Method and apparatus for obtaining weibull distribution parameters

Abstract

The application relates to the technical field of reliability evaluation, and provides a Weibull distribution parameter acquisition method and device. The method comprises the following steps: performing parameter estimation according to fatigue failure life of each mechanical component to obtain initial Weibull distribution parameters; after determining initial life unavailability of each mechanical component according to the initial Weibull distribution parameters, the initial life unavailability is optimized according to a first genetic algorithm to obtain corrected life unavailability of each mechanical component; and target Weibull distribution parameters of the mechanical component are obtained according to the corrected life unavailability. The Weibull distribution parameter acquisition method provided in the application can improve the accuracy of the obtained Weibull parameters.

Description

Technical Field

[0001] This application relates to the field of reliability assessment technology, specifically to a method and apparatus for obtaining Weibull distribution parameters. Background Technology

[0002] The Weibull distribution is the most commonly used distribution model to describe the lifespan of mechanical products. The three-parameter Weibull distribution, in particular, is highly adaptable due to its ability to describe the minimum fatigue life of a sample, and is widely used in fatigue life distribution models for mechanical components. In related technologies, the parameters of the Weibull distribution are mainly obtained using the right approximation estimation method. This method transforms the target distribution function into a linear function based on a specific transformation, and then uses the least squares method to fit the optimal parameters. By iteratively calculating and approximating the optimal target parameters from the right side, it can effectively improve the problem of non-convergence in computation.

[0003] However, since the right approximation estimation method requires the median rank as the initial iterative value of lifetime unreliability during the iteration process, and the median rank, as an empirical parameter, may differ significantly from the true value, it is easy for the iterative algorithm to obtain a local optimum rather than a global optimum. This results in the Weibull parameters being inaccurate and thus affecting the accuracy of the reliability assessment of mechanical components. Summary of the Invention

[0004] This application aims to address at least one of the technical problems existing in related technologies. To this end, this application proposes a method for obtaining Weibull distribution parameters, which can improve the accuracy of the obtained Weibull parameters.

[0005] This application also proposes a device for obtaining Weibull distribution parameters.

[0006] This application also proposes an electronic device.

[0007] This application also proposes a computer-readable storage medium.

[0008] The method for obtaining Weibull distribution parameters according to the first aspect of this application includes: Based on the obtained fatigue failure life of each mechanical component, parameter pre-estimation is performed to obtain the initial Weibull distribution parameters; After determining the initial lifetime unreliability of each mechanical component based on the initial Weibull distribution parameters, the initial lifetime unreliability is optimized according to the first genetic algorithm to obtain the corrected lifetime unreliability of each mechanical component. Based on the unreliability of each correction lifetime, the target Weibull distribution parameters of the mechanical component are obtained.

[0009] By pre-estimating the fatigue failure life of each mechanical component and obtaining the initial Weibull distribution parameters, the initial life unreliability of each mechanical component is determined based on the initial Weibull distribution parameters. A genetic algorithm is then used to optimize each initial life unreliability, resulting in a corrected life unreliability for each mechanical component. This leverages the characteristics of the genetic algorithm to make the obtained life unreliability of each mechanical component closer to the true value. Based on the corrected life unreliability, which is closer to the true value, the target Weibull distribution parameters of the mechanical component are obtained, thus making the obtained Weibull distribution parameters even closer to the true value. This improves the accuracy of the obtained Weibull parameters and consequently enhances the accuracy of the reliability assessment of the mechanical components.

[0010] According to one embodiment of this application, parameter pre-estimation is performed based on the obtained fatigue failure life of each mechanical component to obtain initial Weibull distribution parameters, including: Based on the right approximation estimation method, the fatigue failure life is pre-estimated to obtain the initial Weibull distribution parameters.

[0011] According to one embodiment of this application, determining the initial lifetime unreliability of each of the mechanical components based on the initial Weibull distribution parameters includes: The initial Weibull distribution parameters and each fatigue failure life are input into the initial fatigue life distribution function model to obtain the initial life unreliability of each mechanical component. Wherein, the initial fatigue life distribution function model is ; Indicates mechanical parts fatigue failure life The initial lifetime is unreliable. The location parameter represents the initial Weibull distribution parameter. The size parameter represents the initial Weibull distribution parameters. The shape parameter represents the initial Weibull distribution parameters.

