Inspection area division method and system for crack type damage of turbine blade

Abstract

The invention provides a turbine blade crack type damage inspection area division method and system, and relates to the technical field of turbine blade detection.The method comprises the steps that multiple data points are extracted from the blade surface of a turbine blade through finite element analysis, and a data set is formed; clustering each data point in the data set by using a second-order clustering algorithm, and dividing the blade surface into a plurality of preliminary inspection areas; extracting boundary points corresponding to each preliminary inspection area; through a least square method, curve fitting is carried out on the boundary points corresponding to all the preliminary inspection areas, and smooth boundary curves corresponding to all the preliminary inspection areas are determined; according to the smooth boundary curves corresponding to the preliminary inspection areas, the fine inspection areas of the crack type damage of the turbine blade are divided, curve fitting is conducted through the boundary points of the inspection areas, the corresponding smooth boundary curves are determined, the geometrical characteristics of the area boundaries are accurately captured, and the boundary division accuracy is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of turbine blade detection, and particularly to a method and system for dividing inspection areas of crack-type damage of turbine blades. Background Art

[0002] As the core of modern aircraft, the performance of aero-engines directly affects the safety and efficiency of aircraft. In aero-engines, turbine blades, as one of the key components, bear the important task of converting the high-temperature and high-pressure gas flow generated by fuel combustion into mechanical energy. Turbine blades are in a harsh environment of high temperature, high pressure and high-speed operation for a long time, and their damage tolerance design is of great significance for ensuring the reliability of the engine and flight safety. However, due to the uneven stress distribution of turbine blades under complex working conditions, there are significant differences in the damage tolerance of different regions. Therefore, adopting a scientific and reasonable regional management strategy to manage turbine blades by zoning has become a key technical path to improve their safety and service life.

[0003] Currently, in the field of damage zoning research, scholars have carried out a large number of studies on the damage characteristics and life prediction of various key components through zoning methods. For example, for the head of an ultra-high pressure vessel, stress zoning is carried out based on the actual service state to study the fatigue propagation characteristics of crack-type damage in different stress regions, and zoning is carried out according to the yield strength differences of different parts of the engine cylinder head to predict the small crack propagation life of each region. The zoning method has shown important application value in many fields and provided theoretical support for improving the damage management level of components.

[0004] However, traditional zoning methods usually rely on experience or simplified models in the division of regional boundaries, resulting in ambiguity in boundary division and affecting the accuracy of zoning management; for the differential damage characteristics of turbine blades under complex working conditions, no effective zoning optimization and management scheme has been formed, making it difficult to meet the high requirements of modern aero-engines for safety and economy. Summary of the Invention

[0005] In order to solve the technical problems that traditional zoning methods usually rely on experience or simplified models in the division of regional boundaries, resulting in ambiguity in boundary division and affecting the accuracy of zoning management; for the differential damage characteristics of turbine blades under complex working conditions, no effective zoning optimization and management scheme has been formed, making it difficult to meet the high requirements of modern aero-engines for safety and economy, the present invention provides a method and system for dividing inspection areas of crack-type damage of turbine blades.

[0006] The technical solutions provided by the embodiments of the present invention are as follows:

[0007] In the first aspect:

[0008] A method for dividing inspection regions of crack-type damage of a turbine blade provided by an embodiment of the present invention includes:

[0009] S1: Through finite element analysis, extract a plurality of data points from the blade surface of the turbine blade to form a data set, where the data set includes a plurality of index variables related to the working state of the blade;

[0010] S2: Use the second-order clustering algorithm to cluster each data point in the data set, and divide the blade surface of the turbine blade into a plurality of preliminary inspection regions;

[0011] S3: Use the Alpha-Shapes algorithm to extract boundary points corresponding to each preliminary inspection region;

[0012] S4: Through the least squares method, perform curve fitting on the boundary points corresponding to each preliminary inspection region respectively to determine the smooth boundary curve corresponding to each preliminary inspection region;

[0013] S5: According to the smooth boundary curves corresponding to each preliminary inspection region, divide the fine inspection regions of crack-type damage of the turbine blade.

[0014] Second aspect:

[0015] A system for dividing inspection regions of crack-type damage of a turbine blade provided by an embodiment of the present invention includes:

[0016] A processor;

[0017] A memory, on which computer-readable instructions are stored, and when the computer-readable instructions are executed by the processor, the method for dividing inspection regions of crack-type damage of the turbine blade as described in the first aspect is implemented.

