A reliability evaluation method for the manufacturing and processing of turbine pump blades

Abstract

The present invention discloses a method for evaluating the manufacturing and processing reliability of turbine pump blades, which includes: analyzing and determining the processing methods that affect the life and reliability of turbine pump blades, extracting processing defects and clustering them, using modeling techniques to establish a mechanics-processing defect model and conducting finite element analysis, obtaining the weight parameters corresponding to different defect types through neural network training based on the analyzed data, and finally using the fuzzy comprehensive evaluation method to predict the expected life of the blade under the condition of processing defects and the influence degree of the processing defects on the blade reliability. The present invention has generality and high efficiency, is applicable to the evaluation of the processing reliability of blades and similar parts, has a simple theory, simple steps, and high solution efficiency, and can be widely applied to engineering practice.

Description

Technical Field

[0001] The present invention relates to a method for evaluating the machining reliability in the machining field, and particularly to a method for evaluating the machining reliability of turbine pump blades. Background Art

[0002] The turbine pump system is one of the most important components in modern liquid rocket engines and is applied to the propulsion system of large launch vehicle systems. The turbine pump blade is one of the core components of the turbine pump rotor system. Its function and working characteristics determine that the blade is a part with complex shape, large size span, severe stress, and maximum load in the engine. Therefore, in order to meet the requirements of high performance, working safety, reliability, and life of the engine, it is very necessary to conduct research on the failure probability and reliability of turbine blades from multiple aspects.

[0003] In the research on the failure and reliability of blades, the current work mainly focuses on the design and working environment of the blades, and the actual fault diagnosis work is also carried out according to this thinking. In most cases, only when the data is insufficient to prove that the failure is caused by improper design, operation, or installation, the machining process is investigated as the cause of the failure. However, defects or damages caused by machining may lead to changes in the blade surface, affect the aerodynamic efficiency of the blade, and in severe cases, failures and malfunctions may occur. Therefore, the machining method and manufacturing level of the blade directly determine the machining accuracy and integrity of such parts, that is, to a large extent, they decisively affect the efficiency and reliability of the engine.

[0004] In summary, it is necessary to propose a method for evaluating the machining reliability of turbine pump blades from the perspective of manufacturing and machining. In this method, a reasonable influence model for the mechanical properties and reliability analysis of the blade should be established, and as many influencing factors as possible should be included to make the analysis results closer to the actual situation. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for evaluating the machining reliability of turbine pump blades, which uses comprehensive fuzzy evaluation to evaluate the reliability of the results generated by simulating the constructed blade defect-mechanics model, simplifies the solution process, and can quickly obtain conclusions.

[0006] The present invention proposes a method for evaluating the machining reliability of turbine pump blades, including the following steps:

[0007] Step S101: Analyze and determine the machining methods that affect the reliability of turbine pump blades, extract machining defects, and then perform clustering;

[0008] Step S102: Construct a mechanics-defect model of the turbine pump blade, analyze the influence of defects of different types and positions on the blade state, and based on the analysis results, use neural network theory to train the weight parameters corresponding to different types of defects;

[0009] Step S103: Establish a factor set with processing defects as evaluation factors, divide the influencing factor levels as the evaluation set, and form a factor level set;

[0010] Step S104: Calculate the weight values of each defect factor to establish a factor weight set;

[0011] Step S105: Conduct single-factor evaluation to obtain a single-factor evaluation vector;

[0012] Step S106: Construct a single-factor judgment matrix with each single-factor evaluation vector, and perform fuzzy synthesis on the factor weight set and the single-factor judgment matrix to obtain a fuzzy comprehensive judgment set corresponding to all processing defects.

[0013] Step S107: Determine the fatigue life interval for one start-stop cycle of the blade, perform calculation and processing on the blade fatigue life and the fuzzy comprehensive judgment set, obtain the fuzzy fatigue life value of the blade under the current blade defects, and calculate the reduction amplitude of the blade life under the existence of processing defects, that is, the influence degree of the processing defects on the reliability of the blade.

[0014] Among them, Step S101 specifically includes: determining the processing methods affecting the reliability of the turbine pump blade through investigation and statistical analysis, and determining the representative processing defects of the processing methods. According to the distribution position dimensions, topological characteristics of the representative processing defects, and the resulting blade mechanical properties and reliability results, perform fuzzy clustering analysis on the representative processing defects within a certain distribution and size range according to their topological characteristics.

