CN103150417B - Weapon equipment complete machine measurement technique index determination method - Google Patents
Weapon equipment complete machine measurement technique index determination method Download PDFInfo
- Publication number
- CN103150417B CN103150417B CN201310021107.XA CN201310021107A CN103150417B CN 103150417 B CN103150417 B CN 103150417B CN 201310021107 A CN201310021107 A CN 201310021107A CN 103150417 B CN103150417 B CN 103150417B
- Authority
- CN
- China
- Prior art keywords
- complete machine
- machine measurement
- index
- measuring process
- measurement technology
- 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
Landscapes
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention discloses a kind of weapon equipment complete machine measurement technique index determination method, its concrete steps are as follows: when the input of armament equipment complete machine, output relation can with clear and definite mathematics or physical equation statement time, carry out the determination of complete machine measurement technology index by determinacy measuring process; When the input of armament equipment complete machine, output relation can not with clear and definite mathematics or physical equation statement time, carry out the determination of complete machine measurement technology index by uncertain measuring process.Beneficial effect of the present invention is as follows: the complete machine measurement technology index that can be used for checked object is determined, have simple and convenient, counting yield is high, good stability, order of accuarcy advantages of higher, solve a determination difficult problem for complete machine measurement technology index, the particularly local derviation of the large complicated test macro complete machine index of "black box" model and a Calculation of Sensitivity difficult problem, ingeniously avoids the immesurable difficult problem of each test intermediate link, and the robotization completing checked object global error or uncertainty calculates.
Description
Technical field
The present invention relates to equipment complete machine field of measuring techniques.
Background technology
Its technical index is the synthetic parameters of the whole signal transmission passage of system, and it depends primarily on the technical indicator of test resource itself, simultaneously relevant with the array interface in transmission channel, switch, adapter, connector and tested object.The impact of usual binding signal transmitting procedure, signals transmission provides modified value and carries out uncertainty evaluation, requires to obtain equipment complete machine index according to the evaluation of uncertainty of measurement.
For armament equipment metering, uncertainty will consider a lot of link.Metering signal will inevitably pass through some connecting lines or switch, affects to metric results; May there is electromagnetic interference (EMI) in environment residing during metering, can have a significant impact, will take into full account the various factors affecting metric results during design to the metering of high-frequency microwave signal.In addition, when measuring weaponry, due to the reason of design itself, test Reachability question will inevitably be there is, make parameter can not realize original position metering, measurement and calibration must be carried out in the input/output port of equipment, therefore need to consider various factors and uncertainty is evaluated.
Current, device fabrication producer is relied on to the complete machine calibration of armament equipment, detailed equipment de-sign drawing and principle of work must be had, by clear and definite propagation of error mode, the local derviation of computation and measurement equation and sensitivity coefficient, but when measuring process is "black box" formula measuring process, when measuring process cannot utilize clear and definite funtcional relationship to express, said method often lost efficacy.In addition, often there is a large amount of test resources in modern testing apparatus, test resource and computer for controlling connect into an entirety by bus, all inputs, to output signal through the connection device such as switch and stube cable at attachment unit interface place, be connected with armament equipment by adapter, even if there is propagation of error process each other, but the introducing of the unit such as switch, stube cable, metering adapter, also make to measure equation accurately and be difficult to obtain.In order to make metrology and measurement result truly can reflect practical situations to surveying instrument, equipment complete machine measurement technology index must be considered, consider the impact that each signalling channel is brought into.
The method calculating uncertainty has category-A and category-B two kinds of methods, be called type A standard uncertainty with the standard uncertainty drawn the statistical study evaluation of observation row, be called type B standard uncertainty with the standard uncertainty be different from the statistical study of observation row is evaluated.Uncertainty is divided into " A " class and " B " class, is only and discusses convenient, and do not mean that the difference existed between two classes evaluations in essence.Category-A uncertainty is that the probability density function of being derived by one group of frequency distribution observing obtaining draws; Category-B uncertainty is then the trusting degree based on occurring an event.They are all based on probability distribution, and all characterize by variance or standard deviation.There is not that class problem comparatively reliably in two class uncertainties.In general, category-A is comparatively more objective than category-B, and has severity statistically.The independence measured, whether be in the reliability that state in cont rol and pendulous frequency determine category-A uncertainty; Type B evaluation need rely on abundant equipment prior imformation, and due to generally, the prior imformation of individual module easily obtains, but after module composition complete machine, the priori of complete machine often cannot be known; Therefore, general type A evaluation method carries out uncertain evaluation, because type A evaluation needs the test sample book of a large amount of probability distribution corresponding to measuring process, must seek the random sample generation method under condition of small sample.
