CN111539457B - Fault fusion diagnosis method based on Bayesian network - Google Patents
Fault fusion diagnosis method based on Bayesian network Download PDFInfo
- Publication number
- CN111539457B CN111539457B CN202010255330.0A CN202010255330A CN111539457B CN 111539457 B CN111539457 B CN 111539457B CN 202010255330 A CN202010255330 A CN 202010255330A CN 111539457 B CN111539457 B CN 111539457B
- Authority
- CN
- China
- Prior art keywords
- fault
- diagnosis
- fusion
- diagnosis method
- occurs
- 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
- 238000003745 diagnosis Methods 0.000 title claims abstract description 73
- 238000000034 method Methods 0.000 title claims abstract description 51
- 230000004927 fusion Effects 0.000 title claims abstract description 31
- 238000002405 diagnostic procedure Methods 0.000 claims description 10
- 238000004088 simulation Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000011058 failure modes and effects analysis Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Test And Diagnosis Of Digital Computers (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention relates to a fault fusion diagnosis method based on a Bayesian network, which comprises the following steps: (1) Carrying out {1,2,3, …, N } numbering on faults of a spacecraft control system, and calculating occurrence probability of each fault; (2) For each fault diagnosis method, calculating the false alarm rate, false alarm rate and accuracy rate when the fault q occurs; and (3) calculating to obtain a fault diagnosis fusion result. The method overcomes the difference of diagnosis results caused by the difference of information acquisition paths, information quantity and diagnosis methods adopted in the fault diagnosis process, and greatly improves the fault diagnosis precision of the spacecraft control system.
Description
Technical Field
The invention relates to a fault fusion diagnosis method.
Background
The fault diagnosis of the spacecraft control system consists of an on-board autonomous diagnosis part and a ground diagnosis part. The information acquired by the on-orbit autonomous fault diagnosis unit reflects the original information of the most direct faults and anomalies of the spacecraft, has the characteristics of good real-time performance and no limitation of measurement and control areas, and limits the application of some advanced fault diagnosis methods due to the calculation amount of the on-orbit computer and the limitation of the storage space. The ground fault diagnosis unit has strong capability of computer software and hardware, and can use various complex models and reasoning methods to comprehensively and deeply diagnose the faults of the spacecraft by combining historical data and expert knowledge of the spacecraft.
Therefore, due to the difference of acquired information, the difference of resource allocation and the difference of personnel intervention degree, different diagnosis methods can be adopted. The diversity of fault diagnosis methods and the limitations of application in the diagnosis of faults in an aircraft result in the possible occurrence of conflicts or differences in the fault diagnosis results obtained.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: aiming at the problem of fusion of fault diagnosis results, giving an online self-adaptive updating strategy of fault occurrence probability, false alarm rate and false alarm rate; on the basis, the fault fusion diagnosis method based on the Bayesian network is provided, the diagnosis result difference caused by the difference of the information acquisition path, the information quantity and the diagnosis method adopted in the fault diagnosis process is overcome, and the fault diagnosis precision of the spacecraft control system is greatly improved.
The technical scheme adopted by the invention is as follows: a fault fusion diagnosis method based on a Bayesian network comprises the following steps:
(1) Carrying out {1,2,3, …, N } numbering on faults of a spacecraft control system, and calculating occurrence probability of each fault; n is the total number of faults of the spacecraft control system and is a positive integer;
(2) For each fault diagnosis method, calculating the false alarm rate, false alarm rate and accuracy rate when the fault q occurs;
(3) And calculating to obtain a fault diagnosis fusion result.
In the step (1), the estimated probability P of occurrence of the fault q at the time of k q (k):
Wherein Y is q (j) According to D 0 (j) Obtaining:
wherein: d (D) 0 (j) The fault diagnosis fusion result given at the moment j is represented; q=1, 2,3, …, N, k, j are positive integers.
In the step (1), P of the current period q When (k-1) is known, the fault occurs at time kProbability of q occurrence P q (k) The following iterative form is adopted to obtain:
initial value P q (0) And the fault model influence analysis is carried out.
In the step (2), when the fault q occurs, the failure diagnosis method l has a failure report rate beta at the time k l,q (k) The iterative calculation formula of (2) is as follows:
wherein d l,q (j) Representation d l,q The value at the moment j, the data transmitted to the diagnosis result fusion unit by the diagnosis method l is D l ={d l,q },d l,q ∈{0,1},l=1,2,…,M;d l,q The expression of =1 indicates that the diagnostic method l considers that the fault q occurs, d l,0 =1 means that diagnostic method i considers no fault to occur; m represents the number of fault diagnosis methods;
β l,q (0) Beta is l,q (k) Is obtained by mathematical simulation statistics.
