The corresponding calculations are as follows:

[0079] ;

[0080] ;

[0081] ;

[0082] ;

[0083] The included angle of segment 2 is 18°, the difference is 6°, the stability factor is 0.80, the strain angle is 3°, and the harmony degree is 0.88.

[0084] ;

[0085] ;

[0086] ;

[0087] ;

[0088] Segment 3 has an included angle of 14°, a difference of 2°, a stability factor of 0.82, a strain angle of 4°, and a harmonicity of 0.87.

[0089] ;

[0090] ;

[0091] ;

[0092] ;

[0093] ;

[0094] Right now:

[0095] ;

[0096] Interpretation of results and numerical significance: The change range of angle difference in the differentiated paragraph is 4.83°. The reference angle fluctuation range is set to 0–7°. The current result falls within the fluctuation range, indicating that the degree of angle offset is in the stable zone.

[0097] Explanation of the innovative aspects of the formula:

[0098] The advantages of the formula are that the square and square root operations are used to normalize the difference in magnitude, the absolute value ensures that the directional error is not distorted due to the cancellation of positive and negative directions, the multiplication is used to quantify the coupling effect between factors, the division is average normalization, and it reflects the trend of the overall trajectory disturbance amplitude. The formula enhances the response intensity to nonlinear turning trajectories through a multi-parameter coupling mechanism and improves the detection capability of grain offset paths.

[0099] S203: Based on the migration and offset trajectory of impurities, analyze the path direction of impurities at the outer edge of the grain structure, determine whether they continue to expand along the non-principal axis region, identify the continuity of movement and similar characteristics of the direction during the outward migration process, and obtain the thermally induced offset path of impurities.

[0100] First, the boundary segments of impurity brightness signals are extracted from the thermal imaging sequence. The continuous movement direction of edge pixels as impurities expand outward from the grain is identified. Vector projections along the X and Y coordinate axes of the image are constructed, and their connection trajectories are established. The path direction is identified by the pixel offset values ​​of impurity edge points in consecutive frames. For example, if an impurity point is located at coordinates (118, 84), (120, 86), and (123, 88) in three consecutive frames, then the continuous outward movement direction of the impurity is 45 degrees to the upper right. Subsequently, this directional trajectory is matched with the directional angles within the grain principal axis arrangement area, and the angle range of non-principal axis directions is selected as the 65-90 degree range. Impurity trajectories continuously located in this angle range are classified as non-principal axis trajectories. The process continues to detect any jumps, breaks, or changes in direction in the impurity's movement path. If the trajectory maintains a constant directional trend in any three consecutive frames, the path segment is determined to be a continuous movement segment. For example, between frames 65 and 72, if the change in the movement direction angle of the impurity boundary point does not exceed 3 degrees and the change in the movement path length is within 2 pixels, it is classified as a continuous path. Further, the direction angles of these continuous trajectories are compared, and similar angle segments are selected. Among multiple impurity trajectories, sample paths with continuous directions maintained between 75 and 80 degrees are selected as similar paths. Finally, a set of impurity outward movement paths with consistent directional trends, obvious continuity, and deviation from the grain principal axis are extracted to obtain the impurity thermally induced migration path.

[0101] Please see Figure 4 The specific steps of S3 are as follows:

[0102] S301: Based on the impurity thermally induced migration path, identify the angular change position of the grain arrangement connection region, obtain the grain principal axis direction vectors on both sides of the connection point, calculate the angle between adjacent principal axis vectors, and obtain the connection direction angle dataset.

[0103] The formula for calculating the angle between adjacent principal axis vectors is as follows:

[0104] ;

[0105] in, Representing the The angle between the principal axis vectors of the grains on both sides of the grain connection point Representing the The first connection point The unit vector of the principal axis direction of the left grain in this measurement. Representing the The first connection point The unit vector of the principal axis direction of the right-side grain in this measurement. The perturbation direction correction vector represents the measurement location. Representing the The first connection point Thermal conductivity anisotropy coefficient at the second measurement point Represents the stability constant. This represents the number of measurements taken at each connection point.