[0015] Step S102 specifically includes: using modeling technology to set the type and position of defects on the three-dimensional blade model, establishing a mechanical-defect three-dimensional model of the turbine pump blade, and performing finite element simulation analysis based on the model to obtain the mechanical influence of defects at different positions and types on the blade.

[0016] In Step S103, the specific method of establishing a factor set with processing defects as evaluation factors, dividing the influencing factor levels as the evaluation set, and forming a factor level set is as follows:

[0017] Combine various factors that may affect the judgment of blade failure or reliability into a set U,

[0018] U = (u1 u2 … u n )

[0019] where: The element u i (i = 1, 2 …, n) in the set U is each defect influencing factor,

[0020] Divide each factor into several levels, and these several levels form the evaluation set, then the factor level set:

[0021] u i = (u i1 u i2 …u ip ),

[0022] where: u ij (i = 1, 2…, n; j = 1, 2…, p) is the j-th level of the i-th factor.

[0023] In step S104, the specific method for calculating the weight values of each defect factor to establish the factor weight set is as follows:

[0024] According to the degree to which each factor u i affects the reliability of the blade, a corresponding weight a i (i = 1, 2…, n) is assigned to each factor u i to obtain the factor weight set A, then A = (a1 a2 … a n ).

[0025] The calculation method for the weight values of factors at different levels is as follows:

[0026] where: w ij is the weight value of each factor, and w ij is determined by calculation through the following formula:

[0027]

[0028] where, is the stress / strain value under ideal conditions without defects in the blade calculated by finite element simulation analysis, and s ij is the stress / strain value obtained by simulation calculation under different levels of defect severity of each processing factor.

[0029] In step S105, the specific method for obtaining the single-factor evaluation vector through single-factor evaluation is as follows: Obtain the membership degrees of each level of the single factor to obtain the single-factor evaluation vector. When considering the influence of the j-th level u ij of the i-th factor for evaluation, the membership degree of the blade fatigue life to the k-th element in the evaluation set is r ijk (i = 1, 2…, n; j = 1, 2…, p; k = 1, 2,…, m), and its value is equal to the weight parameter obtained from neural network training in this case. Thus, the single-factor evaluation vector R ij corresponding to each level of the i-th factor can be obtained, that is, R ij = (r ij1 r ij2 … r ijm ).

[0030] In step S106, it includes: forming a single-factor evaluation matrix R with each single-factor evaluation vector, and obtaining a fuzzy comprehensive evaluation set B corresponding to all processing defects by fuzzy synthesis of the factor weight set A and the single-factor evaluation matrix R.

[0031] The calculation method of the fuzzy comprehensive evaluation set is:

[0032] Adopt operator, expressed as:

[0033] Where: is the operator symbol, A is the factor weight set, a ij is an element in A, R is the single-factor evaluation vector, r ijk is an element in R, b ik is an element in B.

[0034] Step S107 includes:

[0035] Determine the fatigue life interval for one start-stop cycle of the blade, equally divide it by linear interpolation according to the number of factors, and the result is represented by V, that is, V = (v1 v2 … v n );

[0036] The fuzzy fatigue life value of the blade in the presence of the current blade defect:

[0037]

[0038] The degree of influence of processing defects on the reliability of the blade:

[0039] L p = 1 - L b ,

[0040]

[0041] Where: L p is the reduction amplitude of the blade life, that is, the degree of influence of processing defects on the reliability of the blade, H is the fuzzy fatigue life value of the blade in the presence of the current blade defect, v k is an element in V, b k is an element in B, and V max is the maximum fatigue life value.