In the step (2), when the fault q occurs, the fault diagnosis method l identifies the fault q as the false positive rate ζ of the fault p l,p,q (k) The iterative formula of (2) is:
wherein at time kξ l,p,q (0) Is xi l,p,q (k) Is obtained by mathematical simulation statistics.
In step (2), when a fault q occurs, the accuracy ψ of the diagnostic method l l,q (k) The method comprises the following steps:
ψ l,q (k)=1-β l,q (k)-ξ l,p,q (k)。
in step (3), the diagnosis results d= (D) given for the M fault diagnosis methods 1 ,D 2 ,…,D M ) The fault diagnosis fusion result at the time k+1 is as follows:
wherein,
P 0 (k) The probability that no failure occurs at time k is indicated.
Compared with the prior art, the invention has the beneficial effects that:
(1) The invention utilizes the Bayesian network to fuse the fault diagnosis results, can fully consider the multi-factor influences such as the fault occurrence probability, the failure report rate, the false report rate and the like, and effectively ensures the rationality of the fault diagnosis results.
(2) The invention solves the problem of low accuracy of the fault fusion result caused by the simple criterion of 'minority compliance' at present, and the method is simple and clear and is suitable for engineering design.
Detailed Description
The invention is further illustrated below with reference to examples.
Example 1
A fault fusion diagnosis method based on a Bayesian network comprises the following steps:
(1) The faults of the spacecraft control system are numbered {1,2,3, …, N }, and the occurrence probability of each fault is calculated.
Fault q occurrence probability P estimated at k time q (k) The following formula may be used for description:
wherein Y is q (j) According to D 0 (j) Obtaining:
wherein: d (D) 0 (j) The fault diagnosis fusion result given at the moment j is represented; k. j are positive integers.
P of the current period q When (k-1) is known, the probability P of occurrence of the fault q at time k q (k) The following iterative form can also be used to obtain:
wherein the initial value P q (0) Can be obtained by Failure Mode Effect Analysis (FMEA).
(2) For each fault diagnosis method, the rate of false alarm and the rate of accuracy when the fault q occurs are calculated.
Assuming M fault diagnosis methods, N faults are considered, and the data transmitted to the diagnosis result fusion unit by the diagnosis method I is D l ={d l,q }(d l,q E {0,1}, l=1, 2, …, M, q=1, 2, …, N). If d l,q The expression of =1 indicates that the diagnostic method l considers that the fault q occurs, d l,0 =1 indicates that diagnostic method i considers no fault to occur.
When a fault q occurs, the failure diagnosis method l has a false alarm rate beta at the time k l,q (k) The calculation formula is as follows:
the above is rewritten into an iterative form:
wherein d l,q (j) Representation d l,q The value at time j, beta l,q (0) Beta is l,q (k) Is typically obtained by multiple mathematical simulation statistics.
When a fault q occurs, the fault diagnosis method l considers the fault q as the false alarm rate xi of the fault p l,p,q (k) The method comprises the following steps:
wherein, at the moment j
The above is rewritten into an iterative form:
at time k
In the same way, xi l,p,q (0) Is xi l,p,q (k) Is typically obtained by multiple mathematical simulation statistics.
Further, it is known that the accuracy ψ of the diagnostic method l when the fault q occurs l,q (k) The method comprises the following steps:
ψ l,q (k)=1-β l,q (k)-ξ l,p,q (k)
(3) And calculating to obtain a fusion result according to the fault diagnosis fusion rule.
Diagnostic results d= (D) given for M fault diagnostic methods 1 ,D 2 ,…,D M ) The fault diagnosis fusion result at the time k+1 is as follows:
wherein,
D 0 (k+1) represents a fault diagnosis fusion result obtained at time k+1;
P 0 (k) The probability that no failure occurs at time k is indicated.
Although the present invention has been described in terms of the preferred embodiments, it is not intended to be limited to the embodiments, and any person skilled in the art can make any possible variations and modifications to the technical solution of the present invention by using the methods and technical matters disclosed above without departing from the spirit and scope of the present invention, so any simple modifications, equivalent variations and modifications to the embodiments described above according to the technical matters of the present invention are within the scope of the technical matters of the present invention.
