CN117131784B - Global agent optimization-based accelerated degradation test design method and device - Google Patents
Global agent optimization-based accelerated degradation test design method and device Download PDFInfo
- Publication number
- CN117131784B CN117131784B CN202311361414.2A CN202311361414A CN117131784B CN 117131784 B CN117131784 B CN 117131784B CN 202311361414 A CN202311361414 A CN 202311361414A CN 117131784 B CN117131784 B CN 117131784B
- Authority
- CN
- China
- Prior art keywords
- model
- optimization
- performance degradation
- test
- representing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012360 testing method Methods 0.000 title claims abstract description 187
- 238000006731 degradation reaction Methods 0.000 title claims abstract description 145
- 230000015556 catabolic process Effects 0.000 title claims abstract description 140
- 238000005457 optimization Methods 0.000 title claims abstract description 139
- 238000000034 method Methods 0.000 title claims abstract description 85
- 238000013461 design Methods 0.000 title claims abstract description 80
- 230000001133 acceleration Effects 0.000 claims abstract description 55
- 238000013528 artificial neural network Methods 0.000 claims abstract description 29
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 21
- 230000002068 genetic effect Effects 0.000 claims abstract description 20
- 238000013401 experimental design Methods 0.000 claims abstract description 17
- 238000004458 analytical method Methods 0.000 claims abstract description 8
- 230000006870 function Effects 0.000 claims description 100
- 239000003795 chemical substances by application Substances 0.000 claims description 27
- 238000012549 training Methods 0.000 claims description 19
- 238000009826 distribution Methods 0.000 claims description 13
- 238000006243 chemical reaction Methods 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 12
- 230000000750 progressive effect Effects 0.000 claims description 12
- 238000003860 storage Methods 0.000 claims description 11
- 230000005653 Brownian motion process Effects 0.000 claims description 9
- 238000009792 diffusion process Methods 0.000 claims description 7
- 238000007476 Maximum Likelihood Methods 0.000 claims description 5
- 238000005537 brownian motion Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 5
- 238000005315 distribution function Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 abstract description 12
- 238000004590 computer program Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 238000012545 processing Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Computer Hardware Design (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Geometry (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the disclosure provides a method and a device for designing an accelerated degradation test based on global agent optimization. Belonging to the field of reliability engineering, the method comprises the following steps: establishing an accelerated performance degradation model of the product based on the product performance degradation history data; determining test design variables, and constructing an optimization objective function and constraint conditions by combining acceleration performance degradation modeling and reliability analysis of products so as to establish an optimization model; and constructing a proxy model based on the radial basis function neural network, and carrying out global optimization on experimental design variables in the optimization model by combining a genetic algorithm to obtain an optimization design result. In this way, an optimal test scheme of the accelerated degradation test of the aviation aircraft product can be obtained rapidly, and the most abundant product life information can be obtained by carrying out the test based on the test scheme, so that the effective improvement of the product life evaluation accuracy is realized under a certain test resource constraint.
Description
Technical Field
The disclosure relates to the field of aerospace, and further relates to the field of reliability engineering, in particular to an accelerated degradation test design method and device based on global agent optimization.
Background
Along with the development of scientific technology, many products in the fields of aerospace and the like have the characteristics of long service life and high reliability, and accurate and effective reliability evaluation is a necessary premise for ensuring stable and reliable operation of the products. On the one hand, the long-term service characteristics make it particularly difficult to evaluate the service life and reliability of the product in a limited development period, and failure data are difficult to obtain in the traditional life test and even in the accelerated life test. On the other hand, products which are in long-term service in the aerospace field are usually small in production batch and high in price, and the number of samples put into a test is very limited. Therefore, the optimization design of the accelerated performance degradation test for the product has very important practical significance for acquiring as much product reliability information as possible on the premise of limited test resources.
The design variables of the accelerated degradation test generally comprise stress level, sample number, test times, test interval and the like, and the current optimization design method generally only considers part of the design variables, so that the optimization result has a certain limitation. Meanwhile, the existing method generally adopts an enumeration method or a surface fitting method to determine an optimal test scheme, and the characteristic that part of test design variables can be continuously changed is ignored, so that global optimization cannot be performed; in addition, when the optimization of multiple test design variables is performed, the optimization efficiency is low. Therefore, in a comprehensive view, the current accelerated degradation test optimization design method based on global agent optimization cannot efficiently realize global optimization of multiple test design variables.
Disclosure of Invention
The disclosure provides a global agent optimization-based accelerated degradation test design method and device.
According to a first aspect of the present disclosure, there is provided a method of accelerated degradation testing design based on global agent optimization. The method comprises the following steps:
establishing an accelerated performance degradation model of the product based on the product performance degradation history data;
determining test design variables, and constructing an optimization objective function and constraint conditions by combining acceleration performance degradation modeling and reliability analysis of products so as to establish an optimization model;
and constructing a proxy model based on the radial basis function neural network, and carrying out global optimization on experimental design variables in the optimization model by combining a genetic algorithm to obtain an optimization design result.
Further, the establishing the accelerated performance degradation model of the product based on the product performance degradation history data includes:
and establishing an accelerated performance degradation model of the product according to the statistical characteristics of the product performance degradation historical data.
Further, the expression of the accelerated performance degradation model is as follows:
；
wherein,representing the performance degradation quantity at the time t; />Representing a drift coefficient and representing a degradation rate; />As a diffusion coefficient, characterizing dispersibility; />Represents a standard brownian motion; / >、/>Representing a time-shift scale function,/->Representative ofSThe rate of degradation at stress levels is accelerated,Srepresenting the acceleration stress level.