[0042] Compared with the prior art, the present invention has at least the following beneficial effects:

[0043] The method of the present invention is universal and efficient, and is applicable to the processing reliability evaluation of blades and similar parts. First, the present invention analyzes and determines the processing methods that affect the life and reliability of the turbine pump blades, extracts the processing defects and then clusters them. Then, a mechanical-processing defect model is established using modeling technology and finite element analysis is carried out. Based on the analysis data, the weight parameters corresponding to different types of defects are obtained through neural network training. Finally, the fuzzy comprehensive evaluation method is used to predict the expected life and reliability influence degree of the blades in the case of processing defects. The theory is simple, the steps are convenient, the solution efficiency is high, and it can be widely applied in engineering practice. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flowchart of the present invention;

[0045] Figure 2 is a schematic diagram of the three-dimensional model of the blade in an embodiment of the present invention;

[0046] Figure 3 is a schematic diagram of the position of the processing defect of the blade in an embodiment of the present invention;

[0047] Figure 4 is a schematic diagram of the crack defect constructed at the A-2-y position of the blade in an embodiment of the present invention;

[0048] Description of the reference numerals: the bottom of the blade basin near the exhaust edge 1, the middle of the blade back near the intake edge 2, the cracked blade body 3. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the scope of protection of the present application.

[0050] In a specific embodiment, as Figure 1 shown, a method for evaluating the manufacturing and processing reliability of turbine pump blades provided by the present invention includes the following steps:

[0051] S101: Analyze and determine the processing methods that affect the reliability of the turbine pump blades, extract the processing defects and then cluster them. The specific method is as follows:

[0052] In the embodiment, first, the processing methods for manufacturing and processing this type of blade are investigated and statistically analyzed:

[0053] The current blade processing methods and techniques can be mainly classified into precision forging, precision casting, grinding, milling, and special processing techniques that use chemical, physical, or electrochemical methods to process workpiece materials. Each processing method and technique may leave processing defects on the blade.

[0054] Specifically, they include: (1) Surface cracks, folds, lack of material, misalignment, insufficient die forging, surface pits, surface bubbles, and orange peel-like surfaces and other external defects may occur during forging; internal defects such as shrinkage cavities, porosity, white spots, disordered forging flow lines, segregation, coarse grains, stone-like fractures, and foreign metal inclusions; and performance defects such as unqualified plasticity, toughness, or fatigue performance. (2) Defects such as pores, pinholes, porosity, inclusions, slag inclusions, cracks, and segregation may occur during casting. (3) Grinding burns, grinding residual stresses, grinding cracks, and grinding chatter marks and other defects may occur during grinding. (4) Surface roughness, deep groove marks, scaly burrs, and surface mechanical damage and other defects may occur during milling.

[0055] Secondly, extract representative processing defects: Use the method of fuzzy clustering to conduct fuzzy clustering analysis on various types of defects. According to the distribution position dimensions, topological features, and possible impacts on the mechanical properties and reliability results of the blade, defects within a certain distribution and size range are processed by fuzzy clustering according to their topological features. These defects can be approximately classified as: dimensional accuracy defects or deficiencies of parts; defects distributed on the workpiece surface such as surface roughness; defects close to the surface (subsurface) such as microcracks; defects distributed in the part body such as shrinkage cavities or porosity; and fiber distortion in the part body material.

[0056] Step S102: Construct a mechanical-defect model of the turbine pump blade and analyze the effects of different types and positions of defects on the blade state.

[0057] In this embodiment, it is preferably to use UG or ANSYS software to establish a mechanical-defect model and conduct finite element simulation analysis, and perform neural network training on the results to obtain weight parameters corresponding to different defect types.

[0058] Construction of the processing defect model: Use UG software or ANSYS to construct defects, and import the three-dimensional model of the blade in stp format, as Figure 2 shown. First, set the occurrence positions of blade defects. For surface defects, they are divided into 3 types: the blade suction side is designated as A, the pressure side is B; the blade tip is designated as 1, the middle part is designated as 2, and the root is designated as 3; the side near the exhaust edge is x, the middle part is y, and the side near the intake edge is z. Then, a total of 2×3×3 = 18 position nodes of the three types of position defects can represent the situations of each part of the blade.

[0059] Refer to Figure 3As shown in Table 1 below, the defect code A-3-x corresponds to the position near the exhaust side 1 at the bottom of the blade basin, and the defect code B-2-z corresponds to the position near the intake side 2 in the middle of the blade back. The specific division is shown in Table 1 below.

[0060] Table 1 Serial numbers corresponding to defects at each position

[0061]

[0062]

[0063] Secondly, set the defect form in step S101 in UG or ANSYS. After establishing the defect model at the corresponding position, import it into the Workbench-Static Structural module for subsequent finite element simulation analysis.

