Proxy model method for turbine casing reliability analysis under evidence uncertainty
The structural reliability evaluation model of the turbine receiver is constructed through the agent model method, which solves the problem of turbine receiver reliability analysis under evidence uncertainty, and achieves a robust and rapid failure probability assessment, which improves the accuracy and efficiency of the analysis.
Patent Information
- Application Number
- CN202311359650.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-10-19
- Publication Date
- 2024-11-26
AI Technical Summary
Under the uncertainty of evidence, traditional optimization methods are difficult to effectively analyze the reliability of turbine receivers, and are prone to local optimization or inefficiency, and lack a robust and efficient analysis plan.
The proxy model method is used to construct a structural reliability evaluation model by obtaining input variables, generate an initial proxy model based on functional functions, and update the model until converges the model by training the sample set to converge, and determine the limit on the failure probability of the turbine receiver.
The critical value of the failure probability of the turbine receiver structure is achieved under evidence uncertainty, and the reliability level of the turbine receiver is reasonably evaluated, providing a reliability analysis solution that takes into account evidence uncertainty.
Smart Images

Figure CN119026256A_ABST
Abstract
Description
Background Art
[0002] Uncertainty widely exists in the structural design process of complex mechanical products. For example, the uncertainty factors in mechanical product design include design variable uncertainty, parameter uncertainty, model uncertainty, measurement uncertainty, etc. In the absence of sample information, it is difficult to characterize the uncertainty of structural input using random theory, while evidence theory is a more universal uncertainty characterization theory under small sample information.
[0003] For complex structures such as turbine casings, the amount of input sample information is relatively small. When solving the reliability limit of the turbine casing structure, multiple optimization problems need to be solved. Traditional optimization methods are prone to fall into local optimality or have low efficiency problems, and are difficult to be directly applied to the reliability analysis of turbine casings under evidence uncertainty.
[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention
[0005] The purpose of the present disclosure is to provide a proxy model method for turbine case reliability analysis under evidential uncertainty, a proxy model device, an electronic device and a computer-readable storage medium for turbine case reliability analysis under evidential uncertainty, thereby at least to a certain extent overcoming the problem of the lack of a robust and efficient turbine case reliability analysis scheme under evidential uncertainty in related schemes.
[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present invention.
[0007] According to a first aspect of the present disclosure, a proxy model method for turbine casing reliability analysis under evidence uncertainty is provided, comprising: obtaining input variables corresponding to the turbine casing, constructing a structural reliability assessment model of the turbine casing based on the input variables, the structural reliability assessment model being constructed based on a functional function of the input variables; constructing an initial proxy model of the functional function, and generating a training sample set of the initial proxy model based on function extreme value samples of the functional function; updating the initial proxy model based on the training sample set until the initial proxy model converges to obtain a proxy model of the functional function; and determining a failure probability limit of the turbine casing being in a failure state according to the proxy model.
[0008] In an exemplary embodiment of the present disclosure, constructing the structural reliability assessment model of the turbine casing based on the input variables includes: acquiring test data corresponding to the input variables; constructing an evidence characterization model of the input variables based on the test data, the evidence characterization model including a set of variable focus elements corresponding to the input variables, and a basic probability distribution corresponding to each variable focus element; constructing a reliability boundary function of the turbine casing based on a plurality of the variable focus elements and the respective corresponding basic probability distributions; and generating the structural reliability assessment model according to the reliability boundary function.
[0009] In an exemplary embodiment of the present disclosure, the reliability boundary function of the turbine case is constructed based on the multiple variable focal elements and the respective corresponding basic probability allocations, including: constructing a functional function of the turbine case based on the input variables, the functional function being used to perform reliability analysis on the turbine case; determining a failure domain corresponding to the input variables based on the functional function; and constructing the reliability boundary function based on the failure domain, the multiple variable focal elements and the respective corresponding basic probability allocations.
[0010] In an exemplary embodiment of the present disclosure, the reliability boundary function includes a reliability upper bound function and a reliability lower bound function, and the reliability boundary function is constructed based on the failure domain, the plurality of variable focus elements and the respective corresponding basic probability allocations, including: determining a first variable focus element from the variable focus element set, the first variable focus element is a variable focus element included in the failure domain; constructing the reliability lower bound function based on the first variable focus element and the basic probability allocation corresponding to the first variable focus element; determining a second variable focus element from the variable focus element set, the second variable focus element is a variable focus element whose intersection with the failure domain is a non-empty set; constructing the reliability upper bound function based on the second variable focus element and the basic probability allocation corresponding to the second variable focus element.
[0011] In an exemplary embodiment of the present disclosure, constructing the initial proxy model of the functional function includes: determining the initial training samples using a low-bias sampling method based on the variable value space of the input variables; and constructing the initial proxy model based on the initial training samples and model construction instructions.
[0012] In an exemplary embodiment of the present disclosure, the function extreme value samples of the functional function are used to generate a training sample set of the initial proxy model, including: based on the variable focus set of the input variables, using a random sampling method to determine the function extreme value samples of the functional function; based on the determined function extreme value samples, constructing the training sample set.
[0013] In an exemplary embodiment of the present disclosure, the initial proxy model is updated based on the training sample set until the initial proxy model tends to converge to obtain the proxy model of the functional function, including: dividing the training sample set into a first sample subset and a second sample subset, the first sample subset consisting of sample data determined by sample output symbols; obtaining the current state of the model, and if the current state of the model is in an unconverged state, selecting new training sample points through a sample classification learning function to obtain an updated sample point set; updating the initial proxy model based on the updated sample point set until the initial proxy model tends to converge to obtain the proxy model.