Further, the expression of the objective function is as follows:
；
wherein,representing the progressive variance of the 90% reliable lifetime estimate,Ha vector representing the deviation of the failure life distribution to each model parameter,Irepresenting the Fisher information matrix, which is a matrix formed by solving second order partial derivatives of model parameters by likelihood functions, wherein ∈>The probability density function representing lifetime is +.>A value at.
Further, the optimization model includes: optimizing targets and constraint conditions;
the optimization objective is that the progressive variance of the 90% reliable lifetime estimate be minimal;
constraints include total cost of test, number of tests, time interval, sample size, and acceleration stress level.
Further, the constructing the proxy model based on the radial basis function neural network includes:
obtaining a training data set according to the value range of each experimental design variable in the optimization model;
training the radial basis function neural network by utilizing the training data set;
and stopping training to obtain the proxy model when the prediction accuracy index reaches a preset threshold value.
Further, the constructing the proxy model based on the radial basis function neural network, and performing global optimization on the experimental design variables in the optimization model by combining a genetic algorithm to obtain an optimization design result comprises:
And determining the mapping relation between each test scheme in the optimization model and the objective function value according to the agent model.
According to a second aspect of the present disclosure, an accelerated degradation testing design device based on global agent optimization is provided. The device comprises:
the accelerated performance degradation model determining module is used for establishing an accelerated performance degradation model of the product based on the product performance degradation historical data;
the optimization model construction module is used for determining test design variables, constructing an optimization objective function and constraint conditions by combining acceleration performance degradation modeling and reliability analysis of the product, and thus constructing an optimization model;
the optimal design result determining module is used for constructing a proxy model based on the radial basis function neural network, and carrying out global optimization on experimental design variables in the optimal model by combining a genetic algorithm to obtain an optimal design result.
According to a third aspect of the present disclosure, an electronic device is provided. The electronic device includes: a memory and a processor, the memory having stored thereon a computer program, the processor implementing the method as described above when executing the program.
According to a fourth aspect of the present disclosure, there is provided a computer readable storage medium having stored thereon a computer program which when executed by a processor implements a method as described above.
The method and the device take multiple experimental design variables into consideration simultaneously, construct a proxy model by means of a neural network, and complete the objective of global optimization of multiple design variables by combining a genetic algorithm, so that the objective function is prevented from being repeatedly calculated in the optimization process, and the optimization efficiency is greatly improved. According to the method, the optimal test scheme can be obtained under a certain test resource constraint, so that the service life evaluation efficiency of the product can be improved, and the most accurate evaluation of the service life of the product is realized.
It should be understood that what is described in this summary is not intended to limit the critical or essential features of the embodiments of the disclosure nor to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
The above and other features, advantages and aspects of embodiments of the present disclosure will become more apparent by reference to the following detailed description when taken in conjunction with the accompanying drawings. For a better understanding of the present disclosure, and without limiting the disclosure thereto, the same or similar reference numerals denote the same or similar elements, wherein:
FIG. 1 illustrates a flow chart of an accelerated degradation testing design method based on global agent optimization in accordance with an embodiment of the present disclosure;
FIG. 2 illustrates a block diagram of an accelerated degradation testing design device based on global agent optimization in accordance with an embodiment of the present disclosure;
fig. 3 illustrates a block diagram of an exemplary electronic device capable of implementing embodiments of the present disclosure.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present disclosure more apparent, the technical solutions of the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present disclosure, and it is apparent that the described embodiments are some embodiments of the present disclosure, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments in this disclosure without inventive faculty, are intended to be within the scope of this disclosure.
In addition, the term "and/or" herein is merely an association relationship describing an association object, and means that three relationships may exist, for example, a and/or B may mean: a exists alone, A and B exist together, and B exists alone. In addition, the character "/" herein generally indicates that the front and rear associated objects are an "or" relationship.
In the disclosure, an accelerated degradation test optimization design method based on global efficient optimization of agent optimization is provided. On the basis of reasonably establishing an acceleration performance degradation model of a product, determining multiple types of test design variables, and further constructing an optimization objective function and constraint conditions; constructing a proxy model based on a radial basis function neural network, and optimizing a test scheme by applying a genetic algorithm to achieve the purposes of searching an optimal test scheme and improving optimization efficiency in the whole field of a feasible solution space; meanwhile, the method and the device consider various experimental design variables, construct a proxy model by means of a neural network, and complete the objective of global optimization of the multiple design variables by combining a genetic algorithm, so that the objective function is prevented from being repeatedly calculated in the optimization process, and the optimization efficiency is greatly improved. According to the method, the optimal test scheme can be obtained under a certain test resource constraint, so that the service life evaluation efficiency of the product can be improved, and the most accurate evaluation of the service life of the product is realized.
FIG. 1 illustrates a flow chart of an accelerated degradation testing design method based on global agent optimization in accordance with an embodiment of the present disclosure;
the disclosure provides an accelerated degradation test optimization design method based on global agent optimization. The method comprises the following steps:
s1: and establishing an accelerated performance degradation model of the product based on the product performance degradation history data.
Specifically, time-varying statistical characteristics of product performance degradation data are analyzed, the relation between the performance degradation amount and the acceleration stress level is comprehensively considered, and an acceleration performance degradation model capable of reasonably describing the product performance degradation rule and the acceleration relation is established.
S2: and determining test design variables, and constructing an optimization objective function and constraint conditions by combining acceleration performance degradation modeling and reliability analysis of the product, so as to establish an optimization model.
Specifically, three test design variables of stress level, sample size and test time are determined, an optimization target with minimum progressive variance of reliable life estimation is constructed, life estimation accuracy under working conditions can be effectively represented, constraint conditions considering test cost are established, and a final optimization model is established.