Summary of the invention
The technical problem to be solved in the present invention is to provide a kind of weapon equipment complete machine measurement technique index determination method, the complete machine measurement technology index that can be used for checked object is determined, have simple and convenient, counting yield is high, good stability, order of accuarcy advantages of higher, solve a determination difficult problem for complete machine measurement technology index, the particularly local derviation of the large complicated test macro complete machine index of "black box" model and a Calculation of Sensitivity difficult problem, ingeniously avoid the immesurable difficult problem of each test intermediate link, the robotization completing checked object global error or uncertainty calculates.
For solving the problems of the technologies described above, the technical solution used in the present invention is: a kind of weapon equipment complete machine measurement technique index determination method, and its method step is as follows:
(1) when the input of armament equipment complete machine, output relation can with clear and definite mathematics or physical equation statement time, carry out the determination of complete machine measurement technology index by determinacy measuring process; When the input of armament equipment complete machine, output relation can not with clear and definite mathematics or physical equation statement time, carry out the determination of complete machine measurement technology index by uncertain measuring process;
(2) refer to that calibration method is as follows by uncertain measuring process determination complete machine measurement technology:
1. utilize Simulation of Neural Network modeling technique, complete the measurement equation Modeling between complete machine input variable and output variable;
2. the probability density function of information Entropy Method to input variable is utilized to estimate;
3. utilize pseudorandom sample technique to generate the random sample of q.s, in conjunction with measurement equation Modeling, the complete machine measurement technology index completing uncertain measuring process is estimated.
Determinacy measuring process described in step (1) is divided into indirect inspection process and direct measuring process two class, for direct measuring process, utilizes category-B uncertainty method to export complete machine measurement technology index; For indirect inspection process, utilize indirect inspection equation to calculate local derviation and sensitivity coefficient, thus export complete machine measurement technology index.
Step 2. in utilize the probability density function of information Entropy Method to input variable to estimate concrete grammar as follows:
A, input variable flexibility transformed to [0,1] interval;
B, utilize sample information entropy to calculate each rank square of input variable sample, set up the functional relation between characteristic parameter and each rank square;
C, by arranging discrimination threshold, setting up residual sequence, using non-linear two multiplication algorithms, realizing the ART network of probability density function.
Beneficial effect of the present invention is as follows:
Randomization sample method is adopted in the present invention, checked object each several part technical indicator being comprehensively mapped to can test port, develop complete machine measurement technology Index, the complete machine measurement technology index that can be used for checked object by the method is determined, have simple and convenient, counting yield is high, good stability, order of accuarcy advantages of higher, solve a determination difficult problem for complete machine measurement technology index, the particularly local derviation of the large complicated test macro complete machine index of "black box" model and a Calculation of Sensitivity difficult problem, ingeniously avoid the immesurable difficult problem of each test intermediate link, the robotization completing checked object global error or uncertainty calculates, the method also can be used for the R&D process of complicated testing apparatus or system, reduces the complete machine index deterministic process of very complicated, greatly shortens the construction cycle of project.
Embodiment
Embodiment
When measuring a kind of complete machine measurement technology index of armament equipment, first judge that it inputs, can output relation with clear and definite mathematics or physical equation statement, when stating, carry out the determination of complete machine measurement technology index by determinacy measuring process; When not stating, carry out the determination of its technical index by uncertain measuring process.
