Method for constructing three-dimensional model of turbocharger shell based on data twinning

Abstract

The invention relates to the technical field of 3D modeling, in particular to a turbocharger shell three-dimensional model construction method based on data twinning, and the method comprises the steps: obtaining a three-dimensional point cloud of a turbocharger shell and temperature data at each point cloud data; determining local structure parameters by analyzing the distribution condition of all the point cloud data in each point cloud data neighborhood and combining the quantity of all the point cloud data in the neighborhood and the size of the neighborhood; and respectively analyzing the average distribution condition and the dispersion degree of the temperature influence factors of all the point cloud data, comparing the differences between the temperature influence factors of all the point cloud data and the average distribution condition, determining temperature abnormal parameters, and combining with the local structure parameters to construct the three-dimensional model of the turbocharger shell. The method aims at improving the construction precision of the three-dimensional model of the turbocharger shell.

Description

Technical Field

[0001] This application relates to the field of 3D modeling technology, and specifically to a method for constructing a three-dimensional model of a turbocharger housing based on digital twin. Background Art

[0002] In the automotive engine industry, turbochargers have the advantages of improving automotive engine exhaust emission pollution, increasing the power and mass ratio of the engine, improving the torque characteristics of the engine, reducing fuel consumption and engine noise, etc., and are the main components of modern automobiles and construction machinery. During the casting process and the subsequent finishing process, defects such as internal pores and external multi-material, lack of material or damage often occur. These defects will affect the quality and service life of the cast parts. Therefore, the defect detection technology for key cast parts of automobiles is particularly important. By establishing an accurate digital model of the turbocharger, the digital twin system can be used to collect the housing data of the turbocharger in real time by means of intelligent remote 3D digital twin display technology. Through remote operation, online detection management can be implemented to realize functions such as real-time transmission, storage, analysis and processing of data, and improve work efficiency and detection accuracy.

[0003] The point cloud data of the turbocharger housing has a complex geometric structure. When using the point cloud data to construct a three-dimensional model for digital twin technology analysis, traditional point cloud clustering methods are often limited when dealing with these complex structures and cannot effectively consider the deformation and thermal expansion effects of the turbocharger under actual working conditions, resulting in inaccurate clustering results and affecting the construction effect of the three-dimensional model of the turbocharger housing. Summary of the Invention

[0004] To solve the above technical problems, this application provides a method for constructing a three-dimensional model of a turbocharger housing based on digital twin to solve the existing problems.

[0005] The method for constructing a three-dimensional model of a turbocharger housing based on digital twin in this application adopts the following technical solutions:

[0006] An embodiment of this application provides a method for constructing a three-dimensional model of a turbocharger housing based on digital twin, and this method includes the following steps:

[0007] S1: Obtain the three-dimensional point cloud of the turbocharger housing and the temperature data at each point cloud data;

[0008] S2: Through analyzing the correlation between the distribution of the three-dimensional point cloud of the turbocharger housing and the temperature data, perform three-dimensional modeling on the turbocharger housing, and the specific process is as follows:

[0009] S201: Determine the local structure parameters of each point cloud data by analyzing the distribution of all point cloud data within the neighborhood of each point cloud data, and combining the quantity of all point cloud data in the neighborhood and the size of the neighborhood.

[0010] S202: Measure the distances between all point cloud data of the turbocharger housing, and combine the local structure parameters of all point cloud data to cluster all point cloud data, obtaining all clustering clusters of the turbocharger housing.

[0011] S203: Analyze the differences in distances and temperature data between each point cloud data and all its adjacent point cloud data, and determine the first temperature difference coefficient of each point cloud data; determine the second temperature difference coefficient of each point cloud data according to the differences in temperature data between each point cloud data and all point cloud data within its neighborhood, and combine the first temperature difference coefficient to determine the temperature influence factor of each point cloud data.

[0012] S204: In each clustering cluster, respectively analyze the average distribution and dispersion degree of the temperature influence factors of all point cloud data, and compare the differences between the temperature influence factors of all point cloud data and the average distribution to determine the temperature anomaly parameters of each clustering cluster.