S3: and constructing a proxy model based on the radial basis function neural network, and carrying out global optimization on experimental design variables in the optimization model by combining a genetic algorithm to obtain an optimization design result.
Specifically, aiming at the problem of huge optimization calculation amount caused by a double-layer optimization process in optimization, an optimization objective function proxy model is constructed based on a radial basis function neural network.
The proxy model construction process is as follows:
(1) Obtaining a training data set according to the value range of each experimental design variable in the optimization model;
(2) Dividing the training data set into a training set and a testing set according to the proportion of 7:3; training the radial basis function neural network by utilizing the training data set;
(3) And stopping training to obtain the proxy model when the prediction accuracy index reaches a preset threshold value. The preset threshold of the prediction accuracy index is generally set to 0.8.
The agent model constructed through the process is used for determining the mapping relation between each test scheme and the objective function value in the optimization model, so that the larger calculated amount and calculation time caused by double-layer optimization are reduced.
Based on the method, a genetic algorithm is applied to globally optimizing test design variables in the optimization model, and an optimal test scheme is determined.
One or more acceleration performance degradation models are provided for the products, and different products correspond to different acceleration performance degradation models. And determining a corresponding pre-constructed optimization model according to the target product parameters, performing global optimization from the optimization model by an optimization method based on the agent model to obtain an optimal test scheme in the product acceleration performance degradation test scheme, and evaluating the service life of the aircraft product to be evaluated based on the optimal test scheme.
Based on the optimal test scheme, the service life of each product of the aviation aircraft is estimated, so that the service life estimation efficiency is higher, and the estimation result is more accurate.
[ embodiment one ]
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be clearly and completely described below by taking an optimized design of an accelerated degradation test of an O-type rubber ring of a typical seal of an aircraft as an example, and referring to the drawings in the embodiments of the present disclosure.
The method for designing the accelerated degradation test optimization by considering the global efficient optimization of the multiple test design variables based on the agent optimization comprises the following steps:
firstly, analyzing the performance degradation rule of the O-shaped rubber ring, and establishing a performance degradation model based on a Wiener process. Taking temperature as acceleration stress, introducing an Arrhenius model (Arrhenius model) to construct an acceleration performance degradation model, wherein the acceleration performance degradation model is specifically as follows:
by analyzing the statistical characteristics of the performance degradation data of the O-shaped rubber ring, the mean value and the variance of the performance degradation data are nonlinear and can be described by the same time scale, and in addition, the degradation process does not strictly monotonically show a certain randomness, so that a single time scale model based on the wiener process is obtained, the performance degradation rule of the O-shaped rubber ring can be better described, and the performance degradation model of the O-shaped rubber ring can be established as follows: ；
Wherein,representing the performance degradation quantity at the time t; />Representing a drift coefficient and representing a degradation rate; />As a diffusion coefficient, characterizing dispersibility; />Represents a standard brownian motion; />Is an exponential form of the time-shift scale function, which converts the nonlinear problem into a linear problem.
ConsiderThe acceleration stress is temperature, an Arrhenius model is introduced, and the data rule under each stress level shows that the acceleration stress mainly influences the degradation rate, namelyThe accelerated degradation model of an O-ring rubber can therefore be given as follows:
；
wherein S is the acceleration stress level, i.e., the temperature level;is the degradation rate at the S stress level; a, b are unknown parameters related to the characteristics of the product itself in the Arrhenius model.
Wherein the model parameters to be estimated are a, b,，/>；/>by ∈>Obtaining the product.
Based on the historical test data of the O-shaped rubber ring, the estimated values of all parameters are obtained by adopting a maximum likelihood estimation method as follows:
secondly, determining test design variables, constructing an optimization objective function considering life evaluation precision, and establishing constraint conditions specifically as follows:
assuming that the constant stress accelerated degradation test of the O-shaped rubber ring was performed at 2 stress levels, it is noted that The number of samples tested was arranged at each level +.>The number of performance index tests at each level was +.>And the test interval is f. Assuming that the total number of samples available for testing is 15, therefore when m 1 When determined, can be according to 15-m 1 Determining m 2 . Thus, in this example there are 6 test design variables, respectively +.>。
The optimized objective function of the test design can be constructed based on the accelerated degradation model established in the first step, so as to estimate the progressive variance of 90% reliable life of more concern in engineeringAs an optimization objective function, the objective function characterizes life prediction precision, and the specific expression is: />；
Wherein,Hthe model parameters are biased to form a vector for the failure life distribution,Ithe matrix is formed by solving a second-order partial derivative of model parameters by likelihood functions,is the probability density function of lifetime +.>A value at. Model parameters in the performance degradation model established in the first step +.>，/>Is a modelMaximum likelihood estimation of parameters, thereforeH、IAnd->The expression of (2) will be specifically set forth below.
Assuming that the preliminary test in the first step is performed at p stress levels, M is set at the p-th stress level p Samples are taken of N p The log likelihood function lnL is the samples with sub-equal time interval f:
；
Wherein lnL is a log-likelihood function; p is the number of stress levels; m is M p Sample size at p stress levels; n (N) p Number of tests at the p-th stress level;time after time-scale conversion for jth test, +.>The time after the time scale conversion is the j-1 th test time; />For the amount of performance degradation obtained at the j-th test moment for the i-th sample at the p-th stress level,/>The performance degradation amount obtained at the j-1 test moment of the ith sample under the p stress level is obtained;，/>，/>；a，b，/>，/>the method comprises the steps of (1) obtaining model parameters to be estimated in an acceleration performance degradation model; s is S p Is the p-th stress level.
Assuming that the actual working stress level of the O-shaped rubber ring in the second step is S 0 Failure level of D f The lifetime distribution F (t) and the probability density function of lifetime are thenValue of->The method comprises the following steps:
；
；
wherein,is a standard normal distribution function.