1, the concrete grammar of determinacy measuring process is as follows:
Utilize limited input variable sample
with output variable sample
, measuring equation is:
;
(1) synthesizing uncertainty of measurement is:
(2) wherein
for the uncertainty of each component,
for the related coefficient of input quantity, undertaken evaluating by the category-A of determinacy measuring process or type B evaluation method, in GJB3756 " expression of uncertainty of measurement and evaluation ", have detailed introduction.
2, the concrete steps of uncertain measuring process are as follows:
(1) limited input variable sample is utilized
with output variable sample
, use information neural network theoretical, equation measured by the "black box" set up between constrained input:
;
(2) information Entropy Method is used to estimate
in each component
probability density function:
;
(3) rejection technique is utilized, to each component probability density function estimating gained
carry out pseudorandom sampling, obtain a large amount of pseudo-random patterns
;
(4) by the sample of pseudorandom sampling acquisition
, substitute into the "black box" measuring process set up, obtain a large amount of output samples
sample estimates
;
(5) estimation average is generated
; Type A evaluation method is utilized to evaluate output variable
uncertainty
,
with
as system tractor parameter
overall measurement technology index.
Information Entropy Method is utilized to carry out the method for PDF estimation in above-mentioned steps (2) as follows:
In order to avoid matching Problem of Failure, utilize between the field of definition of change of variable method converted variable x to [0,1].
If when stochastic variable x is changed to z, corresponding interval transforms to interval [c, d] from [a, b]
[0,1], arranges conversion parameter
with
, then the relation between new and old variable has:
；
By setting up the maximum entropy of input variable
, the probability density function analytic expression that can obtain variable x and z is:
, wherein
;
, wherein
;
Utilize the relational expression of infinitely small event establishment Two Variables probability function, namely exist
in
event should be with
probability of happening is equal simultaneously for interior z event, therefore:
after above formula is changed, can obtain:
Therefore can obtain:
Utilize binomial expansion, can obtain:
Coefficient of comparisons is equal to be obtained:
，
；
By improving the probability density function approximating method of maximum entropy method, the stochastic variable in Multilayer networks being transformed between [0,1], efficiently solving the problem that model overflows risk, expanding the algorithm scope of application.
Probability density function analytic expression method of estimation is as follows:
Utilize analytic expression
and constraint condition
can obtain:
above two formulas pair
differential, can obtain:
in formula
be
rank initial point distance, relatively above two formulas can obtain:
Set up by above formula and solve
's
individual system of equations, obtains
after, utilize formula
solve
, thus obtain variable
the characteristic parameter of probability density function
, for the ease of solving, order
Wherein ri is residual error, and available values computing method make it close to zero, solve residual sum of squares (RSS) with nonlinear least square method
minimum value, thus obtain characteristic parameter
.
Above-mentioned whole process can use software matlab+VC hybrid programming to realize, sample data input and interactive click mode is adopted to operate, main point of two parts composition, Part I is that the complete machine measurement technology index of determinacy measuring process is determined, this part is subdivided into the index estimation module of direct measuring process and the index estimation module of indirect inspection process, and wherein the index estimation module of straight directly measuring process utilizes the uncertainty of input variable
carry out gentle synthesis; The index estimation module of indirect inspection process utilizes the uncertainty of input variable
, measurement functions equation, measure equation local derviation and, input variable related coefficient etc., complete complete machine index robotization calculate.Part II is that the complete machine measurement technology index of uncertain course is determined, divided data typing module, input-output equation training module, input amendment PDF estimation module, input variable pseudorandom decimation blocks, output sample generation module, output variable Composite Seismogram estimation module form, each module organically combines, and the robotization completing complete machine index calculates.
The implementation of concrete software program is as follows:
Whether select is uncertain measuring process, if, start the complete machine measurement technology index determination module of uncertain measuring process, typing input amendment X and output sample Y, start the probability estimate module of neural metwork training model and X, obtain black box measuring process Y=f(X) and probability density function p (x) of input variable, utilize probability density function p (x) that step 1-4 estimates, carry out the desampling of input amendment, obtain the random sample of a large amount of input variables, substitute in step 1-4 and obtain measuring process equation Y=f(X), obtain a large amount of output random samples, the average of metering output variable sample Y
, utilize type B evaluation method, evaluation
, export average
and uncertainty
as complete machine measurement technology index, EOP (end of program).
