A life-limited part probability failure risk analysis method based on scattered point fitting response surface

Abstract

The present application relates to a kind of based on scatter point fitting response surface limited life component probability failure risk analysis method, belong to the life management technical field of aero-engine limited life component, solve the error caused by the characterization of risk in prior art is large, weighted processing part is prone to overestimate the risk, the analysis precision of probability failure risk framework under complex environment is insufficient, thereby affecting the safety of aero-engine and on-board personnel problem.The present application is based on the method of fitting response surface to characterize the probability failure risk of limited life component universe, compared with the method of most dangerous point characterization, can greatly improve the accuracy of analysis;The present application is based on time domain to analyze crack propagation, has the ability to analyze stress nephogram change at different time, to further ensure the safety of aero-engine and on-board personnel.

Description

Technical Field

[0001] This invention relates to the field of life management technology for life-limited components of aero-engines, and specifically to a method for probabilistic failure risk analysis of life-limited components based on scatter-point fitting response surface. Background Technology

[0002] The safety of aircraft engines is crucial to flight safety. The primary failure of life-limited components can have catastrophic consequences; therefore, ensuring the safety of these components is a key aspect of engine safety. To address the safety of life-limited components, the aviation industry has developed a life-limited component assessment process centered on probabilistic failure risk analysis.

[0003] Characterizing and weighting the risks at various points on the roulette wheel is a crucial step in the probabilistic failure risk assessment and analysis of life-limited components. Traditional risk characterization and processing methods, such as optimal automatic partitioning, pre-partitioning, and stress similarity partitioning, all first select a typical moment that represents all flight trajectories for partitioning, then use the most dangerous point within the partition to characterize the risk value of the entire partition, and finally obtain the total risk of the life-limited component through weighted calculation. However, in actual complex flight conditions, firstly, there is no typical moment that can represent all flight trajectories, and using a single moment for full life-cycle characterization will introduce errors into the analysis. Secondly, using the most dangerous point to characterize the risk value within a partition will inevitably lead to an overestimation of risk. All of the above processing methods will reduce the accuracy of the analysis within the original processing framework.

[0004] In summary, existing technologies suffer from several problems: the risk characterization leads to large errors; the weighted processing component is prone to overestimating risks; the probabilistic failure risk framework lacks analytical accuracy in complex environments; and these issues affect the safety of aero engines and onboard personnel. Summary of the Invention

[0005] In view of the above problems, the present invention provides a method for probabilistic failure risk analysis of life-limited components based on scatter-point fitting response surface, which solves the problems of large errors caused by risk characterization in the prior art, easy overestimation of risk in the weighted processing part, insufficient analysis accuracy of probabilistic failure risk framework in complex environments, and impact on the safety of aero engines and onboard personnel.

[0006] This invention provides a method for probabilistic failure risk analysis of life-limited components based on scatter-point fitting response surfaces, comprising the following steps:

[0007] Step S1. Perform finite element analysis on the life-limited component to obtain finite element results including stress and temperature at each finite element node of the life-limited component;

[0008] Step S2. Based on the finite element results, divide the life-limited component into regions to obtain an initial partition including finite element nodes;

[0009] Step S3. Select the scattered points to be analyzed in the finite element nodes of each initial partition; for each scattered point to be analyzed, obtain the stress history of each scattered point according to the finite element results, and use it to perform probabilistic failure risk analysis to obtain the probabilistic failure risk on each scattered point.

[0010] Step S4. Fit the probabilistic failure risk of the finite element nodes other than the scattered points selected for analysis in step S3 to obtain the probabilistic failure risk response surface distribution of the life-limited component; wherein, the probabilistic failure risk response surface distribution of the life-limited component includes the risk value of each finite element node in each partition.

[0011] Step S5. Set a risk threshold based on accuracy requirements, which will be used to perform the following optimization processing on each partition:

[0012] If the difference between the maximum and minimum risk values ​​of each finite element node in a partition exceeds the risk threshold, then the partition is split; otherwise, the partition is retained.

[0013] Repeat steps S3 and S4 for the split partitions until the difference between the maximum and minimum values ​​of all partitions is less than the risk threshold, and obtain the optimized probability failure risk response surface distribution of the life-limited component.

[0014] Step S6. Weight the optimized probabilistic failure risk response surface distribution of the life-limited component to obtain the probabilistic failure risk of the life-limited component.

[0015] Furthermore, step S1 specifically includes:

[0016] Obtain the parameter conditions of the life-limiting component under different boundary conditions; the parameter conditions include working fluid information and rotational speed data;

[0017] Fluid analysis was performed on the life-limited component based on the working fluid information to obtain the volume temperature distribution data of the life-limited component;

[0018] Stress analysis was performed on the life-limiting component based on body temperature distribution data and rotation speed data to obtain stress values ​​at all finite element nodes of the life-limiting component.