[0106] By connection point Number of measurements For example:

[0107] Example sample value:

[0108] , ;

[0109] , ;

[0110] , ;

[0111] , , ;

[0112] , , ;

[0113] Substitute the above data into the calculation:

[0114] Dot product (difference vector dot product with correction vector):

[0115] ;

[0116] The value in the second group is approximately 0.17;

[0117] The value in group 3 is approximately 0.0222;

[0118] Summing after taking the absolute value: ;

[0119] Calculation of the denominator:

[0120] multiplied by 1.2;

[0121] The value of the second group is approximately 2 × 0.7 = 1.4;

[0122] The value of the third group is approximately 2 × 0.8 = 1.6;

[0123] Summation: 1.2 + 1.4 + 1.6 = 4.2;

[0124] Square root handling: ;

[0125] Adding the stabilizing term: 2.049 + 0.01 = 2.059;

[0126] Finally, substitute into the formula:

[0127] ;

[0128] Interpretation of results and numerical significance: The angle between the principal axes of the grains on both sides at the first grain connection point is 1.411 rad, which is a slight orientation deflection compared with the preset grain boundary orientation change threshold (1.570 rad = 90°).

[0129] Explanation of the innovative aspects of the formula:

[0130] The advantage of the formula lies in the introduction of the thermally induced perturbation direction vector. and thermal conductivity anisotropy weighting parameters This not only enhances the response capability to thermal field disturbances in the calculation of grain angle, but also makes the value of the angle distinguishable from the local thermal conductivity asymmetry at the grain boundary, thereby improving the accuracy of microstructure modeling.

[0131] S302: Based on the connection direction angle dataset, analyze the change of the included angle value in the grain connection path, filter the angle segments where the direction changes, determine whether the direction trend is discontinuous with the adjacent area, and obtain the boundary connection direction jump interval.

[0132] First, the included angle sequence of each connecting line segment is extracted according to the arrangement order of continuous boundary nodes in the grain connection path. By judging the difference in included angle values ​​between any two adjacent line segments, the location of the direction change in the path is identified. For example, if the included angle of a certain grain connection line segment is 32 degrees, and the included angle of its adjacent next segment changes to 68 degrees, this can be marked as an angle change node. Further, all nodes in the path with included angle changes greater than 30 degrees are grouped together and identified as segments with direction change angles. The relative positions of the nodes with continuous angle change in the entire path are judged with the direction of adjacent paths. To identify areas where the directional trend of connecting segments is inconsistent, for example, if the angle of segment 7 is 45 degrees, segment 8 becomes 75 degrees, and segment 9 drops back to 44 degrees and is parallel to the direction of segment 7, then segment 8 can be determined to have a brief directional discontinuity. This segment is considered a directional jump segment. By statistically analyzing the segments with such brief directional jumps along the entire path, and using the coordinates of their starting and ending nodes as the boundary range, these segments are classified into the directional discontinuity segment range. Finally, these connecting path segments with significant directional abrupt changes and continuity interruption characteristics are screened out to obtain the boundary connecting directional jump intervals.

[0133] S303: Based on the boundary connection direction jump interval, detect the grain arrangement path, extract the connection line segment corresponding to the continuous direction change, track the corresponding directional trend in the structural connection path, and obtain the grain boundary misalignment path direction.

[0134] First, the grain boundary segments containing the identified transition nodes are extracted, and the connection order of these boundary segments in space is identified segment by segment. Path segments with consecutive directional abrupt changes are located. During the location process, the correspondence between the angle transition position and the grain arrangement boundary nodes is relied upon. The corresponding line segment direction is plotted in the actual grain structure diagram, tracing the arrangement trend of adjacent boundary segments before and after the angle transition. For example, when a connection path is continuously composed of segments 11, 12, and 13, with segment 12 showing a significant directional deflection and its starting point located at the node where the grain connection direction angle abruptly changes from 45 degrees to 81 degrees, such paths will be identified. Within the scope of structural misalignment judgment, the analysis examines whether there are repeated bends or continuous reversals in the grain arrangement within the selected multiple angle jump segments. If the angle change of every three line segments in the path shows an inconsistent trend or does not maintain a unified direction, the path segment is marked as a candidate misalignment segment. By connecting the node trajectories of continuous misalignment segments on the two-dimensional projection map, each misalignment segment is sequentially connected to form a continuous path band. This further depicts the specific orientation area occupied by directional misalignment in the grain boundary and records the spatial displacement order relationship of the corresponding nodes on the grain map, ultimately obtaining the grain boundary misalignment path orientation.