[0018] Third aspect:

[0019] A computer-readable storage medium provided by an embodiment of the present invention, on which a computer program is stored, and when the program is executed by a processor, the method for dividing inspection regions of crack-type damage of the turbine blade as described in the first aspect is implemented.

[0020] The beneficial effects brought by the technical solution provided by the embodiment of the present invention at least include:

[0021] (1) In the present invention, the Alpha-Shapes algorithm is used to extract boundary points corresponding to each preliminary inspection region, and through the least squares method, curve fitting is performed on the boundary points corresponding to each preliminary inspection region respectively to determine the smooth boundary curve corresponding to each preliminary inspection region, accurately capturing the geometric features of the region boundary and effectively improving the accuracy of boundary division.

[0022] (2) In the present invention, data points are extracted through finite element analysis, the second-order clustering algorithm is used to achieve differential partitioning of the preliminary inspection area, the Alpha-Shapes algorithm and the least squares method are adopted to determine the smooth boundary curve of the partitioned area, and the crack-type damage area is finely partitioned in combination with the smooth boundary curve, thereby constructing a systematic partition optimization and management scheme, which effectively meets the high requirements of modern aero-engines for safety and economy. Brief Description of the Drawings

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0024] Figure 1 It is a schematic flow chart of a method for partitioning the inspection area of crack-type damage of a turbine blade provided by an embodiment of the present invention;

[0025] Figure 2 It is a schematic structural diagram of a system for partitioning the inspection area of crack-type damage of a turbine blade provided by an embodiment of the present invention. Detailed Embodiments

[0026] The following will describe the technical solutions in the present invention in conjunction with the drawings.

[0027] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "example" in the present invention should not be construed as being more preferred or more advantageous than other embodiments or design solutions. Exactly speaking, the use of the word "example" is intended to present concepts in a specific manner. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one of the two can be selected.

[0028] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when their differences are not emphasized, the meanings they express are the same. "(of)", "corresponding" and "corresponding" can sometimes be used interchangeably. It should be noted that when their differences are not emphasized, the meanings they express are the same.

[0029] In the embodiments of the present invention, sometimes subscripts such as W1 may be written in a non-subscript form such as W1. When their differences are not emphasized, the meanings they express are the same.

[0030] To make the technical problems, technical solutions, and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0031] Refer to the attached Figure 1 , which shows a schematic flowchart of a method for dividing inspection areas of crack-type damage of turbine blades provided by an embodiment of the present invention.

[0032] An embodiment of the present invention provides a method for dividing inspection areas of crack-type damage of turbine blades. This method can be implemented by an inspection area division device for crack-type damage of turbine blades, and the inspection area division device for crack-type damage of turbine blades can be a terminal or a server. The processing flow of the method for dividing inspection areas of crack-type damage of turbine blades can include the following steps:

[0033] S1: Through finite element analysis, extract multiple data points from the blade surface of the turbine blade to form a data set, where the data set includes multiple index variables related to the working state of the blade.

[0034] Among them, finite element analysis (Finite Element Analysis, FEA) is a numerical calculation method used to solve physical field and structural problems in complex engineering problems. It divides a continuum (such as solids, fluids, electromagnetic fields, etc.) into a finite number of small elements, analyzes the behavior of each element, and then combines the overall behavior to solve the problem.

[0035] In a possible implementation manner, the data points are specifically:

[0036] P i ={(x i ,y i ,z i ),σ i ,T i ,…}

[0037] Among them, P i represents a data set composed of i data points, (x i ,y i ,z i ) represents the three-dimensional space coordinates of the i-th data point, σ i represents the stress value of the i-th data point, and T i represents the temperature value of the i-th data point.

[0038] Specifically, in order to achieve area division, a large number of data points are extracted from the blade surface as sample points through the finite element method. These sample points not only record the three-dimensional space coordinates on the blade surface but also contain multiple key indicators related to the working state.

[0039] In the present invention, the data points not only contain spatial coordinates but also incorporate key indicators related to the working state of the blade, which can comprehensively reflect the force and heat distribution characteristics of the blade under complex working conditions and provide a scientific basis for subsequent regional division.

[0040] It should be noted that before performing the regional division, it is necessary to conduct a correlation test on the key influencing indicators and use the Pearson correlation coefficient to eliminate highly correlated indicators to avoid interfering with the analysis results.