It is further possible to obtain:
；
；
wherein,indicating that the content in brackets is desired.
And then establishing constraint conditions, comprehensively considering test equipment conditions and engineering experience, and determining the following cost of the test: price of individual sampleThe cost of manpower and public resources per hour during the test is +.>The primary and single test cost is +.>Meta, test total cost constraint- >And (5) a meta.
In order to improve the efficiency of test optimization design, the value range of each test design variable can be constrained according to engineering experience, so that a final optimization model of the O-shaped rubber ring can be obtained as follows:
；
wherein,representing the objective function, i.e. the progressive variance of the 90% reliable lifetime estimate,Hthe model parameters are biased to form a vector for the failure life distribution,Ithe matrix is formed by solving a second order partial derivative of model parameters by likelihood functions, and is a Fisher information matrix>Is the probability density function of lifetime +.>The value at, cost represents a Cost constraint, C u For single sample price, C o C for the cost of labor and public resources per hour during the test m For single test cost, C b Constraint for total cost of test; m is m 1 Represents the sample size, m, at stress level 1 2 Represents the sample size at stress level 2; n is n 1 Represents the number of tests at stress level 1, n 2 Representing the number of tests at stress level 2; f represents a test interval; s is S 1 Represents the 1 st acceleration stress level, S 2 Representing the 2 nd acceleration stress level.
The optimization model is divided into two parts of an optimization target and constraint conditions. The first part is an optimization objective, requiring a progressive variance of 90% reliable lifetime estimation And the service life prediction precision is ensured at the minimum. The second part is the constraint condition, and the total cost requirement of the test cannot exceed C b In addition, there are often certain constraints on the number of tests, time intervals, sample size, and acceleration stress levels in engineering practice.
And thirdly, constructing an optimized objective function proxy model based on the radial basis function neural network. On the basis, a genetic algorithm is applied to globally optimizing the design variables, and an optimal test scheme is determined. The method comprises the following steps:
firstly, constructing a proxy model, uniformly selecting 4096 test schemes from a feasible solution space according to the value range of each test design variable, and calculating a corresponding objective function. And randomly selecting 2867 groups (about 70% of the total test scheme) as a training set of the radial basis function neural network, and the rest 1229 groups (about 30% of the total test scheme) as a test set, so as to obtain the prediction precision index of the radial basis function neural network shown in the following table:
RMSE and R 2 The root mean square error of the predicted value of the objective function obtained by using the radial basis function neural network reflects the deviation between the predicted value and the true value, and the decision coefficient reflects the proportion of the variation of the objective function which can be interpreted by the experimental design variable through the neural network. General R 2 A model exceeding 0.8 is considered to have a relatively good predictive effect. R calculated in the present disclosure 2 Has a value of 0.877>0.8, so that the proxy model constructed by the method can be considered to be capable of predicting the objective function better.
On the basis, global optimization can be performed by utilizing a genetic algorithm, so that a final optimal design result is obtained as follows:
wherein,for the optimized 1 st acceleration stress level, < ->For the optimized 2 nd acceleration stress level, < ->For the number of tests at optimized 1 st acceleration stress level, +.>The test times under the optimized 2 nd acceleration stress level are obtained; />The optimized test interval is obtained; />For the sample size at the optimized 1 st acceleration stress level; />The objective function value after optimization; />To be an optimized total cost.
[ example two ]
In addition, the present disclosure may be applied to the field of electronic products, and the following description will take an example of an accelerated degradation test optimization design of a Light Emitting Diode (LED) of a typical electronic product, and with reference to the drawings in the embodiments of the present disclosure, clearly and completely describe the technical solutions in the embodiments of the present disclosure.
The global efficient optimizing accelerated degradation test design method considering multiple test design variables based on agent optimization in the embodiment comprises the following steps:
First, analyzing the performance degradation rule of the LED, and establishing a performance degradation model based on a Wiener process. Taking current as acceleration stress, introducing inverse power rate to construct an acceleration performance degradation model, wherein the acceleration performance degradation model is specifically as follows:
by analyzing the statistical characteristics of the performance degradation data of the LED, the mean value and the variance of the performance degradation data are nonlinear, the nonlinear rules of the mean value and the variance are different, and in addition, the degradation process does not strictly monotonically show a certain randomness, so that the nonlinear double-time-scale model based on the wiener process is obtained, the performance degradation rule of the model can be better described, and the performance degradation model of the LED can be established as follows:
。
wherein,the performance degradation quantity at the time t; />Characterizing the degradation rate as a drift coefficient; />As a diffusion coefficient, characterizing dispersibility; />Is a standard Brownian motion; />，/>Is an exponential form of the time-shift scale function, which converts the nonlinear problem into a linear problem.
Taking the acceleration stress as current, introducing an inverse power rate model, and displaying data rules under each stress level to ensure that the acceleration stress mainly influences the degradation rateWherein S is the acceleration stress level, i.e. the current level; />Is the degradation rate at the S stress level; a and b are unknown parameters related to the characteristics of the product in the inverse power rate model.
The accelerated degradation model of an LED can thus be given as follows:。
wherein the model parameters to be estimated are a, b,，/>，/>；/>by ∈>Obtaining the product.