[0064] In this embodiment, referring to Figure 4 , establish a crack defect model. After importing the model into Workbench, create an arbitrary-shaped crack on the blade back or blade basin surface in the 3D sketch in its SCDM; form a sheet body, i.e., a crack sheet body, through the shear between surfaces and name it; assign a value to the sheet body thickness, assuming Thickness = 1 mm, and construct a crack defect at position A-2-y, i.e., in the middle of the blade basin near the middle.

[0065] Conduct finite element simulation analysis on the machining defect model: define the material parameters, elastic modulus, Poisson's ratio, density, and thermal expansion coefficient of the blade model; perform mesh division on the main body; establish a local coordinate system for the defect at the corresponding position; insert the crack module Fracture in Model, select Arbitrary Crack, set the dimension parameters in this module, select the object and the local coordinate system, and after setting, perform mesh division on the local area; set the centrifugal tensile stress as the main working load and given a rotational speed of 18,000 rpm, set constraints at the tenon, and then solve. The result output form is the overall deformation, maximum stress, and maximum strain of different machining defect models. List the data obtained from multiple groups of simulation analyses. Table 2 shows the simulation results of cracks at different positions at the same level.

[0066] Table 2 Finite element simulation analysis results of the blade

[0067]

[0068] Based on the analysis of experiments and results, establish a machining defect model corresponding to the blade material characteristics and manufacturing and processing technologies, thereby establishing a reliability database. Use the neural network theory to train the weight parameters corresponding to different types of defects, obtain a discrimination model for machining defects that may occur in the blade material and manufacturing and processing technologies, and then conduct subsequent analysis on this basis.

[0069] Step S103: Establish a factor set with the processing defects as evaluation factors and establish an evaluation set to form a factor grade set.

[0070] In this embodiment, the factor set affecting the service performance and failure life of the blade can be taken as:

[0071] U = (u1 u2 … u 17 u 18 )

[0072] Where: u1 - u 18 represents the factor set of surface defects (cracks).

[0073] The defect degrees of the blade influencing factors are successively divided into 5 grades: very large, relatively large, general or medium, relatively small, and very small. These five grades constitute the evaluation set. To provide reference values for the subsequent judgment set, fuzzy function values are assigned to each grade. Among them, very large v1 indicates that the defect degree is very serious, that is, such defects will necessarily cause failures, and its corresponding fuzzy function value is 1; very small v5 indicates that the defect degree is slight, that is, the failure probability caused by such defects can be ignored; the other defect grades are successively divided into relatively large v2 corresponding to the fuzzy value of 0.75, general or medium v3 corresponding to the fuzzy value of 0.5, and relatively small v4 corresponding to the fuzzy value of 0.25. Then the factor grade set is shown in Table 3.

[0074] Table 3 Design of the influencing factor set of blade defects

[0075]

[0076] Step S104: Calculate the weight values of each defect factor to establish a factor weight set.

[0077] In this embodiment, according to the blade defect grades set in the foregoing analysis, mechanical analysis is carried out to obtain its mechanical characteristics, and its mechanical life characteristics are calculated according to the mechanical characteristics. The calculation basis of the weight value is:

[0078]

[0079] Where: w ij is the weight value of each factor. Here, the weight value is determined according to the stress / strain value calculated from the blade defects.

[0080] That is:

[0081] Where: is the stress / strain value of the blade in the ideal defect-free state; s ijThe stress / strain values calculated under different levels of defect severity for each processing factor. Through simulation, the maximum stress in the defect-free state is 617.6 MPa, and the maximum strain is 0.0034874. The factor weight set A is calculated as shown in Table 4.

[0082] Table 4 Weight values for each defect location

[0083]

[0084] Step S105: Conduct single-factor evaluation to obtain a single-factor evaluation vector.

[0085] In this embodiment, when considering the influence of the j-th level u ij of the i-th factor for evaluation, the membership degree of the blade fatigue life to the k-th element in the evaluation set is r ijk (i = 1, 2, …, n; j = 1, 2, …, p; k = 1, 2, …, m), and its value is equal to the weight parameter value obtained from the neural network training in Step S102. Thus, the single-factor evaluation vector R ij .