[0041] In a possible implementation manner, the index variables include stress and temperature; after S1, it further includes:

[0042] Calculate the Pearson correlation coefficient values between each index variable, and retain any one of the index variables in multiple groups of index variable pairs with Pearson correlation coefficient values greater than the preset Pearson correlation coefficient value.

[0043] Among them, the Pearson Correlation Coefficient (PCC) is a statistical index used to measure the linear correlation degree between two variables, and its value ranges from -1 to 1. This coefficient was proposed by the British statistician Karl Pearson and is widely used in the data analysis fields of natural science and social science.

[0044] In a possible implementation manner, the calculation formula for the Pearson correlation coefficient value is:

[0045]

[0046] Among them, r represents the Pearson correlation coefficient value, n represents the total number of samples, x i represents the i-th observation value of variable x, y i represents the i-th observation value of variable y, represents the mean value of variable x, represents the mean value of variable y.

[0047] In the present invention, by calculating the Pearson correlation coefficient, the redundancy between index variables can be identified. For variables with high correlation, only one of them is retained, thereby reducing the data dimension, lowering the computational complexity, and enhancing the efficiency of subsequent clustering and analysis. At the same time, the redundant terms in strongly correlated variables are removed, avoiding the influence of multicollinearity on the model, making the subsequent clustering algorithm more stable, and the partitioning result more representative and scientific.

[0048] S2: Use the second-order clustering algorithm to cluster each data point in the dataset and divide the blade surface of the turbine blade into multiple preliminary inspection regions.

[0049] Among them, the second-order clustering algorithm is a clustering method that combines statistical characteristics and recursive aggregation strategies. Its core idea is to gradually establish a hierarchical clustering structure by calculating the similarity or distance between data points, and select the optimal grouping result by evaluating the clustering scheme.

[0050] In a possible implementation, S2 specifically includes:

[0051] S201: taking the first data point in the data set as the first leaf node of the clustering feature tree.

[0052] Among them, Clustering Feature Tree (CF Tree) is a hierarchical data structure, which is often used to store and manage intermediate results in clustering algorithms. It aggregates data points level by level to form a tree structure, providing efficient support for fast clustering, feature extraction and partitioning.

[0053] S202: Calculate the log-likelihood value between the second data point in the data set and the first leaf node using the log-likelihood distance formula:

[0054] d(i,j)=ζ i +ζ j -ζ (i,j) ;

[0055] Where d(i,j) represents the log-likelihood distance between the i-th data point and the j-th data point, ζ i represents the log-likelihood value of the ith data point, ζ j represents the log-likelihood value of the jth data point, ζ (i,j) Represents the log-likelihood value of the merging of the i-th data point and the j-th data point.

[0056] S203: Determine whether the log-likelihood value is less than a preset log-likelihood value. If so, merge the second data point into the first leaf node, otherwise, use the second data point as the second leaf node of the cluster feature tree.

[0057] S204: Recursively process the remaining data points in the data set step by step to form a clustering feature tree, wherein each leaf node of the clustering feature tree represents a clustering scheme.

[0058] In the present invention, through the recursive optimization of the second-order clustering algorithm, combined with the log-likelihood distance and the evaluation value, it is possible to effectively capture the slight differences between the data points on the blade surface, and to divide the areas with similar characteristics together, thus providing a high-precision preliminary partitioning scheme for subsequent damage assessment and inspection management.

[0059] S205: Calculate the log-likelihood evaluation value of each clustering scheme:

[0060]

[0061] Among them, ζ v represents the log-likelihood evaluation value of the v-th clustering scheme, N v represents the number of data points in the v-th clustering scheme, K A represents the total number of continuous variables, K B represents the number of categorical variables, represents the global variance of the k-th continuous variable, represents the estimated variance of the k-th continuous variable in the v-th clustering scheme, represents the entropy of the k-th categorical variable in the v-th clustering scheme, L k represents the number of categories of the k-th categorical variable, N vkl represents the number of data points of the k-th categorical variable classified in the l-th category.

[0062] In the present invention, by evaluating the variances of continuous variables (such as stress and temperature) and the entropies of categorical variables (such as material type or damage type), multiple indicators can be comprehensively integrated to ensure that the divided regions are more in line with the actual working conditions.

[0063] S206: Use the agglomerative clustering algorithm to combine each leaf node of the clustering feature tree to generate multiple updated clustering schemes with different numbers of leaf nodes.