[0013] S205: Measure the distances from all point cloud data in each clustering cluster to the cluster center, combine the local structure parameters to determine the structure index of each clustering cluster, and combine the temperature anomaly parameters to determine the structure change degree of each clustering cluster.

[0014] S206: Based on the structure change degrees of all clustering clusters, construct a three-dimensional model of the turbocharger housing.

[0015] Preferably, the expression of the local structure parameter of each point cloud data is: In the formula, z i represents the local structure parameter of the i-th point cloud data; λ i represents the absolute value of the curvature of the i-th point cloud data; n i represents the quantity of all point cloud data within the neighborhood of the i-th point cloud data; r represents the radius of the spherical neighborhood constructed with the i-th point cloud data as the center.

[0016] Preferably, the method for obtaining all clustering clusters of the turbocharger housing is:

[0017] Calculate the mean value of the local structure parameters of any two point cloud data, and take the ratio of the distance between any two point cloud data to the mean value as the weighted distance between any two point cloud data.

[0018] Cluster all the point cloud data in the three-dimensional point cloud of the turbocharger housing. Among them, the weighted distance between all the point cloud data is used as the metric distance in the clustering algorithm, and the radius of the preset spherical neighborhood is used as the neighborhood radius in the clustering algorithm to obtain all the clustering clusters.

[0019] Preferably, the method for determining the first temperature difference coefficient of each point cloud data is as follows:

[0020] The difference in temperature data between the i-th point cloud data and its adjacent u-th point cloud data is divided by the distance between the i-th point cloud data and its adjacent u-th point cloud data as the temperature difference value between the i-th point cloud data and its adjacent u-th point cloud data;

[0021] The average value of the temperature difference values between each point cloud data and all its adjacent point cloud data is used as the first temperature difference coefficient of each point cloud data.

[0022] Preferably, the second temperature difference coefficient of each point cloud data is the cumulative result of the difference in temperature data between each point cloud data and all the point cloud data within its neighborhood.

[0023] Preferably, the temperature influence factor of each point cloud data is the normalized value of the product of the first temperature coefficient and the second temperature coefficient of each point cloud data.

[0024] Preferably, the expression for the temperature anomaly parameter of each clustering cluster is: In the formula, δ j represents the temperature anomaly parameter of the j-th clustering cluster; a j,i represents the temperature influence factor of the i-th point cloud data in the j-th clustering cluster; represents the average value of the temperature influence factors of all the point cloud data in the j-th clustering cluster; σ j represents the standard deviation of the temperature influence factors of all the point cloud data in the j-th clustering cluster; m j represents the number of all the point cloud data in the j-th clustering cluster; ε represents a preset constant greater than 0.

[0025] Preferably, the expression for the structure index of each clustering cluster is: In the formula, D j represents the structure index of the j-th clustering cluster; C j represents the average value of the local structure parameters of all the point cloud data in the j-th clustering cluster; p j represents the degree of dispersion of the distances from all the point cloud data in the j-th clustering cluster to the clustering center; represents a preset constant greater than 0.

[0026] Preferably, the method for determining the structural change degree of each clustering cluster is:

[0027] For the turbocharger housing, calculate the mean value of the temperature anomaly parameters of all clustering clusters, denoted as the anomaly mean value, and use the ratio of the temperature anomaly parameters of each clustering cluster to the anomaly mean value as the temperature anomaly value of each clustering cluster;

[0028] Use the ratio of the temperature anomaly value of each clustering cluster to the structure index as the structural change degree of each clustering cluster.

[0029] Preferably, the process of constructing the three-dimensional model of the turbocharger housing is as follows:

[0030] Use the structural change degrees of all clustering clusters of the turbocharger housing as the input of the threshold segmentation algorithm, output the segmentation threshold, and use the clustering clusters with structural change degrees greater than the segmentation threshold as the set to be processed;

[0031] In each set to be processed, calculate the product of the mean value of the temperature influence factors of any two point cloud data and the distance between the corresponding two point cloud data as the weighted distance between any two point cloud data. Use all the point cloud data in each set to be processed as the input of the clustering algorithm. Among them, use the weighted distance between all the point cloud data as the metric distance in the clustering algorithm, output all clustering clusters, denoted as segmented clustering clusters, and denote all the segmented clustering clusters in all sets to be processed and all clustering clusters except the clustering clusters to be processed as the final clustering clusters;

[0032] Extract the edge curves in each final clustering cluster, use all the edge curves in each final clustering cluster as the input of the surface construction algorithm, output the surfaces corresponding to all the point cloud data in each clustering cluster, and merge all the surfaces to obtain the three-dimensional model of the turbocharger housing.