[0019] Furthermore, step S2 specifically includes:

[0020] Based on the finite element results, the moment of maximum stress of the life-limited component is selected as the basis for initial partitioning. The temperature, stress, and geometric characteristics of the life-limited component are used as partitioning criteria to divide the life-limited component into regions, obtaining an initial partition including finite element nodes.

[0021] Furthermore, in step S3, the scattered points to be analyzed are selected from the finite element nodes of each initial partition. Specifically, the node containing the maximum stress value in each initial partition is selected as the scattered point to be analyzed.

[0022] Furthermore, step S4 specifically includes:

[0023] By selecting a fitting method, the probabilistic failure risk on each finite element node, except for the scatter points selected for analysis in step S3, is fitted to obtain the risk value of the finite element node without probabilistic failure risk calculation, thereby obtaining the probabilistic failure risk response surface distribution of the life-limited component.

[0024] Furthermore, the fitting method in step S4 is specifically a triangle-based linear interpolation method.

[0025] Furthermore, step S6 specifically includes:

[0026] By establishing quadrilateral cells for adjacent elements in the optimized probabilistic failure risk response surface distribution of the life-limited component, weighting is performed on the quadrilateral cells, and the risk values ​​of each quadrilateral cell are summarized to obtain the probabilistic failure risk value of the entire life-limited component.

[0027] Furthermore, the life-limiting component is a life-limiting component for aero engines.

[0028] Furthermore, the life-limiting component of the aero-engine is the compressor wheel of the aero-engine.

[0029] Furthermore, the compressor disk of the aero-engine is made of titanium alloy.

[0030] Compared with the prior art, the present invention has at least the following beneficial effects:

[0031] (1) The present invention uses a method based on fitting response surface to characterize the probabilistic failure risk of the entire life-limited component. Compared with the method of characterizing the most dangerous point, it can significantly improve the accuracy of the analysis.

[0032] (2) The present invention analyzes crack propagation based on the time domain and has the ability to analyze the changes in stress cloud diagrams at different times, thereby further ensuring the safety of aero engines and onboard personnel. Attached Figure Description

[0033] The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of the invention.

[0034] Figure 1 This is a flowchart of the method for probabilistic failure risk analysis of life-limited components based on scatter-point fitting response surface disclosed in this invention;

[0035] Figure 2 This is a schematic diagram of the finite element nodes and initial partitioning features of the probabilistic failure risk analysis method for life-limited components based on scattered point fitting response surfaces disclosed in this invention.

[0036] Figure 3The distribution of the probability failure risk response surface on all nodes of the life-limited component is the initial fitted of the life-limited component probability failure risk analysis method based on scattered point fitting response surface disclosed in this invention.

[0037] Figure 4 This is the optimized distribution of the probabilistic failure risk response surface on all nodes of a life-limited component, based on the scattered point fitting response surface-based probabilistic failure risk analysis method disclosed in this invention. Detailed Implementation

[0038] To better understand the above-described objectives, features, and advantages of the present invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments of the present invention and the features thereof can be combined with each other. Furthermore, the present invention can be implemented in other ways different from those described herein; therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0039] This invention discloses a method for probabilistic failure risk analysis of life-limited components based on scattered point fitting response surface; wherein, life-limited components include, but are not limited to, life-limited components of aero-engines; exemplarily, specifically, the compressor disk of an aero-engine, made of titanium alloy.

[0040] The method for probabilistic failure risk analysis of life-limited components based on scattered point fitting response surfaces disclosed in this invention includes the following steps:

[0041] Step S1. Perform finite element analysis on the life-limited component to obtain finite element results including stress and temperature at each finite element node of the life-limited component.

[0042] Specifically, the parameters of the component with limited lifespan under different boundary conditions are obtained; the parameters include rotational speed, working fluid information and rotational speed data, and the working fluid information includes working fluid temperature and working fluid pressure.

[0043] Fluid analysis is performed on the life-limited component based on the working fluid information. For example, the working fluid information is substituted into ANSYS CFX to perform fluid analysis on the life-limited component and obtain the volume temperature distribution data of the life-limited component.

[0044] Stress analysis is performed on the life-limiting component based on body temperature distribution data and rotation speed data. For example, the body temperature distribution data and rotation speed data are substituted into the Transient Structural module of ANSYS WORKBENCH to perform stress analysis on the life-limiting component and obtain the stress values ​​on all finite element nodes of the life-limiting component.

[0045] Step S2. Based on the finite element results, divide the finite element into regions to obtain an initial partition including finite element nodes.