[0012] According to one embodiment of this application, optimizing the initial lifetime unreliability of each of the mechanical components using a first genetic algorithm to obtain the corrected lifetime unreliability of each of the mechanical components includes: Based on the mathematical model of the initial lifetime unreliability, determine the unknown parameter set corresponding to each of the initial lifetime unreliability; The unknown parameter set is selected, crossed, and mutated according to the first genetic algorithm to obtain a target unknown parameter set that satisfies the first objective function of the first genetic algorithm. Input the target unknown parameter set into the mathematical model to determine each corrected lifetime unreliability corresponding to each initial lifetime unreliability; The mathematical model includes ; ; Let represent the first objective function. Indicates an unknown parameter set. Indicates mechanical parts j The set of unknown parameters corresponding to the initial lifetime unreliability.

[0013] According to one embodiment of this application, the mathematical model further includes: .

[0014] According to one embodiment of this application, obtaining the target Weibull distribution parameters of the mechanical component based on each of the said corrected lifetime unreliability includes: Based on the input of each of the corrected life unreliability values ​​into the target fatigue life distribution function model, obtain the set of unknown Weibull distribution parameters that correspond one-to-one with each of the corrected life unreliability values. The second genetic algorithm is used to select, crossover, and mutate each of the unknown Weibull distribution parameter sets to obtain the target Weibull distribution parameter set that satisfies the second objective function of the second genetic algorithm. Determine the target Weibull distribution parameters based on the target Weibull distribution parameter set; The target fatigue life distribution function model is as follows: ; Indicates mechanical parts fatigue life, Indicates mechanical parts The calibration lifetime is unreliable. This represents the location parameter in the unknown Weibull distribution parameter set. This represents the size parameter in an unknown Weibull distribution parameter set. This represents the shape parameter in an unknown Weibull distribution parameter set; The second objective function includes the target Weibull distribution parameter set input to the target fatigue life distribution function model, and the absolute value of the difference between the target Weibull distribution parameter set and each of the corrected life unreliability values ​​is less than a preset value.

[0015] According to one embodiment of this application, the second objective function further includes the linear correlation between the dependent variable and the independent variable in the linear equation determined in the preset model by the target Weibull distribution parameter set being greater than a preset value; Wherein, the linear equation is The preset model includes ; Indicates the independent variable. Indicates the dependent variable. This represents the mechanical component obtained by inputting the target Weibull distribution parameter set into the target fatigue life distribution function model. The target lifetime is unreliable.

[0016] The Weibull distribution parameter acquisition apparatus according to the second aspect of this application includes: The parameter prediction module is used to perform parameter prediction based on the obtained fatigue failure life of each mechanical component and obtain the initial Weibull distribution parameters. The data processing module is used to determine the initial life unreliability of each mechanical component based on the initial Weibull distribution parameters, and then optimize the initial life unreliability based on the first genetic algorithm to obtain the corrected life unreliability of each mechanical component. The parameter acquisition module is used to acquire the target Weibull distribution parameters of the mechanical component based on the unreliability of each correction life.

[0017] An electronic device according to a third aspect of this application includes a processor and a memory storing a computer program, wherein the processor executes the computer program to implement the Weibull distribution parameter acquisition method described in any of the above embodiments.

[0018] A computer-readable storage medium according to a fourth aspect of this application stores a computer program thereon, which, when executed by a processor, implements the Weibull distribution parameter acquisition method described in any of the above embodiments.

[0019] A computer program product according to a fifth aspect of this application includes: when the computer program is executed by a processor, it implements the Weibull distribution parameter acquisition method as described in any of the above embodiments.

[0020] The above-described one or more technical solutions in the embodiments of this application have at least one of the following technical effects: By pre-estimating the fatigue failure life of each mechanical component and obtaining the initial Weibull distribution parameters, the initial life unreliability of each mechanical component is determined based on the initial Weibull distribution parameters. A genetic algorithm is then used to optimize each initial life unreliability, resulting in a corrected life unreliability for each mechanical component. This leverages the characteristics of the genetic algorithm to make the obtained life unreliability of each mechanical component closer to the true value. Based on the corrected life unreliability, which is closer to the true value, the target Weibull distribution parameters of the mechanical component are obtained, thus making the obtained Weibull distribution parameters even closer to the true value. This improves the accuracy of the obtained Weibull parameters and consequently enhances the accuracy of the reliability assessment of the mechanical components.