[0033] This application has at least the following beneficial effects:

[0034] By analyzing the distribution characteristics of the point cloud data in the local area of the turbocharger housing, this application constructs local structure parameters, which reflect the density and complexity of the point cloud data in the local area of the turbocharger housing, helps to more accurately describe the geometric characteristics of the point cloud data, can better identify points with similar joint structures and cluster them together, thereby improving the accuracy of three-dimensional model construction; further, by analyzing the influence of temperature data on the point cloud distribution, this application constructs temperature anomaly parameters, which helps to identify abnormal point cloud data in the clustering clusters, thereby improving the accuracy of the clustering results; further, according to the local structure parameters and temperature anomaly parameters, this application constructs the structural change degree, which helps to identify the structural anomalies of the clustering clusters, thereby improving the accuracy of the clustering results and further improving the accuracy of three-dimensional modeling. By analyzing the influence of temperature on the point cloud distribution, this application optimizes the clustering method, improves the accuracy of the clustering results, and further improves the accuracy of the three-dimensional model construction of the turbocharger housing. Description of the Drawings

[0035] To more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0036] Figure 1 It is a flowchart of the steps of a method for constructing a three-dimensional model of a turbocharger housing based on digital twin provided by an embodiment of the present application;

[0037] Figure 2 It is a schematic diagram of the process of extracting the degree of structural change provided by an embodiment of the present application;

[0038] Figure 3 It is a schematic diagram of the process of constructing a three-dimensional model provided by an embodiment of the present application. Detailed implementation manners

[0039] To further elaborate on the technical means and effects adopted by the present application to achieve the intended invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, details the specific implementation manners, structures, features, and effects of the method for constructing a three-dimensional model of a turbocharger housing based on digital twin proposed according to the present application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0040] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present application belongs.

[0041] The following specifically describes the specific solution of the method for constructing a three-dimensional model of a turbocharger housing based on digital twin provided by the present application with reference to the accompanying drawings.

[0042] A method for constructing a three-dimensional model of a turbocharger housing based on digital twin provided by an embodiment of the present application. Specifically, the following method for constructing a three-dimensional model of a turbocharger housing based on digital twin is provided. Please refer to Figure 1 and this method includes the following steps:

[0043] S1: Obtain the three-dimensional point cloud of the turbocharger housing and the temperature data at each point cloud data.

[0044] Digital twin refers to creating a virtual copy of a physical entity, which can reflect the state, performance, and behavior of the object. In this embodiment, 3D point cloud of the turbocharger housing is obtained by using laser scanning technology. The 3D point cloud not only contains the geometric shape information of the surface of the turbocharger housing, but also contains the structural features. Meanwhile, combined with the high-sensitivity thermal imager technology, the temperature data corresponding to each point cloud data of the turbocharger housing is synchronously obtained.

[0045] Step S2: Perform 3D modeling on the turbocharger housing by analyzing the correlation between the distribution of the 3D point cloud of the turbocharger housing and the temperature data.

[0046] The point cloud data can accurately reflect the spatial position data of the turbocharger housing. However, when using the point cloud data to construct a 3D model, with the change of production conditions during the production process, the state of the generated turbocharger housing will also change accordingly. These changes are manifested as tiny geometric deformations in the point cloud data. Directly constructing a 3D model based on the point cloud data cannot identify these tiny regions. By performing feature extraction and preliminary classification on the original data, and combining the temperature data at the point cloud data during the production process, establishing data association, and reconstructing the clustering results, real-time sensor data is embedded in the 3D model to generate a digital twin model. Therefore, the specific process of constructing the 3D model of the turbocharger housing in this embodiment is as follows:

[0047] S201: Determine the local structure parameters of each point cloud data by analyzing the distribution of all point cloud data within the neighborhood of each point cloud data, and combining the quantity of all point cloud data in the neighborhood and the size of the neighborhood.

