CN109829551A - A kind of testability allocation method based on weighted difference distribution - Google Patents
A kind of testability allocation method based on weighted difference distribution Download PDFInfo
- Publication number
- CN109829551A CN109829551A CN201811651514.8A CN201811651514A CN109829551A CN 109829551 A CN109829551 A CN 109829551A CN 201811651514 A CN201811651514 A CN 201811651514A CN 109829551 A CN109829551 A CN 109829551A
- Authority
- CN
- China
- Prior art keywords
- impact factor
- distribution
- cost
- factor
- failure
- 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.)
- Pending
Links
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The present invention provides a kind of testability index distribution methods based on weighted difference distribution, can be realized reasonable distribution.The present invention is using the failure rate of system as distribution principle, it is different from failure rate distribution method, five broad aspects such as failure rate situation, failure rate influence situation, average time for repair of breakdowns, the complexity of fault detection isolation and cost are comprehensively considered, so that distribution is more reasonable；The apportioning cost of impact factor can be determined according to the requirement and actually detected situation of equipment after the characteristic of overall analysis system composition and each system unit using the thought of level Distribution Indexes；To avoid the case where there may be unreasonable distributions in traditional aggregative weighted distribution method, on the basis of aggregative weighted distribution method, testability index distribution is carried out by the principle that difference is distributed, keeps distribution more reasonable, also more objectivity.
Description
Technical field
The invention belongs to testability index distribution technique fields, and in particular to a kind of testability based on weighted difference distribution
Allocation method.
Background technique
In recent years, many distribution methods are proposed in terms of distribution principle, both at home and abroad, including failure rate distribution principle, again
Distribution principle, Reliability Distribution principle and test method distribution principle are spent, to be that failure rate is distributed former using relatively broad
Then, but often the fault diagnosis rate and Percent Isolated of system and each LRU unit are only considered in modeling and assigning process.Dividing
With in terms of algorithm, main distribution method include equivalence distribution method, experience distribution method, failure rate distribution method, aggregative weighted distribution method,
Difference distribution method etc., using it is more be failure rate distribution method and aggregative weighted distribution method.Failure rate distribution method exist consider because
The single problem of son, assigning process often only consider the fault detection rate and Percent Isolated of each LRU unit of system, to neglect
Depending on factors such as failure influence and costs.And aggregative weighted distribution method is although it is contemplated that more impact factor, but can be because some
The Distribution Indexes result that will lead to module when the weight of a certain item impact factor is excessive in system is unreasonable.Existing testability point
Method of completing the square is substantially the distribution method based on failure rate, aggregative weighted and multi-constraint condition, for quantitative analysis and reality
System verifying modified distribution method substantially without reference to.It is mentioned in standard GJB 2547-1995 equipment Test outline
The concept and some general requirements of testability model and distribution, but not to the tool of the class model and construction method
Body description.
Testability index distribution utilizes certain algorithm by the testability index of system in the solution formulation stage of equipment
It is top-down to be successively assigned to each component units.Although traditional distribution method assigning process based on failure rate is relatively simple
It is single, but the influence due to only considering failure rate, when the failure rate difference between each module is larger, the mould with larger failure rate
Block its allocation result may be unreasonable, need to it is modified and be verified, be only applicable between module failure rate difference less and
The negligible system of the other influences factor.And common Method of Weighted Assignment Based is although it is contemplated that the factor is more, but works as the influence of a certain item
The Distribution Indexes result that will lead to module when the weight of the factor is excessive is unreasonable, also need to be modified allocation result and verify with
Guarantee the reasonability of result.
Summary of the invention
In view of this, the present invention provides a kind of testability index distribution method based on weighted difference distribution, Neng Goushi
Now reasonable distribution.