[0021] Furthermore, by performing selection, crossover, and mutation processing on each unknown parameter set corresponding to each initial lifetime unreliability based on the first genetic algorithm, a target unknown parameter set satisfying the first objective function of the first genetic algorithm is obtained. This target unknown parameter set is then input into the mathematical model of the initial lifetime unreliability, and a new initial lifetime unreliability is determined as the corrected lifetime unreliability. This allows the new initial lifetime unreliability to effectively approximate the true value, further reducing the error between the lifetime unreliability and the true value, and thus further improving the accuracy of the subsequently obtained Weibull parameters.

[0022] Furthermore, the unreliability of each corrected lifetime is optimized by a second genetic algorithm, thereby using two genetic algorithms to estimate the Weibull distribution parameters of the mechanical component, improving the iteration speed of the Weibull distribution parameter acquisition process, and further improving the accuracy of the Weibull distribution parameter estimation. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating the method for obtaining Weibull distribution parameters provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of the Weibull distribution parameter acquisition device provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] The Weibull distribution parameter acquisition method and apparatus provided in this application will be described in detail and explained through several specific embodiments.

[0027] In one embodiment, a method for obtaining Weibull distribution parameters is provided. This method is applied to a server to obtain Weibull distribution parameters. The server can be a standalone server or a server cluster consisting of multiple servers. It can also be a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence sampling point devices.

[0028] like Figure 1 As shown in the figure, the method for obtaining Weibull distribution parameters provided in this embodiment includes: Step 101: Based on the obtained fatigue failure life of each mechanical component, perform parameter pre-estimation to obtain the initial Weibull distribution parameters; Step 102: After determining the initial lifetime unreliability of each mechanical component based on the initial Weibull distribution parameters, optimize the initial lifetime unreliability according to the first genetic algorithm to obtain the corrected lifetime unreliability of each mechanical component. Step 103: Obtain the target Weibull distribution parameters of the mechanical component based on the unreliability of each correction lifetime.

[0029] By pre-estimating the fatigue failure life of each mechanical component and obtaining the initial Weibull distribution parameters, the initial life unreliability of each mechanical component is determined based on the initial Weibull distribution parameters. A genetic algorithm is then used to optimize each initial life unreliability, resulting in a corrected life unreliability for each mechanical component. This leverages the characteristics of the genetic algorithm to make the obtained life unreliability of each mechanical component closer to the true value. Based on the corrected life unreliability, which is closer to the true value, the target Weibull distribution parameters of the mechanical component are obtained, thus making the obtained Weibull distribution parameters even closer to the true value. This improves the accuracy of the obtained Weibull parameters and consequently enhances the accuracy of the reliability assessment of the mechanical components.

[0030] In one embodiment, each mechanical component can undergo a fatigue endurance bench test beforehand. The number of each mechanical component is... The fatigue failure life of each mechanical component can be obtained through fatigue durability bench testing. Then, the lifespans of each part are arranged from shortest to longest, and combined to form the lifespan of the entire batch of parts. And the fatigue failure life of each mechanical component. Send to the server. The server obtains the fatigue failure life of each mechanical component. Then, the fatigue failure life of each mechanical component can be determined. The initial Weibull distribution parameters are obtained by estimating the Weibull distribution parameters. These parameters include shape, size, and position.

[0031] Based on the fatigue failure life of each mechanical component Pre-estimation of Weibull distribution parameters can be performed using statistical estimation, gray-scale method, maximum likelihood estimation, and right approximation estimation. To ensure convergence in the pre-estimation process, in one embodiment, the right approximation estimation method can be used to pre-estimate the fatigue failure life of each mechanical component. This method transforms the target distribution function into a linear function using a specific transformation, and then uses the least squares method to fit the optimal parameters. By iteratively calculating and approximating the optimal target parameters from the right, the initial Weibull distribution parameters are obtained, effectively improving the problem of non-convergence.

[0032] In one embodiment, the fatigue life distribution function of the mechanical component is: ,in Also known as lifetime unreliability, For the fatigue life of mechanical components, For shape parameters, For dimensional parameters, Here, represents the positional parameters. Therefore, when using the right approximation estimation method for parameter pre-estimation, the fatigue life distribution function must first be converted into a linear function. That is, by taking the natural logarithm twice on the three-parameter Weibull distribution of the fatigue life distribution function, we can obtain: (Equation 1) Then, the median rank is used. The initial unreliability value is determined by the following expression: , This refers to the initial lifetime unreliability. .

[0033] When transforming the target distribution function into a linear function using a specific transformation method in the right approximation estimation method, let (Equation 1), where, Then, the target distribution function can be transformed into a linear function through a specific transformation method in the right approximation estimation method: .

