A method for testing the time-varying reliability of a phononic crystal, a computing device, and a storage medium

Abstract

The present invention relates to the field of phononic crystals, and discloses a method for testing the time-varying reliability of a phononic crystal, a computing device, and a storage medium, and includes the steps of: determining a sample space according to the parameters of the phononic crystal; constructing a reliability test model according to the sample space and the failure conditions of the phononic crystal; constructing a neural network model according to the reliability test model; and predicting the failure rate of the phononic crystal during the service time according to the neural network model and the sample space. By constructing a reliability test model of the phononic crystal and then constructing a neural network model through the reliability test model, the present invention can predict the failure rate of the phononic crystal during the service time, avoiding the inability to judge the performance of the phononic crystal due to the lack of parameters of the phononic crystal, and realizing the determination of the performance of the phononic crystal during the service time.

Description

Technical Field

[0001] The present invention relates to the field of phononic crystals, and particularly to a method for testing the time-varying reliability of phononic crystals, a computing device, and a storage medium. Background Art

[0002] Phononic crystals generally refer to functional materials with elastic (acoustic) wave bandgaps, in which materials with different density and elastic parameters are compounded in a periodic structure. However, the self-structure and material properties of phononic crystals will cause the phononic crystals to exhibit time-varying uncertainty. Time-varying uncertainty means that one or more parameters of the phononic crystal will change over time, resulting in fluctuations in the performance of the phononic crystal. Therefore, it becomes particularly important to test the time-varying reliability of phononic crystals.

[0003] In the prior art, when testing phononic crystals, the characteristic parameters of the phononic crystal are often set through sufficient sample data, and then a test model of the phononic crystal is constructed according to the set characteristic parameters to test the performance of the phononic crystal. However, in actual work and research, it is often difficult to obtain the sample data of the phononic crystal, and accurate characteristic parameters cannot be constructed, resulting in inaccurate test results.

[0004] Therefore, a new method for testing the time-varying reliability of phononic crystals is needed. Summary of the Invention

[0005] Therefore, the present invention provides a method for testing the time-varying reliability of phononic crystals in an attempt to solve or at least alleviate the problems existing above.

[0006] According to one aspect of the present invention, there is provided a method for testing the time-varying reliability of phononic crystals, which is suitable for execution in a computing device. The method includes the steps of: determining a sample space according to the parameters of the phononic crystal; constructing a reliability test model according to the sample space and the failure conditions of the phononic crystal; constructing a neural network model according to the reliability test model; and predicting the failure rate of the phononic crystal during the service time according to the neural network model and the sample space.

[0007] Optionally, in the method according to the present invention, the parameters of the phononic crystal include: random variable parameters, random process parameters, interval variable parameters, and interval process parameters; constructing a reliability test model according to the sample space and the failure conditions of the phononic crystal includes the steps of: converting the random process parameters in the crystal parameters to obtain equivalent random variable parameters; converting the interval process parameters in the crystal parameters to obtain equivalent interval variable parameters; converting the time parameters of the service time of the phononic crystal to equivalent distribution time parameters; and constructing a reliability test model according to the random variable parameters, equivalent random variable parameters, interval variable parameters, equivalent interval variable parameters, equivalent distribution time parameters, and failure conditions.

[0008] Optionally, in the method according to the present invention, the failure condition includes: during the service time, when the lower limit of the bandgap of the phononic crystal is greater than the preset frequency, the phononic crystal fails.

[0009] Optionally, in the method according to the present invention, constructing a neural network model according to the reliability test model includes the steps of: determining a training set according to a set of sample points; training the neural network model according to the reliability test model and the training set.

[0010] Optionally, in the method according to the present invention, determining a training set according to a set of sample points includes the steps of: transforming the sample points in the sample space to obtain a transformed set of sample points; determining a training set according to the transformed set of sample points.

[0011] Optionally, in the method according to the present invention, determining a training set according to the transformed set of sample points includes the steps of: determining the sample weight of each sample point in the transformed set of sample points; determining the eigenvalue of each sample point according to the sample weight; determining a training set from the transformed set of sample points according to the eigenvalue.

[0012] Optionally, in the method according to the present invention, transforming the sample points in the sample space includes the steps of: performing equivalent uncertainty transformation on each sample point in the sample space to obtain a set of sample points independent of time.

[0013] Optionally, in the method according to the present invention, it further includes the steps of: determining whether the failure rate of the instantaneous reliability calculated by the neural network model satisfies a stop rule; if it does not satisfy the stop rule, setting the number of iterations, and determining incremental sample points according to the number of iterations; determining a new set of sample points according to the incremental sample points and the set of sample points; training a new neural network model according to the new set of sample points until a neural network model that satisfies the stop rule is trained.

[0014] Optionally, in the method according to the present invention, determining incremental sample points according to the number of iterations includes the steps of: determining a set of candidate points according to the training set, and determining a first set of points according to the set of candidate points; determining a second set of points from the first set of points by weighted sampling; determining incremental sample points from the second set of points according to an active learning function.

[0015] Optionally, in the method according to the present invention, determining incremental sample points from the second set of points according to an active learning function includes the steps of: determining the uncertainty of each sample point in the second set of points; determining the Euclidean distance between each sample point in the second set of points and the training set; inputting the uncertainty and Euclidean distance of each sample point into the active learning function to obtain the function value of each sample point; taking the sample point with the smallest function value in the second set of points as the incremental sample point.

[0016] Optionally, in the method according to the present invention, determining the uncertainty of each sample point in the second point set includes the steps of: generating a plurality of training complementary sets according to the sample set generated last time; generating complementary neural network models according to each training complementary set; determining the uncertainty of each sample point according to the complementary neural network models and the neural network model generated last time.

[0017] Optionally, in the method according to the present invention, predicting the failure rate of a phononic crystal during the service time according to the neural network model and the sample space includes the steps of: calculating the failure rate of the mixed time-varying reliability of the phononic crystal according to the neural network model and the sample points in the sample space.

[0018] Optionally, in the method according to the present invention, it further includes the steps of: calculating the coefficient of variation of the failure rate according to the failure rate of the mixed time-varying reliability of the phononic crystal; determining whether the coefficient of variation is greater than a preset coefficient threshold; if the coefficient of variation is not greater than the preset coefficient threshold, adding new sample points to the sample space to obtain a new sample space; training the neural network model according to the new sample space until the coefficient of variation of the neural network model is greater than a preset minimization threshold.

[0019] According to another aspect of the present invention, there is provided a computing device, including: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include those for executing the time-varying reliability test method of the phononic crystal according to the present invention.