[0048] Since the structure of the turbocharger housing is complex and there may be subtle defects on the surface of the housing, in order to accurately describe the geometric features of the point cloud data in the local area of the turbocharger housing, so as to better construct the 3D model of the turbocharger housing and improve the construction accuracy of the 3D model, the local structure parameters of each point cloud data are constructed by analyzing the distribution characteristics of the local point cloud data. Specifically:

[0049] Neighborhoods are divided with each point cloud data as the center. In this embodiment, the neighborhoods of each point cloud data are spherical neighborhoods, and the radius is set to 7. Regarding the setting of the radius size, the implementer can also set it according to the specific situation by himself / herself, and this embodiment does not make special restrictions.

[0050] Furthermore, based on the curvature of the 3D point cloud where the point cloud data is located, and analyzing the size of the neighborhood of the point cloud data and the quantity of the point cloud data in the neighborhood, the local structure parameters of each point cloud data are constructed. Specifically:

[0051] The local structure parameter z of the i-th point cloud data i The expression is: In the formula, λ i represents the absolute value of the curvature of the i-th point cloud data; n i represents the number of all point cloud data within the neighborhood of the i-th point cloud data; r represents the radius of the spherical neighborhood constructed with the i-th point cloud data as the center.

[0052] Among them, in this embodiment, the normal vector estimation method is used to calculate the curvature at each point cloud data. The normal vector estimation method is a well-known technology, and the specific process of calculating the curvature of the point cloud data will not be elaborated here.

[0053] It can be understood from the local structure parameters of each point cloud data that if the ratio of the number of all point cloud data within the neighborhood of the point cloud data to the neighborhood size is larger, it indicates that the point cloud data is more densely distributed within the neighborhood of the point cloud data. And when the curvature is higher, it indicates that the neighborhood of the point cloud data corresponds to a more complex part of the structure such as the protrusion or depression of the turbocharger housing. Therefore, the obtained local structure parameter is larger; on the contrary, if the ratio of the number of all point cloud data within the neighborhood of the point cloud data to the neighborhood size is smaller, it indicates that the point cloud data is more dispersed within the neighborhood of the point cloud data. And when the curvature is lower, it indicates that the neighborhood of the point cloud data corresponds to the flat area of the turbocharger housing. Therefore, the obtained local structure parameter is smaller.

[0054] S202: Measure the distances between all point cloud data of the turbocharger housing, and combine the local structure parameters of all point cloud data to cluster all point cloud data, so as to obtain all clustering clusters of the turbocharger housing.

[0055] According to the local structure parameters obtained in S201, the point cloud data can be preliminarily clustered. The specific clustering process is as follows:

[0056] In the three-dimensional point cloud of the turbocharger, calculate the mean value of the local structure parameters of any two point cloud data, and use the ratio of the distance between any two point cloud data to the mean value as the weighted distance between any two point cloud data;

[0057] Furthermore, cluster all point cloud data in the three-dimensional point cloud of the turbocharger housing. Among them, use the weighted distance between all point cloud data as the metric distance in the clustering algorithm, and use the radius of the preset spherical neighborhood as the neighborhood radius in the clustering algorithm to obtain all clustering clusters. It should be noted that there are many common clustering algorithms. In this embodiment, the DBSCAN clustering algorithm is used. In the actual application process, as other implementation manners, the implementer can also use other clustering methods such as the DPC density peak clustering. There is no special limitation on the selection of the clustering algorithm in this embodiment.

[0058] Among them, the DBSCAN clustering algorithm is a well-known technology, and its specific clustering principle will not be elaborated here.

[0059] S203: Analyze the differences in distance and temperature data between each point cloud data and all its adjacent point cloud data, and determine the first temperature difference coefficient of each point cloud data; determine the second temperature difference coefficient of each point cloud data according to the differences in temperature data between each point cloud data and all point cloud data within its neighborhood, and combine the first temperature difference coefficient to determine the temperature influence factor of each point cloud data.