[0086] In this embodiment, a defect of the crack type is involved. In this case, the evaluation set for the three-level crack defect is:

[0087] R = [0.75 0.5 0.25]

[0088] Step S106: Combine each single-factor evaluation vector to form a single-factor evaluation matrix, and perform fuzzy synthesis on the factor weight set and the single-factor evaluation matrix to obtain the fuzzy comprehensive evaluation set corresponding to all processing defects, that is, conduct fuzzy comprehensive evaluation.

[0089] In this embodiment, since only crack defects are involved, the single-factor evaluation matrix R here is the same as the single-factor evaluation vector in Step S105. Considering the influence of all levels of processing factors in the evaluation, the fuzzy comprehensive evaluation set corresponding to all processing factors is obtained:

[0090]

[0091] Using the operator, the first-level comprehensive evaluation result is:

[0092]

[0093] Step 107: Determine the fatigue life interval for one start-stop cycle of the blade. Calculate and process the blade fatigue life and the fuzzy comprehensive evaluation set to obtain the fuzzy fatigue life value of the blade under the current existence of blade defects, and calculate the reduction amplitude of the blade life under the existence of machining defects, that is, the influence degree of machining defects on the reliability of the blade.

[0094] In this embodiment, the fatigue life interval for one start-stop cycle of this type of blade is:

[0095] V = [1600 2000 2400] h

[0096] After adopting the weighted average evaluation method, the fuzzy fatigue life value H of the blade can be obtained under the condition that all the assumed blade defects exist currently.

[0097]

[0098] Explore the reduction amplitude L of the life p , that is, the influence degree of reliability caused by machining defects, and its calculation formula is:

[0099]

[0100] L p = 1 - L b = 22.18%

[0101] In summary, in the present invention, after determining the machining methods affecting the life and reliability of the turbine pump blade and extracting machining defects, the ANSYS software is used to establish a mechanics-defect model and perform finite element analysis. The weight parameters corresponding to different types of defects are obtained by neural network training. Finally, the fuzzy comprehensive evaluation method is used to predict the expected life and the influence degree of reliability of the blade in the case of machining defects. The theory is simple, the steps are few, the solution efficiency is high, and it has high engineering application value.

[0102] The specific implementation manners of the present application have been described above, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0103] In the description of the present application, it should be understood that the orientation or positional relationship indicated by terms such as "upper", "lower", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation on the present application. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to improve. Any reference signs in the claims should not be construed as limiting the scope.

Claims

1. A reliability evaluation method for the manufacturing and processing of turbine pump blades, characterized in that The steps include the following: Analyze and determine the processing methods affecting the reliability of turbine pump blades, extract processing defects, and then perform clustering. The specific method for analyzing and determining the processing methods affecting the reliability of turbine pump blades, extracting processing defects, and then performing clustering is as follows: Determine the processing methods affecting the reliability of turbine pump blades through investigation and statistical analysis, and determine the representative processing defects of the processing methods. According to the distribution position dimensions, topological features, and the resulting blade mechanical properties and reliability results of the representative processing defects, perform fuzzy clustering analysis on the representative processing defects within a certain distribution and size range according to their topological features; Construct a mechanical-processing defect model of the turbine pump blade, analyze the influence of defects of different types and positions on the blade state, and based on the analysis results, use neural network theory to train the weight parameters corresponding to different types of defects; Establish a factor set with processing defects as evaluation factors, and divide the influence factor levels as the evaluation set to form a factor level set; Calculate the weight values of each processing defect factor to establish a factor weight set; Conduct single-factor evaluation to obtain a single-factor evaluation vector; Form a single-factor judgment matrix with each of the single-factor evaluation vectors, and perform fuzzy synthesis of the factor weight set and the single-factor judgment matrix to obtain a fuzzy comprehensive judgment set corresponding to all processing defects; Determine the fatigue life interval for one start-stop cycle of the blade, perform calculation processing on the blade fatigue life and the fuzzy comprehensive judgment set, obtain the fuzzy fatigue life value of the blade under the current presence of blade defects, and calculate the reduction amplitude of the blade life under the presence of processing defects, that is, the degree of influence of the processing defects on the reliability of the blade.