[0020] According to still another aspect of the present invention, there is provided a computer-readable storage medium storing one or more programs, and the one or more programs include instructions which, when executed by a computing device, cause the computing device to execute the time-varying reliability test method of the phononic crystal according to the present invention.

[0021] The present invention discloses a time-varying reliability test method for a phononic crystal, which is suitable for being executed in a computing device. The method includes the steps of: determining a sample space according to the parameters of the phononic crystal; constructing a reliability test model according to the sample space and the failure conditions of the phononic crystal; constructing a neural network model according to the reliability test model; predicting the failure rate of the phononic crystal during the service time according to the neural network model and the sample space. By constructing a reliability test model of the phononic crystal and then constructing a neural network model through the reliability test model, the present invention can predict the failure rate of the phononic crystal during the service time, avoiding the inability to judge the performance of the phononic crystal due to the lack of parameters of the phononic crystal, and realizing the determination of the performance of the phononic crystal during the service time. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] To achieve the above and related purposes, certain illustrative aspects are described herein in conjunction with the following description and drawings, which indicate various ways in which the principles disclosed herein can be practiced, and all aspects and their equivalent aspects are intended to fall within the scope of the claimed subject matter. By reading the following detailed description in conjunction with the drawings, the above and other objects, features, and advantages of the present disclosure will become more apparent. Throughout the present disclosure, like reference numerals generally refer to like components or elements.

[0023] Figure 1 FIG. shows a schematic diagram of a time-varying reliability test method 100 for a phononic crystal according to an exemplary embodiment of the present invention;

[0024] Figure 2 FIG. shows a structural block diagram of a computing device 200 according to an exemplary embodiment of the present invention;

[0025] Figure 3a FIG. shows a schematic diagram of the propagation of sound waves in a phononic crystal according to an exemplary embodiment of the present invention;

[0026] Figure 3b FIG. shows a schematic structural diagram of a phononic crystal according to an exemplary embodiment of the present invention;

[0027] Figures 4a to 4d FIG. shows a schematic diagram of equivalent uncertainty transformation according to an exemplary embodiment of the present invention;

[0028] Figure 5 FIG. shows a schematic diagram of time-varying reliability analysis of a phononic crystal according to an exemplary embodiment of the present invention. Detailed Description of the Invention

[0029] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art. Like reference numerals generally refer to like components or elements.

[0030] The present invention proposes a time-varying reliability test method for a phononic crystal. The time-varying reliability test method for a phononic crystal is suitable for execution in a computing device. Figure 2 FIG. shows a structural block diagram of a computing device 200 according to an exemplary embodiment of the present invention.

[0031] In a basic configuration, computing device 200 includes at least one processing unit 220 and system memory 210. According to one aspect, depending on the configuration and type of the computing device, system memory 210 includes, but is not limited to, volatile storage (e.g., random access memory), non-volatile storage (e.g., read only memory), flash memory, or any combination of such memories. According to one aspect, system memory 210 includes an operating system 211.

[0032] According to one aspect, the operating system 211, for example, is suitable for controlling the operation of computing device 200. Additionally, the examples are practiced in conjunction with a graphics library, other operating systems, or any other application programs, and are not limited to any particular application or system. In Figure 2 the basic configuration is illustrated by those components within dashed line 215. According to one aspect, computing device 200 has additional features or functionality. For example, according to one aspect, computing device 200 includes additional data storage devices (removable and / or non-removable), such as magnetic disks, optical disks, or magnetic tapes.

[0033] As stated above, according to one aspect, program modules 212 are stored in system memory 210. According to one aspect, program modules 212 may include one or more application programs, and the present invention does not limit the type of application programs. For example, applications also include: email and contact applications, word processing applications, spreadsheet applications, database applications, slide show applications, painting or computer-aided applications, web browser applications, etc.

[0034] According to one aspect, the examples can be practiced in a circuit including discrete electronic components, a packaged or integrated electronic chip containing logic gates, a circuit utilizing a microprocessor, or on a single chip containing electronic components or a microprocessor. For example, the examples can be practiced via a system on a chip (SOC) in which each or many of the components shown in Figure 2 can be integrated on a single integrated circuit. According to one aspect, such an SOC device may include one or more processing units, graphics units, communication units, system virtualization units, and various application functions, all integrated (or "burned") onto a chip substrate as a single integrated circuit. When operating via an SOC, the functions described herein can be operated via dedicated logic integrated with other components of computing device 200 on a single integrated circuit (chip). Embodiments of the present invention can also be practiced using other technologies capable of performing logical operations (such as AND, OR, and NOT), including but not limited to mechanical, optical, fluidic, and quantum technologies. Additionally, embodiments of the present invention can be practiced within a general purpose computer or in any other circuit or system.

[0035] According to one aspect, the computing device 200 may also have one or more input devices 231, such as a keyboard, a mouse, a pen, a voice input device, a touch input device, etc. An output device 232, such as a display, a speaker, a printer, etc., may also be included. The foregoing devices are examples and other devices may also be used. The computing device 200 may include one or more communication connections 233 that allow communication with other computing devices 200. Examples of suitable communication connections 233 include, but are not limited to: RF transmitter, receiver, and / or transceiver circuits; Universal Serial Bus (USB), parallel, and / or serial ports.

[0036] Embodiments of the present invention also provide a non-transitory readable storage medium storing instructions for causing the computing device to execute the method according to the embodiments of the present invention. The readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information may be computer-readable instructions, data structures, program modules, or other data. Examples of the readable storage medium include, but are not limited to: Phase Change Memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technologies, Compact Disc Read Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, magnetic cassette tapes, magnetic disk storage or other magnetic storage devices, or any other non-transitory readable storage medium.

[0037] According to one aspect, the communication medium is implemented by computer-readable instructions, data structures, program modules, or other data in a modulated data signal (e.g., a carrier wave or other transmission mechanism), and includes any information delivery medium. According to one aspect, the term "modulated data signal" describes a signal having one or more characteristic sets or a signal that has been altered in a manner that encodes information in the signal. By way of example and not limitation, the communication medium includes wired media such as a wired network or a direct wired connection, and wireless media such as acoustic, radio frequency (RF), infrared, and other wireless media.

[0038] It should be noted that although the foregoing computing device only shows the processing unit 220, the system memory 210, the input device 231, the output device 232, and the communication connection 233, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, those skilled in the art can understand that the foregoing device may also only include the components necessary to implement the solution of the embodiments of this specification, and do not have to include all the components shown in the figure.