[0135] Please see Figure 5 The specific steps of S4 are as follows:

[0136] S401: Based on the grain boundary misalignment path, monitor the directional trend of adjacent pores in the associated area of ​​the path, identify the connection trend of the two ends of the pore in space, determine the angle distribution between the extension direction and the grain misalignment path, and obtain the pore extension direction vector set.

[0137] First, the coordinates of the endpoints of the pores within the region corresponding to the grain misalignment path are retrieved. The spatial connection between each pair of pore endpoints is traced segment by segment. Pore connection segments are extracted sequentially from the 2D or 3D grain arrangement diagram. The direction vector of each segment is analyzed, and the angle change between it and the direction of the grain misalignment path is compared. In determining the angle trend, several representative pore paths, such as segments AB, CD, and EF, can be selected. Their respective direction vectors are calculated and compared with the vector relationship of the corresponding segment GH in the grain misalignment path. The angle for segment AB is 72 degrees, for segment CD it is 74 degrees, and for segment EF it is 68 degrees. All of them are offset between 60 and 80 degrees, indicating that there is a trend of angular deflection in the overall direction of the pores in this region. At the same time, the spatial distance between the two ends of each pore path and the direction of the connecting line are statistically analyzed to determine whether the continuous directional extension is maintained. During the tracking process, abnormal pore segments with obvious directional changes due to bifurcation or intersection need to be removed, and the reference coordinates of the extracted segments in space are uniformly adjusted to form a set of directional vectors under a stable connecting path. The angle states of all extension directions and grain misalignment paths are summarized in the dataset in the form of angle sequences, and finally the pore extension direction vector set is obtained.

[0138] S402: Based on the pore extension direction vector set, according to the angle sequence between the pore extension line segment and the adjacent grain arrangement direction, the connecting segments with continuous changes in direction and consistent direction are screened to separate the spatial path region with consistent direction and obtain the region with consistent direction.

[0139] First, the sequence of angle values ​​between the direction vector corresponding to each pore connection segment and its adjacent grain arrangement segment is read sequentially. All direction pairs with consecutive angles within the range of 20 to 30 degrees are extracted. During the operation, pore paths P1, P2, and P3 can be extracted from sample region A, with corresponding angle values ​​of 22 degrees, 26 degrees, and 24 degrees, respectively. The trend of change is judged to be continuous and consistent, without abrupt jumps. Further screening is performed on direction segments with angle differences not exceeding 5 degrees in this sequence. Pore paths meeting the criteria are marked as candidate connection segments. Then... In the 3D grain arrangement diagram, the spatial orientation of these candidate segments is checked segment by segment to confirm whether they are consistent. If there is a sudden change in orientation, the segment needs to be removed from the candidate path and the path extension boundary range needs to be readjusted. In the process of confirming the effective path, the starting point and ending point of each path segment need to be marked in the grain image coordinates to determine whether there are obvious backsliding, sudden bends or path intersections. If not, the segment is retained as a path segment with consistent orientation, and the 3D coordinate block where it is located is classified into a segment region with a stable orientation. Finally, the set of all such path blocks with consistent orientation is extracted to obtain the distribution area with consistent orientation.

[0140] S403: Based on the directional distribution region, the transverse continuity of the pore path in the grain arrangement space is analyzed, the continuous extension of the structural segment without path breakpoint is identified, the set of spatially connected pore paths is divided, and the pore connectivity offset trend structure is obtained.