Based on historical test data of the LEDs, the estimated values of all parameters are obtained by adopting a maximum likelihood estimation method as follows:
secondly, determining test design variables, constructing an optimization objective function considering life evaluation precision, and establishing constraint conditions specifically as follows:
consider the LED to develop a step stress accelerated degradation test, 3 acceleration stress levelsThe sample size is m, each water is given in advanceThe number of performance index tests under the flat condition is +.>And the test interval is f, so that in this example there are 5 test design variables, respectively +.>。
The optimized objective function of the test design can be constructed based on the accelerated degradation model established in the first step, so as to estimate the progressive variance of 90% reliable life of more concern in engineeringAs an optimization objective function, the objective function characterizes life prediction precision, and the specific expression is:
；
wherein,Hthe model parameters are biased to form a vector for the failure life distribution,Ithe matrix is formed by solving a second-order partial derivative of model parameters by likelihood functions,is the probability density function of lifetime +. >A value at. Model parameters in the performance degradation model established in the first step +.>，/>Is a maximum likelihood estimation of model parameters, and thereforeH、IAnd->The expression of (2) will be specifically set forth below.
Suppose that the preliminary test in the first step proceeds at P stress levelsLine, P stress level set M P Samples are taken of N P The log likelihood function lnL is the samples with sub-equal time interval f:
；
wherein lnL is a log-likelihood function; p is the number of stress levels; m is the sample size; n (N) p For the number of tests at the p-th stress level,number of tests at the p-1 th stress level; />，/>Time after time-scale conversion for jth test, +.>，/>The time after the time scale conversion is the j-1 th test time; />For the performance degradation amount obtained at the j-th test time of the i-th sample,/th sample>The performance degradation amount obtained at the j-1 test moment for the ith sample;，/>the method comprises the steps of carrying out a first treatment on the surface of the A, b, & gt>，/>，/>The method comprises the steps of (1) obtaining model parameters to be estimated in an acceleration performance degradation model; s is S p Is the p-th stress level.
Assume that the actual operating stress level of the LED in the second step is S 0 Failure level of D f The lifetime distribution F (t) and the probability density function of lifetime are thenValue of->The method comprises the following steps:
；
；
wherein,is a standard normal distribution function.
It is further possible to obtain:
；
；
wherein,indicating that the content in brackets is desired.
And then establishing constraint conditions, comprehensively considering test equipment conditions and engineering experience, and determining the following cost of the test: price of individual sampleThe cost of manpower and public resources per hour during the test is +.>The primary and single test cost is +.>Meta, test total cost constraint->And (5) a meta.
In order to improve the efficiency of the test optimization design, the value range of each test design variable can be constrained according to engineering experience, so that a final optimization model of the LED can be obtained as follows:
；
wherein,progressive variance ++representing objective function, i.e. 90% reliable lifetime estimation>，HThe model parameters are biased to form a vector for the failure life distribution,Ithe matrix is formed by solving a second order partial derivative of model parameters by likelihood functions, and is a Fisher information matrix>Is the probability density function of lifetime +.>The value at, cost represents a Cost constraint, C u For single sample price, C o C for the cost of labor and public resources per hour during the test m For single test cost, C b Constraint for total cost of test; m represents the total sample amount; n is n 1 Represents the number of tests at stress level 1, n 2 Represents the 2 nd oneNumber of tests at force level, n 3 Representing the number of tests at stress level 3; f represents the test interval.
The optimization model is divided into two parts of an optimization target and constraint conditions. The first part is an optimization objective, requiring a progressive variance of 90% reliable lifetime estimationAnd the service life prediction precision is ensured at the minimum. The second part is constraint condition, the total cost requirement of the test cannot exceed +.>In addition, there are often certain constraints on the number of tests, time intervals, and sample size in engineering practice.
And thirdly, constructing an optimized objective function proxy model based on the radial basis function neural network. On the basis, a genetic algorithm is applied to globally optimizing the design variables, and an optimal test scheme is determined. The method comprises the following steps:
firstly, constructing a proxy model, uniformly selecting 3125 test schemes from a feasible solution space (a feasible solution is a solution which enables an optimized objective function to be minimum under cost constraint) according to the value range of each test design variable, and calculating the corresponding objective function. Thereafter, 2188 groups (about 70% of the total test scheme) are randomly selected as training sets of the radial basis function neural network, and the other 937 groups (about 30% of the total test scheme) are selected as test sets, so that the prediction accuracy indexes of the radial basis function neural network shown in the following table can be obtained:
Wherein, RMSE and R 2 The root mean square error of the predicted value of the objective function obtained by using the radial basis function neural network reflects the deviation between the predicted value and the true value, and the decision coefficient reflects the proportion of the variation of the objective function which can be interpreted by the experimental design variable through the neural network. General R 2 A model exceeding 0.8 is considered to have a relatively good predictive effect. This publicR calculated in the opening 2 Has a value of 0.832>0.8, so that the proxy model constructed by the method can be considered to be capable of predicting the objective function better.
On the basis, global optimization can be performed by utilizing a genetic algorithm, so that a final optimal design result is obtained as follows:
wherein,for the number of tests at optimized 1 st acceleration stress level, +.>For the number of tests at optimized acceleration stress level 2, +.>The test times under the optimized 3 rd acceleration stress level are obtained; />The optimized test interval is obtained; />The total sample amount after optimization; />The objective function value after optimization; />To be an optimized total cost.
The multi-design-variable global optimizing accelerated degradation test design method based on the agent optimization is not only suitable for the accelerated degradation test optimization design of the O-shaped sealing ring and the LEDs, but also can be applied to a plurality of other electronic products and mechanical products.
The accelerated degradation test optimization method can be applied to the life assessment of the aviation aircraft product. The specific process is as follows:
(1) And establishing a life assessment method of the aircraft product to be assessed.