To achieve the above object, a kind of testability index distribution method based on weighted difference distribution of the invention, including
Following steps:
Step 1, using the failure rate of examining system as distribution principle, building examining system carries out testability index and distributes mould
Type, the model include 4 layers, are from top to bottom followed successively by system, subsystem, LRU unit and impact factor, the impact factor
Including failure rate impact factor, failure impact factor, mean time to repair impact factor, fault detection and isolation difficulty or ease
Degree impact factor and cost factor；
Step 2, the distribution of impact factor is determined according to the requirement of equipment and actually detected situation for each LRU unit
Value；The apportioning cost value range is 1-5；
Step 3, the fault diagnosis rate apportioning cost of each LRU unit is calculated, wherein the fault diagnosis rate of i-th of LRU unit point
With value are as follows:
Pi=Αi·(Ki-Kri)+Pi
Wherein, i=1,2 ... N, N are the LRU unit sum of examining system, KiFor the underlying parameter of i-th of LRU unit, institute
It states underlying parameter and acquisition is calculated by the apportioning cost of the impact factor；KriIt is required underlying parameter, A to be knowniFor difference
The factor:
Wherein Pmax iIt is known quantity, K for fault diagnosis rate apportioning cost maximum valuemax iFor the maximum underlying parameter of setting,
PriIt is required Distribution Indexes value to be known；
Step 4, by the fault diagnosis rate apportioning cost of all LRU units, the fault detection rate of examining system is calculated, is judged
Whether the fault detection rate of calculated examining system meets the requirements, if meeting the requirements, the apportioning cost of the impact factor is
For Distribution Indexes as a result, if being unsatisfactory for requiring, the apportioning cost of the impact factor is added and subtracted in setting range, is redefined
The apportioning cost of impact factor executes step 3-4, until meeting the requirements, hit the target distribution.
Wherein, in the step 2, the apportioning cost method of determination of the impact factor are as follows:
Failure Mode Effective Analysis is carried out to system model first, obtains fault mode, influence and the crash rate of the system
Data, or related data is obtained by fail-safe analysis, then according to the composition mechanism characteristic of system, incorporation engineering analysis,
The experience of expertise and similar products determines the apportioning cost of impact factor.
Wherein, the size of the failure rate impact factor depends on the failure rate height of allocation unit；
The failure impact factor depends on the size that failure influences, failure influence the biggish failure of bigger correspondence influence because
Son；
The mean time to repair impact factor is certainly in the size of mean time to repair, when mean failure rate is repaired
Between the shorter biggish mean time to repair impact factor of correspondence；
The complexity impact factor of the fault detection and isolation depends on the fault detection and isolation hardly possible of component units
Easily, the trouble unit of more easy detected isolation corresponds to biggish complexity impact factor；
The cost factor depends on realizing the cost spent needed for the fault detection and isolation of component units, realizes failure
Detect and be isolated cost it is lower correspond to biggish cost factor.
The utility model has the advantages that
The present invention is different from failure rate distribution method using the failure rate of system as distribution principle, has comprehensively considered failure rate feelings
It is generous that condition, failure rate influence situation, average time for repair of breakdowns, the complexity of fault detection isolation and cost etc. five
Face, so that distribution is more reasonable；Using the thought of level Distribution Indexes, in the spy of overall analysis system composition and each system unit
After property, the apportioning cost of impact factor can be determined according to the requirement and actually detected situation of equipment；To avoid traditional synthesis from adding
The case where there may be unreasonable distributions in distribution method is weighed, on the basis of aggregative weighted distribution method, is distributed by difference
Principle carries out testability index distribution, keeps distribution more reasonable, also more objectivity.
Detailed description of the invention
Fig. 1 is testability index assigning process schematic diagram of the present invention.
Fig. 2 is the functional block diagram in testability index of the present invention.
Fig. 3 is aggregative weighted difference distribution method flow chart of the present invention.
Fig. 4 is the hierarchical relationship schematic diagram of search radar system of the present invention.
Fig. 5 is the index parameter figure of each LRU unit.
Fig. 6 is the present invention and the final allocation result comparison diagram of conventional weight distribution method.