[0014] In an exemplary embodiment of the present disclosure, the method of selecting new training sample points through a sample classification learning function to obtain an updated sample point set includes: determining a symbol prediction error probability of an output result of the functional function based on the sample classification learning function; obtaining a predicted mean of the functional function; and determining the updated sample point set based on the predicted mean of the functional function, the symbol prediction error probability and a maximum independent variable point set function.
[0015] In an exemplary embodiment of the present disclosure, determining the failure probability limit of the turbine casing in a failure state based on the proxy model includes: determining the failure domain indicator function of the input variable in each variable focal element based on the proxy model; constructing a failure probability boundary function of the turbine casing based on the failure domain indicator function; and determining the failure probability limit based on the failure probability boundary function.
[0016] In an exemplary embodiment of the present disclosure, the failure domain indication function includes a minimum failure domain indication function and a maximum failure domain indication function, and constructing the failure probability boundary function of the turbine casing according to the failure domain indication function includes: constructing the failure probability upper bound function according to the minimum failure domain indication function; and constructing the failure probability lower bound function according to the maximum failure domain indication function.
[0017] According to a second aspect of the present disclosure, there is provided a proxy model device for reliability analysis of a turbine casing under evidence uncertainty, comprising: a reliability model construction module, for acquiring input variables corresponding to the turbine casing, and constructing a structural reliability assessment model of the turbine casing based on the input variables, wherein the structural reliability assessment model is constructed based on a functional function of the input variables; a sample set construction module, for constructing an initial proxy model of the functional function, and generating a training sample set of the initial proxy model based on function extreme value samples of the functional function; a model updating module, for updating the initial proxy model based on the training sample set until the initial proxy model converges to obtain a proxy model of the functional function; and a reliability analysis module, for determining, based on the proxy model, a failure probability limit of the turbine casing being in a failure state.
[0018] According to a third aspect of the present disclosure, there is provided an electronic device, comprising: a processor; and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, a proxy model method for turbine casing reliability analysis under evidential uncertainty as described in any one of the above items is implemented.
[0019] According to a fourth aspect of the present disclosure, there is provided a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the proxy model method for turbine casing reliability analysis under evidential uncertainty as described in any one of the above items is implemented.
[0020] The technical solution provided by the present disclosure may have the following beneficial effects:
[0021] The proxy model method for turbine case reliability analysis under evidential uncertainty in the exemplary embodiment of the present disclosure provides a proxy model scheme for reliability analysis of the turbine case mechanism taking into account evidential uncertainty based on the uncertainty of the input variables, thereby robustly and quickly solving the critical value of the turbine case structure failure probability under evidential uncertainty, and further reasonably evaluating the reliability level of the turbine case structure.
[0022] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present disclosure, and together with the specification, are used to explain the principles of the present disclosure. Obviously, the drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:
[0024] Figure 1 A flowchart of a proxy model method for turbine case reliability analysis under evidential uncertainty according to an exemplary embodiment of the present disclosure is schematically shown.
[0025] Figure 2 A schematic structural diagram of a turbine case according to an exemplary embodiment of the present disclosure is schematically shown.
[0026] Figure 3 A finite element model diagram of a turbine case according to an exemplary embodiment of the present disclosure is schematically shown.
[0027] Figure 4 A stress contour diagram of a turbine case according to an exemplary embodiment of the present disclosure is schematically shown.
[0028] Figure 5 A block diagram schematically illustrates a proxy model apparatus for turbine case reliability analysis under evidential uncertainty according to an exemplary embodiment of the present disclosure.
[0029] Figure 6 A block diagram of an electronic device according to an exemplary embodiment of the present disclosure is schematically shown.
[0030] Figure 7 A schematic diagram of a computer-readable storage medium according to an exemplary embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION
[0031] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be comprehensive and complete and will fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.
[0032] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known structures, methods, devices, implementations, materials or operations are not shown or described in detail to avoid blurring the various aspects of the present disclosure.
[0033] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or these functional entities or parts of functional entities may be implemented in one or more software hardened modules, or these functional entities may be implemented in different networks and / or processor devices and / or microcontroller devices.
[0034] When solving the limit of turbine case structure reliability, it is necessary to establish the limit model of turbine case reliability analysis in advance and solve multiple optimization problems based on the limit model. Traditional optimization methods are prone to fall into local optimality or have low efficiency problems, and are difficult to be directly applied to turbine case reliability analysis under evidence uncertainty. Therefore, there is currently a lack of robust and efficient turbine case reliability analysis methods under evidence uncertainty.
[0035] Based on this, in this example embodiment, a proxy model method for turbine casing reliability analysis under evidence uncertainty is first provided. The proxy model method for turbine casing reliability analysis under evidence uncertainty of the present disclosure can be implemented by a server, and the method described in the present disclosure can also be implemented by a terminal device, wherein the terminals described in the present disclosure may include mobile terminals such as mobile phones, tablet computers, laptops, PDAs, personal digital assistants (PDAs), and fixed terminals such as desktop computers. Figure 1 The schematic diagram schematically shows the process of the proxy model method for turbine case reliability analysis under evidence uncertainty according to some embodiments of the present disclosure. Figure 1 The surrogate model method for turbine case reliability analysis under evidential uncertainty may include the following steps:
[0036] Step S110, obtaining input variables corresponding to the turbine casing, and constructing a structural reliability assessment model of the turbine casing based on the input variables, wherein the structural reliability assessment model is constructed based on a functional function of the input variables.