[0064] Among them, agglomerative clustering is a bottom-up clustering method. Starting from each data point, regarding it as a separate cluster, and then according to a certain similarity or distance criterion, gradually merge the most similar clusters until the stopping condition is reached (such as forming a single cluster or reaching the specified number of clusters).

[0065] S207: Combine the log-likelihood evaluation values of each updated clustering scheme, and determine the optimal number of clusters through the Bayesian Information Criterion.

[0066] Among them, the Bayesian Information Criterion (BIC) is a standard for statistical model selection, used to select the optimal model among multiple candidate models. It comprehensively considers the goodness of fit of the model (the ability to explain the data) and the complexity of the model (the number of parameters) to avoid overfitting problems. BIC is particularly suitable for scenarios such as clustering analysis, regression analysis, and probability model selection.

[0067] In a possible implementation manner, in a possible implementation manner, the calculation formula of the Bayesian Information Criterion is:

[0068]

[0069] Among them, B(j) represents the scoring value of the clustering scheme when the number of clusters is j, and ζ v represents the log-likelihood evaluation value of the v-th clustering scheme, j represents the number of clusters, and m j represents the total number of free parameters when the number of clusters is j, N represents the total number of data points, and K A represents the total number of continuous variables, and K B represents the total number of categorical variables, and L k represents the number of categories of the k-th categorical variable.

[0070] S208: Divide the blade surface of the turbine blade into inspection areas equal to the optimal number of clusters as the preliminary inspection areas.

[0071] Specifically, place the first record in the dataset on a leaf node initiated by the root node, which contains all variable information of this record. Using distance measurement as the similarity criterion, according to the similarity between the new record and the existing nodes, merge it with the existing nodes to generate new nodes. Through this recursive method, gradually establish a clustering feature tree, and finally provide a summary of the variable information of the dataset. The distance measurement model uses logarithmic similarity. Use the agglomerative clustering algorithm to combine the leaf nodes to generate a set of schemes with different numbers of clusters, and screen out the optimal number of clusters through the Bayesian information criterion.

[0072] In the present invention, through the second-order clustering algorithm and the clustering feature tree, group the data points on the blade surface, divide the data points into multiple preliminary inspection areas according to similarity, effectively reduce the repetition and omission of the inspection scope, and lay a foundation for subsequent refined management. At the same time, combined with the log-likelihood evaluation value and the Bayesian information criterion (BIC), it can dynamically determine the optimal number of clusters, avoid the division error that may be caused by manually setting the number of clusters, and improve the scientificity and rationality of the area division.

[0073] It should be noted that when using the clustering algorithm to divide the inspection area of the turbine blade, it is crucial to accurately identify the clustering boundary. Using the clustering algorithm to process the characteristic sample points of the turbine blade, although it can complete the classification of the sample points, the final result is usually the scattered sample points after classification. Therefore, in order to effectively divide the inspection area of the turbine blade, it is necessary to further identify the boundaries of these scattered sample points

[16] . To achieve a relatively regular and smooth curve for the regional boundary.

[0074] S3: Use the Alpha-Shapes algorithm to extract the boundary points corresponding to each preliminary inspection area.

[0075] Among them, the Alpha-Shapes algorithm is a geometric algorithm used to describe the shape characteristics of a point set, especially to extract the boundary or contour of the point set. It can be regarded as a generalization of the Convex Hull, but is more flexible than the Convex Hull. The level of detail of the shape can be controlled by the parameter α, so as to adapt to the extraction of non-convex shapes or irregular boundaries.

[0076] In a possible implementation, S3 specifically includes:

[0077] S301: Set the adjustment parameter of the Alpha-Shapes algorithm.

[0078] S302: Calculate the distance values between each point in the dataset and the remaining data points.

[0079] S303: Combine the data points corresponding to the distance values less than the preset distance value to form a first set.

[0080] S304: Randomly select a target point from the first set.

[0081] S305: Calculate the center of the first sphere and the center of the second sphere with the target point as the reference.

[0082] S306: Calculate the first distance value from the remaining data points in the first set to the center of the first sphere and the second distance value of the distance to the center of the second sphere. Determine whether both the first distance value or the second distance value is greater than the adjustment parameter. If so, determine the target point as a boundary point. Otherwise, randomly select a target point from the remaining data points in the first set and enter S405.

[0083] In the present invention, by screening the data points whose distance from the target point is less than the preset distance, the Alpha-Shapes algorithm effectively excludes the isolated points and noise data outside the boundary range. This can not only ensure the accuracy of the boundary, but also improve the reliability of the division result.