[0039] Next, according to Figure 1 the specific implementation process of the time-varying reliability test method for the phononic crystal of the present invention will be described in detail. Figure 1 FIG. shows a schematic diagram of a time-varying reliability test method 100 for a phononic crystal according to an exemplary embodiment of the present invention. As Figure 1 shown, first, step S110 is executed to determine the sample space according to the parameters of the phononic crystal.

[0040] In a phononic crystal, materials with different density and elastic parameters related to the propagation of elastic waves are periodically compounded together in a structure similar to that of a natural crystal. The materials distributed at lattice points and not connected to each other are called scatterers, and the background medium material connected as a whole is called the matrix. The so-called elastic wave bandgap means that there are no elastic wave eigenmodes within a certain frequency range, that is, elastic waves within this frequency range are prohibited from propagating.

[0041] According to the shape of the scatterer and its periodic distribution form in the matrix, phononic crystals can be divided into three types: one-dimensional (layered), two-dimensional, and three-dimensional phononic crystals. According to the periodicity of the lattice, common two-dimensional phononic crystals are further divided into: square lattice, triangular lattice, hexagonal lattice, etc. According to an embodiment of the present invention, the phononic crystal for which the present invention performs time-varying reliability testing is specifically a two-dimensional phononic crystal.

[0042] Figure 3a FIG. shows a schematic diagram of the propagation of sound waves in a phononic crystal according to an exemplary embodiment of the present invention.

[0043] The phononic crystal has various parameters in terms of its own structure and material properties. These parameters can be classified into random variable parameters, random process parameters, interval variable parameters, and interval process parameters according to continuity and time correlation.

[0044] According to an embodiment of the present invention, the random variable parameters include the Young's modulus and density of the scatterer and the matrix; the random process parameters include the side length of the matrix; the interval variable parameters include the Poisson's ratio of the scatterer and the matrix; the interval process parameters include the diameter of the scatterer.

[0045] Figure 3b FIG. shows a schematic diagram of the structure of a phononic crystal according to an exemplary embodiment of the present invention. As Figure 3b shown, the phononic crystal includes a scatterer and a matrix; wherein, the diameter of the scatterer is R C , and the side length of the matrix is a.

[0046] According to the value ranges of the various parameters of the phononic crystal, the sample space for taking values of the parameters of the phononic crystal can be determined. The sample space includes multiple sample points. Each sample point includes multiple parameters, and the value of each parameter falls within the value range of the corresponding parameter of the phononic crystal. Each sample point in the sample space represents a possible parameter value situation of the phononic crystal.

[0047] According to an embodiment of the present invention, in order to simulate the parameter value situations of the phononic crystal, Monte Carlo simulation can be used, and the sample points are randomly distributed in the sample space. At this time, each sample point in the sample space is a Monte Carlo sample point. The number of sample points in the sample space is N mc ones, and the set of sample points composed of N mc sample points is the sample point set S MCS .

[0048] Subsequently, step S120 is executed to construct a reliability test model according to the sample space and the failure condition of the phononic crystal.

[0049] According to an embodiment of the present invention, in order to explore the reliability of the phononic crystal, a reliability test model needs to be constructed to test the phononic crystal. The basic rule of the reliability test model is that when the lower limit of the band gap of the elastic wave band gap of the phononic crystal is less than the preset frequency during the service time of the phononic crystal, it is considered that the structure of the phononic crystal fails. When the lower limit of the band gap of the elastic wave band gap of the phononic crystal is less than the preset frequency, the blocking performance of the phononic crystal for elastic waves does not meet the expectation, and the phononic crystal fails. Therefore, the failure condition of the phononic crystal is: during the service time, when the lower limit of the band gap of the phononic crystal is greater than the preset frequency, the phononic crystal fails. According to the failure condition of the phononic crystal, a reliability test model of the phononic crystal can be established.

[0050] According to an embodiment of the present invention, the lower limit of the band gap of the phononic crystal can be calculated by the finite element method. The lower limit of the band gap is calculated from random variable parameters, random process parameters, interval variable parameters, and interval process parameters, and its specific calculation result is: wherein, X = (X1, X2,... X m ), represents an m-dimensional vector composed of random variable parameters; Y = (Y1, Y2,... Y n ), represents an n-dimensional vector composed of interval variable parameters; S(t) = [S1(t), S2(t),... S k (t)], represents a vector composed of k random process parameters; I(t) = [I1(t), I2(t),... I l (t)], represents a vector composed of l interval process parameters; t is the preset service time of the phononic crystal.

[0051] The preset frequency f thIt can be set as needed. In the present invention, there is no limitation on the specific value of the preset frequency f th According to an embodiment of the present invention, the preset frequency f th can be set to 268.

[0052] According to the failure condition of the phononic crystal, the limit state equation for calculating whether the phononic crystal fails can be expressed as:

[0053]

[0054] Among them, g(X, Y, S(t), I(t), t) represents the interpolation between the preset frequency and the lower limit of the bandgap of the phononic crystal. When g(X, Y, S(t), I(t), t) is less than 0, the phononic crystal fails.

[0055] The time-varying statement period of the phononic crystal is [0, T L , and the service time of the phononic crystal within the time-varying statement period is T, where 0 ≤ T ≤ T L . The failure probability of the phononic crystal during the service time can be expressed as:

[0056]

[0057] When the parameters of the phononic crystal take specific values, the maximum and minimum values of the lower limit of the bandgap can be calculated. Correspondingly, substituting the lower limit of the bandgap with the maximum value into the limit state equation, the lower bound of the failure probability of the phononic crystal can be calculated; substituting the lower limit of the bandgap with the minimum value into the limit state equation, the upper bound of the failure probability of the phononic crystal can be calculated. The lower and upper bounds of the failure rate of the phononic crystal during the service time can be expressed as:

[0058]

[0059]

[0060] In order to facilitate the reliability test of the phononic crystal and reduce the influence of the service time on the random process parameters and interval process parameters, therefore, the random process parameters and interval process parameters are applied with equivalent uncertainty transformation to obtain the equivalent random variable parameters and equivalent interval variable parameters independent of time, so as to transform the time-varying reliability analysis into a time-invariant reliability analysis.

[0061] According to an embodiment of the present invention, the equivalent uncertainty transformation can be specifically realized through the following formula:

[0062]

[0063] Among them, P t = [S(t), I(t)], P t represents the time-varying uncertain parameters, including S(t) and I(t). Pt ′ = [S′, I′], P t ′ represents the transformed time-invariant uncertain parameters, including S′ and I′. S′ is the equivalent random variable parameter after the transformation of the random process parameter, and I′ is the equivalent interval variable parameter after the transformation of the interval process parameter.