Specific embodiment
With reference to the accompanying drawing, develop simultaneously embodiment, and the present invention will be described in further detail.
Fig. 1 is testability index assigning process of the present invention.Firstly the need of the distribution target of the system of determination, according to specific system
Structure determination distributes level, and then according to distribution target selection distribution principle, the present invention is allocated according to failure rate, by comprehensive
It closes weighted difference distribution method and testability allocation is carried out to each LRU unit, it is final to determine and verify allocation result.
Testability allocation method based on weighted difference distribution of the invention, includes the following steps:
Step 1, testability index distribution modeling；
The present embodiment carries out testability index distribution modeling for search radar system, is point with the failure rate of examining system
With principle, construct examining system and carry out testability index distribution model, the model includes 4 layers, be from top to bottom followed successively by system,
Subsystem, LRU unit and impact factor, the impact factor include failure rate impact factor, failure impact factor, average event
Hinder the complexity impact factor and cost factor of repairing time impact factor, fault detection and isolation.Fig. 2 is that the present invention surveys
Functional block diagram in examination property index, can be divided into system-level, subsystem grade, LRU grades and impact factor grade from top to bottom.
Step 2, the distribution of impact factor is determined according to the requirement of equipment and actually detected situation for each LRU unit
Value；The apportioning cost value range is 1-5；The apportioning cost method of determination of the impact factor are as follows:
Failure Mode Effective Analysis is carried out to system model first, obtains fault mode, influence and the crash rate of the system
Data, or related data is obtained by fail-safe analysis.Then according to the characteristics such as the composition mechanism of system, incorporation engineering point
The experience of analysis, expertise and similar products, determines the apportioning cost of impact factor.Impact factor is generally divided into 5 kinds, main shadow
Ringing the factor includes failure rate impact factor, failure impact factor, mean time to repair (MTTR) impact factor, fault detection
And the complexity impact factor and cost factor of isolation.The apportioning cost value range of impact factor is 1 to 5, only when
When a certain impact factor is not considered, which can be set 0.
Wherein, the size of failure rate impact factor depend on allocation unit failure rate height, the higher correspondence of failure rate compared with
Big failure rate impact factor.Under normal circumstances, the scale of system total failare rate is accounted for by the failure rate of the unit come really
It is fixed.
Failure impact factor depends on the size that failure influences, and failure influences the bigger biggish failure impact factor of correspondence.
Under normal circumstances, each I class of LRU unit and the event of II class are calculated by the result of fault mode, influence and severity analysis (FMECA)
Barrier accounts for the scale of the total fault mode of system to determine.
Mean time to repair (MTTR) impact factor certainly in the size of mean time to repair, repair by mean failure rate
The biggish mean time to repair impact factor of time shorter correspondence.The group of long period is needed with isolated fault for detecting
At unit, Ying Jinliang makes them using built-in test system (BIT), to reduce the mean time to repair of the unit, phase
When in making the factor take the larger value.
The complexity impact factor of fault detection and isolation depends on the fault detection and isolation difficulty or ease of component units, more
The trouble unit for being easy to be detected isolation corresponds to biggish complexity impact factor.
Cost factor depends on realizing the cost spent needed for the fault detection and isolation of component units, realizes fault detection
And be isolated cost it is lower correspond to biggish cost factor.