[0060] The performance of the turbocharger housing and the morphological performance of the point cloud are closely related to the surface temperature of the turbocharger housing. The turbocharger housing will undergo thermal expansion during operation, and the temperature distribution on different parts of the turbocharger housing is uneven. Some areas may expand and deform due to thermal load, resulting in changes in the positions in the point cloud data. This thermal deformation effect is very important for the geometric reconstruction of the point cloud data. Therefore, it is necessary to dynamically adjust the clustering process of DBSCAN by combining temperature. By combining temperature data with point cloud data, compensate or correct the geometric changes caused by temperature to obtain a more accurate three-dimensional model. The specific process is as follows:

[0061] Take the ratio of the difference in temperature data between the i-th point cloud data and its adjacent u-th point cloud data to the distance between the i-th point cloud data and its adjacent u-th point cloud data as the temperature difference value between the i-th point cloud data and its adjacent u-th point cloud data;

[0062] Furthermore, take the average value of the temperature difference values between each point cloud data and all its adjacent point cloud data as the first temperature difference coefficient of each point cloud data;

[0063] Furthermore, take the accumulated result of the differences in temperature data between each point cloud data and all point cloud data within its neighborhood as the second temperature difference coefficient of each point cloud data;

[0064] The temperature influence factor of each point cloud data is the normalized value of the product of the first temperature coefficient and the second temperature coefficient of each point cloud data.

[0065] It can be understood from the temperature influence factor of each point cloud data that if the first temperature coefficient of the point cloud data is larger and the second temperature coefficient is larger, it means that the temperature difference between the point cloud data and its adjacent point cloud data is larger, and the temperature influence factor is larger, indicating that the temperature has a greater impact on the construction effect of the three-dimensional model; conversely, if the first temperature coefficient of the point cloud data is smaller and the second temperature coefficient is smaller, it means that the temperature difference between the point cloud data and its adjacent point cloud data is smaller, and the temperature influence factor is smaller, indicating that the temperature has a smaller impact on the construction of the three-dimensional model.

[0066] S204: In each clustering cluster, analyze the average distribution and dispersion degree of the temperature influence factors of all point cloud data respectively, compare the differences between the temperature influence factors of all point cloud data and the average distribution, and determine the temperature anomaly parameters of each clustering cluster.

[0067] Based on S203, the temperature influence factors of the point cloud data are obtained. The temperature anomaly parameters of the clustering clusters are determined according to the temperature influence factors, specifically as follows:

[0068] The temperature anomaly parameter δ of the j-th clustering cluster j has the following expression: In the formula, a j,i represents the temperature influence factor of the i-th point cloud data in the j-th clustering cluster; represents the mean value of the temperature influence factors of all point cloud data in the j-th clustering cluster; σ j represents the standard deviation of the temperature influence factors of all point cloud data in the j-th clustering cluster; m j represents the number of all point cloud data in the j-th clustering cluster; ε represents a preset constant greater than 0, which is used to prevent the denominator from being 0. In this embodiment, the value of ε is 0.01. On the premise of ensuring that the denominator is not 0 and does not overly affect the calculation result, the implementer can also set it according to the specific situation by himself / herself, and this embodiment does not make special restrictions.

[0069] It can be understood from the temperature anomaly parameters of each clustering cluster that under normal circumstances, the temperatures in the local areas of the turbocharger housing should be similar. Therefore, the differences between the temperatures of the point cloud data in each clustering cluster should also be smaller. If the ratio of the difference between the temperature influence factor of the current point cloud data and the mean value of the temperature influence factors in its current clustering cluster to the standard deviation is larger, it means that the current point cloud data is less likely to belong to its current clustering cluster, the larger the temperature anomaly parameter of the clustering cluster, the more necessary it is to re-cluster the point cloud data in the current clustering cluster;

[0070] On the contrary, if the ratio of the difference between the temperature influence factor of the current point cloud data and the mean value of the temperature influence factors in its current clustering cluster to the standard deviation is smaller, it means that the current point cloud data belongs to its current clustering cluster, the smaller the temperature anomaly parameter of the clustering cluster, and the more accurate the current clustering result.