[0034] Since only need to be given The value can be used to obtain the corresponding value. Then, based on the linear function, linear regression is performed using the least squares method, and the coefficient expression for the fit is as follows: (Equation 2); at this time and correlation coefficient The expression is: (Equation 3).

[0035] Among them, the correlation coefficient This reflects the degree of linear correlation between the independent and dependent variables. The larger the value, the more it indicates and The higher the degree of linear correlation, the better. As a position parameter The initial value is selected, and the calculation step size is chosen. , in turn , , , ..., Substituting into equations 1 and 3 above, we can obtain the corresponding correlation coefficients. At this point, the maximum value is... Corresponding That is, the desired position parameter. The estimated value, i.e. After obtaining the estimated location parameters, Substituting the values ​​into Equations 1 and 2, the estimated values ​​of the dimensional parameters can be determined. and shape parameter estimates Thus, the initial Weibull distribution parameters are obtained. .

[0036] In one embodiment, the initial Weibull distribution parameters are obtained through a right approximation estimation method. Then, the initial Weibull distribution parameters can be... By inputting an initial fatigue life distribution function model constructed from the fatigue life distribution function of mechanical components, the initial life unreliability of mechanical components can be accurately obtained.

[0037] Specifically, based on the initial Weibull distribution parameters, the initial life unreliability of each mechanical component is determined, including: The initial Weibull distribution parameters and each fatigue failure life are input into the initial fatigue life distribution function model to obtain the initial life unreliability of each mechanical component. Wherein, the initial fatigue life distribution function model is ; Indicates mechanical parts fatigue failure life The initial lifetime is unreliable. The location parameter represents the initial Weibull distribution parameter. The size parameter represents the initial Weibull distribution parameters. The shape parameter represents the initial Weibull distribution parameters.

[0038] In the initial Weibull distribution parameters Input an initial fatigue life distribution function model constructed from the fatigue life distribution functions of mechanical components, and obtain the initial life unreliability of each mechanical component. Subsequently, since the initial lifetime unreliability is based on the median rank as the initial iterative value of lifetime unreliability, and the median rank, as an empirical parameter, may differ significantly from the true value, the initial lifetime unreliability will have a large error compared to the true value.

[0039] At this point, a genetic algorithm can be used to optimize the initial lifetime unreliability to reduce the error between the initial lifetime unreliability and the true value, thereby obtaining the corrected lifetime unreliability. To further reduce the error between the initial lifetime unreliability and the true value, in one embodiment, the initial lifetime unreliability is optimized according to a first genetic algorithm to obtain the corrected lifetime unreliability of each mechanical component, including: Based on the mathematical model of the initial lifetime unreliability, determine the unknown parameter set corresponding to each of the initial lifetime unreliability; The unknown parameter set is selected, crossed, and mutated according to the first genetic algorithm to obtain a target unknown parameter set that satisfies the first objective function of the first genetic algorithm. Input the target unknown parameter set into the mathematical model to determine each corrected lifetime unreliability corresponding to each initial lifetime unreliability; The mathematical model includes ; ; Let represent the first objective function. Indicates an unknown parameter set. Indicates mechanical parts j The set of unknown parameters corresponding to the initial lifetime unreliability.

[0040] In one embodiment, after obtaining the initial lifetime unreliability Then, the unreliability of the initial lifetime is determined using a genetic algorithm. The expression is:

[0041] Among them, parameters and The undetermined coefficients were determined using a genetic algorithm. This indicates the corresponding mechanical component.

[0042] At this time, use This represents a vector composed of unknown parameters in each initial lifetime unreliability, establishing a set of unknown parameters. That is, for any initial lifetime unreliability Each of them has a corresponding set of unknown parameters. .

[0043] Then, based on each unknown parameter group Establish an unknown parameter set The mathematical model for the estimation method is: ; in, Let represent the first objective function of the first genetic algorithm.