[0037] According to some exemplary embodiments of the present disclosure, the input variables may be variable factors that affect the reliability of the turbine casing structure. For example, the input variables may be generated based on uncertain factors such as the material properties of the turbine casing and the external environment in which the turbine casing is located. The structural reliability assessment model may be a mathematical model for analyzing the change in the structural reliability of the turbine casing under the influence of multiple uncertain input variables. The functional function may be a function constructed based on the correspondence between the input variables of the turbine casing structure and the reliability boundary value.
[0038] Before analyzing the reliability of the turbine casing structure, multiple input variables that affect the reliability of the turbine casing structure can be determined, such as the material properties of the turbine casing, the external environment in which the turbine casing is located, and other uncertain factors as input variables for reliability analysis. After determining multiple input variables, a structural reliability assessment model of the turbine casing can be constructed based on the input variables. Before constructing the structural reliability assessment model, a functional function for characterizing the reliability of the turbine casing can be established based on the input variables, and a structural reliability assessment model can be constructed based on the functional function, so as to determine the boundary value for analyzing the failure probability of the turbine casing through the constructed reliability assessment model.
[0039] Step S120, constructing an initial proxy model of the performance function, and generating a training sample set of the initial proxy model based on the function extreme value samples of the performance function.
[0040] According to some exemplary embodiments of the present disclosure, function extreme value samples may be samples consisting of extreme values of a function function solved in multiple focal elements of an input variable. An initial proxy model may be an initially constructed analytical model for regression prediction of a function function, for example, an initial proxy model may be a Kriging model of the function function. A training sample set may be a set consisting of determined function extreme value samples.
[0041] The constructed reliability assessment model includes variable focal units corresponding to multiple input variables and relevant data of basic probability distribution corresponding to each variable focal unit. For the constructed reliability assessment model, sample extraction operation can be performed within the coverage range of the variable value interval to obtain the corresponding initial training samples. Then, the initial proxy model corresponding to the functional function is constructed based on the initial training samples, which is used to predict and analyze the functional function.
[0042] In addition, based on each variable focal element, the function extreme value corresponding to the functional function is solved. For the solved function extreme value, a random sampling technique can be used to extract function extreme value samples from the function extreme value, and a training sample set of the initial proxy model is constructed based on the function extreme value samples, so as to use the training sample set to train and update the initial proxy model.
[0043] Step S130, updating the initial proxy model based on the training sample set until the initial proxy model converges to obtain a proxy model of the functional function.
[0044] According to some exemplary embodiments of the present disclosure, the proxy model may be an analysis model in a converged state obtained after the initial proxy model is trained.
[0045] After selecting a training sample set, the initial proxy model can be updated based on the training sample set. Before performing the model update operation, the samples in the training sample set can be classified. For example, the training sample set can be divided into two categories: sign-determined and sign-uncertain sample output results.
[0046] For the training sample set, a sample classification learning function is used to continuously select new training sample points, and the new training sample points are used to update the initial proxy model until the initial proxy model reaches a convergence condition and tends to converge, thereby obtaining a proxy model of the functional function.
[0047] Step S140: determining the failure probability limit of the turbine casing being in a failure state according to the proxy model.
[0048] According to some exemplary embodiments of the present disclosure, the failure probability limit may be a critical value for determining whether the turbine casing is in a failure state.
[0049] After obtaining the proxy model of the functional function, the failure domain indicator function for determining whether the turbine casing structure is in a failure state can be solved based on the converged proxy model, and then the failure probability boundary function of the turbine casing can be constructed based on the failure domain indicator function to determine the failure probability limit of the turbine casing failure.
[0050] According to the proxy model method for turbine casing reliability analysis under evidential uncertainty in the present example embodiment, a proxy model scheme for reliability analysis of the turbine casing mechanism taking into account evidential uncertainty is provided based on the uncertainty of the input variables, thereby robustly and quickly solving the critical value of the turbine casing structure failure probability under evidential uncertainty, and further reasonably evaluating the reliability level of the turbine casing structure.
[0051] Next, the proxy model method for turbine case reliability analysis under evidence uncertainty in this example embodiment will be further described.
[0052] In an exemplary embodiment of the present disclosure, for step S110, a structural reliability assessment model of a turbine casing is constructed based on input variables, including: obtaining test data corresponding to the input variables; constructing an evidence characterization model of the input variables based on the test data, the evidence characterization model including a set of variable focus elements corresponding to the input variables, and a basic probability distribution corresponding to each variable focus element; constructing a reliability boundary function of the turbine casing based on multiple variable focus elements and their corresponding basic probability distributions; and generating a structural reliability assessment model based on the reliability boundary function.
[0053] Among them, the test data can be relevant data on the effective state of the turbine casing under different variable values of the above-mentioned input variables. The evidence characterization model can be a mathematical model constructed using the test data corresponding to the input variables. The variable focus element set can be a set composed of variable focus elements corresponding to the input variables. The variable focus element can be all value intervals corresponding to the input variables. The reliability boundary function can be a function used to calculate the upper reliability bound and the lower reliability bound of the turbine casing.
[0054] The input variables of the turbine casing may include n input variables x={x1, x2, ..., x n}, for example, the input variables may include elastic modulus E (GPa), Poisson's ratio μ, linear expansion coefficient α (1 / ℃), external pressure P1 (MPa) and internal pressure P2 (MPa), etc. Taking the above variables as input variables as an example, after determining the input variables that affect the reliability of the turbine casing, the failure probability of the turbine casing structure under the above input variables can be determined to obtain the corresponding test data. Based on the above test data or expert experience, x={x1,x2,…,x n}Evidence representation model, the evidence representation model includes n E focal element k (k=1,2,…,n E ) and the corresponding basic probability distribution m E (Α k ).