[0071] S205: Measure the distances from all point cloud data in each clustering cluster to the clustering center, combine the local structure parameters to determine the structure index of each clustering cluster, and combine the temperature anomaly parameters to determine the structure change degree of each clustering cluster.

[0072] The structure index D of the j-th clustering cluster j has the following expression: In the formula, C jrepresents the mean of the local structure parameters of all point cloud data in the j-th cluster; p j represents the degree of dispersion of the distances from all point cloud data in the j-th cluster to the cluster center; represents a preset constant greater than 0, used to prevent the denominator from being 0. In this embodiment, the value of is 0.01. On the premise of ensuring that the denominator is not 0 and not overly affecting the calculation results, the implementer can also set it according to the specific situation by himself / herself. This embodiment does not make special restrictions.

[0073] It should be noted that there are many methods to measure the degree of dispersion of a set of data. In this embodiment, the variance of the distances from all point cloud data in the j-th cluster to the cluster center is used as the degree of dispersion of the distances from all point cloud data in the j-th cluster to the cluster center; in the actual application process, as other implementation manners, the implementer can also adopt other methods to measure the degree of dispersion of data, such as the standard deviation or the coefficient of variation. Regarding the selection of the method to measure the degree of dispersion of data, this embodiment does not make special restrictions.

[0074] It can be understood from the structure indices of each cluster that if the variance of the distances from the point cloud data in the cluster to the cluster center is smaller and the mean of the local structure parameters of the point cloud data is larger, the structure index of the cluster is larger, indicating that the point cloud data in the current cluster is more likely to belong to the same cluster, that is, the clustering effect of the current cluster is better; on the contrary, if the variance of the distances from the point cloud data in the cluster to the cluster center is larger and the mean of the local structure parameters of the point cloud data is smaller, the structure index of the cluster is smaller, indicating that the point cloud data in the current cluster is less likely to belong to the same cluster, that is, the clustering effect of the current cluster is poor, and it is necessary to further divide and cluster the point cloud data in the cluster.

[0075] Further, for the turbocharger housing, calculate the mean of the temperature anomaly parameters of all clusters, denoted as the anomaly mean, and use the ratio of the temperature anomaly parameter of each cluster to the anomaly mean as the temperature anomaly value of each cluster;

[0076] Use the ratio of the temperature anomaly value and the structure index of each cluster as the structure change degree of each cluster.

[0077] It can be understood from the structure change degrees of each cluster that if the temperature anomaly value of the current cluster is larger and the structure index is smaller, the obtained structure change degree is larger, indicating that the point cloud data in the current cluster is less likely to belong to the same cluster, that is, the clustering effect of the current cluster is poor, and it is necessary to further divide and cluster the point cloud data in the cluster; on the contrary, if the temperature anomaly value of the current cluster is smaller and the structure index is larger, the obtained structure change degree is smaller, indicating that the point cloud data in the current cluster is more likely to belong to the same cluster, that is, the clustering effect of the current cluster is better.

[0078] Preferably, the schematic diagram of the structural change degree extraction process provided in this embodiment is as shown in Figure 2 the following figure.

[0079] S206: Construct a three-dimensional model of the turbocharger housing based on the structural change degrees of all clustering clusters.

[0080] Use the structural change degrees of all clustering clusters of the turbocharger housing as the input of the threshold segmentation algorithm, output the segmentation threshold, and use the clustering clusters with structural change degrees greater than the segmentation threshold as the set to be processed;

[0081] It should be noted that there are many common threshold segmentation algorithms. In this embodiment, the Otsu threshold segmentation algorithm is used. In the actual application process, as other implementation manners, implementers can also use other threshold segmentation methods, and this embodiment does not make special restrictions.

[0082] Among them, the Otsu threshold segmentation algorithm is a well-known technology, and its specific principle will not be elaborated here.