[0044] Then, group the unknown parameters. Selection, crossover, and mutation are performed using the first genetic algorithm. Specifically, each unknown parameter group... As the initial population P(t), each individual in P(t) is encoded and substituted into the first objective function to evaluate its fitness. Then, based on the fitness of each individual, selection, crossover, and mutation operations are performed to obtain the next population P(t+1). This next population P(t+1) is then used as the initial population P(t), and the above operations are repeated until the preset number of generations is reached. The individual with the highest fitness is then output as the optimal solution. In the first genetic algorithm, the encoding type can be double-precision real number encoding, the fitness scale is set to rank, and the selection operation uses roulette wheel selection to randomly select individuals. The crossover operation is set to distributed, randomly generating genetic binary vectors and crossing them according to 0 to 1. Thus, through the first genetic algorithm, an optimal set of individuals can be found. , so that the corresponding first objective function The value is minimized, and at this time... This is used as the target unknown parameter set, and then the target unknown parameter set is input into the mathematical model. Each mechanical component can then be obtained. New initial lifetime unreliability At this point, the new initial lifetime is unreliable. This refers to correcting lifetime unreliability.

[0045] By performing selection, crossover, and mutation processing on each unknown parameter set corresponding to each initial lifetime unreliability based on the first genetic algorithm, a target unknown parameter set satisfying the first objective function of the first genetic algorithm is obtained. This target unknown parameter set is then input into the mathematical model of the initial lifetime unreliability, and a new initial lifetime unreliability is determined as the corrected lifetime unreliability. This allows the new initial lifetime unreliability to effectively approximate the true value, further reducing the error between the lifetime unreliability and the true value, and thus further improving the accuracy of the subsequently obtained Weibull parameters.

[0046] Considering that in the mathematical model, if only setting This could cause the first genetic algorithm to get stuck in a local optimum at the beginning, and the probability of mechanical parts failing early in their lifespan is low. Data points with values ​​near 1 are scarce. Therefore, to prevent the first genetic algorithm from getting trapped in a local optimum at the outset, in one embodiment, the mathematical model further includes: .

[0047] In one embodiment, after obtaining each corrected lifetime unreliability through a first genetic algorithm, the Weibull distribution parameters of each mechanical component are determined based on each corrected lifetime unreliability. For example, any corrected lifetime unreliability can be used as the lifetime unreliability. Substituting the fatigue life distribution function and the probability density function expression of the three-parameter Weibull distribution function: Thus, the final target Weibull distribution parameters are obtained.

[0048] To make the final Weibull distribution parameters more accurate, in one embodiment, the target Weibull distribution parameters of the mechanical component are obtained based on the respective corrected lifetime unreliability, including: Based on the input of each of the corrected life unreliability values ​​into the target fatigue life distribution function model, obtain the set of unknown Weibull distribution parameters that correspond one-to-one with each of the corrected life unreliability values. The second genetic algorithm is used to select, crossover, and mutate each of the unknown Weibull distribution parameter sets to obtain the target Weibull distribution parameter set that satisfies the second objective function of the second genetic algorithm. Determine the target Weibull distribution parameters based on the target Weibull distribution parameter set; The target fatigue life distribution function model is as follows: ; Indicates mechanical parts fatigue life, Indicates mechanical parts The calibration lifetime is unreliable. This represents the location parameter in the unknown Weibull distribution parameter set. This represents the size parameter in an unknown Weibull distribution parameter set. This represents the shape parameter in an unknown Weibull distribution parameter set; The second objective function includes the target Weibull distribution parameter set input to the target fatigue life distribution function model, and the absolute value of the difference between the target Weibull distribution parameter set and each of the corrected life unreliability values ​​is less than a preset value.

[0049] In one embodiment, after obtaining each corrected lifetime unreliability, any corrected lifetime unreliability is taken as the lifetime unreliability. Substitute into the target fatigue life distribution function model constructed from the fatigue life distribution function. Determine the unknown Weibull distribution parameter set corresponding to the unreliability of the correction lifetime. ;in, This represents the location parameter in the unknown Weibull distribution parameter set. This represents the size parameter in an unknown Weibull distribution parameter set. This represents the shape parameter in the unknown Weibull distribution parameter set.

[0050] Then, the unknown Weibull distribution parameter sets corresponding to each corrected lifetime unreliability are subjected to selection, crossover, and mutation processing using a second genetic algorithm. Specifically, the unknown Weibull distribution parameter sets... As the initial population P(t), each individual in P(t) is encoded and substituted into the second objective function to evaluate its fitness. Then, based on the fitness of each individual, selection, crossover, and mutation operations are performed to obtain the next population P(t+1). This next population P(t+1) is then used as the initial population P(t), and the above operations are repeated until the preset number of generations is reached. The individual with the highest fitness is then output as the optimal solution. In the second genetic algorithm, the encoding type can be double-precision real number encoding, the fitness scale is set to rank, and the selection operation uses roulette wheel selection to randomly select individuals. The crossover operation is set to distributed, randomly generating genetic binary vectors and crossing them according to 0 to 1. Thus, through the first genetic algorithm, an optimal set of individuals can be found. This ensures that the absolute value of the difference between the target Weibull distribution parameter input to the target fatigue life distribution function model and the value obtained from each corrected life unreliability is less than a preset value. This makes the life unreliability obtained through the target Weibull distribution parameter as close as possible to each corrected life unreliability, thus avoiding divergence in the iteration process.