[0055] On the basis of the evidence characterization model, the reliability boundary function of the turbine casing can be constructed according to the determined focal element variables and the basic probability distribution corresponding to each focal element variable, and then the structural reliability assessment model of the turbine casing can be constructed according to the obtained reliability boundary function. The structural reliability assessment model constructed through the above steps can determine the boundary value of the failure probability of the turbine casing based on the structural reliability assessment model.
[0056] In an exemplary embodiment of the present disclosure, the process of constructing the reliability boundary function is as follows: based on the input variables, a functional function of the turbine casing is constructed, and the functional function is used to perform reliability analysis on the turbine casing; based on the functional function, the failure domain corresponding to the input variable is determined; based on the failure domain, multiple variable focal elements and their corresponding basic probability distributions, a reliability boundary function is constructed.
[0057] The failure domain may be a value range of the input variable when the turbine casing is in a failure state.
[0058] When constructing the reliability boundary function, we can first construct a functional function g(x) for reliability analysis of the turbine casing based on the input variables, and determine the failure domain F of the input variables of the turbine casing based on the functional function. Specifically, F={x:g(x)≤0}. It can be seen from the calculation formula of the failure domain that when the value of g(x) is less than or equal to 0, the turbine casing is in a failure state.
[0059] After calculating the failure domain of the input variables, the reliability boundary function can be constructed based on the failure domain of the input variables, multiple variable focal elements, and their corresponding basic probability distribution. The reliability boundary function obtained in the above way constitutes the structural reliability assessment model of the turbine casing, as shown in Formula 1:
[0060]
[0061] Among them, Α k (k=1,2,…,n E ) can represent the variable focal element corresponding to the input variable; F can represent the failure domain, F = {x:g(x)≤0}, g(x) is the functional function in the reliability analysis of the turbine casing; m E (Α k ) can represent the basic probability distribution corresponding to each variable focal element; Bel(F) can represent the lower bound of the reliability of the turbine casing; Pl(F) can represent the upper bound of the reliability of the turbine casing.
[0062] In an exemplary embodiment of the present disclosure, a reliability boundary function is constructed based on a failure domain, multiple variable focus elements and their respective corresponding basic probability assignments, including: determining a first variable focus element from a variable focus element set, the first variable focus element is a variable focus element included in the failure domain; constructing a reliability lower bound function based on the first variable focus element and the basic probability assignment corresponding to the first variable focus element; determining a second variable focus element from the variable focus element set, the second variable focus element is a variable focus element whose intersection with the failure domain is a non-empty set; constructing a reliability upper bound function based on the second variable focus element and the basic probability assignment corresponding to the second variable focus element.
[0063] The failure probability may be the probability that the turbine casing structure cannot complete the predetermined function within a specified time and under specified conditions. The reliability lower bound function may be a calculation function for characterizing the lower bound of the failure probability of the turbine casing. The reliability upper bound function may be a calculation function for characterizing the upper bound of the failure probability of the turbine casing.
[0064] For the reliability lower bound function, it can be constructed based on the variable focus element included in the failure domain, and the variable focus element included in the failure domain is obtained from the variable focus element set of the input variable as the first variable focus element, that is, the first variable focus element can be expressed as A k∈F. For the first variable focal element obtained, the basic probability distribution corresponding to each first variable focal element can be obtained respectively, so as to construct the reliability lower bound function based on the first variable focal element and the basic probability distribution corresponding to it. The reliability lower bound function is obtained by summing the basic probability distribution corresponding to the first variable focal element, that is, the reliability lower bound function is expressed as
[0065] Similarly, when constructing the reliability upper bound function, it can be based on the second variable focal element. The second variable focal element can be a variable focal element whose intersection with the failure domain is a non-empty set, that is, the second variable focal element is A k ∩F≠φ; then, determine the basic probability distribution corresponding to each second variable focal element, and construct the reliability upper bound function based on the second variable focal element and its corresponding basic probability distribution. The reliability upper bound function is obtained by summing the basic probability distribution corresponding to the second variable focal element, that is, the reliability upper bound function is expressed as
[0066] Through the above processing steps, a reliability boundary function of a proxy model for turbine casing reliability analysis under evidence uncertainty can be constructed, and the reliability boundary functions can be combined to obtain a structural reliability assessment model of the turbine casing for analyzing the reliability of the turbine casing structure.
[0067] In an exemplary embodiment of the present disclosure, for step S120, an initial proxy model of the functional function is constructed, including: determining initial training samples based on the variable value space of the input variables using a low-bias sampling method; and constructing an initial proxy model based on the initial training samples and model construction instructions.
[0068] Among them, the variable value space can be the value space corresponding to the input variable, and the variable value space is composed of the variable focus elements corresponding to the input variable. The low-bias sampling method can be a sampling method that minimizes the sampling process due to a series of factors that do not conform to the random sampling principle. The initial training sample can be a sample obtained by sampling from the variable focus elements corresponding to different input variables using a low-bias sampling method. The model building instruction can be an instruction statement for building an initial proxy model. The initial proxy model can be used to model and predict the failure probability of a turbine casing under the influence of uncertain input variables.
[0069] After the structural reliability assessment model of the turbine case is constructed, a low-bias sampling method can be used to determine the initial training samples used to construct the initial proxy model. The number of initial training samples can be a pre-configured number of samples, for example, the number of initial training samples can be 50, 100, etc. For example, the low-bias sampling method can include sampling methods such as uniformly distributed random number sequence (Sobol sequence) sampling and Latin hypercube sampling.