[0141] Extract the start and end coordinates of all pore paths in the 3D grain arrangement diagram. Project each pore path onto the distribution plane of the grain arrangement principal axis and the transverse intersecting region according to its endpoint position. Track the change amplitude of the direction vector between path segments one by one. Analyze the direction of stable and continuous paths to determine whether they maintain line segment connection without interruption during path extension. If, in the example, pore paths P5, P6, and P7 are observed to extend at a 40-degree angle in the xy plane and maintain the same direction vector as the subsequent P8 without any breakpoints between endpoints, then merge P5 to P8 into one extended path. Subsequently, according to... The lateral arrangement of paths is used to define multiple locally connected path groups. It is confirmed whether the paths in each group penetrate between adjacent grains without spatial jumps. At the same time, the included angle values ​​between the path directions are compared one by one. Under the premise that the included angle value is less than 10 degrees and the distance between the path endpoints is less than 1.5 units, they are classified into the same connectivity set. The area where the path group is located is marked by referring to the distribution map of the pore orientation in the grain arrangement space. Finally, the number of continuous connected paths, the stability of the directional trend, and the lateral crossing range in each connectivity set are used as classification indicators to sort out the spatial segments corresponding to various path sets and obtain the pore connectivity offset trend structure.

[0142] Please see Figure 6The specific steps of S5 are as follows:

[0143] S501: Based on the pore connectivity offset trend structure, analyze the correspondence between the grain misalignment path and the impurity offset trajectory in space, identify the segments in the grain alignment direction jump segment that intersect with the impurity offset trajectory, and obtain the grain-impurity overlap region.

[0144] First, the spatial coordinate range of each pore connectivity path segment is extracted. The grain number covered by each path segment and its misalignment path identifier are obtained. The grain misalignment path direction is segmented and labeled. All grain arrangement direction jump segments are used as the retrieval benchmark based on the bend feature identifier. The start and end positions of each jump segment are defined. At the same time, the outward migration trajectory points and direction trends in the impurity thermally induced migration path are retrieved and superimposed onto the grain spatial layout map. Matching and identification are performed based on the degree of overlap between the spatial grid points where the trajectory points are located and the grain coordinates. In the example area, if there is a jump between grain numbers G23 and G24 where the grain boundary direction changes from 45 degrees to 15 degrees... In the case of a variable angle segment, if the impurity path number M12 happens to pass through the variable angle segment, then the intersection is marked as an overlapping segment. Further, the intersection positions with a spatial overlap rate greater than 50% are screened from all variable angle segments and impurity path sets, and their start and end path numbers and intersection angle values ​​are recorded. Under the condition that the angle is less than 30 degrees, they are included in the initial selection set of overlapping regions. Then, the path continuity length of each overlapping segment is counted to determine whether it exceeds 3 grain spans. If it meets the requirements, it is determined to be a valid overlapping segment and added to the target set. Finally, all valid segments that meet the requirements of spatial intersection, angle conditions and path span are integrated to obtain the grain impurity overlapping region.

[0145] S502: Based on the overlapping region of grain impurities, the spatial intersection position of the pore extension path and the abrupt segment of grain connection is compared, and the position segment where the connection direction changes and intersects with the pore path is extracted to obtain the path intersection structure segment.

[0146] First, extract the grain connection jump segment numbers and their start and end coordinate ranges for all overlapping segments. Simultaneously, obtain the endpoint and corner positions of each pore extension path in the pore connectivity offset trend structure. Call the spatial positioning data of both, and define the position where the connection direction changes as the corner coordinates in the grain connection segment. Mark the sequence of turning points and compare them spatially with the intersection points of line segments in the pore path one by one. In the example, if the grain connection abrupt segment D07 to D08 changes from a northwest to southeast direction to a south direction, and the pore path number P16 forms a traversing pattern in the same area, then this intersection position is judged as a candidate segment for path intersection. Further limit the intersection judgment conditions to intersection angle less than 40 degrees and intersection distance within 5 micrometers. Filter all segment numbers that meet the conditions and extract the corresponding path number and grain connection number to form a cross-matching mapping relationship table. After removing path items with no continuous connection features or a connection span of less than two grain units, integrate the remaining intersection node set to obtain the path intersection structure segment.

[0147] S503: Based on path intersection structural segments, identify the structural locations where grain jumps, impurity orientations and pore extensions interact in space, locate the intersecting path states, and obtain the structural interference trend of the solidification process.