Step 3, the fault diagnosis rate apportioning cost of each LRU unit is calculated, wherein the fault diagnosis rate of i-th of LRU unit point
With value are as follows:
Pi=Αi·(Ki-Kri)+Pi
Wherein, i=1,2 ... N, N are the LRU unit sum of examining system, KiFor the underlying parameter of i-th of LRU unit, institute
It states underlying parameter and acquisition is calculated by the apportioning cost of the impact factor；KriIt is required underlying parameter, A to be knowniFor difference
The factor:
Wherein Pmax iIt is known quantity, K for fault diagnosis rate apportioning cost maximum valuemax iFor the maximum underlying parameter of setting,
PriIt is required Distribution Indexes value to be known；
Step 4, by the fault diagnosis rate apportioning cost of all LRU units, the fault detection rate of examining system is calculated, is judged
Whether the fault detection rate of calculated examining system meets the requirements, if meeting the requirements, the apportioning cost of the impact factor is
For Distribution Indexes as a result, if being unsatisfactory for requiring, the apportioning cost of the impact factor is added and subtracted in setting range, is redefined
The apportioning cost of impact factor executes step 3-4, until meeting the requirements, hit the target distribution.
Fig. 3 is aggregative weighted difference distribution method flow chart of the present invention.
In traditional aggregative weighted distribution method, the weight of each impact factor can be distributed according to demand before dispensing, by upper
State the Distribution Indexes situation that five impact factors calculate each LRU unit.I-th of LRU unit of traditional aggregative weighted distribution method
Distribution Indexes situation are as follows:
Wherein, i=1,2 ... N；PFDiFor the Distribution Indexes value of i-th of unit, PFDsrTarget value, λ are distributed for i-th of uniti
I-th of cell failure factor, KiFor the underlying parameter of i-th of unit, i.e., after five impact factors normalization of i-th unit
Weighted sum.
In the step 3, to avoid the case where there may be unreasonable distributions in traditional aggregative weighted distribution method,
In the case where comprehensively considering each impact factor, the method distributed using difference finally makes distribution more reasonable, also more objective
Property.The core of difference distribution method is Maximum Index apportioning cost and is required apportioning cost difference and maximum underlying parameter and is required base
The ratio of plinth parameter difference should be equal to each LRU unit Distribution Indexes value and be required apportioning cost difference and each LRU unit basis
Parameter and the ratio for being required underlying parameter difference, wherein underlying parameter is calculated by the apportioning cost of the impact factor and obtained.
System-level and LRU grades for search radar using the method for the present invention, the experiment of being tested property Distribution Indexes is tested
Card: Fig. 4 is the hierarchical relationship for the search radar system that experimental verification uses, and the LRU grade of search radar system includes antenna, antenna
Seat, feeder system, C transmitter cabinet, X transmitter cabinet, processing cabinet, servo cabinet, aobvious control cabinet, water-cooled cabinet, distribution box, electricity
Source group, transformer and dried-air drier, each LRU grades comprising signal regulating panel, receive channel plate, channel board, Universal Power
Deng.Fig. 5 is the index parameter of each LRU unit, according to the regulation of finished product factory, is determined in the figure respectively in conjunction with the parameter of specific finished product
Failure rate impact factor, failure impact factor, mean time to repair (MTTR) impact factor, fault detection of LRU unit
And the complexity impact factor and cost factor of isolation.According to the rules, the apportioning cost of the system requires to be 0.85, to each
After the item factor is normalized, pass through the aggregative weighted distribution method meter of traditional aggregative weighted distribution method and this programme respectively
Calculate final allocation result.Fig. 6 is the present invention and the final allocation result comparison diagram of conventional weight distribution method, can be with after comparison
It was found that aggregative weighted in the case where not setting the distribution upper limit, since weight is overweight, leads to the fault detection rate point of certain units
It is excessive with result, objective circumstances are not met, and allocation result polarization is more serious.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (3)