[0064] Figures 4a to 4d Shows a schematic diagram of the equivalent uncertain transformation according to an exemplary embodiment of the present invention. Figures 4a to 4d Taking the transformation of the interval process parameter into the equivalent interval variable parameter as an example, the process of the equivalent uncertain transformation is described.

[0065] As Figure 4a shown, the interval process parameter I(t) is related to the time t; the upper bound function I U (t) and the lower bound function I L (t) determine the value range of the interval process parameter I(t) at time t. Where the time t is continuous time, and its value range is [0, T].

[0066] As Figure 4b shown, first take N discrete times on the continuous time t: t1 ~ t N . Determine the value of the upper bound function I U (t) at the discrete time, and obtain the discrete upper bound function I U (t i ) of the interval process parameter I(t). Determine the value of the lower bound function I L (t) at the discrete time ti, and obtain the discrete lower bound function I L (t i ). Where i is a positive integer between 1 and N, i = 1, 2,..., N.

[0067] As Figure 4c shown, the values of the discrete upper bound function and the discrete lower bound function at the discrete time t i determine the value range of the interval process parameter I(t) at the discrete time t i . Subsequently, take the value range of the interval process parameter I(t) at the discrete time t i as the interval variable I i , where i is a positive integer between 1 and N, i = 1, 2,..., N.

[0068] As Figure 4d shown, count the interval variables I1 ~ I N , and determine the value probability function f PDF (I) of the interval process parameter I(t) in the value space. The value probability function f PDF (I) is the equivalent interval variable parameter independent of time.

[0069] According to an embodiment of the present invention, the time parameter t describing the service time of the phononic crystal is also transformed into an equivalent distributed time parameter t' evenly distributed within the time-varying life cycle [0, T L of the phononic crystal, and t' ~ U(0, T L ).

[0070] According to the random variable parameters, interval variable parameters, as well as the transformed equivalent random variable parameters, equivalent interval variable parameters, and equivalent distributed time parameter, a reliability test model can be constructed.

[0071] In the finally obtained reliability test model, the limit state function used to calculate whether the phononic crystal fails is as follows:

[0072]

[0073] Correspondingly, the failure probability of the phononic crystal is calculated through the reliability test model, and can be calculated by the following formula:

[0074] P f = Pr(g(X, Y, S', I', t') < 0)

[0075] Next, step S130 is executed to construct a neural network model according to the reliability test model.

[0076] When the failure probability of the phononic crystal needs to be predicted according to the reliability test model, the neural network model is trained according to the reliability test model to predict the failure probability of the phononic crystal. According to an embodiment of the present invention, the neural network model adopted can specifically be a DNN neural network model, and the present invention does not limit the specific type of the neural network model.

[0077] When training the neural network model, first perform an equivalent uncertainty transformation on the sample point set S MCS constituting the sample space, and transform the time-related parameters of the sample points in the sample point set into time-independent parameters according to the formula of the equivalent uncertainty transformation to obtain the transformed sample point set so as to reduce the influence of time on the random process parameters and interval process parameters during training. Subsequently, sample points are selected from the generated sample point set to obtain the training set S. The number of sample points included in the training set S is M, and M ≤ N mc , N mc is the number of sample points included in the sample point set . The present invention does not limit the specific number of sample points included in the training set S, and can be determined as needed by comprehensively considering factors such as training time.

[0078] According to an embodiment of the present invention, when selecting sample points from the sample space to obtain the training set S, the sample points can be selected by weighted sampling. Weighted sampling is used to solve the problem of unbalanced sampling. Since random sampling collects more samples in the area with a large probability in the sampling space, and the sample points near the limit state surface are generally in the area with a small probability in the sampling space, in order to ensure that the sampled samples are as uniform as possible, large weights are given to the sample points with small probabilities through weighted sampling to ensure that the sampled samples are as uniform as possible.

[0079] When obtaining the training set S from the sample space by weighted sampling, first, for the sample points V in the sample space t (i) Calculate their sample weights, which can be specifically calculated by the following formula:

[0080]

[0081] where w (i) is the sample weight of the sample point, and f(V t (i) ) is the probability density of the sample point V t (i) in the sample space. The parameters of the sample point V t (i) are V t = [X, Y, S′, I′, t′].

[0082] Subsequently, calculate the eigenvalue of each sample point according to the sample weight, specifically calculated by the following formula:

[0083]

[0084] where u (i) can be set as a random number in the range of (0, 1). The present invention does not limit the specific value range of u (i) , and it can be specifically set according to needs.

[0085] Finally, sort each sample point in ascending order according to the eigenvalue, and select the first M sample points to obtain the training set S.

[0086] Sample point set Among them, the unselected sample points, that is, the sample points not in the training set S, are used as the candidate point set S * . The candidate point set S * Combined with the training set S can obtain the sample point set Sample point set Among them, the candidate point set S * is the complement of the training set S.

[0087] After obtaining the training set S, the neural network model is trained according to the training set S. After the neural network model is trained, the failure rate of the phononic crystal is predicted according to the neural network model. According to an embodiment of the present invention, when predicting the failure rate of the phononic crystal, the transformed sample point set is input into the trained neural network model to obtain the failure rate of the instantaneous reliability of the phononic crystal.

[0088] According to an embodiment of the present invention, in order to more accurately test the reliability of the phononic crystal, the present invention uses an iterative learning method to update the trained neural network model to obtain a more accurate failure rate of the phononic crystal.

[0089] According to an embodiment of the present invention, a stopping rule is set to determine when to end the iteration according to the stopping rule and stop updating the neural network model. When each iteration is performed and a new neural network model is determined, the failure rate of the phononic crystal is measured according to the new neural network model, and then it is judged whether to stop the iteration according to the stopping rule. If it is judged that the stopping rule is not satisfied, the iteration continues to update the application network model. If it is judged that the stopping rule is satisfied, the iteration stops and the loop exits.

[0090] The stopping rule includes:

[0091]

[0092]

[0093]

[0094] where k represents the k-th iteration process, n represents the n-th iteration process, and 1 ≤ n ≤ k. represents the failure rate of the phononic crystal calculated by the neural network model in the i-th iteration process, where n ≤ i ≤ k. ε th is a user-defined stopping threshold and can be determined as needed considering factors such as test time. According to an embodiment of the present invention, ε th can be set to 0.01. n is a user-defined iteration range that needs to be calculated; when judging the stopping rule, it is judged whether the stopping rule is satisfied according to the failure rate of the phononic crystal from the n-th to the k-th iteration.