Specifically, based on the service environment and working principle of the aircraft product to be evaluated, the key failure mode and failure mechanism of the product are analyzed, and the acceleration stress type is determined. And (3) combining performance degradation characteristics or historical test data of the product, establishing a performance degradation model of the product, introducing a reasonable acceleration model according to the acceleration stress type, and further establishing an acceleration performance degradation model of the product. On the basis, the performance degradation prediction of the product under the normal working condition can be realized, so that the failure life distribution function of the product is constructed, and the life assessment method of the product under the normal working condition is established.
(2) The optimal test scheme is obtained based on the agent model-based accelerated performance degradation test optimization design method, the accelerated performance degradation data of the product is obtained by carrying out the test under the scheme, and the service life evaluation of the product is realized by combining the service life evaluation method.
Specifically, according to an acceleration performance degradation model and a reliability analysis result of a product, an optimized objective function of the product is constructed, a constraint condition of test cost and test design variables is combined, an optimized model of the product is established, a mapping relation construction between a test scheme and the objective function is realized by using a radial basis neural network, global optimization is performed by combining a genetic algorithm, so that an optimal test scheme is efficiently obtained, and the service life of the aircraft product to be evaluated is evaluated based on the optimal test scheme. Based on the optimal test scheme, the service life of each product of the aviation aircraft is estimated, so that the service life estimation efficiency is higher, and the estimation result is more accurate.
One or more acceleration performance degradation models are provided for the products, and different products correspond to different acceleration performance degradation models. And determining a corresponding pre-constructed optimization model and agent model according to the target product parameters, obtaining an optimal test scheme in the product acceleration performance degradation test scheme from the optimization model, and evaluating the service life of the aircraft product to be evaluated based on the optimal test scheme. Based on the optimal test scheme, the service life of each product of the aviation aircraft is estimated, so that the service life estimation efficiency is higher, and the estimation result is more accurate.
In the disclosure, an accelerated degradation test optimization design method based on global efficient optimization of agent optimization is provided. On the basis of reasonably establishing an acceleration performance degradation model of a product, determining multiple types of test design variables, and further constructing an optimization objective function and constraint conditions; constructing a proxy model based on a radial basis function neural network, and optimizing a test scheme by applying a genetic algorithm to achieve the purposes of searching an optimal test scheme and improving optimization efficiency in the whole field of a feasible solution space; meanwhile, the method and the device consider various experimental design variables, construct a proxy model by means of a neural network, and complete the objective of global optimization of the multiple design variables by combining a genetic algorithm, so that the objective function is prevented from being repeatedly calculated in the optimization process, and the optimization efficiency is greatly improved.
According to the embodiment of the disclosure, the optimal test scheme can be obtained under a certain test resource constraint by the optimal design method, so that the service life evaluation efficiency of the product can be improved, and the most accurate evaluation of the service life of the product is realized.
It should be noted that, for simplicity of description, the foregoing method embodiments are all described as a series of acts, but it should be understood by those skilled in the art that the present disclosure is not limited by the order of acts described, as some steps may be performed in other orders or concurrently in accordance with the present disclosure. Further, those skilled in the art will also appreciate that the embodiments described in the specification are all alternative embodiments, and that the acts and modules referred to are not necessarily required by the present disclosure.
The foregoing is a description of embodiments of the method, and the following further describes embodiments of the present disclosure through examples of apparatus.
FIG. 2 illustrates a block diagram of an accelerated degradation testing design device 200 based on global agent optimization in accordance with an embodiment of the present disclosure. The apparatus 200 comprises:
an accelerated performance degradation model determination module 210 for establishing an accelerated performance degradation model of the product based on the product performance degradation history data;
The optimization model construction module 220 is used for determining test design variables, constructing an optimization objective function and constraint conditions by combining acceleration performance degradation modeling and reliability analysis of the product, and thus constructing an optimization model;
the optimal design result determining module 230 is configured to construct a proxy model based on the radial basis function neural network, and perform global optimization on the experimental design variables in the optimal model in combination with a genetic algorithm to obtain an optimal design result.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the described modules may refer to corresponding procedures in the foregoing method embodiments, which are not described herein again.
According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.
Fig. 3 shows a schematic block diagram of an electronic device 300 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.
The electronic device 300 includes a computing unit 301 that can perform various appropriate actions and processes according to a computer program stored in a ROM302 or a computer program loaded from a storage unit 308 into a RAM 303. In the RAM303, various programs and data required for the operation of the electronic device 300 may also be stored. The computing unit 301, the ROM302, and the RAM303 are connected to each other by a bus 304. I/O interface 305 is also connected to bus 304.
Various components in the electronic device 300 are connected to the I/O interface 305, including: an input unit 306 such as a keyboard, a mouse, etc.; an output unit 307 such as various types of displays, speakers, and the like; a storage unit 308 such as a magnetic disk, an optical disk, or the like; and a communication unit 309 such as a network card, modem, wireless communication transceiver, etc. The communication unit 309 allows the electronic device 300 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunication networks.
The computing unit 301 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 301 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 301 performs the various methods and processes described above, such as an accelerated degradation trial design method based on global agent optimization. For example, in some embodiments, a method of accelerated degradation testing design based on global agent optimization may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 308. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 300 via the ROM302 and/or the communication unit 309. When the computer program is loaded into RAM303 and executed by computing unit 301, one or more of the steps of a global agent optimization-based accelerated degradation testing design method described above may be performed. Alternatively, in other embodiments, the computing unit 301 may be configured to perform an aircraft product lifetime assessment method by any other suitable means (e.g. by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.
The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel, sequentially, or in a different order, provided that the desired results of the disclosed aspects are achieved, and are not limited herein.
The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.