1. a kind of testability index distribution method based on weighted difference distribution, which comprises the steps of:
Step 1, using the failure rate of examining system as distribution principle, building examining system carries out testability index distribution model, institute
Stating model includes 4 layers, is from top to bottom followed successively by system, subsystem, LRU unit and impact factor, the impact factor includes
Failure rate impact factor, failure impact factor, mean time to repair impact factor, fault detection and isolation complexity
Impact factor and cost factor；
Step 2, the apportioning cost of impact factor is determined according to the requirement of equipment and actually detected situation for each LRU unit；
The apportioning cost value range is 1-5；
Step 3, the fault diagnosis rate apportioning cost of each LRU unit is calculated, wherein the fault diagnosis rate apportioning cost of i-th of LRU unit
Are as follows:
Pi=Αi·(Ki-Kri)+Pi
Wherein, i=1,2 ... N, N are the LRU unit sum of examining system, KiFor the underlying parameter of i-th of LRU unit, the base
Plinth parameter is calculated by the apportioning cost of the impact factor and is obtained；Kr iIt is required underlying parameter, A to be knowniFor difference because
Son:
Wherein Pmax iIt is known quantity, K for fault diagnosis rate apportioning cost maximum valuemax iFor the maximum underlying parameter of setting, PriFor
It is known to be required Distribution Indexes value；
Step 4, by the fault diagnosis rate apportioning cost of all LRU units, the fault detection rate of examining system is calculated, judges to calculate
Whether the fault detection rate of examining system out meets the requirements, if meeting the requirements, the apportioning cost of the impact factor as refers to
Mark allocation result adds and subtracts the apportioning cost of the impact factor in setting range, redefines influence if being unsatisfactory for requiring
The apportioning cost of the factor executes step 3-4, until meeting the requirements, hit the target distribution.
2. a kind of testability index distribution method based on weighted difference distribution as described in claim 1, which is characterized in that institute
It states in step 2, the apportioning cost method of determination of the impact factor are as follows:
Failure Mode Effective Analysis is carried out to system model first, obtains fault mode, influence and the crash rate data of the system,
Or related data is obtained by fail-safe analysis, then according to the composition mechanism characteristic of system, incorporation engineering analysis, Zhuan Jiajing
It tests and the experience of similar products, determines the apportioning cost of impact factor.
3. a kind of testability index distribution method based on weighted difference distribution as claimed in claim 1 or 2, feature exist
In the size of the failure rate impact factor depends on the failure rate height of allocation unit；
The failure impact factor depends on the size that failure influences, and failure influences the bigger biggish failure impact factor of correspondence；
The mean time to repair impact factor certainly in the size of mean time to repair, get over by mean time to repair
The short biggish mean time to repair impact factor of correspondence；
The complexity impact factor of the fault detection and isolation depends on the fault detection and isolation difficulty or ease of component units, more
The trouble unit for being easy to be detected isolation corresponds to biggish complexity impact factor；
The cost factor depends on realizing the cost spent needed for the fault detection and isolation of component units, realizes fault detection
And be isolated cost it is lower correspond to biggish cost factor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811651514.8A CN109829551A (en) | 2018-12-31 | 2018-12-31 | A kind of testability allocation method based on weighted difference distribution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811651514.8A CN109829551A (en) | 2018-12-31 | 2018-12-31 | A kind of testability allocation method based on weighted difference distribution |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109829551A true CN109829551A (en) | 2019-05-31 |
Family
ID=66861480
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811651514.8A Pending CN109829551A (en) | 2018-12-31 | 2018-12-31 | A kind of testability allocation method based on weighted difference distribution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109829551A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110287578A (en) * | 2019-06-20 | 2019-09-27 | 中国人民解放军国防科技大学 | External field replaceable unit dividing method considering development and maintenance |
| CN110570117A (en) * | 2019-09-04 | 2019-12-13 | 湖南联诚轨道装备有限公司 | Reliability allocation method |