[0148] First, the numbers and corresponding spatial coordinates of the transition segments in the grain connection path are extracted. Then, path segments with significant directional changes are selected from the impurity turning paths, and the endpoints of the two consecutive directional vectors before and after the turning point are recorded. This further locates the spatial relative relationship between each turning node and the grain connection inflection point. Simultaneously, the path numbers and endpoint coordinates of the pore extension paths with staggered endpoint distribution characteristics are retrieved. The data of the three types of paths are uniformly arranged in the same spatial reference system. By establishing a spatial grid to divide the region, the grain path bends, impurity deflection segments, and pore path crossing points are simultaneously assigned to the spatial interaction monitoring area. The existence of overlapping areas of the three types of paths in these interaction monitoring areas is then statistically analyzed. In the example, grain segment L12, impurity trajectory T08, and pore path P05 form a fan-shaped intersection between blocks M5 and M6, with spatial angles ranging from 25 to 60 degrees. Further, the path segment lengths within a 10-micrometer range of the above intersection points are extracted and labeled to determine whether direction switching, path turning, and spatial reversal phenomena exist simultaneously in the intersection areas of the three types of paths. All intersection node numbers that meet the above characteristics and their connection path types are counted and classified as structural interaction interference sources. Then, combined with the grain segment number and impurity migration number associated with each intersection node, an interference trend segment index table is formed, and finally, the structural interference trend performance of the solidification process is obtained.

[0149] Please see Figure 7 A multi-parameter integrated intelligent evaluation system for the solidification effect of ductile iron, including:

[0150] The grain offset recognition module obtains the initial interface of graphite sphere crystallization and the liquid phase cooling path, extracts the grain propulsion direction vector, compares it with the spatial relationship of heat flow guidance, detects whether the grain boundary in the propulsion path has a directional offset, filters the position segments of structural extension and turning, and obtains the grain cooling direction offset segment.

[0151] The impurity migration extraction module extracts the impurity precipitation starting point based on the grain cooling direction offset segment, tracks the direction change in the grain path, identifies the offset trend of impurities in the abrupt arrangement region, locates the position segment that causes offset along the grain boundary, and obtains the thermally induced offset path of impurities.

[0152] The structural misalignment identification module extracts the position of angle change in the grain connection region based on the impurity thermally induced offset path, determines whether there is a direction switch in the grain boundary connection, detects the segments of extension and change in the connection path, identifies the structural segments of grain misalignment, and obtains the direction of the grain boundary misalignment path.

[0153] The pore extension identification module extracts the extension direction of adjacent pores based on the grain boundary misalignment path, determines whether the pores are connected along the misalignment boundary region, compares the overlap between the path and the grain arrangement direction, extracts the directional features of the pores extending inside the structure, and obtains the pore connectivity offset trend structure.

[0154] The interference trend assessment module is based on the pore connectivity offset trend structure. It compares the spatial positions of grain misalignment, impurity offset and pore penetration path, identifies the intersection relationship between behavioral paths, and extracts the regional performance of paths superimposed and adjacent extension in the structure to obtain the structural interference trend performance of the solidification process.