Claims (9)
1. The method for optimizing the design of the accelerated degradation test is characterized by comprising the following steps of:
establishing an accelerated performance degradation model of the product based on the product performance degradation history data;
the expression of the accelerated performance degradation model is as follows:the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Represents->Performance degradation amount at time; />Representing a drift coefficient and representing a degradation rate; />As a diffusion coefficient, characterizing dispersibility; />Represents a standard brownian motion; / >、/>Representing a time conversion scale function, and being capable of representing and describing the condition that the mean value and variance of the product performance degradation process under a certain stress level show a nonlinear time-varying rule; />Represents->Rate of degradation at stress level, +.>Representing an acceleration stress level;
determining test design variables, and constructing an optimization objective function and constraint conditions by combining acceleration performance degradation modeling and reliability analysis of products so as to establish an optimization model;
the expression of the objective function is:the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Progressive variance representing 90% reliable lifetime estimate,/-j>Vector comprising partial derivatives of the model parameters representing the failure life distribution +.>Representing the Fisher information matrix, which is a matrix formed by solving second order partial derivatives of model parameters by likelihood functions, wherein ∈>The probability density function representing lifetime is as followsA value at;
constructing a proxy model based on a radial basis function neural network, and carrying out global optimization on experimental design variables in an optimization model by combining a genetic algorithm to obtain an optimization design result; the training data set of the agent model is obtained according to the value range of each experimental design variable in the optimization model;
the failure life distribution is:
，
where phi is a standard normal distribution function, As an unknown parameter related to the characteristics of the product itself in the inverse power rate model,for the diffusion coefficient, characterize the dispersibility, +.>For the actual working stress level +.>For failure level, ++>The method is an exponential time conversion scale function, and represents the condition that the mean value and the variance of the product performance degradation process show different nonlinear time-varying rules; />Unknown parameters for representing the nonlinear degree of the mean value and variance time-varying rule of the product performance degradation rule;
the specific expression is as follows:，
wherein,for model parameters +.>，/>For maximum likelihood estimation of model parameters, F is failure life distribution, +.>Representing a deviation derivative;
the likelihood function is:，
wherein lnL is a log-likelihood function; p is the number of stress levels; m is the sample size; n (N) p For the number of tests at the p-th stress level, N p-1 Number of tests at the p-1 th stress level;time after time-scale conversion for jth test, +.>The time after the time scale conversion is the j-1 th test time; x is x ij The performance degradation amount, x, obtained at the j-th test time for the i-th sample i（j-1） The performance degradation amount obtained at the j-1 test moment for the ith sample; i=1, 2, … ,M，j= N p-1 +1, … ,N p ；/>the method comprises the steps of (1) obtaining model parameters to be estimated in an acceleration performance degradation model; s is S p Is the p-th stress level.
2. The method of claim 1, wherein the modeling accelerated performance degradation of the product based on the product performance degradation history data comprises:
and establishing an accelerated performance degradation model of the product according to the statistical characteristics of the product performance degradation historical data.
3. The method of claim 1, wherein the optimization model comprises: optimizing targets and constraint conditions;
the optimization objective is that the progressive variance of the 90% reliable lifetime estimate be minimal;
constraints include total cost of test, number of tests, time interval, sample size, and acceleration stress level.
4. The method of claim 1, wherein constructing a proxy model based on a radial basis function network comprises:
dividing the training data set into a training set and a testing set according to the proportion of 7:3; training the radial basis function neural network by utilizing the training data set;
and stopping training to obtain the proxy model when the prediction accuracy index reaches a preset threshold value.
5. The method of claim 1, wherein the constructing a proxy model based on the radial basis function network, and performing global optimization on the experimental design variables in the optimization model in combination with the genetic algorithm to obtain the optimal design result comprises:
And determining the mapping relation between each test scheme in the optimization model and the objective function value according to the agent model.
6. An accelerated degradation test optimization design device, characterized in that the device comprises:
an accelerated performance degradation model determination module for establishing accelerated performance degradation of a product based on product performance degradation history dataModeling; the expression of the accelerated performance degradation model is as follows:the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representative ofPerformance degradation amount at time; />Representing a drift coefficient and representing a degradation rate; />As a diffusion coefficient, characterizing dispersibility; />Represents a standard brownian motion; />、/>Representing a time conversion scale function, and being capable of representing and describing the condition that the mean value and variance of the product performance degradation process under a certain stress level show a nonlinear time-varying rule; />Represents->Rate of degradation at stress level, +.>Representing an acceleration stress level;
the optimization model construction module is used for determining test design variables, and combining accelerated performance degradation modeling and reliability of productsPerforming sex analysis, and constructing an optimization objective function and constraint conditions so as to establish an optimization model; the expression of the objective function is:the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Progressive variance representing 90% reliable lifetime estimate,/-j >Vector comprising partial derivatives of the model parameters representing the failure life distribution +.>Representing the Fisher information matrix, which is a matrix formed by solving second order partial derivatives of model parameters by likelihood functions, wherein ∈>The probability density function representing lifetime is +.>A value at; the failure life distribution is:wherein Φ is a standard normal distribution function, +.>Is an unknown parameter related to the characteristics of the product itself in the inverse power rate model, ++>For the diffusion coefficient, characterize the dispersibility, +.>For the actual working stress level +.>For failure level, ++>The method is an exponential time conversion scale function, and represents the condition that the mean value and the variance of the product performance degradation data show different nonlinear time-varying rules;unknown parameters for representing the nonlinear degree of the mean value and variance time-varying rule of the product performance degradation rule; said->The specific expression of (2) is: />Wherein->As a parameter of the model, it is possible to provide,，/>for maximum likelihood estimation of model parameters, F is failure life distribution, +.>Representing a deviation derivative; the likelihood function is:wherein lnL is a log-likelihood function; p is the number of stress levels; m is the sample size; n (N) p For the number of tests at the p-th stress level, N p-1 Number of tests at the p-1 th stress level; / >For the time after time-scale conversion at the jth test,the time after the time scale conversion is the j-1 th test time; x is x ij The performance degradation amount, x, obtained at the j-th test time for the i-th sample i（j-1） The performance degradation amount obtained at the j-1 test moment for the ith sample; i=1, 2, … ,M，j= N p-1 +1, … ,N p ；/>the method comprises the steps of (1) obtaining model parameters to be estimated in an acceleration performance degradation model; s is S p Is the p-th stress level.