[0046] Specifically, based on the finite element results, the moment of maximum stress of the life-limited component is selected as the basis for initial partitioning. The temperature, stress, and geometric characteristics of the life-limited component are used as partitioning criteria to divide the life-limited component into regions, obtaining initial partitions including finite element nodes. Among them, the stress used as the partitioning criterion is the stress gradient, the temperature used as the partitioning criterion is the temperature gradient, and the geometric characteristics used as the partitioning criterion are divided into edges, corners, and interiors.

[0047] Preferably, the stress gradient is 85 MPa and the temperature gradient is 50 K.

[0048] Finite element node and initial partition characteristics such as Figure 2 As shown, areas of different gray levels represent different partitions, and white dots represent finite element nodes.

[0049] Step S3. Select the scattered points to be analyzed in the finite element nodes of each initial partition; for each scattered point to be analyzed, obtain the stress history of each scattered point according to the finite element results, and use it to perform probabilistic failure risk analysis to obtain the probabilistic failure risk on each scattered point.

[0050] Specifically, the scattered points to be analyzed are selected from the finite element nodes of each initial partition. Specifically, the node containing the maximum stress value in each initial partition is selected as the scattered point to be analyzed.

[0051] For example, the process of probabilistic failure risk analysis is as follows:

[0052] First, the transcendence curve of the initial defect of TC4 titanium alloy in FAA Airworthiness Advisory Circular AC33.14 is selected as the transcendence curve of the initial crack distribution. Monte Carlo sampling of the initial crack distribution is performed using Matlab software to obtain sample points. For each sample point, crack propagation calculation is performed using the finite element nodal stress process obtained in step S1 to obtain the final crack propagation value of each sample point after a specified number of cycles; optionally, 20,000 cycles. The stress intensity factor K corresponding to the final crack propagation value under each sample is calculated. final .

[0053]

[0054] Where G is the shape factor, s is the stress value at the scatter point, π is pi, and a final This represents the final value of crack propagation.

[0055] Secondly, select the fracture toughness K. c As a criterion for judgment, for example, the fracture toughness is taken as the fracture toughness of titanium alloy. For the stress intensity factor K at each sample point final The percentage of items exceeding the fracture toughness is counted, and this percentage is the probability failure risk value of that scatter point.

[0056] Step S4. Fit the probabilistic failure risk of the finite element nodes other than the scattered points selected for analysis in step S3 to obtain the probabilistic failure risk response surface distribution of the life-limited component; wherein, the probabilistic failure risk response surface distribution of the life-limited component includes the risk value of each finite element node in each partition.

[0057] In Matlab software, triangle-based linear interpolation is used as the fitting method to fit the probabilistic failure risk at each finite element node except for the scattered points selected for analysis in step S3. This yields the risk values ​​of the finite element nodes without probabilistic failure risk calculation, thus obtaining the probabilistic failure risk response surface distribution at all finite element nodes of the life-limited component. Figure 3 As shown.

[0058] Step S5. Set a risk threshold based on accuracy requirements, which will be used to perform the following optimization processing on each partition:

[0059] If the difference between the maximum and minimum risk values ​​of each finite element node in a partition exceeds the risk threshold, the partition is split; otherwise, the partition is retained.

[0060] Repeat steps S3 and S4 for the split partitions until the difference between the maximum and minimum values ​​of all partitions is less than the risk threshold, thereby obtaining the optimized probability failure risk response surface distribution of the life-limited component.

[0061] Optionally, a risk threshold of 0.1 is set. The probabilistic failure risk values ​​of nodes within each partition are statistically analyzed. If the difference between the maximum and minimum risk values ​​of a partition exceeds the set risk threshold, the original partition is split into two partitions based on stress magnitude. Steps S3 and S4 are repeated for the split partitions: establishing new scatter plots, calculating the risk of the scatter plots, and fitting risk values ​​to the uncalculated finite element results based on the scatter plot results. This continues until the difference between the maximum and minimum values ​​of all partitions is less than the risk threshold. At this point, the optimized probabilistic failure risk response surface distribution on all nodes of the life-limited component is obtained, as shown below. Figure 4 As shown.

[0062] Step S6. Weight the optimized probabilistic failure risk response surface distribution of the life-limited component to obtain the probabilistic failure risk of the life-limited component.

[0063] Specifically, by establishing quadrilateral cells for adjacent elements in the optimized probability failure risk response surface distribution of the life-limited component, weighting is performed on the quadrilateral cells, and the risk values ​​of each quadrilateral cell are summarized to obtain the probability failure risk value of the entire life-limited component.

[0064] For example, after obtaining the optimized probability failure risk distribution on all nodes of the life-limited component, four adjacent finite element nodes are arranged into a quadrilateral element, and the failure risk value P of the entire life-limited component is obtained. f,total It can be calculated by weighted addition using the following formula.

[0065]

[0066] Where elem is the number of quadrilateral units, and n is the maximum value of the quadrilateral units. P f,elem1 P f,elem2 P f,elem3 and P f,elem4 These are the risk values ​​of the four finite element nodes that make up the quadrilateral element.