[0155] 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 multi-parameter fusion-based intelligent method for evaluating the solidification effect of ductile iron, characterized in that, Includes the following steps: S1: Obtain the initial interface of graphite sphere crystallization and the liquid phase cooling path, identify the grain propagation direction and the spatial offset trend of heat flow guidance, locate the turning point region in the grain expansion path, and obtain the grain cooling direction offset segment. S2: Based on the grain cooling direction offset segment, extract the impurity precipitation start point, track the outward movement direction in the grain structure, identify the impurity offset performance in the grain direction abrupt change region, and obtain the impurity thermally induced offset path. S3: Based on the impurity thermally induced migration path, analyze the trend of grain connection region direction change, determine the structural state at the point of continuous abrupt change in connection angle, identify the misalignment boundary path in grain arrangement, and obtain the grain boundary misalignment path direction. S4: Based on the grain boundary misalignment path, track the extension direction of the pores in the misalignment region, determine the relationship between the extension characteristics of the pore path along the grain boundary and the arrangement direction, and obtain the pore connectivity offset trend structure. S5: Based on the pore connectivity offset trend structure, analyze the intersection of grain misalignment path, impurity offset trajectory and pore extension direction, identify the path superposition and direction convergence area segments in the structure, and obtain the structural interference trend of the solidification process. The specific steps of S4 are as follows: S401: Based on the grain boundary misalignment path, monitor the directional trend of adjacent pores in the area associated with the path, identify the connection trend of the two ends of the pores in space, determine the angle distribution between the extension direction and the grain misalignment path, and obtain the pore extension direction vector set. S402: Based on the set of pore extension direction vectors, according to the angle sequence between the pore extension line segment and the adjacent grain arrangement direction, filter the connecting segments with continuously changing directional trends and consistent orientation, separate the spatial path regions with consistent orientation, and obtain the regions with consistent orientation. S403: Based on the directionally consistent distribution region, analyze the transverse continuity of the pore path in the grain arrangement space, identify the continuous extended structural segments without path breaks, divide the set of spatially connected pore paths, and obtain the pore connectivity offset trend structure. The specific steps of S5 are as follows: S501: Based on the pore connectivity offset trend structure, analyze the correspondence between the grain misalignment path and the impurity offset trajectory in space, identify the segments in the grain arrangement direction jump segment that intersect with the impurity offset trajectory, and obtain the grain-impurity overlap region. S502: Based on the grain impurity overlap region, compare the spatial intersection position of the pore extension path and the grain connection abrupt segment, extract the position segment where the connection direction turns and intersects with the pore path, and obtain the path intersection structure segment. S503: Based on the path intersection structure segment, identify the structural positions where grain jumps, impurity turning and pore extension interact in space, locate the intersecting path states, and obtain the structural interference trend of the solidification process.

2. The intelligent evaluation method for solidification effect of ductile iron based on multi-parameter fusion according to claim 1, characterized in that, The grain cooling direction offset segment includes the grain propulsion direction vector, the spatial distribution relationship of heat flow guidance, the boundary deviation trend, and the structural turning section of the propulsion path. The impurity thermally induced offset path includes the distribution of precipitation initiation points, migration path turning characteristics, the movement trajectory of the abrupt arrangement region, and the non-principal axis diffusion region. The grain boundary misalignment path includes the arrangement connection angle variation point, multi-directional boundary intersection structure, grain angle abrupt change mode, and the continuous jump region of the connection direction. The pore connectivity offset trend structure includes the pore structure extension trajectory, the expansion trend along the boundary direction, the grain region connection mode, and the convergence of pore and arrangement directions. The solidification process structural interference trend includes the overlapping area of ​​grain jump and impurity turning, the boundary zone of pore extension and connection abrupt change, and the state of the interference behavior co-aggregation region.

3. The intelligent evaluation method for solidification effect of ductile iron based on multi-parameter fusion according to claim 1, characterized in that, The grain growth path refers to the continuous line segment of the grain growth path in space extracted during the solidification process of ductile iron, and the growth trend and directional offset characteristics under the influence of the main heat flow axis are analyzed. The impurity precipitation start point refers to the spatial location point in the grain cooling direction offset region where the impurity first precipitates from the melt. Based on the relationship between the temperature gradient direction at the location and the main axis of grain arrangement, it is determined whether the impurity is located in the position segment where the grain arrangement trend changes.

4. The intelligent evaluation method for solidification effect of ductile iron based on multi-parameter fusion according to claim 1, characterized in that, The misaligned boundary path refers to the set of spatial line segments that detect continuously jumping angle change segments in the grain connection section based on the grain connection angle dataset, track the position sequence of boundary direction abrupt change, and identify the spatial line segments in the grain arrangement structure that show discontinuous direction and multi-directional connection phenomena. The impurity migration trajectory refers to tracing the migration path direction in the grain structure, screening continuous migration segments with consistent orientation, and identifying the spatial migration path morphology of impurities under thermal induction in the non-principal axis region.