The invention, in part not described in detail, is within the skill of those skilled in the art.
Claims (6)
1. A fault fusion diagnosis method based on a Bayesian network is characterized by comprising the following steps:
(1) Carrying out {1,2,3, …, N } numbering on faults of a spacecraft control system, and calculating occurrence probability of each fault; n is the total number of faults of the spacecraft control system and is a positive integer;
(2) For each fault diagnosis method, calculating the false alarm rate, false alarm rate and accuracy rate when the fault q occurs;
(3) Calculating to obtain a fault diagnosis fusion result;
in step (3), the diagnosis results d= (D) given for the M fault diagnosis methods 1 ,D 2 ,…,D M ) The fault diagnosis fusion result at the time k+1 is as follows:
wherein,
P 0 (k) The probability of failure at time k is represented;
P q (k) The occurrence probability of the fault q estimated at the moment k;
β l,q (k) The failure diagnosis method l has a missing report rate at the time k when the failure q occurs;
d l,q (j) Representation d l,q The value at the moment j, the data transmitted to the diagnosis result fusion unit by the diagnosis method l is D l ={d l,q },d l,q ∈{0,1},l=1,2,…,M;
ξ l,p,q (k) When the fault q occurs, the fault diagnosis method l determines the fault q as the false alarm rate of the fault p;
ψ l,q (k) Indicating the accuracy of the diagnostic method i when a fault q occurs.
2. The bayesian network-based fault fusion diagnosis method according to claim 1, wherein in the step (1), the probability P of occurrence of the fault q estimated at the time k is determined q (k):
Wherein Y is q (j) According to D 0 (j) Obtaining:
wherein: d (D) 0 (j) The fault diagnosis fusion result given at the moment j is represented; q=1, 2,3, …, N, k, j are positive integers.
3. The bayesian network-based fault fusion diagnosis method according to claim 2, wherein in the step (1), P of the current period q When (k-1) is known, the probability P of occurrence of the fault q at time k q (k) The following iterative form is adopted to obtain:
initial value P q (0) And the fault model influence analysis is carried out.
4. A bayesian network-based fault fusion diagnosis method according to claim 3, wherein in step (2), when a fault q occurs, the failure diagnosis method i has a failure report rate β at time k l,q (k) The iterative calculation formula of (2) is as follows:
wherein d l,q (j) Representation d l,q The value at the moment j, the data transmitted to the diagnosis result fusion unit by the diagnosis method l is D l ={d l,q },d l,q ∈{0,1},l=1,2,…,M;d l,q The expression of =1 indicates that the diagnostic method l considers that the fault q occurs, d l,0 =1 means that diagnostic method i considers no fault to occur; m represents the number of fault diagnosis methods;
β l,q (0) Beta is l,q (k) Is obtained by mathematical simulation statistics.
5. The bayesian network-based fault fusion diagnosis method according to claim 4, wherein step (2)) In the method, when a fault q occurs, the fault diagnosis method l considers the fault q as the false alarm rate xi of the fault p l,p,q (k) The iterative formula of (2) is:
wherein at time kξ l,p,q (0) Is xi l,p,q (k) Is obtained by mathematical simulation statistics.