[0070] After the sample extraction is completed, the model building instruction is used to build the initial proxy model corresponding to the performance function. For example, the model building instruction can be generated based on the Design and Analysis of Computer Experiments (DACE) toolbox, and the DACE toolbox is used to build the proxy model corresponding to the performance function. For example, the proxy model can be a Kriging model. The Kriging model of the performance function g(x) can be expressed as The basis for constructing the proxy model through the above steps is to replace the original function and approach the original function as closely as possible in terms of accuracy.
[0071] In an exemplary embodiment of the present disclosure, for step S120, a training sample set of the initial proxy model is generated based on the function extreme value samples of the functional function, including: based on the variable focus set of the input variables, using a random sampling method to determine the function extreme value samples of the functional function; based on the determined function extreme value samples, constructing a training sample set.
[0072] The random sampling method may be a completely random sampling method for obtaining samples from the overall sample without any grouping, classification, queuing, etc. The function extreme value sample may be a sample composed of the extreme values of the function. The training sample set may be a sample set composed of multiple function extreme value samples.
[0073] For the variable focus set of the input variable, random sampling can be used to select the variable in each focus variable Α. k (k=1,2,…,n E ) determines the function extreme value samples of the functional function, such as N that generates the solution to the extreme value of the functional function k Samples After determining the function extreme value samples, we can build a scale of N based on the function extreme value samples. k Update Kriging model candidate sample pool Through the above processing steps, a training sample set corresponding to the initial proxy model can be constructed, and then a model update operation can be performed based on the training sample set.
[0074] In an exemplary embodiment of the present disclosure, for step S130, the initial proxy model is updated based on the training sample set until the initial proxy model tends to converge to obtain a proxy model of the functional function, including: dividing the training sample set into a first sample subset and a second sample subset, the first sample subset consisting of sample data determined by the sample output symbol; obtaining the current state of the model, if the current state of the model is in an unconverged state, selecting new training sample points through a sample classification learning function to obtain an updated sample point set; updating the initial proxy model based on the updated sample point set until the initial proxy model tends to converge to obtain the proxy model of the functional function.
[0075] The first sample subset may be a set of sample data with determined sample output signs. The second sample subset may be a set of sample data with uncertain sample output signs. The sample classification learning function may be a function for classifying samples.
[0076] Before using the training sample set to update the model, the training sample set is classified using the sample classification learning function, which is defined as in, and Represents the Kriging model For example, the training sample set is classified according to whether the sign of the sample output result is determined, and the sample data with determined sample output signs are classified into a set as the first sample set. The first sample set is The sample data with uncertain sample output signs is used as the second sample set
[0077] After the sample set is divided, it is determined whether the current state of the initial proxy model is in a convergent state. The criterion for the convergence of the Kriging model is: the first sample set contains both failed samples and safe samples, that is, There are both failed samples and safe samples, or the basic sample classification learning function determines that the sample output symbols are all in a certain state, that is,
[0078] If the current model is not in a convergent state, new training sample points are selected through the sample classification learning function to obtain an updated sample point set, denoted as x new ; Based on the updated sample point set x new The initial proxy model is updated. After the update operation is completed, whether the model has converged is determined based on a convergence criterion. If the model has not yet converged, the model update step is repeated until the initial proxy model converges to obtain a proxy model of the functional function.
[0079] The training sample points for updating the Kriging model are selected according to the following two stages. First, the sample pool S is updated according to the proxy model. (k) The samples in are divided into two categories, namely and Then, according to The number of elements in the sample classification learning function is used to select new training sample points, denoted as x new Through the above processing steps, a proxy model in a converged state can be obtained, and the failure domain indicator function can be solved based on the converged model.
[0080] In an exemplary embodiment of the present disclosure, new training sample points are selected through a sample classification learning function to obtain an updated sample point set, including: determining the symbol prediction error probability of the output result of the functional function based on the sample classification learning function; obtaining the predicted mean of the functional function; and determining the updated sample point set based on the predicted mean of the functional function, the symbol prediction error probability and the maximum independent variable point set function.
[0081] The symbol prediction error probability can be obtained by performing a value operation on the sample learning function. The predicted mean of the performance function can be the mean of the predicted values obtained by performing a reliability analysis on the input variables. The maximum independent variable point set function can be a function used to find the parameters (set) of the function. The specific content of updating the sample point set is shown in Formula 2.
[0082]
[0083] in, Can represent sample classification learning function; Can be expressed The predicted mean of Can be expressed The probability of a symbol prediction error; argmax() can be a maximum independent variable point set function; for In the case where there are only failed samples, Samples with larger values and greater probability of symbol prediction errors will be selected as new training samples; In the case where there are only safe samples, Samples with smaller values and greater probability of symbol prediction errors will be selected as new training samples; For the empty set, The samples with a higher probability of symbol prediction error will be selected as new training samples. The above operation gives the specific process of selecting new training sample points for the sample learning classification function.
[0084] In an exemplary embodiment of the present disclosure, for step S140, the failure probability limit of the turbine casing in a failure state is determined according to the proxy model, including: determining the failure domain indicator function of the input variable in each variable focal element based on the proxy model; constructing the failure probability boundary function of the turbine casing according to the failure domain indicator function; and determining the failure probability limit based on the failure probability boundary function.
[0085] The failure domain indicator function may be an indicator function for determining whether the turbine casing is in a failure state under the current input variable value. The failure probability boundary function may be a function for indicating the boundary value of the failure probability.