2. The reliability evaluation method for manufacturing and processing of turbine pump blades according to claim 1, characterized in that The specific method for constructing a mechanical-processing defect model of the turbine pump blade and analyzing the influence of defects of different types and positions on the blade state is as follows: Use modeling technology to set the types and positions of defects on the three-dimensional model of the blade, establish a three-dimensional mechanical-processing defect model of the turbine pump blade, and based on the model, perform finite element simulation analysis to obtain the mechanical influence of processing defects of different positions and types on the blade.

3. A reliability evaluation method for the manufacturing and processing of turbine pump blades according to claim 1, characterized in that The specific method for establishing an evaluation factor set with processing defects as evaluation factors and establishing levels as the evaluation set to form a factor level set is as follows: Form a set U by combining various factors that may affect the judgment of blade failure or reliability; U = (u1 u2 … u n ) Among them: the element u in the set U i is each defect influencing factor, i = 1, 2..., n Each factor is further divided into several levels, and the several levels form the evaluation set, so the factor level set: u i = (u i1 u i2 … u ip ), where: u ij is the j-th level of the i-th factor, where i = 1, 2, …, n; j = 1, 2, …, p.

4. A reliability evaluation method for manufacturing and processing of turbine pump blades according to claim 3, characterized in that, The specific method for calculating the weight values of each defect factor to establish a factor weight set is as follows: According to each factor u i To the extent of affecting the reliability of the blade, for each of the factors u i A corresponding weight ai is assigned i , i = 1, 2…, n, to obtain the factor weight set A, then A=(a1 a2…a n ), Calculation method for factor weight values at different levels: where: w ij is the weight value of each factor, w ij is determined by calculation through the following formula: Among them, is the stress / strain value calculated by finite element simulation analysis under the ideal state of the blade without defects, s ij is the stress / strain value obtained by simulation calculation under different levels of defect degrees of each processing factor.

5. A reliability evaluation method for manufacturing and processing of turbine pump blades according to claim 4, characterized in that, The specific method for conducting single-factor evaluation to obtain a single-factor evaluation vector is as follows: Obtain the membership degrees of each level of the single factor to obtain a single-factor evaluation vector.

6. The reliability evaluation method for manufacturing and processing of turbine pump blades according to claim 5, wherein, The specific method for obtaining the membership degree of each level of a single factor to obtain a single-factor evaluation vector is as follows: When considering the influence of the j-th level u of the i-th factor for evaluation, the membership degree of the blade fatigue life to the k-th element in the evaluation set is r ij , (i = 1, 2..., n; j = 1, 2..., p; k = 1, 2,..., m), and its value is equal to the weight parameter obtained from neural network training in this case. Thus, the single-factor evaluation vector R ijk corresponding to each level of the i-th factor can be obtained, that is, R ij = (r ij r ij1 r ij2 ... r ijm ).

7. A reliability evaluation method for manufacturing and processing of turbine pump blades according to claim 6, characterized in that, The method for performing fuzzy synthesis of the factor weight set and the single-factor judgment matrix to obtain a fuzzy comprehensive judgment set corresponding to all processing defects is as follows: Fuzzy comprehensive evaluation set: The M(+, ○) operator is adopted, which is specifically represented as: Among them: ○ is the operator symbol, A is the factor weight set, and a ij is an element in A, R is the single-factor evaluation matrix, and r ijk is an element in R, b ik is an element in B, 8. A reliability evaluation method for manufacturing and processing of turbine pump blades according to claim 7, characterized in that, The specific method for determining the fatigue life interval of one start-stop cycle of the blade, calculating and processing the blade fatigue life and the fuzzy comprehensive evaluation set, and obtaining the fuzzy fatigue life value of the blade under the current blade defect is as follows: Determine the fatigue life interval of one start-stop cycle of the blade, linearly interpolate and equally divide according to the number of factors, and the result is represented by V, that is, V = (v1 v2 … v n ) Fuzzy fatigue life value of the blade: where: v k is an element in V, b k is an element in B.

9. A reliability evaluation method for manufacturing and processing of turbine pump blades according to claim 8, characterized in that, The calculation method for the degree of influence of the processing defects on the reliability of the blade is as follows: L p = 1 - L b , Where: L p is the reduction amplitude of the blade life, that is, the influence degree of the processing defect on the blade reliability. H is the fuzzy fatigue life value of the blade under the current blade defect, and V max is the maximum fatigue life value.

Images