6. The bayesian network-based fault fusion diagnosis method according to claim 5, wherein in the step (2), when the fault q occurs, the accuracy ψ of the diagnosis method i is set l,q (k) The method comprises the following steps:
ψ l,q (k)=1-β l,q (k)-ξ l,p,q (k)。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010255330.0A CN111539457B (en) | 2020-04-02 | 2020-04-02 | Fault fusion diagnosis method based on Bayesian network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010255330.0A CN111539457B (en) | 2020-04-02 | 2020-04-02 | Fault fusion diagnosis method based on Bayesian network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111539457A CN111539457A (en) | 2020-08-14 |
CN111539457B true CN111539457B (en) | 2023-12-15 |
Family
ID=71974915
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010255330.0A Active CN111539457B (en) | 2020-04-02 | 2020-04-02 | Fault fusion diagnosis method based on Bayesian network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111539457B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117169717A (en) * | 2023-09-11 | 2023-12-05 | 江苏微之润智能技术有限公司 | Motor health assessment method and device based on single chip microcomputer and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107544460A (en) * | 2017-09-05 | 2018-01-05 | 北京控制工程研究所 | Consider the diagnosticability quantization method of spacecraft control non-fully failure of removal |
CN109117353A (en) * | 2018-08-20 | 2019-01-01 | 中国石油大学(北京) | The fusion method and device of fault diagnosis result |
CN109961027A (en) * | 2019-03-13 | 2019-07-02 | 东北大学 | Subspace multi-angle of view electric melting magnesium furnace fault monitoring method is shared based on similarity measurement |
-
2020
- 2020-04-02 CN CN202010255330.0A patent/CN111539457B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107544460A (en) * | 2017-09-05 | 2018-01-05 | 北京控制工程研究所 | Consider the diagnosticability quantization method of spacecraft control non-fully failure of removal |
CN109117353A (en) * | 2018-08-20 | 2019-01-01 | 中国石油大学(北京) | The fusion method and device of fault diagnosis result |
CN109961027A (en) * | 2019-03-13 | 2019-07-02 | 东北大学 | Subspace multi-angle of view electric melting magnesium furnace fault monitoring method is shared based on similarity measurement |
Non-Patent Citations (6)
Title |
---|
Fault Diagnosis Based on Deep Learning;Feiya Lv等;《2016 American Control Conference (ACC)》;第6851-6856页 * |
Principal Polynomial Analysis for Fault Detection and Diagnosis of Industrial Processes;XINMIN ZHANG等;《IEEEAccess》;第52298-52307页 * |
Satellite Fault Diagnosis U sing Support Vector Machines Based on a Hybrid Voting Mechanism;Hong Yin等;《Hindawi》;第1-11页 * |
基于人工免疫系统的机组复合故障诊断技术研究;岑健;《中国博士学位论文全文数据库工程科技Ⅱ辑》;C029-10 * |
基于贝叶斯网络和AMESim仿真的液压系统故障诊断方法;姚成玉;刘文静;赵静一;冯中魁;;机床与液压(13);第172-177页 * |
数据驱动的在线MW-MSPCA故障诊断;胡友强等;《重庆大学学报》;第100-106页 * |
Also Published As
Publication number | Publication date |
---|---|
CN111539457A (en) | 2020-08-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP2332017B1 (en) | Estimating a state of at least one target | |
CN109709934B (en) | Fault diagnosis redundancy design method for flight control system | |
CN110008253B (en) | Industrial data association rule mining and abnormal working condition prediction method | |
Liu et al. | Adaptive sensor allocation strategy for process monitoring and diagnosis in a Bayesian network | |
CN104865956B (en) | Sensor fault diagnosis method in a kind of complication system based on Bayesian network | |
CN109635318B (en) | Intelligent analysis redundancy design method for aero-engine sensor | |
US8977575B2 (en) | Confidence level generator for bayesian network | |
CN106934242B (en) | The health degree appraisal procedure and system of equipment under multi-mode based on Cross-Entropy Method | |
EP1630635A2 (en) | Method and apparatus for improved fault detection in power generation equipment | |
CN106776288A (en) | A kind of health metric method of the distributed system based on Hadoop | |
CN111539457B (en) | Fault fusion diagnosis method based on Bayesian network | |
JP2021018821A (en) | Data intelligent prediction method, device, computer device, and storage medium | |
JP2022502737A (en) | Boiler operation data monitoring method and equipment by Bayesian network | |
CN116519021B (en) | Inertial navigation system fault diagnosis method, system and equipment | |
CN111680398B (en) | Single machine performance degradation prediction method based on Holt-windows model | |
CN110262954A (en) | Method based on the automatic learning system reliability model of Condition Monitoring Data | |
CN114578792A (en) | Multi-agent fault diagnosis method and system | |
CN117390498B (en) | Flight capability assessment method of fixed wing cluster unmanned aerial vehicle based on Transformer model | |
CN110146084B (en) | Distributed collaborative navigation filtering method for multi-satellite formation system for satellite faults | |
CN116772867A (en) | Multi-AUV self-adaptive co-location method and system based on node optimization of factor graph | |
CN106874531B (en) | Method for automatically recovering abnormal measurement value data of atmospheric data system in case of failure | |
CN113377595B (en) | Fault diagnosis method, device, electronic equipment and storage medium | |
CN113852515B (en) | Node state management and control method and system for digital twin network | |
CN113569384B (en) | Digital-analog-linkage-based online adaptive prediction method for residual service life of service equipment | |
Rigatos et al. | Distributed state estimation for condition monitoring of nonlinear electric power systems |
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 |