[0051] The second genetic algorithm is used to optimize the unreliability of each corrected lifetime, thereby using two genetic algorithms to estimate the Weibull distribution parameters of the mechanical component, improving the iteration speed of the Weibull distribution parameter acquisition process, and further improving the accuracy of Weibull distribution parameter estimation.

[0052] In one embodiment, the second objective function further includes the linear correlation between the dependent and independent variables in the linear equation determined in the preset model by the target Weibull distribution parameter set being greater than a preset value; Wherein, the linear equation is The preset model includes ; Indicates the independent variable. Indicates the dependent variable. This represents the mechanical component obtained by inputting the target Weibull distribution parameter set into the target fatigue life distribution function model. The target lifetime is unreliable.

[0053] In one embodiment, in addition to ensuring that the absolute value of the difference between the target Weibull distribution parameter set input to the target fatigue life distribution function model and each corrected life unreliability value is less than a preset value, it is also necessary to set the independent and dependent variables transformed using Equation 2. To satisfy a high degree of linear correlation, i.e., the correlation coefficient in Equation 3. The larger the value, the better, in order to further avoid divergence in the iterative process of the second genetic algorithm.

[0054] The device for obtaining Weibull distribution parameters provided in this application is described below. The device for obtaining Weibull distribution parameters described below can be referred to in correspondence with the method for obtaining Weibull distribution parameters described above.

[0055] In one embodiment, such as Figure 2 As shown, a device for obtaining Weibull distribution parameters is provided, comprising: The parameter prediction module 210 is used to perform parameter prediction based on the obtained fatigue failure life of each mechanical component and obtain the initial Weibull distribution parameters. The data processing module 220 is used to determine the initial life unreliability of each mechanical component based on the initial Weibull distribution parameters, and then optimize the initial life unreliability based on the first genetic algorithm to obtain the corrected life unreliability of each mechanical component. The parameter acquisition module 230 is used to acquire the target Weibull distribution parameters of the mechanical component based on the unreliability of each of the correction lifespans.

[0056] By pre-estimating the fatigue failure life of each mechanical component and obtaining the initial Weibull distribution parameters, the initial life unreliability of each mechanical component is determined based on the initial Weibull distribution parameters. A genetic algorithm is then used to optimize each initial life unreliability, resulting in a corrected life unreliability for each mechanical component. This leverages the characteristics of the genetic algorithm to make the obtained life unreliability of each mechanical component closer to the true value. Based on the corrected life unreliability, which is closer to the true value, the target Weibull distribution parameters of the mechanical component are obtained, thus making the obtained Weibull distribution parameters even closer to the true value. This improves the accuracy of the obtained Weibull parameters and consequently enhances the accuracy of the reliability assessment of the mechanical components.

[0057] In one embodiment, the parameter estimation module 210 is specifically used for: Based on the right approximation estimation method, the fatigue failure life is pre-estimated to obtain the initial Weibull distribution parameters.

[0058] In one embodiment, the data processing module 220 is specifically used for: The initial Weibull distribution parameters and each fatigue failure life are input into the initial fatigue life distribution function model to obtain the initial life unreliability of each mechanical component. Wherein, the initial fatigue life distribution function model is ; Indicates mechanical parts fatigue failure life The initial lifetime is unreliable. The location parameter represents the initial Weibull distribution parameter. The size parameter represents the initial Weibull distribution parameters. The shape parameter represents the initial Weibull distribution parameters.

[0059] In one embodiment, the data processing module 220 is specifically used for: Based on the mathematical model of the initial lifetime unreliability, determine the unknown parameter set corresponding to each of the initial lifetime unreliability; The unknown parameter set is selected, crossed, and mutated according to the first genetic algorithm to obtain a target unknown parameter set that satisfies the first objective function of the first genetic algorithm. Input the target unknown parameter set into the mathematical model to determine each corrected lifetime unreliability corresponding to each initial lifetime unreliability; The mathematical model includes ; ; Let represent the first objective function. Indicates an unknown parameter set. Indicates mechanical parts j The set of unknown parameters corresponding to the initial lifetime unreliability.