[0084] In a possible implementation, the calculation formulas for the center of the first sphere and the center of the second sphere are:

[0085]

[0086] S=(x - x1) 2 +(y - y1) 2 +(z - z1) 2 ;

[0087] Among them, (x2, y2, z2) represent the three-dimensional coordinates of the center of the sphere, (x1, y1, z1) represent the three-dimensional coordinates of the current target point, (x, y, z) represent the three-dimensional coordinates of another data point in the set where the current target point is located, α represents the adjustment parameter, and S represents the square of the distance between the current target point and another data point in the set where the current target point is located.

[0088] S307: Traverse the first set to determine the boundary points corresponding to each preliminary inspection area.

[0089] In the present invention, the algorithm calculates boundary points point by point to ensure that the extracted boundary points can represent the geometric characteristics of the preliminary inspection area and provide a high-quality initial point set for subsequent boundary optimization (such as curve fitting).

[0090] Specifically, the selected α value defines a sphere with a radius of α that rolls along the point cloud in three-dimensional space. The points it touches constitute boundary points, which are connected to form the surface boundary of the entire point set. [17-18] . In the three-dimensional point cloud data, for each point P(x,y,z), calculate the distance to its adjacent points, and include the point set whose distance does not exceed 2α into the set Q. Then, select any point P1(x1,y1,z1) from the set Q, and calculate the center of the circle according to the parameter α. Calculate the distances from other points in the set Q to the two centers of the circle. If all distances are greater than α, then point P is a boundary point; otherwise, replace point P1 and recalculate until the entire set Q is traversed.

[0091] In summary, the Alpha-Shapes algorithm is used to extract the boundary points of the preliminary inspection area, which can not only accurately capture the boundary features of complex geometric shapes, but also effectively remove noise and improve the flexibility and scientificity of boundary division. At the same time, combined with the adjustment of parameter α, this method can adapt to the regional division requirements of complex working conditions on the surface of turbine blades, and provide high-quality basic data for subsequent smooth boundary fitting and fine regional division, thereby improving the efficiency and accuracy of inspection and management.

[0092] S4: performing curve fitting on the boundary points corresponding to each preliminary inspection area respectively through the least square method to determine the smooth boundary curve corresponding to each preliminary inspection area.

[0093] Among them, the least squares method (LSM) is a mathematical optimization method used to solve a system of equations or fit a model by minimizing the sum of squares of errors. It is widely used in fields such as data fitting, parameter estimation, and signal processing.

[0094] In a possible implementation, S4 specifically includes:

[0095] S401: Determine the fitting curves for each preliminary inspection area through the least squares method. The fitting curves are specifically:

[0096] f(x) = a0 + a1x + a2x 2 + … + a m x m ;

[0097] where f(x) represents the fitting function value with respect to the x-axis coordinate of the boundary point, a0 represents the intercept of the curve, and a t represents the t-th power coefficient, t ∈ (1, m), and m represents the highest power of the polynomial of the fitting curve.

[0098] In the present invention, by adjusting the order of the fitting polynomial, it is possible to flexibly adapt to different complexities of the boundary shape. For simple areas, low-order polynomials can be used for fitting; for complex areas, high-order polynomials can be selected to achieve fine fitting, thus meeting the needs of differential management.

[0099] S402: Determine the set of boundary points corresponding to each preliminary inspection area.

[0100] S403: Set an objective function to reduce the error between each boundary point in the set of boundary points and the corresponding fitting curve:

[0101]

[0102] where E represents the objective function value of the fitting curve error, n represents the total number of boundary points, x i represents the x-axis coordinate of the i-th boundary point, and y i represents the y-axis coordinate of the i-th boundary point.

[0103] S404: Determine the optimal fitting coefficients for each preliminary inspection area by minimizing the objective function.

[0104] S405: Determine the smooth boundary curves corresponding to each preliminary inspection area according to the optimal fitting coefficients of each inspection area.

[0105] In the present invention, the least squares method can effectively remove the noise and minor fluctuations in the boundary points by minimizing the sum of the squares of the errors between the boundary points and the fitting curve, making the boundary smoother while maintaining the overall consistency between the curve and the boundary points. This smoothing process can avoid unnecessary sharp corners or discontinuities in the boundary curve, enhancing the geometric quality of the boundary. At the same time, the smooth boundary curve generated by fitting can accurately describe the geometric shape of the preliminary inspection area, eliminating the subjectivity in manual or empirical division, providing a scientific basis for the division of the turbine blade area, and ensuring the rationality and consistency of the division results.