[0083] Furthermore, in each set to be processed, calculate the product of the mean value of the temperature influence factors of any two point cloud data and the distance between the corresponding two point cloud data as the weighted distance between any two point cloud data. Use all the point cloud data in each set to be processed as the input of the clustering algorithm. Among them, use the weighted distance between all the point cloud data as the metric distance in the clustering algorithm, output all clustering clusters, and denote them as segmentation clustering clusters. Denote all the segmentation clustering clusters in all sets to be processed and all clustering clusters except the clustering clusters to be processed as the final clustering clusters;

[0084] Furthermore, use the edge detection method based on the normal vector to extract the edge curves in each final clustering cluster. Use all the edge curves in each final clustering cluster as the input of the surface construction algorithm, output the surfaces corresponding to all the point cloud data in each clustering cluster, and merge all the surfaces to obtain the three-dimensional model of the turbocharger housing.

[0085] It should be noted that in this embodiment, the nurbs surface fitting method is used to construct the surface for the edge curves, and the Boolean operation is used to merge all the surfaces. Implementers can also use other surface construction methods and surface merging methods, and this embodiment does not make special restrictions.

[0086] Among them, the edge detection method based on the normal vector, the nurbs surface fitting method, and the Boolean operation are all well-known technologies, and their specific principles will not be elaborated here.

[0087] Preferably, the schematic diagram of the three-dimensional model construction process provided in this embodiment is as shown in Figure 3 the following figure.

[0088] It should be noted that the above-mentioned sequence of embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. In addition, the specific embodiments of this specification have been described. Further, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0089] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments.

[0090] The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Modifying the technical solutions recorded in the foregoing embodiments, or equivalently replacing some of the technical features, does not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of each embodiment of the present application, and should all be included within the protection scope of the present application.

Claims

1. A method for constructing a three-dimensional model of a turbocharger housing based on data twinning, characterized in that: The method comprises the following steps: S1: Acquire the three-dimensional point cloud of the turbocharger housing and the temperature data at each point cloud data; S2: By analyzing the correlation between the distribution of the 3D point cloud of the turbocharger housing and the temperature data, the 3D modeling of the turbocharger housing is performed. The specific process is as follows: S201: determining the local structure parameters of each point cloud data by analyzing the distribution of all point cloud data in the neighborhood of each point cloud data and combining the number of all point cloud data in the neighborhood and the size of the neighborhood; S202: measuring the distances between all point cloud data of the turbocharger housing, and combining the local structural parameters of all point cloud data to cluster all point cloud data to obtain all clusters of the turbocharger housing; S203: Analyze the distance and temperature data difference between each point cloud data and all adjacent point cloud data to determine the first temperature difference coefficient of each point cloud data; determine the second temperature difference coefficient of each point cloud data according to the temperature difference between each point cloud data and all point cloud data in its neighborhood, and determine the temperature influence factor of each point cloud data in combination with the first temperature difference coefficient; S204: In each cluster, respectively analyzing the average distribution and the discreteness of the temperature influencing factors of all point cloud data, and comparing the difference between the temperature influencing factors of all point cloud data and the average distribution, to determine the temperature anomaly parameters of each cluster; S205: measuring the distances from all point cloud data in each cluster to the cluster center, determining the structure index of each cluster in combination with the local structure parameter, and determining the structure change degree of each cluster in combination with the temperature anomaly parameter; S206: Constructing a three-dimensional model of the turbocharger housing based on the structural variation of all clusters.

2. The method for constructing a three-dimensional model of a turbocharger housing based on data twinning according to claim 1, characterized in that: The expression of the local structure parameter of each point cloud data is: In the formula, z i Represents the local structure parameter of the i-th point cloud data; i Represents the absolute value of the curvature of the i-th point cloud data; n i represents the number of all point cloud data in the neighborhood of the i-th point cloud data; r represents the radius of the spherical neighborhood constructed with the i-th point cloud data as the center.