[0067] Compared with existing technologies, this invention uses a method based on fitted response surfaces to characterize the probabilistic failure risk of life-limited components across the entire domain. Compared with the method of characterizing the most dangerous point, it can significantly improve the accuracy of the analysis. This invention analyzes crack propagation based on the time domain and has the ability to analyze stress cloud map changes at different times, thereby further ensuring the safety of aero engines and onboard personnel.

[0068] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for probabilistic failure risk analysis of life-limited components based on scatter-point fitting response surface, comprising the following steps: Step S1. Perform finite element analysis on the life-limited component to obtain finite element results including stress and temperature at each finite element node of the life-limited component; Step S2. Based on the finite element results, divide the life-limited component into regions to obtain an initial partition including finite element nodes; Step S3. Select the scattered points to be analyzed from the finite element nodes of each initial partition; For each scatter point to be analyzed, the stress history of each scatter point is obtained based on the finite element results, which is used to perform probabilistic failure risk analysis and obtain the probabilistic failure risk at each scatter point. Step S4. Fit the probabilistic failure risk of the finite element nodes other than the scattered points selected for analysis in step S3 to obtain the probabilistic failure risk response surface distribution of the life-limited component; wherein, the probabilistic failure risk response surface distribution of the life-limited component includes the risk value of each finite element node in each partition. Step S5. Set a risk threshold based on accuracy requirements, which will be used to perform the following optimization processing on each partition: If the difference between the maximum and minimum risk values ​​of each finite element node in a partition exceeds the risk threshold, then the partition is split; otherwise, the partition is retained. Repeat steps S3 and S4 for the split partitions until the difference between the maximum and minimum values ​​of all partitions is less than the risk threshold, and obtain the optimized probability failure risk response surface distribution of the life-limited component. Step S6. Weight the optimized probabilistic failure risk response surface distribution of the life-limited component to obtain the probabilistic failure risk of the life-limited component; In step S3, the scattered points to be analyzed are selected from the finite element nodes of each initial partition. Specifically, the node containing the maximum stress value in each initial partition is selected as the scattered point to be analyzed.

2. The method for probabilistic failure risk analysis of life-limited components based on scattered point fitting response surface according to claim 1, characterized in that, Step S1 specifically includes: Obtain the parameter conditions of the life-limiting component under different boundary conditions; the parameter conditions include working fluid information and rotational speed data; Fluid analysis was performed on the life-limited component based on the working fluid information to obtain the volume temperature distribution data of the life-limited component; Stress analysis was performed on the life-limiting component based on body temperature distribution data and rotation speed data to obtain stress values ​​at all finite element nodes of the life-limiting component.

3. The method for probabilistic failure risk analysis of life-limited components based on scattered point fitting response surface according to claim 1, characterized in that, Step S2 specifically includes: Based on the finite element results, the moment of maximum stress of the life-limited component is selected as the basis for initial partitioning. The temperature, stress, and geometric characteristics of the life-limited component are used as partitioning criteria to divide the life-limited component into regions, obtaining an initial partition including finite element nodes.

4. The method for probabilistic failure risk analysis of life-limited components based on scattered point fitting response surface according to claim 1, characterized in that, Step S4 specifically includes: By selecting a fitting method, the probabilistic failure risk on each finite element node, except for the scatter points selected for analysis in step S3, is fitted to obtain the risk value of the finite element node without probabilistic failure risk calculation, thereby obtaining the probabilistic failure risk response surface distribution of the life-limited component.

5. The method for probabilistic failure risk analysis of life-limited components based on scattered point fitting response surface according to claim 1, characterized in that, The fitting method in step S4 is specifically a triangle-based linear interpolation method.

6. The method for probabilistic failure risk analysis of life-limited components based on scattered point fitting response surface according to claim 1, characterized in that, Step S6 specifically includes: By establishing quadrilateral cells for adjacent elements in the optimized probabilistic failure risk response surface distribution of the life-limited component, weighting is performed on the quadrilateral cells, and the risk values ​​of each quadrilateral cell are summarized to obtain the probabilistic failure risk value of the entire life-limited component.

7. The method for probabilistic failure risk analysis of life-limited components based on scattered point fitting response surface according to claim 1, characterized in that, The life-limiting component mentioned is a life-limiting component for aero engines.

8. The method for probabilistic failure risk analysis of life-limited components based on scattered point fitting response surface according to claim 7, characterized in that, The life-limiting component of the aero-engine is the compressor disc of the aero-engine.

9. The method for probabilistic failure risk analysis of life-limited components based on scattered point fitting response surfaces according to claim 8, characterized in that, The compressor disk of the aero-engine is made of titanium alloy.

Images