[0086] After obtaining the convergent proxy model, the failure domain indicator function of the input variable in each variable focal element can be solved based on the convergent proxy model. For example, the failure domain indicator function can include the minimum failure domain indicator function I min (Α k ) and the maximum failure domain indicator function I max (Α k ).
[0087]
[0088]
[0089] in, It can be the Kriging model corresponding to the performance function; It can represent the minimum value obtained based on the Kriging model; It can indicate that the turbine case is in a failed state; It can indicate that the turbine case is in an effective state; It can represent the maximum value obtained based on the Kriging model; It can indicate that the turbine case is in a failed state; It can indicate that the turbine case is in a valid state.
[0090] According to the above failure domain indicator function, the failure probability boundary function of the turbine casing can be constructed. min (Α k ) and the maximum failure domain indicator function I max (Α k ), the failure probability boundary function of the turbine casing can be constructed based on the failure domain indicator function, and then based on the failure probability boundary function, the failure probability limit can be determined to analyze the reliability of the turbine casing.
[0091] In an exemplary embodiment of the present disclosure, a failure probability boundary function of a turbine casing is constructed according to a failure domain indicator function, including: constructing a failure probability upper bound function according to a minimum failure domain indicator function; and constructing a failure probability lower bound function according to a maximum failure domain indicator function.
[0092] Among them, the minimum failure domain indicator function can be a function for determining the minimum boundary value of the turbine case failure probability. The failure probability upper limit function can be a calculation function for calculating the upper limit value of the turbine case failure probability. The maximum failure domain indicator function can be a function for determining the maximum boundary value of the turbine case failure probability. The failure probability lower limit function can be a calculation function for calculating the lower limit value of the turbine case failure probability.
[0093] When constructing the failure probability boundary function of the turbine casing, according to the minimum failure domain indicator function I min (Α k ), construct the upper bound function of failure probability Pl(F), the upper bound function of failure probability Pl(F) is specifically shown in Formula 5.
[0094]
[0095] Among them, n E Can represent the number of variable focal elements; m E (Α k ) can represent the basic probability distribution corresponding to each variable focal element; I min (Α k ) can be a weight value obtained based on the minimum failure domain indicator function and the basic probability distribution calculation.
[0096] Similarly, according to the obtained maximum failure domain indicator function I max (Α k ), construct the lower bound function of the failure probability Bel(F) of the turbine casing. Specifically, the lower bound function of the failure probability Bel(F) is shown in Formula 6.
[0097]
[0098] Among them, n E Can represent the number of variable focal elements; m E (Α k ) can represent the basic probability distribution corresponding to each variable focal element; I max (Α k ) can be a weight value obtained based on the minimum failure region indicator function and the basic probability distribution calculation. Through the above processing steps, the specific upper limit of failure probability and the lower limit of failure probability can be calculated based on the boundary function of failure probability.
[0099] Next, the surrogate model method for turbine case reliability analysis under evidence uncertainty will be explained based on a specific example. Figure 2 , Figure 2 The schematic diagram of the structure of the turbine casing according to the exemplary embodiment of the present disclosure is shown schematically. Figure 2 The turbine casing structure shown in the figure is made of GH4169 and the operating temperature is assumed to be 150°C. The corresponding elastic modulus, Poisson's ratio and linear expansion coefficient are 162 GPa, 0.355 and 12.4×10 6 / ℃. After finite element analysis of the turbine casing structure, setting of material properties, application of boundary conditions and meshing, the following results can be obtained: Figure 2 The finite element model is shown.
[0100] refer to Figure 3 , Figure 3 The finite element model diagram of the turbine casing according to the exemplary embodiment of the present disclosure is schematically shown. Considering that the turbine casing is subjected to an external pressure of 0.101 MPa and an internal pressure of 1.6299 MPa during operation, a static analysis is performed on it using finite element analysis software, and the stress cloud diagram obtained is as follows: Figure 3 As shown. Figure 3 It can be seen that the maximum stress of the turbine casing structure is 850.54Mpa.
[0101] Due to various uncertainties in design, processing, and production, the material properties and external environment of the turbine casing are also uncertain. Figure 2 The turbine casing structure shown in the figure considers three material property uncertainty input variables, namely the elastic modulus E, Poisson's ratio μ and linear expansion coefficient α, and two external environment uncertainty input variables, namely the external pressure P1 and the internal pressure P2, and uses the evidence theory to establish its uncertainty characterization model.
[0102] The focal elements of each input variable and the corresponding basic probability distribution are shown in Table 1. max It is related to the values of the input variables in Table 1, so it can be expressed as a function of the 5-dimensional input variables x = {x1, x2, x3, x4, x5} = {E, μ, α, P1, P2}, that is, σ max (x). Considering the failure mode that the maximum structural stress is greater than the structural strength, the functional function in the reliability analysis of the turbine casing can be established as shown below:
[0103] g(x)=σ P0.2 -σ max (x) (Formula 7)
[0104] Among them, σ P0.2 It can be the yield strength of the material, which is 995Mpa;max (x) can represent the maximum stress function of the turbine casing.
[0105] Table 1. Focal elements and corresponding basic probability distribution of turbine casing structure input variables
[0106]
[0107] For the five evidence variables in Table 1, each evidence variable has two focal elements, so there will be two 5 = 32 input space combinations, the focal element and basic probability distribution corresponding to each combination are shown in Table 2.
[0108] Table 2 Combination of focal elements in the input variable space of turbine casing structure
[0109]
[0110]
[0111]
[0112] Before executing step S120, 50 initial training samples are generated using Sobol sequence sampling, and the Kriging model of the functional function g(x) is constructed using the DACE toolbox.