[0106] S5: Divide the fine inspection areas of the crack-type damage of the turbine blade according to the smooth boundary curves corresponding to the respective preliminary inspection areas.

[0107] It should be noted that the area enclosed by the smooth boundary curve for each preliminary inspection area is the fine inspection area of the crack-type damage of the turbine blade.

[0108] In the present invention, by dividing the fine inspection areas of the crack-type damage of the turbine blade through smooth boundary curves, the inspection efficiency and accuracy can be significantly improved, while resource waste is reduced, providing scientific and reliable technical support for the damage management and life optimization of the turbine blade.

[0109] The beneficial effects brought by the technical solution provided by the embodiment of the present invention at least include:

[0110] (1) In the present invention, using the Alpha-Shapes algorithm, extract the boundary points corresponding to each preliminary inspection area, and through the least squares method, perform curve fitting on the boundary points corresponding to each preliminary inspection area respectively to determine the smooth boundary curves corresponding to each preliminary inspection area, accurately capturing the geometric features of the area boundary and effectively improving the accuracy of boundary division.

[0111] (2) In the present invention, extract data points through finite element analysis, use the second-order clustering algorithm to achieve differential division of the preliminary inspection areas, use the Alpha-Shapes algorithm and the least squares method to determine the smooth boundary curves of the divided areas, and combine the smooth boundary curves to finely divide the crack-type damage areas, constructing a systematic partition optimization and management scheme, effectively meeting the high requirements of modern aero-engines for safety and economy.

[0112] Refer to the attached Figure 2 illustrates a schematic structural diagram of an inspection area division system for crack-type damage of a turbine blade provided by the present invention.

[0113] The present invention also provides an inspection area division system 20 for crack-type damage of a turbine blade, which is applied to the above-mentioned inspection area division method for crack-type damage of a turbine blade, and includes:

[0114] A processor 201.

[0115] A memory 202, on which computer-readable instructions are stored. When the computer-readable instructions are executed by the processor 201, the inspection area division method for crack-type damage of a turbine blade as in the method embodiment is implemented.

[0116] The inspection area division system 20 for crack-type damage of turbine blades provided by the present invention can execute the above-mentioned inspection area division method for crack-type damage of turbine blades and achieve the same or similar technical effects. To avoid repetition, the present invention will not elaborate further.

[0117] The beneficial effects brought by the technical solution provided by the embodiments of the present invention at least include:

[0118] (1) In the present invention, by using the Alpha-Shapes algorithm, the boundary points corresponding to each preliminary inspection area are extracted, and the boundary points corresponding to each preliminary inspection area are respectively curve-fitted by the least square method to determine the smooth boundary curves corresponding to each preliminary inspection area, accurately capturing the geometric features of the area boundary and effectively improving the accuracy of boundary division.

[0119] (2) In the present invention, data points are extracted through finite element analysis, the differential division of preliminary inspection areas is realized by using the second-order clustering algorithm, the smooth boundary curves of the divided areas are determined by using the Alpha-Shapes algorithm and the least square method, and the crack-type damage area is finely divided in combination with the smooth boundary curves, constructing a systematic partition optimization and management scheme, effectively meeting the high requirements of modern aero-engines for safety and economy.

[0120] It should be understood that the processor in the embodiments of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0121] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0122] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that contains one or more collections of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0123] It should be understood that the term "and / or" in this document is merely a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. Additionally, the character " / " in this document generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood by referring to the context.

[0124] In the present invention, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0125] It should be understood that in various embodiments of the present invention, the sequence numbers of the above processes do not indicate the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0126] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention.

[0127] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the devices, apparatuses, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0128] In several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0129] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0130] In addition, the functional units in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0131] When the above-mentioned functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs.

[0132] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for dividing the inspection area of the crack-type damage of the turbine blade as described in the method embodiment.

[0133] The computer-readable storage medium provided by the present invention can implement the steps and effects of the method for dividing the inspection area of the crack-type damage of the turbine blade in the above method embodiment. To avoid repetition, the present invention will not elaborate further.