7. The optimization design result determining module is used for constructing a proxy model based on the radial basis function neural network, and carrying out global optimization on test design variables in the optimization model by combining a genetic algorithm to obtain an optimization design result; the training data set of the agent model is obtained according to the value range of each experimental design variable in the optimization model.
8. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-5.
9. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311361414.2A CN117131784B (en) | 2023-10-20 | 2023-10-20 | Global agent optimization-based accelerated degradation test design method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311361414.2A CN117131784B (en) | 2023-10-20 | 2023-10-20 | Global agent optimization-based accelerated degradation test design method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117131784A CN117131784A (en) | 2023-11-28 |
| CN117131784B true CN117131784B (en) | 2024-02-23 |
Family
ID=88863050
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311361414.2A Active CN117131784B (en) | 2023-10-20 | 2023-10-20 | Global agent optimization-based accelerated degradation test design method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117131784B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018076475A1 (en) * | 2016-10-26 | 2018-05-03 | 广东产品质量监督检验研究院 | Photovoltaic assembly accelerated degradation model established based on deep approach of learning, and method for predicting photovoltaic assembly lifetime |
| CN111580389A (en) * | 2020-05-21 | 2020-08-25 | 浙江工业大学 | Three-degree-of-freedom helicopter explicit model prediction control method based on deep learning |
| CN112884234A (en) * | 2021-03-04 | 2021-06-01 | 电子科技大学 | Method for searching optimal working parameters of power module of high-power millimeter wave gyrotron traveling wave tube |
| CN115186729A (en) * | 2022-05-30 | 2022-10-14 | 西安石油大学 | Stratum identification and classification method based on cuckoo algorithm optimized elastic BP neural network |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170193460A1 (en) * | 2015-12-31 | 2017-07-06 | General Electric Company | Systems and methods for predicting asset specific service life in components |
-
2023
- 2023-10-20 CN CN202311361414.2A patent/CN117131784B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018076475A1 (en) * | 2016-10-26 | 2018-05-03 | 广东产品质量监督检验研究院 | Photovoltaic assembly accelerated degradation model established based on deep approach of learning, and method for predicting photovoltaic assembly lifetime |
| CN111580389A (en) * | 2020-05-21 | 2020-08-25 | 浙江工业大学 | Three-degree-of-freedom helicopter explicit model prediction control method based on deep learning |
| CN112884234A (en) * | 2021-03-04 | 2021-06-01 | 电子科技大学 | Method for searching optimal working parameters of power module of high-power millimeter wave gyrotron traveling wave tube |
| CN115186729A (en) * | 2022-05-30 | 2022-10-14 | 西安石油大学 | Stratum identification and classification method based on cuckoo algorithm optimized elastic BP neural network |
Non-Patent Citations (1)
| Title |
|---|
| 《基于多变量优化的恒定应力加速退化试验设计》;刘根等;《系统工程与电子技术》;第43卷(第1期);267-271 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117131784A (en) | 2023-11-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111667054B (en) | Method, device, electronic equipment and storage medium for generating neural network model | |
| CN107392259B (en) | Method and device for constructing unbalanced sample classification model | |
| CN115454706A (en) | System abnormity determining method and device, electronic equipment and storage medium | |
| CN114090601B (en) | Data screening method, device, equipment and storage medium | |
| CN114037059A (en) | Pre-training model, model generation method, data processing method and data processing device | |
| CN117131784B (en) | Global agent optimization-based accelerated degradation test design method and device | |
| CN117573507A (en) | Test case generation method and device, electronic equipment and storage medium | |
| CN113612777B (en) | Training method, flow classification method, device, electronic equipment and storage medium | |
| CN116125279A (en) | Method, device, equipment and storage medium for determining battery health state | |
| CN114998649A (en) | Training method of image classification model, and image classification method and device | |
| CN113887101A (en) | Visualization method and device of network model, electronic equipment and storage medium | |
| CN114037060A (en) | Pre-training model generation method and device, electronic equipment and storage medium | |
| CN113361621A (en) | Method and apparatus for training a model | |
| CN113052325A (en) | Method, device, equipment, storage medium and program product for optimizing online model | |
| CN114037057B (en) | Pre-training model generation method and device, electronic equipment and storage medium | |
| CN115034388B (en) | Determination method and device for quantization parameters of ranking model and electronic equipment | |
| CN114037058B (en) | Pre-training model generation method and device, electronic equipment and storage medium | |
| CN112989797B (en) | Model training and text expansion methods, devices, equipment and storage medium | |
| CN117707899A (en) | Micro-service abnormality detection method, device, equipment and storage medium | |
| CN116111562A (en) | Method, device, equipment and storage medium for determining reliability of power distribution network | |
| CN117575553A (en) | Post matching method, device, equipment and storage medium | |
| CN118133103A (en) | Time sequence data anomaly detection model generation method and device and electronic equipment | |
| CN114037061A (en) | Pre-training model generation method and device, electronic equipment and storage medium | |
| JP2022095895A (en) | Traffic data prediction method, traffic data prediction device, electronic device, storage medium, computer program product, and computer program | |
| CN116484237A (en) | Method and device for determining electric carbon factor, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |