CN111680450A - Structural system reliability analysis method based on uncertainty Bayesian network - Google Patents

Structural system reliability analysis method based on uncertainty Bayesian network Download PDF

Info

Publication number
CN111680450A
CN111680450A CN202010407015.5A CN202010407015A CN111680450A CN 111680450 A CN111680450 A CN 111680450A CN 202010407015 A CN202010407015 A CN 202010407015A CN 111680450 A CN111680450 A CN 111680450A
Authority
CN
China
Prior art keywords
distribution
system reliability
bayesian network
fuzzy
probability
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.)
Granted
Application number
CN202010407015.5A
Other languages
Chinese (zh)
Other versions
CN111680450B (en
Inventor
张建国
游令非
叶楠
吴洁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beihang University
Original Assignee
Beihang University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beihang University filed Critical Beihang University
Priority to CN202010407015.5A priority Critical patent/CN111680450B/en
Publication of CN111680450A publication Critical patent/CN111680450A/en
Application granted granted Critical
Publication of CN111680450B publication Critical patent/CN111680450B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
    • 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]

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Mathematics (AREA)
  • Algebra (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computing Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • Mathematical Analysis (AREA)
  • Probability & Statistics with Applications (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • Complex Calculations (AREA)

Abstract

The invention provides a structural system reliability analysis method based on fuzzy random uncertainty Bayesian network, which comprises the following steps: 1) establishing a Bayesian network of system objects; 2) according to the main and objective information, the normal fuzzy random prior distribution of the system reliability is assumed; 3) sampling according to the boundary envelope of prior distribution, and fitting an equivalent Beta distribution expression of the boundary envelope by adopting maximum likelihood estimation; 4) establishing a system likelihood function expression according to the existing data; 5) multiplying the likelihood function by the result of 4) to obtain a boundary expression of the posterior distribution of the system reliability; 6) carrying out Metropolis-Hastings sampling after a normalization constant part of the result in the step 5) is removed, and carrying out system edge sample point statistics on the result to obtain a boundary probability density function PDF of system reliability posterior distribution; 7) obtaining a boundary accumulative probability density function CDF of the posterior distribution of the system reliability; the reliability analysis method provided by the invention is scientific, has good manufacturability and has wide popularization and application values.

Description

Structural system reliability analysis method based on uncertainty Bayesian network
Technical Field
The invention provides a structural system reliability analysis method based on an uncertainty Bayesian network, relates to a structural system reliability analysis method based on a fuzzy random uncertainty Bayesian network, aims to perform reliability analysis on a structural system under fuzzy randomness information containing inherent uncertainty and recognition uncertainty at the same time, and belongs to the field of structural reliability analysis.
Background
In the first aspect, a Bayesian network method (BN) is used as a member of a graph model and an analysis framework, is an important tool for processing uncertainty problems in the fields of system science, reliability science, artificial intelligence and the like, and has been developed vigorously in more than twenty years, and the graphical expression function of the Bayesian network method enables the relationship and state expression of a system and elements to be more intuitive and clear. The problem of poor information and small samples in system analysis can be solved well. The system analysis based on the Bayesian network in the reliability engineering is mainly completed according to the structure composition of the product and the establishment of a fault tree.
Various currently widely used bayesian networks (dynamic bayesian network DBN, object-oriented bayesian network OOBN, qualitative bayesian network, etc.) essentially start from grasped prior knowledge, and use the actually measured data and information as "evidence" to update and correct the existing cognition, thereby obtaining the posterior cognition of the problem after considering the actual factors. From the point of view of physique, the method is based on a Bayesian formula, utilizes a conditional probability table from bottom to top to construct a likelihood function, performs iterative computation, comprehensively utilizes various effective information to obtain a posterior probability reasoning result, and is widely applied to the field of system reliability modeling and analysis at present. The accuracy of the system reliability analysis result mainly depends on three parts of prior information, conditional probability and experimental data serving as evidence. In the traditional research on the Bayesian method, the selection of prior distribution, the acquisition of conditional probability and the influence of experimental data information as evidence on the overall reliability analysis are often focused, while the uncertainty research on input information is relatively less, and most of the existing Bayesian methods only process probability information, fuzzy information or are based on evidence theory.
On the other hand, at present, the types of uncertainty of input information are mainly divided into two main types, namely inherent uncertainty information (represented by probability information) and cognitive uncertainty information (represented by fuzzy information and interval information), and in consideration of the fact that uncertainty information of a complex structure system is mixed, namely, the inherent uncertainty (probability information) and the cognitive uncertainty (poor information or fuzzy information) exist at the same time, so that errors and complexity of a traditional Bayesian network analysis method taking probability or fuzzy information as input are increased, and a Bayesian network inference method depending on single inherent uncertainty and cognitive uncertainty has limitations and cannot guarantee accuracy and confidence. The fuzzy random variable is an uncertain variable with both inherent and cognitive uncertainty characteristics, is an uncertain information description method with a probability distribution form and simultaneously distributed parameters as fuzzy numbers, integrates the characteristics of random distribution and fuzzy mathematics, can represent information containing two uncertainties simultaneously, and improves the cognitive confidence to a certain extent. The fuzzy random variable has been widely applied in the fields of structural reliability analysis, structural reliability evaluation, life models and the like. The method comprises a fuzzy first-order Reliability method FFORM (fuzzy first order Reliability method) based on fuzzy random variables, fuzzy Monte Carlo simulation based on fuzzy random variable theory, a life prediction model based on fuzzy random variables, an interval finite element method based on fuzzy random variables and the like. However, the related algorithm for applying fuzzy random variables to the reliability analysis of the structural system still needs to be expanded.
Based on the two aspects, the invention provides a structural system reliability analysis method based on a fuzzy random uncertainty Bayesian network.
Disclosure of Invention
Objects of the invention
The invention aims to search a new system reliability analysis framework under inherent and cognitive mixed uncertainty by adopting fuzzy random distribution as system reliability prior distribution and combining a Bayesian network to carry out reliability analysis on a structural system aiming at a small subsample structural system with insufficient sample size.
The method comprises the steps of firstly sampling the upper (lower) boundary of the envelope of fuzzy random prior distribution of a Bayesian network subsystem, fitting an equivalent Beta distribution expression by adopting maximum likelihood estimation, then sampling by adopting a Halstein sampling method, namely Metropolis-Hastings sampling method, according to the likelihood function of the Bayesian network, and continuously sampling the system variable edge of a high-dimensional sampling result to obtain the upper (lower) boundary of the equivalent posterior distribution of the system.
(II) technical scheme
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention discloses a structural system reliability analysis method based on an uncertain Bayesian network, namely a structural system reliability analysis method based on a fuzzy random uncertain Bayesian network, which relates to a Bayesian network structure reliability output upper (lower) boundary solving model based on fuzzy random parameters, and comprises the following implementation steps:
step (1), establishing a Bayesian network of a system object;
step (2), according to the main and objective information, the normal fuzzy random prior distribution of the system reliability is assumed;
sampling according to the upper (lower) boundary envelope of the prior distribution, and fitting the equivalent Beta distribution parameters of the envelope by adopting maximum likelihood estimation to obtain an upper (lower) boundary envelope equivalent Beta distribution expression;
step (4), establishing a system likelihood function expression according to the existing data;
multiplying the likelihood function and the equivalent Beta distribution expression of the envelope based on a Bayesian network model to obtain an upper bound (lower bound) expression of the posterior distribution of the system reliability;
step (6), carrying out Metropolis-Hastings sampling on the part with the normalization constant omitted according to an upper bound (lower bound) expression of the posterior distribution of the system, and carrying out system edge sample point statistics on the result to obtain an approximate upper bound (lower bound) probability Density function PDF (probability Density function) of the posterior distribution of the system reliability;
step (7), obtaining an upper bound (lower bound) accumulative probability density function CDF (cumulative Distribution function) of the posterior Distribution of the system reliability;
wherein, the "bayesian network for establishing system objects" in step (1) is implemented by:
the Bayesian network is formed by directed acyclic graphs formed by connecting directed edges of event relations; directed acyclic graphs are defined as G ═ V, E>(ii) a Where the set of nodes is defined as V ═ X1,X2,...,XnAn oriented edge E between the nodes represents the connection relation between events and also corresponds to a causal relation; x connected by directed edgesi→XjIn, XiIs a parent node, and XjIs a child node;
Xican use parent (X) in the parent node seti) Or pa (X)i) To represent;
the directed graph G expresses the conditional independence assumption between events, and if the prior probability value of the parent node and the conditional probability distribution between the parent node and the child node can be determined, the joint probability distribution of all nodes can be obtained by the following formula;
Figure BDA0002491713900000041
given the assumption S and the evidence set E ═ E1,E2,...,EnThe bayes theorem can be expressed as:
Figure BDA0002491713900000042
wherein the content of the first and second substances,
p (S): the probability that H is true is expressed and can be called prior probability;
P(S|Ei): the conditional probability that E is true for a given evidence is denoted as H, also called a posterior probability;
P(Ei| S): representing evidence E given that assumption H is trueiConditional probability of occurrence, also known as likelihood probability;
(1) prior probability: the probability of occurrence of various events is determined according to expert data and objective facts;
(2) posterior probability: the probability of occurrence of the event A, B and the probability of occurrence of A when B occurs are known, and the probability of occurrence of B under the condition that A is determined to occur is obtained by combining a Bayesian formula;
Figure BDA0002491713900000043
(3) the total probability formula: if (A)1,A2,...,An) A set of events that is E, and satisfies: p (A)i) If the total probability formula is more than or equal to 0, the total probability formula is as follows:
Figure BDA0002491713900000044
a typical simple Bayesian network is shown in FIG. 1(a), where X is2And X3Is a parent node, X1As child nodes, directed edge X2→X1，X3→X1A set E is formed, and the set E is a Bayesian network connection relation;
the Bayesian network can be directly provided or derived from a system fault tree, and FIG. 1(a) can describe the top event as X in FIG. 1(b)1The bottom event is X2，X3Fault tree (X)2And X3The relationship is either an or gate/and gate/not gate).
In the step (2), the normal fuzzy random prior distribution of the system reliability is assumed according to the subjective and objective information, wherein the 'objective and subjective information' is based on limited known data and then the fuzzy random prior distribution of the system is determined according to subjective judgment;
the fuzzy random prior distribution refers to distribution based on fuzzy random variables, wherein the distribution is random variables, but the distribution parameters are uncertain variables of fuzzy numbers; i.e. shaped as
Figure BDA0002491713900000051
Wherein f (-) is a random probability density function, and the parameters of the distribution
Figure BDA0002491713900000052
Is a fuzzy number; take a fuzzy random normal distribution as an example: is provided with
Figure BDA0002491713900000053
And
Figure BDA0002491713900000054
fuzzy mean and fuzzy standard deviation of fuzzy random variable respectively, then the fuzzy random normal distribution can be expressed as
Figure BDA0002491713900000055
The "normal fuzzy random prior distribution assuming system reliability according to the main and objective information" in step (2) is implemented as follows:
determining a triangular fuzzy number of a mean value and a standard deviation of fuzzy random distribution of the system reliability according to the existing data (test times T and success times N); maximum possible value of standard deviation triangular ambiguity number is available
Figure BDA0002491713900000056
Upper bound of the table is available
Figure BDA0002491713900000057
Lower bound retrievable
Figure BDA0002491713900000058
Maximum possible value of mean triangular blur number can be taken
Figure BDA0002491713900000059
Upper bound of the table is available
Figure BDA00024917139000000510
Lower bound retrievable
Figure BDA00024917139000000511
Then, the expert or experienced engineering technician modifies the parameters properly to obtain the assumed prior distribution of the system reliability.
Wherein, in the step (3), "sampling is performed according to the upper (lower) boundary envelope of the prior distribution, and the maximum likelihood estimation is adopted to fit the equivalent Beta distribution parameter of the envelope, so as to obtain the upper (lower) boundary envelope equivalent Beta distribution expression", the specific method is as follows:
"envelope" refers to the envelope consisting of the upper and lower boundaries of a fuzzy randomly distributed CDF, the envelope constructed and sampled as follows:
is provided with
Figure BDA00024917139000000512
And
Figure BDA00024917139000000513
fuzzy mean and fuzzy standard deviation of fuzzy random variable respectively; all membership functions are assumed to be triangular fuzzy numbers; thus, the fuzzy mean and standard deviation can be expressed as
Figure BDA00024917139000000514
And
Figure BDA00024917139000000515
where superscripts L, M and U are lower bound, median, and upper bound, respectively; the fuzzy random distribution CDF upper bound envelope and the sample points are composed of two parts: in that
Figure BDA00024917139000000516
To the left side of
Figure BDA00024917139000000517
For distribution sampling, in
Figure BDA00024917139000000518
To the right side of
Figure BDA00024917139000000519
For distribution sampling, an upper bound set of sampling points is defined as
Figure BDA00024917139000000520
Correspondingly, the lower bound envelope and sample points of the fuzzy random CDF are
Figure BDA0002491713900000061
To the left side of
Figure BDA0002491713900000062
For distribution sampling, in
Figure BDA0002491713900000063
To the right side of
Figure BDA0002491713900000064
For distribution sampling, a lower bound set of sampling points is definedX α＝{x 1,x 2，…,x n}; in particular, for compliance with x to N (mu)M,σM) The distribution of (a) is called a nominal distribution; figure 2 shows the shape of
Figure BDA0002491713900000065
And
Figure BDA0002491713900000066
schematic of the envelope and nominal distribution of the fuzzy random variables of (1);
"Beta distribution" means
The probability density function is a function represented by:
Figure BDA0002491713900000067
wherein (. cndot.) is a gamma distribution,
Figure BDA0002491713900000068
for variables x shaped as having a f (x; α) probability density function, called obedience parametersα Beta distribution (Chinese translation: Beta distribution);
maximum likelihood estimation α refers to the use of a set of sample points
Figure BDA0002491713900000069
And Beta distribution probability density function constructing likelihood function:
Figure BDA00024917139000000610
and solving the following equation set to obtain alpha and Beta, and then obtaining the equivalent Beta distribution expression
Figure BDA00024917139000000611
Wherein, the step (4) of establishing the system likelihood function expression according to the existing data is implemented as follows:
according to each node XiN (E) existing data P (1, 2)iH), constructing a likelihood function p (eh) in the bayesian network:
Figure BDA00024917139000000612
wherein E ═ E (E)1,E2,...,En)。
Wherein, in the step (5), "based on the bayesian network model, the likelihood function is multiplied by the equivalent Beta distribution expression of the envelope to obtain an upper bound (lower bound) expression of the posterior distribution of the system reliability", which is specifically performed as follows:
"Bayesian network model-based" means: solving the model description according to the Bayesian network in the step (1),
"multiplying the likelihood function by the equivalent Beta distribution expression of the envelope" means: beta distribution equivalent expression of prior distribution envelope obtained in step (3)
Figure BDA0002491713900000071
And the Bayesian network likelihood function expression constructed based on the current data obtained in the step (4)
Figure BDA0002491713900000072
Multiplication, upper bound (lower bound) expression of the system reliability posterior distribution
Figure BDA0002491713900000073
Wherein E ═ E (E)1,E2,...,En)。
Wherein, in the step (6), "Metropolis-Hastings sampling is performed on the part with the normalization constant being cut off according to the upper bound (lower bound) expression of the system posterior distribution, and the statistics of the system edge sample points is performed on the result to obtain the approximate upper bound (lower bound) probability density function PDF of the system reliability posterior distribution; ", the concrete method is as follows:
"truncation of normalization constant" means to truncate the denominator part of the P (HE) obtained in step (5) and retain the molecular expression; then carrying out Metropolis-Hastings sampling on the molecular expression;
"Metropolis-Hastings sampling" means that sampling is performed as follows:
setting a molecular expression obtained after the normalization constant is omitted as pi (x), and selecting a sampling frequency N;
1) from the proposed distribution density function x-N (x)i0.1) extracting the sample point xi+1Wherein x isiInitial value x1＝0.5，σ1＝0.5；
2) Calculating the transition probability P (x)i；xi+1)＝P(xi|x～N(xi+1,0.1))，P(xi+1；xi)＝P(xi+1|x～N(xi,0.1))
3) Calculating an acceptance probability αi
Figure BDA0002491713900000081
4) From uniform distributionGet uiU (0,1) if Ui＜αiThen σi+1＝xi+1(ii) a Otherwise σi+1＝xi(ii) a Repeating the steps 1) to 3) for N times, and obtaining an N-dimensional sample point set; for the sample point set, sample points in the 'system' dimension are extracted, namely upper bound (lower bound) sample points { sigma } of the posterior distribution of the systemi},i＝1,2,...,N；
"obtaining an approximate upper bound (lower bound) probability density function PDF of a posterior distribution of system reliability" refers to:
according to sample point { sigma i1,2, N, in the interval [0,1 ═ N]The frequency histogram is plotted up to approximate the upper (lower) probability density function PDF of the posterior distribution: will be interval [0,1]Is uniformly divided into m portions { [ dx { [ dX ]0,dx1],(dx1,dx2],...,(dxm-1,dxm]Where dx is0＝0,dxm＝1,
Figure BDA0002491713900000082
i is 0, 1.. times.m, and { σ is calculatedi1,2, N, the number of samples in each interval { σ }Number,i1,2, m, then the histogram is
Figure BDA0002491713900000083
Wherein, the step (7) of obtaining the upper bound (lower bound) cumulative probability density function CDF of the system reliability posterior distribution means that the frequency cumulative calculation is performed on the interval [0,1] according to the frequency histogram obtained in the step (6), and the upper bound (lower bound) approximate CDF histogram of the system reliability posterior distribution is drawn according to the following formula
Figure BDA0002491713900000084
The reliability analysis method of the present invention is illustrated in the flow chart of fig. 3.
(III) advantages and Effect of the invention
The structural system reliability analysis method provided by the invention combines the characteristics of the Bayesian network analysis method and the fuzzy random variables, provides a Bayesian network system reliability calculation framework based on fuzzy random parameters, and has the advantages and effects that:
(1) the Bayesian network solving process under inherent and cognitive mixed uncertainty is constructed, and the reliability analysis of a structural system under incomplete data can be guided;
(2) based on the characteristics of fuzzy random variables, the upper and lower boundaries of the distribution of the system reliability are obtained according to the Bayesian network, and the posterior reference of the system reliability can be provided for engineering/scientific research personnel;
(3) the method is convenient and quick to calculate, can fully utilize the existing data to calculate the reliability of the system, can be implemented under the probability of unknown conditions, and can solve the problems of large numerical calculation amount and low confidence coefficient of the structural system caused by mixed uncertainty information widely existing in engineering;
(4) the reliability analysis method provided by the invention is scientific, has good manufacturability and has wide popularization and application values.
Description of the drawings (in the figures, the numbers, symbols, and symbols are as follows)
Fig. 1 is a typical simple bayesian network and fault tree.
FIG. 2 is a fuzzy random cumulative distribution function.
FIG. 3 is a flow chart of the system reliability analysis of the present invention.
Fig. 4 is a bayesian network in case 1.
Fig. 5 is a case 1 system reliability prior distribution.
Fig. 6 is case 1 system reliability equivalent prior Beta distribution.
FIG. 7 is the acceptance rate (upper bound) 1000 times before sampling of case 1 Metropolis-Hastings.
FIG. 8 is the acceptance rate (lower bound) 1000 times before sampling of case 1 Metropolis-Hastings.
Fig. 9a and 9b are posterior distribution frequency histograms for the case 1 system.
Fig. 10 is the cumulative density of posterior distribution frequencies for the case 1 system.
FIG. 11 is a fault tree for "turbine system faults".
FIG. 12 is a Bayesian network of "turbine system failure".
Fig. 13 is a case 2 system reliability prior distribution.
Fig. 14 is a case 2 system reliability equivalent prior Beta distribution.
FIG. 15 is the acceptance rate (upper bound) 1000 times before sampling of case 2 Metropolis-Hastings.
FIG. 16 is the acceptance rate (lower bound) 1000 times before sampling of case 2 Metropolis-Hastings.
Fig. 17a and 17b are posterior distribution frequency histograms for the case 2 system.
Fig. 18 is the cumulative density of posterior distribution frequencies for the case 2 system.
The numbers, symbols and codes in the figures are explained as follows:
in FIG. 1, X1，X2And X3Is a Bayesian network node; is a fault tree AND gate;
in FIG. 2, CDF is the cumulative distribution function;
in FIG. 5, CDF is the cumulative distribution function;
in fig. 6, PDF is a probability density function, and Beta distribution is Beta distribution;
in fig. 9a and 9b, PDF is a probability density function, and S represents the system of case 1;
in FIG. 10, CDF is the cumulative distribution function, S represents the system for case 1;
in FIG. 13, CDF is the cumulative distribution function;
in fig. 14, PDF is a probability density function, and Beta distribution is Beta distribution;
17a and 17b, PDF is a probability density function, S represents the system of case 2;
in fig. 18, CDF is the cumulative distribution function, and S represents the system of case 2.
The symbols and codes referred to in the present specification are as follows:
beta distribution-Beta distribution
FFORM (fuzzy First Order Reliability method) -fuzzy First Order Reliability method based on fuzzy random variable
PDF (probability Distribution function) -probability density function
CDF (cumulative Distribution function) — cumulative Distribution function
Metropolis-Hastings sampling-Halastine sampling
Detailed Description
The technical scheme of the invention is explained in detail by combining the calculation examples and the attached drawings.
Case 1 (numerical case):
suppose a Bayesian network is shown in FIG. 4, with 3 nodes, node S representing the system and node C1And C2Representing a subsystem. When the edge distribution and the condition distribution are unknown, the posterior distribution P (S | C) of the system is calculated by the method according to the known data1,C2S). Assume that the known data are as shown in table 1.
TABLE 1 known data
Bayesian network node names Known data (number of successes/total number of tests)
Subsystem C1 11/14
Subsystem C2 37/41
System S 8/11
The invention relates to a structural system reliability analysis method based on fuzzy random uncertainty Bayesian network, as shown in FIG. 3, the implementation steps are as follows:
step (1), establishing a Bayesian network of a system object; according to the description, a Bayesian network is established, as shown in FIG. 4 in a Bayesian network schematic diagram 1.
Step (2), according to the main and objective information, the normal fuzzy random prior distribution of the system reliability is assumed;
according to the known data system data of 8/11, the prior distribution of the system is assumed to be in accordance with the normal fuzzy random distribution, and the mean value and the standard deviation of the distribution parameters are both assumed to be triangular fuzzy numbers, which are shown in table 2, and as shown in fig. 5 of the normal fuzzy random cumulative probability diagram 2.
TABLE 2 systematic fuzzy random prior distribution
Figure BDA0002491713900000111
Figure BDA0002491713900000121
Sampling according to the upper (lower) boundary envelope of the prior distribution, and fitting the equivalent Beta distribution parameters of the envelope by adopting maximum likelihood estimation to obtain an upper (lower) boundary envelope equivalent Beta distribution expression;
the fuzzy random variable constructed envelope generated in step (2) is sampled 1000 times according to the method explained in step (3) as shown in fig. 5, and the sample point is marked as xi( i 1, 2.., 1000), the likelihood function in the likelihood estimation is
Figure BDA0002491713900000122
Wherein f is Beta function expression, then α can be obtained by solving the following equation system, and then the equivalent Beta distribution expression is obtained
Figure BDA0002491713900000123
The expression parameters α -2.2510, β -1.4075 of upper boundary envelope equivalent Beta distribution expression, α -1.4969, β -1.5324 of lower boundary envelope equivalent Beta distribution expression are obtained by solution
Figure BDA0002491713900000124
Even the expression corresponding thereto. The upper (lower) envelope equivalent Beta distribution is shown in FIG. 6.
Step (4), establishing a system likelihood function expression according to the existing data;
according to the data in table 1, assuming that all nodes are binomial distributed, the system likelihood function expression is:
L(p1,p2,pS)＝p1 11(1-p1)3p2 37(1-p2)4pS 8(1-pS)3
multiplying the likelihood function and the equivalent Beta distribution expression of the envelope based on a Bayesian network model to obtain an upper bound (lower bound) expression of the posterior distribution of the system reliability;
according to the results of the steps (3) and (4), the upper bound expression of the posterior distribution of the system reliability can be obtained as follows:
Figure BDA0002491713900000125
the lower bound expression of the system reliability posterior distribution is:
Figure BDA0002491713900000131
step (6), carrying out Metropolis-Hastings sampling on the part with the normalization constant omitted according to an upper bound (lower bound) expression of the posterior distribution of the system, and carrying out system edge sample point statistics on the result to obtain an approximate upper bound (lower bound) probability density function PDF of the posterior distribution of the system reliability; (ii) a
The expression after the normalization constant is cut off in the step (5) is as follows:
Figure BDA0002491713900000132
Metropolis-Hastings sampling is carried out, and the sampling time N is 10000 times. For clarity, the acceptance rate of the first 1000 extractions is shown in fig. 7 (upper bound), fig. 8 (lower bound).
Dividing the [0,1] interval into 50 parts, carrying out system edge frequency statistics on Metropolis-Hastings sampling results to finally obtain a fitting posterior distribution PDF frequency table shown in a table 3, wherein the corresponding posterior distribution PDF histogram is shown in a table 9a and a table 9b,
TABLE 3 system posterior distribution PDF frequency table
Figure BDA0002491713900000133
Figure BDA0002491713900000141
Step (7), obtaining an upper bound (lower bound) accumulative probability density function CDF of the posterior distribution of the system reliability;
performing frequency accumulation on each interval according to the result of the step (6) to obtain a system posterior distribution CDF frequency chart shown in a table 4, wherein a corresponding posterior distribution CDF histogram is shown in a table 10,
TABLE 4 posterior distribution CDF frequency table of system
Figure BDA0002491713900000142
Figure BDA0002491713900000151
Case 2 (engineering case):
the aeroengine turbine system fault tree is shown in fig. 11, and the top event is defined as a "turbine system fault" and is composed of 2 fault events which are connected in series, namely a turbine disk fault and a turbine blade fault, wherein 2 possible fault modes exist on the turbine disk, and the occurrence of each fault mode has a corresponding fault reason. Therefore, in order to more accurately find the fundamental events reflecting the rotor system failure, all failure causes that cause the failure mode to occur are defined as the bottom events of the failure tree. Wherein, all child nodes are in a serial relation. The known data are shown in table 5.
TABLE 5 known data
Figure BDA0002491713900000152
Figure BDA0002491713900000161
The invention relates to a structural system reliability analysis method based on fuzzy random uncertainty Bayesian network, as shown in FIG. 3, the implementation steps are as follows:
step (1), establishing a Bayesian network of a system object; according to the case description, a Bayesian network is established, as shown in FIG. 12 in a Bayesian network schematic diagram 1, and each node is marked with known data.
Step (2), according to the main and objective information, the normal fuzzy random prior distribution of the system reliability is assumed;
according to the known data system data of 194/195, the prior distribution of the system is assumed to be in accordance with the normal fuzzy random distribution, and the mean value and the standard deviation of the distribution parameters are both assumed to be triangular fuzzy numbers, which are shown in table 6, as shown in fig. 13 of a normal fuzzy random cumulative probability diagram 2.
TABLE 6 systematic fuzzy random prior distribution
Figure BDA0002491713900000162
Sampling according to the upper (lower) boundary envelope of the prior distribution, and fitting the equivalent Beta distribution parameters of the envelope by adopting maximum likelihood estimation to obtain an upper (lower) boundary envelope equivalent Beta distribution expression;
the fuzzy random variable constructed envelope generated in step (2) is sampled 1000 times according to the method explained in step (3) as shown in fig. 13, and the sample point is marked as xi( i 1, 2.., 1000), the likelihood function in the likelihood estimation is
Figure BDA0002491713900000171
Wherein f is Beta distribution function expression, then α can be obtained by solving the following equation system, and the equivalent Beta distribution expression is obtained
Figure BDA0002491713900000172
The expression parameters α -56.8116, β -1.8496 of upper boundary envelope equivalent Beta distribution expression, α -25.2457, β -1.2317 of lower boundary envelope equivalent Beta distribution expression are obtained by solution
Figure BDA0002491713900000173
Even the expression corresponding thereto. The upper (lower) envelope equivalent Beta distribution is shown in FIG. 14.
Step (4), establishing a system likelihood function expression according to the existing data;
according to the data in table 1, assuming that all nodes are binomial distributed, the system likelihood function expression is:
Figure BDA0002491713900000174
multiplying the likelihood function and the equivalent Beta distribution expression of the envelope based on a Bayesian network model to obtain an upper bound (lower bound) expression of the posterior distribution of the system reliability;
according to the results of the steps (3) and (4), the upper bound expression of the posterior distribution of the system reliability can be obtained as follows:
Figure BDA0002491713900000175
the lower bound expression of the system reliability posterior distribution is:
Figure BDA0002491713900000181
step (6), carrying out Metropolis-Hastings sampling on the part with the normalization constant omitted according to an upper bound (lower bound) expression of the posterior distribution of the system, and carrying out system edge sample point statistics on the result to obtain an approximate upper bound (lower bound) probability density function PDF of the posterior distribution of the system reliability;
the expression after the normalization constant is cut off in the step (5) is as follows:
Figure BDA0002491713900000182
Metropolis-Hastings sampling is carried out, and the sampling time N is 10000 times. For clarity, the acceptance rate of the first 1000 extractions is shown in fig. 15 (upper bound), fig. 16 (lower bound).
Dividing the [0,1] interval into 25 parts, carrying out system edge frequency statistics on Metropolis-Hastings sampling results to finally obtain a fitting posterior distribution PDF frequency table shown in a table 7, wherein the corresponding posterior distribution PDF histogram is shown in a figure 17a and a figure 17b,
TABLE 7 system posterior distribution PDF frequency table
Figure BDA0002491713900000183
Figure BDA0002491713900000191
Step (7), obtaining an upper bound (lower bound) accumulative probability density function CDF of the posterior distribution of the system reliability;
performing frequency accumulation on each interval according to the result of the step (6) to obtain a system posterior distribution CDF frequency chart shown in a table 8, wherein a corresponding posterior distribution CDF histogram is shown in a table 18,
TABLE 8 posterior distribution CDF frequency table of system
Figure BDA0002491713900000201
Figure BDA0002491713900000211

Claims (8)

1. A structural system reliability analysis method based on uncertainty Bayesian network, namely a structural system reliability analysis method based on fuzzy random uncertainty Bayesian network, is characterized in that: the implementation steps are as follows:
step (1), establishing a Bayesian network of a system object;
step (2), according to the main and objective information, the normal fuzzy random prior distribution of the system reliability is assumed;
sampling according to the upper and lower boundary envelopes of the prior distribution, and fitting the equivalent Beta distribution parameters of the envelopes by adopting maximum likelihood estimation to obtain an upper and lower boundary envelope equivalent Beta distribution expression;
step (4), establishing a system likelihood function expression according to the existing data;
multiplying the likelihood function and the equivalent Beta distribution expression of the envelope based on a Bayesian network model to obtain an upper bound expression and a lower bound expression of the posterior distribution of the system reliability;
step (6), carrying out Metropolis-Hastings sampling on the part with the normalization constant omitted according to the upper bound expression and the lower bound expression of the posterior distribution of the system, and carrying out system edge sample point statistics on the result to obtain approximate upper bound probability density function PDF and lower bound probability density function PDF of the posterior distribution of the system reliability;
and (7) obtaining the upper bound and the lower bound of the posterior distribution of the system reliability and accumulating the probability density function CDF.
2. The structural system reliability analysis method based on uncertainty bayesian network according to claim 1, namely a structural system reliability analysis method based on fuzzy random uncertainty bayesian network, characterized in that:
the "bayesian network for establishing system objects" described in step (1) is implemented by:
the Bayesian network is formed by directed acyclic graphs formed by connecting directed edges of event relations; directed acyclic graphs are defined as G ═<V,E>(ii) a Where the set of nodes is defined as V ═ X1,X2,...,XnAn oriented edge E between the nodes represents the connection relation between events and also corresponds to a causal relation; x connected by directed edgesi→XjIn, XiIs a parent node, and XjIs a child node;
Xican use parent (X) in the parent node seti) Or pa (X)i) To represent;
the directed graph G expresses the conditional independence assumption among events, and if the prior probability value of a father node and the conditional probability distribution between the father node and a descendant node can be determined, the joint probability distribution of all the nodes can be obtained by the following formula;
Figure FDA0002491713890000011
given the assumption S and the evidence set E ═ E1,E2,...,EnThe bayes theorem can be expressed as:
Figure FDA0002491713890000021
wherein the content of the first and second substances,
p (S): represents the probability that H is true, also known as the prior probability;
P(S|Ei): the conditional probability that E is true for a given evidence is denoted as H, also called a posterior probability;
P(Ei| S): representing evidence E given that assumption H is trueiConditional probability of occurrence, also known as likelihood probability;
(1) prior probability: the probability of occurrence of various events is determined according to expert data and objective facts;
(2) posterior probability: the probability of occurrence of the event A, B and the probability of occurrence of A when B occurs are known, and the probability of occurrence of B under the condition that A is determined to occur is obtained by combining a Bayesian formula;
Figure FDA0002491713890000022
(3) the total probability formula: if (A)1,A2,...,An) A set of events that is E, and satisfies: p (A)i) If the total probability formula is more than or equal to 0, the total probability formula is as follows:
Figure FDA0002491713890000023
the bayesian network can be given directly and derived from the system fault tree.
3. The structural system reliability analysis method based on uncertainty bayesian network according to claim 1, namely a structural system reliability analysis method based on fuzzy random uncertainty bayesian network, characterized in that:
in the step (2), the normal fuzzy random prior distribution of the system reliability is assumed according to the subjective and objective information, wherein the 'according to the subjective and objective information' is that the fuzzy random prior distribution of the system is determined based on limited known data and according to subjective judgment;
the fuzzy random prior distribution refers to distribution based on fuzzy random variables, wherein the distribution is random variables, but the distribution parameters are uncertain variables of fuzzy numbers; i.e. shaped as
Figure FDA0002491713890000024
Wherein f (-) is a random probability density function, and the parameters of the distribution
Figure FDA0002491713890000025
Is a fuzzy number; take a fuzzy random normal distribution as an example: is provided with
Figure FDA0002491713890000026
And
Figure FDA0002491713890000027
fuzzy mean and fuzzy standard deviation of fuzzy random variable respectively, then the fuzzy random normal distribution can be expressed as
Figure FDA0002491713890000028
The "normal fuzzy random prior distribution assuming system reliability according to the main and objective information" in step (2) is implemented as follows:
according to the existing data: determining the mean value of fuzzy random distribution of the system reliability and the triangular fuzzy number of standard deviation according to the test times T and the success times N; maximum possible value of standard deviation triangular ambiguity number
Figure FDA0002491713890000031
Upper boundary extraction
Figure FDA0002491713890000032
Lower boundary extraction
Figure FDA0002491713890000033
The maximum possible value of the mean triangular blur number can be taken
Figure FDA0002491713890000034
Upper bound of energy can get
Figure FDA0002491713890000035
Lower bound of energy can get
Figure FDA0002491713890000036
Then, experts and experienced engineering technicians modify the parameters appropriately to obtain the assumed prior distribution of the system reliability.
4. The structural system reliability analysis method based on uncertainty bayesian network according to claim 1, namely a structural system reliability analysis method based on fuzzy random uncertainty bayesian network, characterized in that:
in the step (3), "sampling is performed according to the upper and lower envelopes of the prior distribution, and the maximum likelihood estimation is adopted to fit the equivalent Beta distribution parameters of the envelopes, so as to obtain the equivalent Beta distribution expressions of the upper and lower envelopes", which specifically comprises the following steps:
"envelope" refers to the envelope consisting of the upper and lower boundaries of a fuzzy randomly distributed CDF, the envelope constructed and sampled as follows:
is provided with
Figure FDA00024917138900000320
And
Figure FDA00024917138900000321
fuzzy mean and fuzzy standard deviation of fuzzy random variable respectively; all membership functions are assumed to be triangular fuzzy numbers; thus, the fuzzy mean and standard deviation can be expressed as
Figure FDA0002491713890000039
And
Figure FDA00024917138900000310
where superscripts L, M and U are lower bound, median, and upper bound, respectively; the fuzzy random distribution CDF upper bound envelope and the sample points are composed of two parts: in that
Figure FDA00024917138900000311
To the left side of
Figure FDA00024917138900000312
For distribution sampling, in
Figure FDA00024917138900000313
To the right side of
Figure FDA00024917138900000314
For distribution sampling, an upper bound set of sampling points is defined as
Figure FDA00024917138900000315
Corresponding, fuzzy random scoreThe CDF lower bound envelope and sample points are
Figure FDA00024917138900000316
To the left side of
Figure FDA00024917138900000317
For distribution sampling, in
Figure FDA00024917138900000318
To the right side of
Figure FDA00024917138900000319
For distribution sampling, a lower bound set of sampling points is definedX α＝{x 1,x 2，...,x n}; in particular, for compliance with x to N (mu)M,σM) The distribution of (a) is called a nominal distribution;
"Beta distribution" means
The probability density function is a function represented by:
Figure FDA0002491713890000041
wherein (. cndot.) is a gamma distribution,
Figure FDA0002491713890000042
for a variable x shaped as a probability density function with f (x; α), we call the Beta distribution with compliance parameter α;
maximum likelihood estimation α refers to the use of a set of sample points
Figure FDA0002491713890000043
And Beta distribution probability density function constructing likelihood function:
Figure FDA0002491713890000044
solving the following equation system to obtain alpha, Beta and then obtaining the equivalent Beta distribution expression
Figure FDA0002491713890000045
5. The structural system reliability analysis method based on uncertainty bayesian network according to claim 1, namely a structural system reliability analysis method based on fuzzy random uncertainty bayesian network, characterized in that:
the "establishing a system likelihood function expression based on the existing data" in step (4) is implemented as follows:
according to each node XiN existing data P (E), i 1,2iH), constructing a likelihood function P (E | H) in the bayesian network:
Figure FDA0002491713890000046
wherein E ═ E1,E2,...,En。
6. The structural system reliability analysis method based on uncertainty bayesian network according to claim 1, namely a structural system reliability analysis method based on fuzzy random uncertainty bayesian network, characterized in that:
in the step (5), "based on the bayesian network model, the likelihood function is multiplied by the equivalent Beta distribution expression of the envelope to obtain the upper and lower boundary expressions of the posterior distribution of the system reliability", which is specifically performed as follows:
"Bayesian network model-based" means: solving the model description according to the Bayesian network in the step (1),
"multiplying the likelihood function by the equivalent Beta distribution expression of the envelope" means: beta distribution equivalent expression of prior distribution envelope obtained in step (3)
Figure FDA0002491713890000051
And the Bayesian network likelihood function expression constructed based on the current data obtained in the step (4)
Figure FDA0002491713890000052
Multiplication, upper and lower bound expressions of posterior distribution of system reliability
Figure FDA0002491713890000053
Wherein E ═ E1,E2,...,En。
7. The structural system reliability analysis method based on uncertainty bayesian network according to claim 1, namely a structural system reliability analysis method based on fuzzy random uncertainty bayesian network, characterized in that:
in the step (6), "Metropolis-Hastings sampling is performed on the part with the normalization constant omitted according to the upper and lower boundary expressions of the posterior distribution of the system, and system edge sample point statistics is performed on the result to obtain approximate upper and lower boundary probability density functions PDF of the posterior distribution of the system reliability; ", the concrete method is as follows:
"truncation of normalization constant" means to truncate the denominator part of P (H | E) obtained in step (5) and retain the numerator expression; then carrying out Metropolis-Hastings sampling on the molecular expression;
"Metropolis-Hastings sampling" means that sampling is performed as follows:
setting a molecular expression obtained after the normalization constant is omitted as pi (x), and selecting a sampling frequency N;
1) from the proposed distribution density function x-N (x)i0.1) extracting the sample point xi+1Wherein x isiInitial value x1＝0.5，σ1＝0.5；
2) Calculating the transition probability P (x)i；xi+1)＝P(xi|x～N(xi+1,0.1))，
P(xi+1；xi)＝P(xi+1|x～N(xi,0.1))
3) Calculating an acceptance probability αi
Figure FDA0002491713890000061
4) Extracting u from the uniform distributioniU (0,1) if Ui＜αiThen σi+1＝xi+1(ii) a Otherwise σi+1＝xi(ii) a Repeating the steps 1) to 3) for N times to obtain an N-dimensional sample point set; for the sample point set, sample points of 'system' dimension are extracted, namely upper and lower bound sample points { sigma over the posterior distribution of the systemi},i＝1,2,...,N；
"obtaining the upper and lower bound probability density functions PDF of the approximate system reliability posterior distribution" means:
according to sample point { sigmai1,2, N, in the interval [0,1 ═ N]And drawing a frequency histogram to approximate the upper and lower bound probability density functions PDF of the posterior distribution: will be interval [0,1]Is uniformly divided into m portions { [ dx { [ dX ]0,dx1],(dx1,dx2],...,(dxm-1,dxm]}，
Wherein dx is0＝0,dxm＝1,
Figure FDA0002491713890000062
Calculation of { σi1,2, N, the number of samples in each interval { σ }Number,i1,2, m, then the histogram is
Figure FDA0002491713890000063
8. The structural system reliability analysis method based on uncertainty bayesian network according to claim 1, namely a structural system reliability analysis method based on fuzzy random uncertainty bayesian network, characterized in that:
the step (7) of obtaining the upper bound and the lower bound cumulative probability density function CDF of the posterior distribution of the system reliability means that the frequency histogram obtained in the step (6) is subjected to frequency cumulative calculation on the interval [0,1], and the upper bound and the lower bound approximate CDF histogram of the posterior distribution of the system reliability are drawn according to the following formula
Figure FDA0002491713890000064
CN202010407015.5A 2020-05-14 2020-05-14 Structural system reliability analysis method based on uncertainty Bayesian network Active CN111680450B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010407015.5A CN111680450B (en) 2020-05-14 2020-05-14 Structural system reliability analysis method based on uncertainty Bayesian network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010407015.5A CN111680450B (en) 2020-05-14 2020-05-14 Structural system reliability analysis method based on uncertainty Bayesian network

Publications (2)

Publication Number Publication Date
CN111680450A true CN111680450A (en) 2020-09-18
CN111680450B CN111680450B (en) 2023-09-01

Family

ID=72433527

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010407015.5A Active CN111680450B (en) 2020-05-14 2020-05-14 Structural system reliability analysis method based on uncertainty Bayesian network

Country Status (1)

Country Link
CN (1) CN111680450B (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112560283A (en) * 2020-12-24 2021-03-26 北京航空航天大学 Credible Bayes reliability analysis method for interval parameter structure
CN112560271A (en) * 2020-12-21 2021-03-26 北京航空航天大学 Reliability analysis method for non-probabilistic credible Bayes structure
CN112765890A (en) * 2021-01-26 2021-05-07 西安电子科技大学 Dynamic domain adaptive network-based multi-working-condition rotating machine residual life prediction method
CN112836366A (en) * 2021-01-28 2021-05-25 北京科技大学 System reliability parameter estimation method based on dependent life data
CN113033043A (en) * 2021-03-08 2021-06-25 浙江大学 Optimization design method for topology and material collaborative robustness of composite material supporting structure
CN113139247A (en) * 2021-04-19 2021-07-20 北京工业大学 Mechanical structure uncertainty parameter quantification and correlation analysis method
CN113409901A (en) * 2021-06-29 2021-09-17 南华大学 Causal inference method and system for cascade medical observation data
CN113688529A (en) * 2021-08-30 2021-11-23 北京化工大学 Structure reliability calculation method based on non-precise distribution information
CN113962520A (en) * 2021-09-16 2022-01-21 上海核工程研究设计院有限公司 Rapid calculation method for Bayesian update of lognormal distribution of nuclear power plant
CN117113524A (en) * 2023-07-17 2023-11-24 武汉理工大学 Sampling method, system, equipment and terminal integrating design knowledge