5. The intelligent evaluation method for solidification effect of ductile iron based on multi-parameter fusion according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Obtain the grain boundary trajectory and liquid phase region cooling path in the initial interface of graphite sphere crystallization, extract the spatial extension line between the continuous grain propagation direction and the cooling guide, analyze the angle distribution between the grain propagation direction and the heat flow guide, and obtain the grain propagation direction extension relationship; S102: Based on the grain propulsion direction extension relationship, compare the direction of the main axis of heat flow guidance, identify whether the grain propulsion direction deviates from the trend of the main axis of heat flow in a local section, analyze the length change of the angle deflection section and the heat flow direction jump node in the propulsion path, and obtain the grain propulsion direction deviation segment. S103: Based on the grain propulsion direction deviation segment, screen the spatial angle change region within the grain propulsion path, locate the position region where the grain propulsion direction turns from the dominant heat flow direction in the segment, identify the grain boundary direction switching segment, and obtain the grain cooling direction offset segment.

6. The intelligent evaluation method for solidification effect of ductile iron based on multi-parameter fusion according to claim 1, characterized in that, The specific steps of S2 are as follows: S201: Based on the grain cooling direction offset segment, identify the spatial starting point of impurity precipitation, detect the angular difference between the temperature gradient direction of the starting point region and the main axis direction of grain arrangement, determine whether the impurity release position is distributed in the region where the grain arrangement trend changes, and obtain the distribution of impurity precipitation starting position. S202: Based on the distribution of the starting position of the impurity precipitation, track the movement trajectory of the impurity in the grain structure, extract the angle value between the trajectory direction vector and the grain principal axis direction, calculate the change range of the angle difference in the differentiated segment, and locate the continuous segment of the direction change to obtain the direction of the impurity migration offset trajectory. S203: Based on the migration and offset trajectory of the impurities, analyze the path direction of the impurities at the outer edge of the grain structure, determine whether they continue to expand along the non-principal axis region, identify the continuity of movement and similar characteristics of the direction during the outward migration process, and obtain the thermally induced offset path of the impurities.

7. The intelligent evaluation method for solidification effect of ductile iron based on multi-parameter fusion according to claim 1, characterized in that, The specific steps for S3 are as follows: S301: Based on the impurity thermally induced offset path, identify the angular change position of the grain arrangement connection region, obtain the grain principal axis direction vectors on both sides of the connection point, calculate the angle between adjacent principal axis vectors, and obtain the connection direction angle dataset. S302: Based on the connection direction angle dataset, analyze the change of the included angle value in the grain connection path, filter the angle segments where the direction changes, determine whether the direction trend is discontinuous with the adjacent area, and obtain the boundary connection direction jump interval. S303: Based on the boundary connection direction jump interval, detect the grain arrangement path, extract the connection line segments corresponding to the continuous direction change, track the corresponding direction trend in the structural connection path, and obtain the grain boundary misalignment path direction.

8. A multi-parameter integrated intelligent evaluation system for the solidification effect of ductile iron, characterized in that, The system is used to implement the intelligent ductile iron solidification effect evaluation method with multi-parameter fusion as described in any one of claims 1-7, and the system includes: The grain offset recognition module obtains the initial interface of graphite sphere crystallization and the liquid phase cooling path, extracts the grain propulsion direction vector, compares it with the spatial relationship of heat flow guidance, detects whether the grain boundary in the propulsion path has a directional offset, filters the position segments of structural extension and turning, and obtains the grain cooling direction offset segment. The impurity migration extraction module extracts the impurity precipitation start point based on the grain cooling direction offset segment, tracks the direction change in the grain path, identifies the offset trend of impurities in the abrupt arrangement region, locates the position segment that causes offset along the grain boundary, and obtains the impurity thermally induced offset path. Based on the impurity thermally induced offset path, the structural misalignment identification module extracts the position of the angle change in the grain connection region, determines whether there is a direction switch in the grain boundary connection, detects the segments of extension and change in the connection path, identifies the structural segments of grain misalignment, and obtains the direction of the grain boundary misalignment path. The pore extension identification module extracts the extension direction of adjacent pores based on the grain boundary misalignment path, determines whether the pores are connected along the misalignment boundary region, compares the overlap between the path and the grain arrangement direction, extracts the directional features of the pores extending inside the structure, and obtains the pore connectivity offset trend structure. The interference trend assessment module, based on the pore connectivity offset trend structure, compares the spatial positions of grain misalignment, impurity offset and pore penetration path, identifies the intersection relationship between behavioral paths, extracts the regional performance of paths superimposed and adjacently extended in the structure, and obtains the structural interference trend performance of the solidification process.

Images