[0060] In one embodiment, the mathematical model further includes: .

[0061] In one embodiment, the parameter acquisition module 230 is specifically used for: Based on the input of each of the corrected life unreliability values ​​into the target fatigue life distribution function model, obtain the set of unknown Weibull distribution parameters that correspond one-to-one with each of the corrected life unreliability values. The second genetic algorithm is used to select, crossover, and mutate each of the unknown Weibull distribution parameter sets to obtain the target Weibull distribution parameter set that satisfies the second objective function of the second genetic algorithm. Determine the Weibull distribution parameters based on the target Weibull distribution parameter set; The target fatigue life distribution function model is as follows: ; Indicates mechanical parts fatigue life, Indicates mechanical parts The calibration lifetime is unreliable. This represents the location parameter in the unknown Weibull distribution parameter set. This represents the size parameter in an unknown Weibull distribution parameter set. This represents the shape parameter in an unknown Weibull distribution parameter set; The second objective function includes the target Weibull distribution parameter set input to the target fatigue life distribution function model, and the absolute value of the difference between the target Weibull distribution parameter set and each of the corrected life unreliability values ​​is less than a preset value.

[0062] In one embodiment, the second objective function further includes the linear correlation between the dependent and independent variables in the linear equation determined in the preset model by the target Weibull distribution parameter set being greater than a preset value; Wherein, the linear equation is The preset model includes ; Indicates the independent variable. Indicates the dependent variable. This represents the mechanical component obtained by inputting the target Weibull distribution parameter set into the target fatigue life distribution function model. The target lifetime is unreliable.

[0063] Figure 3An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 can call a computer program in the memory 830 to execute a method for obtaining Weibull distribution parameters, such as including: Based on the obtained fatigue failure life of each mechanical component, parameter pre-estimation is performed to obtain the initial Weibull distribution parameters; After determining the initial lifetime unreliability of each mechanical component based on the initial Weibull distribution parameters, the initial lifetime unreliability is optimized according to the first genetic algorithm to obtain the corrected lifetime unreliability of each mechanical component. Based on the unreliability of each correction lifetime, the target Weibull distribution parameters of the mechanical component are obtained.

[0064] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0065] On the other hand, embodiments of this application also provide a storage medium, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the Weibull distribution parameter acquisition method provided in the above embodiments, such as including: Based on the obtained fatigue failure life of each mechanical component, parameter pre-estimation is performed to obtain the initial Weibull distribution parameters; After determining the initial lifetime unreliability of each mechanical component based on the initial Weibull distribution parameters, the initial lifetime unreliability is optimized according to the first genetic algorithm to obtain the corrected lifetime unreliability of each mechanical component. Based on the unreliability of each correction lifetime, the target Weibull distribution parameters of the mechanical component are obtained.

[0066] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0067] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for obtaining Weibull distribution parameters, characterized in that, include: Based on the obtained fatigue failure life of each mechanical component, parameter pre-estimation is performed to obtain the initial Weibull distribution parameters; After determining the initial lifetime unreliability of each mechanical component based on the initial Weibull distribution parameters, the initial lifetime unreliability is optimized according to the first genetic algorithm to obtain the corrected lifetime unreliability of each mechanical component. Based on the unreliability of each correction lifetime, obtain the target Weibull distribution parameters of the mechanical component; Based on the initial Weibull distribution parameters, the initial lifetime unreliability of each mechanical component is determined, including: The initial Weibull distribution parameters and each fatigue failure life are input into the initial fatigue life distribution function model to obtain the initial life unreliability of each mechanical component. Wherein, the initial fatigue life distribution function model is ; Indicates the first Fatigue failure life of individual mechanical components The initial lifetime is unreliable. The location parameter represents the initial Weibull distribution parameter. The size parameter represents the initial Weibull distribution parameters. The shape parameter represents the initial Weibull distribution parameters; The initial lifetime unreliability of each component is optimized using a first genetic algorithm to obtain the corrected lifetime unreliability of each component, including: Based on the mathematical model of the initial lifetime unreliability, determine the unknown parameter set corresponding to each of the initial lifetime unreliability; The unknown parameter set is selected, crossed, and mutated according to the first genetic algorithm to obtain the target unknown parameter set that minimizes the value of the first objective function of the first genetic algorithm. Input the target unknown parameter set into the mathematical model to determine each corrected lifetime unreliability corresponding to each initial lifetime unreliability; The mathematical model includes ; , The initial life unreliability of the i-th mechanical component as characterized in the mathematical model; Let represent the first objective function. Indicates an unknown parameter set. This represents the set of unknown parameters corresponding to the initial life unreliability of the j-th mechanical component; , , represents the vector composed of unknown parameters in each initial lifetime unreliability.

2. The method for obtaining Weibull distribution parameters according to claim 1, characterized in that, Based on the obtained fatigue failure lives of each mechanical component, parameter pre-estimation is performed to obtain initial Weibull distribution parameters, including: Based on the right approximation estimation method, the fatigue failure life is pre-estimated to obtain the initial Weibull distribution parameters.

3. The method for obtaining Weibull distribution parameters according to claim 1, characterized in that, Based on the aforementioned correction lifetime unreliability, the target Weibull distribution parameters of the mechanical component are obtained, including: Based on the input of each of the corrected life unreliability values ​​into the target fatigue life distribution function model, obtain the set of unknown Weibull distribution parameters that correspond one-to-one with each of the corrected life unreliability values. The unknown Weibull distribution parameter sets are selected, crossed, and mutated according to the second genetic algorithm to obtain the target Weibull distribution parameter set that satisfies the second objective function of the second genetic algorithm. Determine the target Weibull distribution parameters based on the target Weibull distribution parameter set; The target fatigue life distribution function model is as follows: ; Indicates mechanical parts fatigue life, Indicates mechanical parts The calibration lifetime is unreliable. This represents the location parameter in the unknown Weibull distribution parameter set. This represents the size parameter in an unknown Weibull distribution parameter set. This represents the shape parameter in an unknown Weibull distribution parameter set; The second objective function includes: the absolute value of the difference between the target Weibull distribution parameter set input to the target fatigue life distribution function model and each of the corrected life unreliability values ​​is less than a preset value.

4. The method for obtaining Weibull distribution parameters according to claim 3, characterized in that, The second objective function further includes: after the target Weibull distribution parameter set is input into the preset model, the linear correlation between the dependent and independent variables of the resulting linear equation is greater than a preset value; Wherein, the linear equation is The preset model includes ; Indicates the independent variable. Indicates the dependent variable. This represents the mechanical component obtained by inputting the target Weibull distribution parameter set into the target fatigue life distribution function model. The target lifetime is unreliable.

5. A device for acquiring Weibull distribution parameters, characterized in that, include: The parameter prediction module is used to perform parameter prediction based on the obtained fatigue failure life of each mechanical component and obtain the initial Weibull distribution parameters. The data processing module is used to determine the initial life unreliability of each mechanical component based on the initial Weibull distribution parameters, and then optimize the initial life unreliability based on the first genetic algorithm to obtain the corrected life unreliability of each mechanical component. The parameter acquisition module is used to acquire the target Weibull distribution parameters of the mechanical component based on the unreliability of each of the correction lifespans. The parameter estimation module is specifically used for: The initial Weibull distribution parameters and each fatigue failure life are input into the initial fatigue life distribution function model to obtain the initial life unreliability of each mechanical component. Wherein, the initial fatigue life distribution function model is ; Indicates the first Fatigue failure life of individual mechanical components The initial lifetime is unreliable. The location parameter represents the initial Weibull distribution parameter. The size parameter represents the initial Weibull distribution parameters. The shape parameter represents the initial Weibull distribution parameters; The data processing module is specifically used for: Based on the mathematical model of the initial lifetime unreliability, determine the unknown parameter set corresponding to each of the initial lifetime unreliability; The unknown parameter set is selected, crossed, and mutated according to the first genetic algorithm to obtain the target unknown parameter set that minimizes the value of the first objective function of the first genetic algorithm. Input the target unknown parameter set into the mathematical model to determine each corrected lifetime unreliability corresponding to each initial lifetime unreliability; The mathematical model includes ; , The initial life unreliability of the i-th mechanical component as characterized in the mathematical model; Let represent the first objective function. Indicates an unknown parameter set. This represents the set of unknown parameters corresponding to the initial life unreliability of the j-th mechanical component; , , represents the vector composed of unknown parameters in each initial lifetime unreliability.

6. An electronic device comprising a processor and a memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the Weibull distribution parameter acquisition method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the Weibull distribution parameter acquisition method according to any one of claims 1 to 4.

Images