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109143972A (en) * 2018-08-28 2019-01-04 大连理工大学 A kind of Reliability Evaluation Methods of CNC Lathes based on Bayes and fault tree
CN110442942A (en) * 2019-07-26 2019-11-12 北京科技大学 A kind of multiechelon system analysis method for reliability based on Bayes's mixing
WO2020041956A1 (en) * 2018-08-28 2020-03-05 大连理工大学 Bayes- and fault tree-based reliability evaluation method for computer numerical control machine tool

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109143972A (en) * 2018-08-28 2019-01-04 大连理工大学 A kind of Reliability Evaluation Methods of CNC Lathes based on Bayes and fault tree
WO2020041956A1 (en) * 2018-08-28 2020-03-05 大连理工大学 Bayes- and fault tree-based reliability evaluation method for computer numerical control machine tool
CN110442942A (en) * 2019-07-26 2019-11-12 北京科技大学 A kind of multiechelon system analysis method for reliability based on Bayes's mixing

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
YOU LINGFEI 等: "Structural reliability analysis based on fuzzy random uncertainty", EKSPLOATACJA I NIEZAWODNOSC – MAINTEN ANCE AND RELIABILITY, vol. 21, no. 4, pages 599 - 609 *
王海朋;段富海;: "复杂不确定系统可靠性分析的贝叶斯网络方法", 兵工学报, no. 01, pages 173 - 184 *

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112560271B (en) * 2020-12-21 2022-07-01 北京航空航天大学 Reliability analysis method for non-probabilistic credible Bayes structure
CN112560271A (en) * 2020-12-21 2021-03-26 北京航空航天大学 Reliability analysis method for non-probabilistic credible Bayes structure
CN112560283A (en) * 2020-12-24 2021-03-26 北京航空航天大学 Credible Bayes reliability analysis method for interval parameter structure
CN112560283B (en) * 2020-12-24 2023-08-11 北京航空航天大学 Credible Bayesian reliability analysis method for interval parameter structure
CN112765890A (en) * 2021-01-26 2021-05-07 西安电子科技大学 Dynamic domain adaptive network-based multi-working-condition rotating machine residual life prediction method
CN112765890B (en) * 2021-01-26 2023-02-10 西安电子科技大学 Dynamic domain adaptive network-based multi-working-condition rotating machine residual life prediction method
CN112836366A (en) * 2021-01-28 2021-05-25 北京科技大学 System reliability parameter estimation method based on dependent life data
CN112836366B (en) * 2021-01-28 2024-04-02 北京科技大学 System reliability parameter estimation method based on component dependent life data
CN113033043A (en) * 2021-03-08 2021-06-25 浙江大学 Optimization design method for topology and material collaborative robustness of composite material supporting structure
CN113139247B (en) * 2021-04-19 2023-04-04 北京工业大学 Mechanical structure uncertainty parameter quantification and correlation analysis method
CN113139247A (en) * 2021-04-19 2021-07-20 北京工业大学 Mechanical structure uncertainty parameter quantification and correlation analysis method
CN113409901A (en) * 2021-06-29 2021-09-17 南华大学 Causal inference method and system for cascade medical observation data
CN113409901B (en) * 2021-06-29 2023-09-29 南华大学 Causal inference method and causal inference system for cascade medical observation data
CN113688529A (en) * 2021-08-30 2021-11-23 北京化工大学 Structure reliability calculation method based on non-precise distribution information
CN113688529B (en) * 2021-08-30 2024-02-02 北京化工大学 Structural reliability calculation method based on inaccurate distribution information
CN113962520A (en) * 2021-09-16 2022-01-21 上海核工程研究设计院有限公司 Rapid calculation method for Bayesian update of lognormal distribution of nuclear power plant
CN113962520B (en) * 2021-09-16 2024-05-17 上海核工程研究设计院股份有限公司 Rapid calculation method for updating logarithmic normal distribution Bayes of nuclear power plant
CN117113524A (en) * 2023-07-17 2023-11-24 武汉理工大学 Sampling method, system, equipment and terminal integrating design knowledge
CN117113524B (en) * 2023-07-17 2024-03-26 武汉理工大学 Sampling method, system, equipment and terminal integrating design knowledge

Also Published As

Publication number Publication date
CN111680450B (en) 2023-09-01

Similar Documents

Publication Publication Date Title
CN111680450A (en) Structural system reliability analysis method based on uncertainty Bayesian network
CN110674604B (en) Transformer DGA data prediction method based on multi-dimensional time sequence frame convolution LSTM
CN111562358B (en) Transformer oil gas content prediction method and system based on combined model
CN111898831B (en) Real-time flood probability forecasting practical method
Jaroszewicz et al. Fast discovery of unexpected patterns in data, relative to a bayesian network
CN101105841A (en) Method for constructing gene controlled subnetwork by large scale gene chip expression profile data
CN104573876A (en) Wind power plant short-period wind speed prediction method based on time sequence long memory model
CN112632794A (en) Power grid reliability evaluation method based on cross entropy parameter subset simulation optimization
CN111401755A (en) Multi-new-energy output scene generation method, device and system based on Markov chain
CN113627685B (en) Wind driven generator power prediction method considering wind power internet load limit
CN116842337A (en) Transformer fault diagnosis method based on LightGBM (gallium nitride based) optimal characteristics and COA-CNN (chip on board) model
CN112434813A (en) Multiple multidimensional language reasoning method based on attribute-oriented language concept lattice
CN117131449A (en) Data management-oriented anomaly identification method and system with propagation learning capability
CN113408138B (en) Lithium battery SOH estimation method and system based on secondary fusion
CN113793227A (en) Human-like intelligent perception and prediction method for social network events
CN114897138A (en) System fault diagnosis method based on attention mechanism and depth residual error network
CN117669656A (en) TCN-Semi PN-based direct-current micro-grid stability real-time monitoring method and device
WO2019134258A1 (en) Reliability approximation calculation method for large-scale multi-state system having cascading structure
Manurung et al. GPD threshold estimation using measure of surprise
CN112820400A (en) Disease diagnosis method, device and equipment based on medical knowledge map knowledge reasoning
Xin et al. Aerospace structural reliability analysis method based on regular vine copula model with the asymmetric tail correlation
CN114330103A (en) Spacecraft anomaly relevance model training method based on-orbit anomaly information
CN115661617B (en) Dynamic self-adaptive distributed cooperative working method for remote sensing big data
CN114462548B (en) Method for improving accuracy of single-cell deep clustering algorithm
CN115410661A (en) Chemical reaction kinetic model intelligent construction method based on data driving

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