[0095] According to an embodiment of the present invention, the initially generated neural network model is recorded as the first iteration, and its iteration number k = 1. Subsequently, the transformed sample point set is input into the trained neural network model to obtain the failure rate of the instantaneous reliability of the phononic crystal. Then it is judged whether the stopping rule is satisfied; if the stopping rule is satisfied, the iteration stops. Subsequently, according to the sample point set S MCSInput the initially generated neural network model and output the failure rate of the hybrid time-varying reliability of the phononic crystal. If the stopping rule is not satisfied, continue the iteration and update the neural network model.

[0096] When updating the neural network model, first determine the first point set from the candidate point set S * among them Specifically: Use the neural network model generated in the previous time to calculate the response value g of each sample point in the candidate point set S * The response value g is calculated according to the following formula for each sample point:

[0097]

[0098] Then, sort according to the absolute value of the response value of each sample point and select N S * sample points as the first point set According to an embodiment of the present invention, The size of can be calculated by the following formula:

[0099]

[0100] The present invention does not limit the specific value of and can be specifically determined according to training needs.

[0101] Subsequently, select M experimental points from the first point set in the way of weighted sampling as the second point set S K . The present invention does not limit the specific value of M and can be determined according to needs.

[0102] Then, use the active learning function to determine the incremental sample points V from the second point set S K among them new .

[0103] Specifically: First, use the K-fold cross-validation method to process the training set. Divide the training set into k training subsets equally, and each training subset includes the same number of sample points. Subsequently, each time during training, select one training subset from the k training subsets as the test subset, and the other training subsets as the training complement set to train the neural network model. Sequentially use each training subset in the k training subsets as the test subset, and then k supplementary neural network models can be trained.

[0104] Input the sample points into the supplementary neural network model to obtain the response value g. Train the supplementary neural network model according to the training complement set where l is a positive integer between 1 and k, l = 1, 2,..., k.

[0105] The response value is calculated according to the following formula:

[0106]

[0107] Input each sample point in the second point set S K into the neural network model trained with the training set in the previous time, and k supplementary neural network models trained with the training complement set, and calculate the uncertainty of each sample point. Specifically, it can be calculated through the following formula:

[0108]

[0109] where, us(V t i ) is the uncertainty of each sample point in the second point set S K . is the response value calculated by inputting the sample point into the neural network model trained with the training set in the previous time, is the response value calculated by inputting the sample point into the l-th supplementary neural network model.

[0110] Subsequently, calculate the Euclidean distance between each sample point in the second point set S K and the existing training set S. Specifically, it can be calculated through the following formula:

[0111]

[0112] where, d(V t i ) is the Euclidean distance between each sample point and the existing training set S, V t i is the sample point in the second point set S K , i is a positive integer between 1 and M, i = 1, 2,..., M. V t j is the sample point in the existing training set S, j is a positive integer between 1 and M, j = 1, 2,..., M.

[0113] Finally, determine the incremental sample point V new according to the Euclidean distance and uncertainty using the active learning function. Specifically, it can be calculated through the following formula:

[0114]

[0115] where, N(·) represents the normalization operation, SLF(V t i ) is to use the second point set S KThe function values obtained by inputting each sample point in [[]] into the active learning function. The value range of β is (0, 1), and the present invention does not limit the specific value of β. According to an embodiment of the present invention, the value of β can be 0.5, indicating that when calculating the active learning function, the Euclidean distance and uncertainty are comprehensively considered with equal importance.

[0116] For the second point set S K Sort the function values obtained by inputting each sample point in [[]] into the active learning function, and use the sample point with the smallest calculated function value as the incremental sample point V new .

[0117] Subsequently, add the incremental sample point to the sample set S to obtain a new sample set, and train a new neural network model.

[0118] According to the above steps, the neural network model is iteratively updated by using the active learning method. Each time, the most suitable incremental sample point is selected and added to the sample set to obtain a new sample set. The incremental sample points calculated in the above manner are far from the existing sample points, preventing problems in the model, ensuring the progress of the model, reducing the number of training sample points, making the sample points as close as possible to the limit state surface and also having high uncertainty.

[0119] When adding the incremental sample point to the sample set S to obtain a new sample set, remove the incremental sample point from the candidate point set S * to obtain a new candidate point set for subsequent iteration use.

[0120] After training the new neural network model, input the sample point set into the new neural network model to obtain the failure rate of the instantaneous reliability of the phononic crystal. Then judge whether the stop rule is satisfied; if the stop rule is satisfied, stop the iteration.

[0121] Finally, execute step S140, and predict the failure rate of the phononic crystal during the service time according to the neural network model and the sample space.

[0122] According to inputting the sample point set S MCS into the new neural network model, output the failure rate of the hybrid time-varying reliability of the phononic crystal. If the stop rule is not satisfied, continue the iteration, and let the iteration number k increase by 1; then continue to update the newly generated neural network model until the output failure rate of the instantaneous reliability satisfies the stop rule.

[0123] According to an embodiment of the present invention, after calculating the failure rate of the hybrid time-varying reliability of the phononic crystal, calculate the coefficient of variation of the failure rate according to the failure rate, which can be specifically calculated by the following formula:

[0124]

[0125] wherein, P f is the failure rate, N is the number of sample points in the sample space, specifically, it can be implemented as the Monte Carlo sampling number N mc .

[0126] Subsequently, compare the calculated coefficient of variation with the preset coefficient threshold. If the coefficient of variation is greater than the preset coefficient threshold, output the failure rate. According to an embodiment of the present invention, the preset coefficient threshold can be set to 0.05, and the present invention does not limit the specific value of the preset coefficient threshold.

[0127] If the coefficient of variation is less than the preset coefficient threshold, increase the number of sample points in the sample space to obtain a new sample space, and perform the above steps S110 to S1n0 according to the new sample space to obtain a new neural network model, calculate the failure rate and the coefficient of variation until the calculated coefficient of compilation is greater than the preset coefficient threshold.

[0128] In the reliability test model, when calculating the failure rate of the phononic crystal during the service time, it includes a lower bound and an upper bound Therefore, the final generated failure rate includes the upper bound and the lower bound of the failure rate, and the failure rate is between the upper bound and the lower bound of the failure rate.

[0129] Figure 5 shows a schematic diagram of the time-varying reliability analysis of a phononic crystal according to an exemplary embodiment of the present invention. As Figure 5 shown, first determine the parameters of the phononic crystal according to the self-structure and material properties of the phononic crystal; these parameters include: random variable parameters, random process parameters, interval variable parameters, and interval process parameters. According to each parameter, the specific value range of each parameter can determine the sample space of the phononic crystal parameter values. The sample space includes multiple sample points. When taking sample points in the sample space, N mc sample points can be generated by Monte Carlo, and the set composed of these sample points is the sample point set S MCS .

[0130] Subsequently, construct a reliability test model to test the phononic crystal. When constructing the reliability test model, when setting the reliability test model, determine the failure condition of the phononic crystal as: during the service time, when the lower bandgap of the phononic crystal is greater than the preset frequency, the phononic crystal fails. Take the failure condition of the phononic crystal as the basic rule for constructing the reliability test model, thereby constructing the reliability test model.

[0131] In the reliability test model: the limit state equation for calculating whether the phononic crystal fails can be expressed as:

[0132]

[0133] According to this limit state equation, the failure probability of the phononic crystal during the service time can be further calculated as follows:

[0134]

[0135] When the parameters of the phononic crystal take specific values, the maximum and minimum values of the lower bound of the band gap can be calculated. Correspondingly, substituting the lower bound of the band gap with the maximum value into the limit state equation, the lower bound of the failure probability of the phononic crystal can be calculated; substituting the lower bound of the band gap with the minimum value into the limit state equation, the upper bound of the failure probability of the phononic crystal can be calculated. The lower bound and upper bound of the failure rate of the phononic crystal during the service time can be expressed as:

[0136]

[0137]

[0138] In order to reduce the influence of the service time on the random process parameters and interval process parameters when testing the reliability of the phononic crystal, the random process parameters and interval process parameters are applied with equivalent uncertainty transformation, and the specific formula is as follows:

[0139]

[0140] After transformation, the equivalent random variable parameters and equivalent interval variable parameters independent of time are obtained, thereby transforming the time-varying reliability analysis into time-invariant reliability analysis.

[0141] The time parameter t describing the service time of the phononic crystal is also transformed into an equivalent distributed time parameter t' evenly distributed within the time-varying life cycle [0, T L of the phononic crystal, t' ∼ U(0, T L ).

[0142] According to the random variable parameters, interval variable parameters, and the transformed equivalent random variable parameters, equivalent interval variable parameters and equivalent distributed time parameters, a reliability test model can be constructed.

[0143] In the reliability test model, the limit state function used to calculate whether the phononic crystal fails is correspondingly modified as:

[0144]

[0145] The formula for calculating the failure probability of the phononic crystal correspondingly becomes:

[0146] P f = Pr(g(X, Y, S', I', t') < 0)

[0147] When it is necessary to predict the failure probability of a phononic crystal according to the reliability test model, a neural network model is trained according to the reliability test model to predict the failure probability of the phononic crystal. In order to more accurately test the reliability of the phononic crystal, the present invention uses an iterative learning method to update the trained neural network model. Therefore, when initially training the neural network model, the number of iterations is set to 1.

[0148] When training the neural network model, first perform an equivalent uncertainty transformation on the sample point set S MCS that constitutes the sample space. According to the formula of the equivalent uncertainty transformation, the time-related parameters of the sample points in the sample point set are transformed into time-independent parameters to obtain the transformed sample point set Subsequently, select sample points from the generated sample point set to obtain the training set S. The number of sample points included in the training set S is M, where M ≤ N mc , N mc is the number of sample points included in the sample point set .

[0149] When selecting sample points from the sample space to obtain the training set S, the sample points are selected by means of weighted sampling. First, calculate the sample weights of the sample points V t (i) in the sample space. Specifically, it can be calculated by the following formula:

[0150]

[0151] Subsequently, calculate the eigenvalue of each sample point according to the sample weight. Specifically, it is calculated by the following formula:

[0152]

[0153] Finally, sort each sample point according to the eigenvalue in ascending order, and select the first M sample points to obtain the training set S.

[0154] In the sample point set , the unselected sample points, the sample points not in the training set S are used as the candidate point set S * .

[0155] After obtaining the training set S, train the neural network model according to the training set S. After training the neural network model, predict the failure rate of the phononic crystal according to the neural network model; input the transformed sample point set into the trained neural network model to obtain the failure rate of the instantaneous reliability of the phononic crystal.

[0156] Subsequently, it is judged according to the stopping rule whether the failure rate of the instantaneous reliability of the phononic crystal satisfies the stopping rule. The stopping rule includes:

[0157]

[0158]

[0159]

[0160] If the stopping rule is satisfied, the sample point set S MCS is input into the neural network model, and the failure rate of the hybrid time-varying reliability of the phononic crystal is output.

[0161] If the stopping rule is not satisfied, the iteration number is updated to increase the iteration number by 1.

[0162] Subsequently, the first point set is determined from the candidate point set S * Among them Using the neural network model generated last time, the response value g of each sample point in the candidate point set S * is calculated, and the response value g is calculated according to the following formula:

[0163]

[0164] Then, sort according to the absolute value of the response value of each sample point, and select sample points as the first point set The size of can be calculated by the following formula:

[0165]

[0166] Subsequently, M experimental points are selected from the first point set in the way of weighted sampling as the second point set S K .

[0167] Next, the active learning function is used to determine the incremental sample points V from the second point set S K . First, the k-fold cross-validation method is used to process the training set. The training set is equally divided into k training subsets, and each training subset includes the same number of sample points. Each training subset in the k training subsets is used as the test subset in turn, and the other training subsets are used as the training complement sets to train the neural network model, and k supplementary neural network models can be trained. new The sample points are input into the supplementary neural network model to obtain the response value g. The supplementary neural network model is trained according to the training complement set

[0168] The response value is calculated according to the following formula: The response value is calculated according to the following formula:

[0169]

[0170] Input each sample point in the second point set S K into the neural network model trained with the training set in the previous time, and k supplementary neural network models trained with the training complement set, and calculate the uncertainty of each sample point. Specifically, it can be calculated by the following formula:

[0171]

[0172] Subsequently, calculate the Euclidean distance between each sample point in the second point set S K and the existing training set S. Specifically, it can be calculated by the following formula:

[0173]

[0174] Finally, determine the incremental sample points V new according to the Euclidean distance and uncertainty using the active learning function. Specifically, it can be calculated by the following formula:

[0175]

[0176] Sort the function values calculated by inputting each sample point in the second point set S K into the active learning function, and take the sample point with the smallest calculated function value as the incremental sample point V new .

[0177] Add the incremental sample points to the sample set S to obtain a new sample set, and train a new neural network model.

[0178] After training a new neural network model, input the sample point set into the new neural network model to obtain the failure rate of the instantaneous reliability of the phononic crystal. Then judge whether the stopping rule is satisfied; if the stopping rule is satisfied, stop the iteration. Subsequently, according to inputting the sample point set S MCS into the new neural network model, output the failure rate of the hybrid time-varying reliability of the phononic crystal. If the stopping rule is not satisfied, continue the iteration, let the iteration number k increase by 1; then continue to update the newly generated neural network model until the output failure rate of the instantaneous reliability satisfies the stopping rule.

[0179] When the stopping rule is satisfied, calculate the coefficient of variation of the failure rate according to the obtained failure rate of the hybrid time-varying reliability of the phononic crystal. Specifically, it can be calculated by the following formula:

[0180]

[0181] Compare the calculated coefficient of variation with a preset coefficient threshold. If the coefficient of variation is greater than the preset coefficient threshold, output the failure rate.

[0182] If the coefficient of variation is less than the preset coefficient threshold, increase the number of sample points in the sample space to obtain a new sample space. Repeat the above steps according to the new sample space to obtain a new neural network model, calculate the failure rate and the coefficient of variation until the calculated coefficient of variation is greater than the preset coefficient threshold.

[0183] When calculating the failure rate of the phononic crystal during the service time, it includes a lower bound and an upper bound The finally generated failure rate includes the upper bound and the lower bound of the failure rate, and the failure rate is between the upper bound and the lower bound of the failure rate.

[0184] The present invention discloses a time-varying reliability test method for a phononic crystal, which is suitable for being executed in a computing device. The method includes the steps of: determining a sample space according to the parameters of the phononic crystal; constructing a reliability test model according to the sample space and the failure conditions of the phononic crystal; constructing a neural network model according to the reliability test model; predicting the failure rate of the phononic crystal during the service time according to the neural network model and the sample space. By constructing a reliability test model of the phononic crystal and then constructing a neural network model through the reliability test model, the present invention can predict the failure rate of the phononic crystal during the service time, avoid being unable to judge the performance of the phononic crystal due to the lack of parameters of the phononic crystal, and realize the determination of the performance of the phononic crystal during the service time.

[0185] In the specification provided herein, a large number of specific details are set forth. It will be understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail so as not to obscure an understanding of this specification.

[0186] A9. The method as described in A8, wherein the determining the incremental sample points according to the number of iterations includes the steps of:

[0187] Determine a candidate point set according to the training set, and determine a first point set according to the candidate point set;

[0188] Determine a second point set from the first point set by weighted sampling;

[0189] Determine the incremental sample points from the second point set according to the active learning function.

[0190] A10. The method as described in A9, wherein the determining the incremental sample points from the second point set according to the active learning function includes the steps of:

[0191] Determine the uncertainty of each sample point in the second point set;

[0192] Determine the Euclidean distance between each sample point in the second point set and the training set;

[0193] Input the uncertainty and Euclidean distance of each sample point into the active learning function to obtain the function value of each sample point;

[0194] Take the sample point with the minimum function value in the second point set as the incremental sample point.

[0195] A11. The method as described in A10, wherein determining the uncertainty of each sample point in the second point set includes the steps of:

[0196] Generate multiple training complementary sets according to the sample set generated last time;

[0197] Generate complementary neural network models according to each training complementary set;

[0198] Determine the uncertainty of each sample point according to the complementary neural network model and the neural network model generated last time.

[0199] A12. The method as described in any one of A1 - A11, wherein predicting the failure rate of the phononic crystal during the service time according to the neural network model and the sample space includes the steps of:

[0200] Calculate the failure rate of the mixed time - varying reliability of the phononic crystal according to the neural network model and the sample points in the sample space.

[0201] A13. The method as described in A12, wherein the method further includes the steps of:

[0202] Calculate the coefficient of variation of the failure rate according to the failure rate of the mixed time - varying reliability of the phononic crystal;

[0203] Determine whether the coefficient of variation is greater than a preset coefficient threshold;

[0204] If the coefficient of variation is not greater than the preset coefficient threshold, add new sample points to the sample space to obtain a new sample space;

[0205] Train the neural network model according to the new sample space until the coefficient of variation of the neural network model is greater than the preset minimization threshold.

[0206] Similarly, it should be understood that, in order to streamline the present disclosure and help understand one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof.

[0207] Those skilled in the art should understand that the modules or units or groups of the devices in the examples disclosed herein can be arranged in the devices as described in the embodiments, or alternatively can be located in one or more devices different from the devices in the examples. The modules in the foregoing examples can be combined into one module or further divided into multiple sub-modules.

[0208] Those skilled in the art can understand that the modules in the devices of the embodiments can be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or groups in the embodiments can be combined into one module or unit or group, and further can be divided into multiple sub-modules or sub-units or sub-groups. Except that at least some of such features and / or processes or units are mutually exclusive, any combination can be adopted to combine all the features disclosed in this specification and all the processes or units of any method or device so disclosed. Unless otherwise explicitly stated, each feature disclosed in this specification can be replaced by an alternative feature that provides the same, equivalent or similar purpose.

[0209] In addition, those skilled in the art can understand that although some of the embodiments described herein include certain features included in other embodiments rather than other features, the combination of the features of different embodiments means that it is within the scope of the present invention and forms different embodiments.

[0210] In addition, some of the embodiments described herein are described as combinations of methods or method elements that can be implemented by a processor of a computer system or by other devices performing the functions. Therefore, a processor having the necessary instructions for implementing the method or method elements forms a device for implementing the method or method elements. In addition, the elements described herein in the device embodiments are examples of the following devices: the device is used to implement the functions performed by the elements for the purpose of implementing the present invention.

[0211] The various technologies described herein can be implemented in combination with hardware or software, or a combination of them. Thus, the methods and devices of the present invention, or certain aspects or parts of the methods and devices of the present invention, can take the form of program code (i.e., instructions) embedded in a tangible medium, such as a floppy disk, CD-ROM, hard disk drive, or any other machine-readable storage medium, where when the program is loaded into a machine such as a computer and executed by the machine, the machine becomes a device for practicing the present invention.

[0212] When the program code is executed on a programmable computer, a computing device generally includes a processor, a processor-readable storage medium (including volatile and non-volatile memories and / or storage elements), at least one input device, and at least one output device. Among them, the memory is configured to store the program code; the processor is configured to execute the time-varying reliability test method of the phononic crystal of the present invention according to the instructions in the program code stored in the memory.

[0213] By way of example and not limitation, computer-readable media include computer storage media and communication media. Computer-readable media include computer storage media and communication media. Computer storage media stores information such as computer-readable instructions, data structures, program modules, or other data. Communication media generally embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and includes any information delivery medium. Combinations of any of the above are also included within the scope of computer-readable media.

[0214] As used herein, unless otherwise specified, the use of ordinal numbers "first", "second", "third", etc. to describe ordinary objects merely indicates different instances of similar objects and is not intended to imply that the objects so described must have a given order in time, space, ranking, or in any other way.

[0215] Although the present invention has been described in terms of a limited number of embodiments, those skilled in the art in this technical field will appreciate that other embodiments can be envisioned within the scope of the present invention as thus described. In addition, it should be noted that the language used in this specification has been principally selected for readability and teaching purposes rather than for the purpose of explaining or limiting the subject matter of the present invention. Accordingly, many modifications and variations will be apparent to those of ordinary skill in the art. The disclosure of the present invention is illustrative, not restrictive, of the scope of the present invention.

Claims

1. A time-varying reliability test method for a phononic crystal, suitable for execution in a computing device, the method comprising the steps of: Determine a sample space according to the parameters of the phononic crystal; Construct a reliability test model according to the sample space and the failure condition of the phononic crystal. The failure condition of the phononic crystal is that within the service time, when the lower limit of the band gap of the phononic crystal is greater than the preset frequency, the phononic crystal fails, where The lower limit of the bandgap is calculated from random variable parameters, random process parameters, interval variable parameters, and interval process parameters, and the specific calculation results are as follows: where X = (X1, X2,... X m ), represents an m-dimensional vector composed of random variable parameters, Y = (Y1, Y2,... Y n ), represents an n-dimensional vector composed of interval variable parameters, S(t) = [S1(t), S2(t),... S k (t)], represents a vector composed of k random process parameters, I(t) = [I1(t), I2(t),... I l (t)], represents a vector composed of l interval process parameters, and t is the preset service time of the phononic crystal; Construct a neural network model according to the reliability test model; Predict the failure rate of the phononic crystal during the service time according to the neural network model and the sample space; Wherein, according to the failure condition of the phononic crystal, the limit state equation for calculating whether the phononic crystal fails is: Wherein, g(X, Y, S(t), I(t), t) represents the difference between the preset frequency and the lower limit of the bandgap of the phononic crystal. When g(X, Y, S(t), I(t), t) is less than 0, the phononic crystal fails.

2. The method according to claim 1, wherein The parameters of the phononic crystal include: random variable parameters, random process parameters, interval variable parameters, and interval process parameters; Constructing a reliability test model according to the sample space and the failure condition of the phononic crystal includes the steps of: Convert the random process parameters in the crystal parameters to obtain equivalent random variable parameters; Convert the interval process parameters in the crystal parameters to obtain equivalent interval variable parameters; Convert the time parameter of the service time of the phononic crystal to an equivalent distribution time parameter; Construct a reliability test model according to the random variable parameters, equivalent random variable parameters, interval variable parameters, equivalent interval variable parameters, equivalent distribution time parameters, and failure conditions.

3. The method according to claim 2, wherein, The failure conditions include: During the service time, when the lower limit of the bandgap of the phononic crystal is greater than the preset frequency, the phononic crystal fails.

4. The method according to any one of claims 1 to 3, wherein, Constructing the neural network model according to the reliability test model includes the steps of: Determine a training set according to the sample point set; Train the neural network model according to the reliability test model and the training set.

5. The method according to claim 4, wherein Determining the training set according to the sample point set includes the steps of: Convert the sample points in the sample space to obtain a converted sample point set; Determine a training set according to the converted sample point set.

6. The method according to claim 5, wherein Determining the training set according to the converted sample point set includes the steps of: Determine the sample weight for each sample point in the converted sample point set; Determine the eigenvalue of each sample point according to the sample weight; Determine a training set from the converted sample point set according to the eigenvalue.

7. The method according to claim 5, wherein The conversion of the sample points in the sample space includes the steps of: Perform equivalent uncertainty conversion on each sample point in the sample space to obtain a sample point set independent of time.

8. The method according to claim 4, wherein The method further includes the steps of: Judge whether the failure rate of the instantaneous reliability calculated by the neural network model satisfies the stop rule; If the stop rule is not satisfied, set the number of iterations, and determine incremental sample points according to the number of iterations; Determine a new sample set according to the incremental sample points and the sample set; Train a new neural network model according to the new sample set until a neural network model that satisfies the stop rule is trained.

9. The method according to claim 8, wherein, Determining the incremental sample points according to the number of iterations includes the steps of: Determine a candidate point set according to the training set, and determine a first point set according to the candidate point set; Determine a second point set from the first point set by weighted sampling; Determine incremental sample points from the second point set according to the active learning function.

10. The method according to claim 9, wherein, Determining incremental sample points from the second point set according to the active learning function includes the steps of: Determining the uncertainty of each sample point in the second point set; Determining the Euclidean distance between each sample point in the second point set and the training set; Inputting the uncertainty and Euclidean distance of each sample point into the active learning function to obtain the function value of each sample point; Taking the sample point with the minimum function value in the second point set as the incremental sample point.

11. The method according to claim 10, wherein, Determining the uncertainty of each sample point in the second point set includes the steps of: Generating multiple training complementary sets according to the sample set generated last time; Generating complementary neural network models according to each training complementary set; Determining the uncertainty of each sample point according to the complementary neural network model and the neural network model generated last time.

12. The method according to claim 11, wherein, Predicting the failure rate of the phononic crystal during the service time according to the neural network model and the sample space includes the steps of: Calculating the failure rate of the mixed time-varying reliability of the phononic crystal according to the neural network model and the sample points in the sample space.

13. The method according to claim 12, wherein, The method further includes the steps of: Calculating the coefficient of variation of the failure rate according to the failure rate of the mixed time-varying reliability of the phononic crystal; determining whether the coefficient of variation is greater than a preset coefficient threshold; If the coefficient of variation is not greater than the preset coefficient threshold, adding new sample points to the sample space to obtain a new sample space; Training the neural network model according to the new sample space until the coefficient of variation of the neural network model is greater than a preset minimization threshold.

14. A computing device, comprising: One or more processors; A memory; And One or more devices, the one or more devices including instructions for performing the method according to any one of claims 1-13.

15. A computer-readable storage medium storing one or more programs, the one or more programs including instructions which, when executed by a computing device, cause the computing device to perform the method according to any one of claims 1-13.

Images