[0134] The beneficial effects brought by the technical solution provided by the embodiment of the present invention at least include:

[0135] (1) In the present invention, by using the Alpha-Shapes algorithm, the boundary points corresponding to each preliminary inspection area are extracted, and through the least squares method, the boundary points corresponding to each preliminary inspection area are respectively curve-fitted to determine the smooth boundary curves corresponding to each preliminary inspection area, accurately capturing the geometric features of the area boundary and effectively improving the accuracy of boundary division.

[0136] (2) In the present invention, data points are extracted through finite element analysis, and the second-order clustering algorithm is used to achieve the differential division of the preliminary inspection areas. The Alpha-Shapes algorithm and the least squares method are used to determine the smooth boundary curves of the divided areas, and the crack-type damage areas are finely divided in combination with the smooth boundary curves, constructing a systematic partition optimization and management scheme, effectively meeting the high requirements of modern aero-engines for safety and economy.

[0137] As described above, this is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

[0138] The following points need to be explained:

[0139] (1) The attached drawings of the embodiments of the present invention only relate to the structures involved in the embodiments of the present invention, and other structures can refer to the general design.

[0140] (2) For clarity, in the attached drawings used to describe the embodiments of the present invention, the thickness of the layer or region is enlarged or reduced, that is, these drawings are not drawn according to the actual scale. It can be understood that when an element such as a layer, film, region or substrate is referred to as being "on" or "under" another element, the element can be "directly" on or under the other element or there can be an intermediate element.

[0141] (3) Without conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0142] As above, this is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for dividing the inspection area of crack-type damage of a turbine blade, characterized in that Including: S1: Extract multiple data points from the blade surface of the turbine blade through finite element analysis to form a data set, where the data set includes multiple metric variables related to the blade working state; S2: Use the second-order clustering algorithm to cluster each data point in the data set, and divide the blade surface of the turbine blade into multiple preliminary inspection areas; S3: Use the Alpha-Shapes algorithm to extract boundary points corresponding to each preliminary inspection area; S4: Through the least squares method, perform curve fitting on the boundary points corresponding to each preliminary inspection area respectively to determine the smooth boundary curve corresponding to each preliminary inspection area; S5: According to the smooth boundary curves corresponding to each preliminary inspection area, divide the fine inspection areas of the turbine blade crack-type damage.

2. The method for dividing the inspection area of the crack-type damage of the turbine blade according to claim 1, wherein The data set specifically is: P i = {(x i , y i , z i ), σ i , T i , …}; Among them, P i represents a data set composed of i data points, (x i , y i , z i ) represents the three-dimensional spatial coordinates of the i-th data point, σ i represents the stress value of the i-th data point, T i represents the temperature value of the i-th data point.

3. The method for dividing the inspection area of crack-type damage of a turbine blade according to claim 1, wherein The metric variables include stress and temperature; After the S1, it further includes: Calculate the Pearson correlation coefficient values between each metric variable, and retain any one of the metric variable pairs in the multiple groups of metric variable pairs whose Pearson correlation coefficient values are greater than the preset Pearson correlation coefficient value.

4. The method for dividing the inspection area of the crack-type damage of the turbine blade according to claim 3, wherein The calculation formula of the Pearson correlation coefficient value is: Among them, r represents the Pearson correlation coefficient value, n represents the total number of samples, x i represents the i-th observation value of variable x, y i represents the i-th observation value of variable y, represents the mean value of variable x, represents the mean value of variable y.

5. The method for dividing the inspection area of the crack-type damage of the turbine blade according to claim 1, wherein, The S2 specifically includes: S201: Take the first data point in the data set as the first leaf node of the clustering feature tree; S202: Use the log-likelihood distance formula to calculate the log-likelihood value between the second data point in the data set and the first leaf node; d(i,j) = ζ i + ζ j - ζ (i,j) ; Among them, d(i,j) represents the log-likelihood distance between the i-th data point and the j-th data point, and ζ i represents the log-likelihood value of the i-th data point, and ζ j represents the log-likelihood value of the j-th data point, and ζ (i,j) represents the log-likelihood value after the i-th data point and the j-th data point are merged; S203: Judge whether the log-likelihood value is less than the preset log-likelihood value; if so, merge the second data point into the first leaf node, otherwise, take the second data point as the second leaf node of the clustering feature tree; S204: In a recursive manner, gradually process the remaining data points in the data set to form a clustering feature tree, where each leaf node of the clustering feature tree represents a clustering scheme; S205: Calculate the log-likelihood evaluation values of each clustering scheme; Among them, ζ v represents the log-likelihood evaluation value of the v-th clustering scheme, N v represents the number of data points of the v-th clustering scheme, K A represents the total number of continuous variables, K B represents the number of categorical variables, represents the global variance of the k-th continuous variable, represents the estimated variance of the k-th continuous variable in the v-th clustering scheme, represents the entropy of the k-th categorical variable in the v-th clustering scheme, L k represents the number of categories of the k-th categorical variable, N vkl represents the number of data points of the k-th categorical variable classified in the l-th category; S206: Use the merging clustering algorithm to combine each leaf node of the clustering feature tree to generate multiple updated clustering schemes with different numbers of leaf nodes; S207: Combine the log-likelihood evaluation values of each updated clustering scheme, and determine the optimal number of clusters through the Bayesian information criterion; S208: Divide the blade surface of the turbine blade into inspection areas with the same number as the optimal number of clusters as the preliminary inspection areas.