[0113] In step S120, a random sampling technique is used to obtain the k Generate 10,000 samples of the extreme value of the solution function within (k=1,2,…,32) And build a candidate sample pool of 10,000 to update the Kriging model
[0114] In step S130, after 202 training sample points are adaptively selected according to sample classification, the Kriging model convergence criterion is met.
[0115] In step S140, the converged Kriging model obtained in step S130 is used to calculate the minimum value g of the functional function in each focal element. min (Α k ) and the maximum value g max (Α k ), and the corresponding failure domain indicator function I min (Α k ) and I max (Α k ) as shown in Table 3.
[0116] Table 3 Statistics of output in each focal element of the turbine casing structure
[0117]
[0118]
[0119] According to the data in Table 3, the upper limit Pl(F) and lower limit Bel(F) of the failure probability of the turbine casing structure can be calculated. The upper limit Pl(F) of the failure probability is: Pl(F) = 0.0108 + 0.0072 + 0.0168 + 0.0162 + 0.0378 + 0.0108 + 0.0252 = 0.1248; the lower limit Bel(F) of the failure probability = 0.
[0120] From the above analysis results, we can see that although the lower limit of the failure probability of the turbine casing structure in this example is 0, its upper limit of failure probability is relatively large, which is 0.1248. Therefore, the upper limit of failure probability should be reduced through optimization design in the future to ensure the reliability of the structure in use. Figure 4 , Figure 4 The stress cloud diagram of the turbine casing according to the exemplary embodiment of the present disclosure is schematically shown. By adopting the above reliability analysis scheme, the reliability of the turbine casing is analyzed, and the internal mechanical state of the turbine casing can be intuitively reflected.
[0121] It should be noted that the terms “first”, “second”, etc. used in the present disclosure are only used to distinguish different sample sets and should not impose any limitations on the present disclosure.
[0122] In summary, the proxy model method for reliability analysis of turbine casing under evidence uncertainty disclosed in the present invention obtains the input variables corresponding to the turbine casing, constructs the structural reliability assessment model of the turbine casing based on the input variables, and the structural reliability assessment model is constructed based on the function function of the input variables; based on the function extreme value samples of the function function, generates a training sample set of the initial proxy model; based on the training sample set, updates the initial proxy model until the initial proxy model tends to converge, and obtains the proxy model of the function function; according to the proxy model, determines the failure probability limit of the turbine casing in the failure state. The present invention can robustly and quickly calculate the failure probability limit of the turbine casing structure under evidence uncertainty, so as to reasonably evaluate the reliability level of the turbine casing. On the one hand, based on the uncertainty of the input variables, a reliability analysis scheme for the turbine casing mechanism is provided, so as to robustly and quickly solve the critical value of the failure probability of the turbine casing structure under evidence uncertainty, and then reasonably evaluate the reliability level of the turbine casing structure. On the other hand, based on the specific values of the upper and lower bounds of the failure probability of the turbine casing, the failure probability boundary value with a larger value should be reduced through optimization design in the future to ensure the reliability of the structure in use.
[0123] It should be noted that, although the steps of the method of the present invention are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.
[0124] In addition, in this exemplary embodiment, a proxy model device for turbine case reliability analysis under evidence uncertainty is also provided. Figure 5 The proxy model device 500 for turbine casing reliability analysis under evidence uncertainty may include: a reliability model building module 510 , a sample set building module 520 , a model updating module 530 , and a reliability analysis module 540 .
[0125] Specifically, the reliability model construction module 510 is used to obtain the input variables corresponding to the turbine casing, and to construct a structural reliability assessment model of the turbine casing based on the input variables, wherein the structural reliability assessment model is constructed based on the functional function of the input variables; the sample set construction module 520 is used to construct an initial proxy model of the functional function, and to generate a training sample set of the initial proxy model based on the function extreme value samples of the functional function; the model update module 530 is used to update the initial proxy model based on the training sample set until the initial proxy model converges to obtain a proxy model of the functional function; the reliability analysis module 540 is used to determine the failure probability limit of the turbine casing in a failure state according to the proxy model.
[0126] The specific details of the virtual module of the proxy model device for turbine case reliability analysis under each of the above-mentioned evidential uncertainties have been described in detail in the corresponding proxy model method for turbine case reliability analysis under evidential uncertainty, so they will not be repeated here.
[0127] It should be noted that, although several modules or units of the proxy model device for turbine case reliability analysis under evidence uncertainty are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiment of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.
[0128] In addition, in an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.
[0129] It will be appreciated by those skilled in the art that various aspects of the present invention may be implemented as systems, methods or program products. Therefore, various aspects of the present invention may be specifically implemented in the following forms, namely: complete hardware embodiments, complete software embodiments (including firmware, microcode, etc.), or embodiments combining hardware and software aspects, which may be collectively referred to herein as "circuits", "modules" or "systems".
[0130] Reference below Figure 6 hereinafter describes an electronic device 600 according to such an embodiment of the present disclosure. Figure 6 The electronic device 600 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.
[0131] like Figure 6 As shown, the electronic device 600 is in the form of a general computing device. The components of the electronic device 600 may include but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different system components (including the storage unit 620 and the processing unit 610), and a display unit 640.
[0132] The storage unit stores program codes, which can be executed by the processing unit 610, so that the processing unit 610 executes the steps according to various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification.
[0133] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 621 and / or a cache memory unit 622 , and may further include a read-only memory unit (ROM) 623 .
[0134] The storage unit 620 may include a program / utility 624 having a set (at least one) of program modules 625, such program modules 625 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.
[0135] Bus 630 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.
[0136] The electronic device 600 may also communicate with one or more external devices 670 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 650. Furthermore, the electronic device 600 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 660. As shown, the network adapter 660 communicates with other modules of the electronic device 600 via a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.
[0137] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiment of the present disclosure.
[0138] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible embodiments, various aspects of the present invention can also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to perform the steps according to various exemplary embodiments of the present invention described in the above "Exemplary Method" section of the present specification.
[0139] refer to Figure 7 As shown, a program product 700 for implementing the above method according to an embodiment of the present invention is described, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.
[0140] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
[0141] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.
[0142] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.
[0143] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).
[0144] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to an exemplary embodiment of the present invention, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.
[0145] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.
[0146] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims
1. A surrogate model approach for turbine case reliability analysis under evidential uncertainty, It is characterized in that include: Acquire input variables corresponding to a turbine casing, and construct a structural reliability assessment model of the turbine casing based on the input variables, wherein the structural reliability assessment model is constructed based on a functional function of the input variables; Constructing an initial proxy model of the performance function, and generating a training sample set of the initial proxy model based on function extreme value samples of the performance function; Updating the initial proxy model based on the training sample set until the initial proxy model converges to obtain a proxy model of the functional function; A failure probability limit of the turbine casing being in a failure state is determined based on the proxy model.
2. The method according to claim 1, It is characterized in that The step of constructing the turbine casing structural reliability assessment model based on the input variables comprises: Acquiring test data corresponding to the input variables; Constructing an evidence representation model of the input variables based on the test data, the evidence representation model including a set of variable focus units corresponding to the input variables and a basic probability distribution corresponding to each variable focus unit; Constructing a reliability boundary function of the turbine casing based on the plurality of variable focal elements and the respective corresponding basic probability distributions; The structural reliability assessment model is generated according to the reliability boundary function.
3. The method according to claim 2, It is characterized in that The step of constructing the reliability boundary function of the turbine casing based on the plurality of variable focal elements and the respective corresponding basic probability distributions includes: Based on the input variables, constructing a functional function of the turbine casing, wherein the functional function is used to perform reliability analysis on the turbine casing; Determine the failure domain corresponding to the input variable based on the functional function; The reliability boundary function is constructed based on the failure domain, the plurality of variable focal elements and the respective corresponding basic probability allocations.
4. The method according to claim 3, It is characterized in that The reliability boundary function includes a reliability upper boundary function and a reliability lower boundary function. The reliability boundary function is constructed based on the failure domain, the plurality of variable focal elements and the respective corresponding basic probability distributions, including: Determine a first variable focus element from the set of variable focus elements, wherein the first variable focus element is a variable focus element included in the failure domain; constructing the reliability lower bound function based on the first variable focus element and the basic probability distribution corresponding to the first variable focus element; Determine a second variable focus element from the variable focus element set, wherein the second variable focus element is a variable focus element whose intersection with the failure domain is a non-empty set; Based on the second variable focus unit and the basic probability distribution corresponding to the second variable focus unit, the reliability upper bound function is constructed.
5. The method according to claim 1, It is characterized in that The constructing of the initial proxy model of the functional function comprises: Based on the variable value space of the input variable, an initial training sample is determined by adopting a low-bias sampling method; The initial proxy model is constructed based on the initial training samples and the model construction instruction.
6. The method according to claim 1, It is characterized in that The step of generating a training sample set of the initial proxy model based on the function extreme value samples of the functional function comprises: Based on the variable focus set of the input variable, a function extreme value sample of the functional function is determined by random sampling; The training sample set is constructed based on the determined function extreme value samples.
7. The method according to claim 1, It is characterized in that The updating of the initial proxy model based on the training sample set until the initial proxy model converges to obtain a proxy model of the functional function includes: Dividing the training sample set into a first sample subset and a second sample subset, wherein the first sample subset consists of sample data determined by sample output symbols; Obtaining the current state of the model, if the current state of the model is in an unconverged state, selecting new training sample points through a sample classification learning function to obtain an updated sample point set; The initial proxy model is updated based on the updated sample point set until the initial proxy model converges to obtain the proxy model.
8. The method according to claim 7, It is characterized in that The method of selecting new training sample points through the sample classification learning function to obtain an updated sample point set includes: Determining a symbol prediction error probability of an output result of the functional function based on a sample classification learning function; Obtaining a predicted mean of the performance function; The updated sample point set is determined according to the predicted mean of the performance function, the symbol prediction error probability and the maximum independent variable point set function.
9. The method according to claim 1, It is characterized in that Determining the failure probability limit of the turbine casing being in a failure state according to the proxy model includes: Based on the proxy model, determining the failure domain indicator function of the input variable in each variable focal element; Constructing a failure probability boundary function of the turbine casing according to the failure domain indicator function; Based on the failure probability boundary function, the failure probability limit is determined.
10. The method according to claim 9, It is characterized in that The failure domain indicator function includes a minimum failure domain indicator function and a maximum failure domain indicator function. The failure probability boundary function of the turbine casing is constructed according to the failure domain indicator function, including: Constructing the failure probability upper bound function according to the minimum failure region indicator function; The failure probability lower bound function is constructed according to the maximum failure domain indicator function.
Citation Information
Cited By
Reliability analysis method for complex structure equipment under measurement condition
CN119670510A
A reliability analysis method for complex-structured equipment under consideration of measurement conditions
CN119670510B