| CN114610645A (en) * | 2022-03-30 | 2022-06-10 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Task reliability and testability joint determination method and device and computer equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106342313B (en) * | 2010-11-30 | 2013-08-28 | 中国人民解放军国防科学技术大学 | A kind of testability index distribution method based on analytic hierarchy process (AHP) |
| CN107316088A (en) * | 2017-07-13 | 2017-11-03 | 中国航空工业集团公司西安飞机设计研究所 | A kind of fault detect rate distribution method |
-
2018
- 2018-12-31 CN CN201811651514.8A patent/CN109829551A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106342313B (en) * | 2010-11-30 | 2013-08-28 | 中国人民解放军国防科学技术大学 | A kind of testability index distribution method based on analytic hierarchy process (AHP) |
| CN107316088A (en) * | 2017-07-13 | 2017-11-03 | 中国航空工业集团公司西安飞机设计研究所 | A kind of fault detect rate distribution method |
Non-Patent Citations (3)
| Title |
|---|
| CHUNLING YANG等: "A testability allocation method based on Analytic Hierarchy Process and comprehensive weighted", 《2014 9TH IEEE CONFERENCE ON INDUSTRIAL ELECTRONICS AND APPLICATIONS》 * |
| 汤文超等: "基于可拓层次分析法的测试性分配方法研究", 《计算机测量与控制》 * |
| 田仲等: "《系统测试性设计分析与验证》", 30 April 2003 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110287578A (en) * | 2019-06-20 | 2019-09-27 | 中国人民解放军国防科技大学 | External field replaceable unit dividing method considering development and maintenance |
| CN110570117A (en) * | 2019-09-04 | 2019-12-13 | 湖南联诚轨道装备有限公司 | Reliability allocation method |
| CN114610645A (en) * | 2022-03-30 | 2022-06-10 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Task reliability and testability joint determination method and device and computer equipment |
| CN114610645B (en) * | 2022-03-30 | 2022-12-23 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Task reliability and testability joint determination method and device and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109829551A (en) | A kind of testability allocation method based on weighted difference distribution | |
| CN101980225B (en) | Method for implementing testability analysis and diagnosis decision system for electronic products | |
| Charrad et al. | NbClust: an R package for determining the relevant number of clusters in a data set | |
| Strader et al. | Globular cluster metallicity subpopulations in NGC 4472 | |
| CN108898311A (en) | A kind of data quality checking method towards intelligent distribution network repairing dispatching platform | |
| CN108489717B (en) | On-load tap changers of transformers machine performance monitoring method and system | |
| Gertler et al. | Detection and diagnosis of plant failures: the orthogonal parity equation approach | |
| CN103886518A (en) | Early warning method for voltage sag based on electric energy quality data mining at monitoring point | |
| CN110348665A (en) | A kind of low-voltage platform area electric power system data quality evaluating method and device | |
| CN111832955B (en) | Contact network state evaluation method based on reliability and multivariate statistics | |
| CN105160149A (en) | Method for constructing demand response scheduling evaluation system of simulated peak-shaving unit | |
| Van Horn et al. | Measurement-based real-time security-constrained economic dispatch | |
| CN109378823A (en) | A kind of comprehensive estimation method of voltage dip level | |
| CN109272190A (en) | A kind of power cable state evaluation method based on momentum BP neural network | |
| CN114926015B (en) | Intelligent electric energy meter quality state evaluation method and system based on D-S evidence theory | |
| CN107300907A (en) | With reference to the flight control system Reliable Evaluating Methods of Their Performance of comprehensive assessment and hypothesis testing | |
| CN1318155C (en) | Method for improving rolling mill thickness control accuracy using data redundance | |
| CN111797530B (en) | Method, device, equipment and medium for determining weak links of power system | |
| CN108427778A (en) | The testable parameter method and device of electronics | |
| CN106342306B (en) | Product test index selection method in undetected situation | |
| CN109840314A (en) | A kind of equipment Maintenance Cost evaluation method | |
| Koya et al. | Terms of trade and the exchange rates of New Zealand and Australia | |
| Jia et al. | The research and application of general item bank automatic test paper generation based on improved genetic algorithms | |
| Hill et al. | Do existing damage scales meet the needs of seismic loss estimation | |
| CN106919800A (en) | The conflicting evidence management method of low similarity collision in a kind of evidence theory |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190531 |
|
| WD01 | Invention patent application deemed withdrawn after publication |