6. The method for dividing the inspection area of the crack-type damage of the turbine blade according to claim 5, wherein The calculation formula of the Bayesian information criterion is: Among them, B(j) represents the scoring value of the clustering scheme when the number of clusters is j, and ζ v represents the log-likelihood evaluation value of the v-th clustering scheme, j represents the number of clusters, and m j represents the total number of free parameters when the number of clusters is j, N represents the total number of data points, and K A represents the total number of continuous variables, and K B represents the total number of categorical variables, and L k represents the number of categories of the k-th categorical variable.

7. The method for dividing the inspection area of the crack-type damage of the turbine blade according to claim 1, wherein, The S3 specifically includes: S301: Set the adjustment parameter of the Alpha-Shapes algorithm; S302: Calculate the distance values between each point in the data set and the remaining data points; S303: Combine the data points corresponding to the distance values less than the preset distance value to form a first set; S304: Randomly select a target point from the first set; S305: Calculate the center of the first sphere and the center of the second sphere based on the target point; S306: Calculate the first distance value from the remaining data points in the first set to the center of the first sphere and the second distance value from the remaining data points in the first set to the center of the second sphere; determine whether both the first distance value and the second distance value are greater than the adjustment parameter; if so, determine the target point as the boundary point; otherwise, randomly select a target point from the remaining data points in the first set and proceed to S405; S307: Traverse the first set to determine the boundary points corresponding to each preliminary inspection area.

8. The method for dividing the inspection area of the crack-type damage of the turbine blade according to claim 7, wherein The calculation formulas for the center of the first sphere and the center of the second sphere are as follows: S = (x - x1) 2 + (y - y1) 2 + (z - z1) 2 ; where (x2, y2, z2) represents the three-dimensional coordinates of the center of the sphere, (x1, y1, z1) represents the three-dimensional coordinates of the current target point, (x, y, z) represents the three-dimensional coordinates of another data point in the set where the current target point is located, α represents the adjustment parameter, and S represents the square of the distance between the current target point and another data point in the set where the current target point is located.

9. The method for dividing the inspection area of the crack-type damage of the turbine blade according to claim 1, wherein, The specific steps of S4 include: S401: Determine the fitting curve of each preliminary inspection area by the least squares method. The fitting curve is specifically: f(x) = a0 + a1x + a2x 2 + … + a m x m ; Among them, f(x) represents the fitting function value of the x-axis coordinate of the boundary point, a0 represents the intercept of the curve, and a t represents the t-th power coefficient, t ∈ (1, m), and m represents the highest power of the polynomial of the fitting curve; S402: Determine the set of boundary points corresponding to each preliminary inspection area; S403: Set an objective function to reduce the error between each boundary point in the set of boundary points and the corresponding fitting curve: Among them, E represents the objective function value of the fitting curve error, n represents the total number of boundary points, and x i represents the x-axis coordinate of the i-th boundary point, and y i represents the y-axis coordinate of the i-th boundary point; S404: Determine the optimal fitting coefficients of each preliminary inspection area by minimizing the objective function; S405: Determine the smooth boundary curve corresponding to each preliminary inspection area according to the optimal fitting coefficients of each inspection area.

10. A system for dividing inspection areas of crack-type damage of turbine blades, characterized in that, including: a processor; a memory, on which computer-readable instructions are stored. When the computer-readable instructions are executed by the processor, the method for dividing the inspection area of the crack-type damage of the turbine blade as described in any one of claims 1 to 9 is implemented.

Images