3. The method for constructing a three-dimensional model of a turbocharger housing based on data twinning according to claim 2, characterized in that: The method for obtaining all clusters of the turbocharger housing is as follows: Calculate the mean of the local structural parameters of any two point cloud data, and use the ratio of the distance between any two point cloud data and the mean as the weighted distance between any two point cloud data; All point cloud data in the three-dimensional point cloud of the turbocharger housing are clustered, wherein the weighted distance between all point cloud data is used as the metric distance in the clustering algorithm, and the radius of the preset spherical neighborhood is used as the neighborhood radius in the clustering algorithm to obtain all cluster clusters.

4. The method for constructing a three-dimensional model of a turbocharger housing based on data twinning according to claim 1, characterized in that: The method for determining the first temperature difference coefficient of each point cloud data is: The difference in temperature data between the i-th point cloud data and its adjacent u-th point cloud data is divided by the distance between the i-th point cloud data and its adjacent u-th point cloud data, as the temperature difference between the i-th point cloud data and its adjacent u-th point cloud data; The temperature difference between each point cloud data and all its adjacent point cloud data is averaged as the first temperature difference coefficient of each point cloud data.

5. The method for constructing a three-dimensional model of a turbocharger housing based on data twinning according to claim 1, characterized in that: The second temperature difference coefficient of each point cloud data is the cumulative result of the difference in temperature data between each point cloud data and all point cloud data in its neighborhood.

6. The method for constructing a three-dimensional model of a turbocharger housing based on data twinning according to claim 1, characterized in that: The temperature influence factor of each point cloud data is a normalized value of a product of a first temperature coefficient and a second temperature coefficient of each point cloud data.

7. The method for constructing a three-dimensional model of a turbocharger housing based on data twinning according to claim 1, characterized in that: The expression of the temperature anomaly parameter of each cluster is: In the formula, δ j represents the temperature anomaly parameter of the jth cluster; a j,i Represents the temperature influence factor of the i-th point cloud data in the j-th cluster; represents the mean value of the temperature influence factor of all point cloud data in the jth cluster; σ j represents the standard deviation of the temperature influence factor of all point cloud data in the jth cluster; m j Represents the number of all point cloud data in the jth cluster; ε represents a constant preset to be greater than 0.

8. The method for constructing a three-dimensional model of a turbocharger housing based on data twinning according to claim 1, characterized in that: The expression of the structural index of each cluster is: Where D j represents the structural index of the jth cluster; C j represents the mean of the local structure parameters of all point cloud data in the jth cluster; p j Indicates the degree of dispersion of the distance from all point cloud data in the jth cluster to the cluster center; Indicates a preset constant greater than 0.

9. The method for constructing a three-dimensional model of a turbocharger housing based on data twinning according to claim 1, characterized in that: The method for determining the structural variation degree of each cluster is as follows: For the turbocharger housing, the mean of the temperature anomaly parameters of all clusters is calculated and recorded as the anomaly mean, and the ratio of the temperature anomaly parameter of each cluster to the anomaly mean is used as the temperature anomaly value of each cluster; The ratio of the temperature anomaly value and the structural index of each cluster is taken as the structural change degree of each cluster.

10. The method for constructing a three-dimensional model of a turbocharger housing based on data twinning according to claim 1, characterized in that: The process of constructing the three-dimensional model of the turbocharger housing is as follows: The structural variation of all clusters of the turbocharger housing is used as the input of the threshold segmentation algorithm, the segmentation threshold is output, and the clusters with a structural variation greater than the segmentation threshold are used as the set to be processed; In each set to be processed, the product of the mean of the temperature influence factor of any two point cloud data and the distance between the corresponding two point cloud data is calculated as the weight distance between any two point cloud data, and all the point cloud data in each set to be processed are used as the input of the clustering algorithm, wherein the weight distance between all point cloud data is used as the metric distance in the clustering algorithm, and all cluster clusters are output and recorded as segmentation cluster clusters, and all segmentation cluster clusters in all sets to be processed and all cluster clusters except the cluster cluster to be processed are recorded as final cluster clusters; The edge curves in each final cluster are extracted, and all the edge curves in each final cluster are used as the input of the surface construction algorithm. The surfaces corresponding to all the point cloud data in each cluster are output, and all the surfaces are merged to obtain the three-dimensional model of the turbocharger housing.

Images