If not uncertain measuring process, start the complete machine measurement technology index determination module of determinacy measuring process, typing input amendment X and output sample Y, select whether directly to measure, computation of mean values
and uncertainty
, export average
and uncertainty
as complete machine measurement technology index, EOP (end of program);
If not directly measure, select indirect inspection assessment module, computation of mean values
and uncertainty
, export average
and uncertainty
as complete machine measurement technology index, EOP (end of program).
The program process of the probability estimate algorithm based on information Entropy Method in above-mentioned uncertain measuring process is as follows:
Input variable and bound, enter step 3, and writing input variable x, to the variable z on interval [0,1], arranges the characteristic parameter of variable z probability density function
initial value, calculate each rank square of variable z
, calculate the characteristic parameter of variable z probability density function
; Set up residual error series
, threshold value is set
, call non-linear square law; Judge whether residual sum of squares (RSS) is less than and threshold value is set
if satisfied being less than arranges threshold value to residual sum of squares (RSS)
time, jump procedure 5 recalculates; If meet, export
and calculate
; The transformation of variable of realization character ginseng obtains
; Output estimation probability density function
, algorithm terminates.
Claims (2)
1. a weapon equipment complete machine measurement technique index determination method, is characterized in that, its method step is as follows:
(1) when the input of armament equipment complete machine, output relation can with clear and definite mathematics or physical equation statement time, carry out the determination of complete machine measurement technology index by determinacy measuring process; When the input of armament equipment complete machine, output relation can not with clear and definite mathematics or physical equation statement time, carry out the determination of complete machine measurement technology index by uncertain measuring process;
(2) refer to that calibration method is as follows by uncertain measuring process determination complete machine measurement technology:
utilize Simulation of Neural Network modeling technique, complete the measurement equation Modeling between complete machine input variable and output variable;
utilize the probability density function of information Entropy Method to input variable to estimate, its concrete grammar is as follows:
A, input variable flexibility transformed to [0,1] interval;
B, utilize sample information entropy to calculate each rank square of input variable sample, set up the functional relation between characteristic parameter and each rank square;
C, by arranging discrimination threshold, setting up residual sequence, using non-linear two multiplication algorithms, realizing the ART network of probability density function;
utilize pseudorandom sample technique to generate the random sample of q.s, in conjunction with measurement equation Modeling, the complete machine measurement technology index completing uncertain measuring process is estimated.
2. weapon equipment complete machine measurement technique index determination method according to claim 1, it is characterized in that, determinacy measuring process described in step (1) is divided into indirect inspection process and direct measuring process two class, for direct measuring process, category-B uncertainty method is utilized to export complete machine measurement technology index; For indirect inspection process, utilize indirect inspection equation to calculate local derviation and sensitivity coefficient, thus export complete machine measurement technology index.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310021107.XA CN103150417B (en) | 2013-01-21 | 2013-01-21 | Weapon equipment complete machine measurement technique index determination method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310021107.XA CN103150417B (en) | 2013-01-21 | 2013-01-21 | Weapon equipment complete machine measurement technique index determination method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103150417A CN103150417A (en) | 2013-06-12 |
| CN103150417B true CN103150417B (en) | 2016-01-13 |
Family
ID=48548493
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310021107.XA Active CN103150417B (en) | 2013-01-21 | 2013-01-21 | Weapon equipment complete machine measurement technique index determination method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103150417B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101520652A (en) * | 2009-03-03 | 2009-09-02 | 华中科技大学 | Method for evaluating service reliability of numerical control equipment |
| CN101718634A (en) * | 2009-11-20 | 2010-06-02 | 西安交通大学 | Equipment state comprehensive dynamic alarming method based on multivariate probability model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101678246B1 (en) * | 2009-06-25 | 2016-11-21 | 아사히 가라스 가부시키가이샤 | Calculation method for physical value, numerical analysis method, calculation program for physical value, numerical analysis program, calculation device for physical value, and numerical analysis device |
-
2013
- 2013-01-21 CN CN201310021107.XA patent/CN103150417B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101520652A (en) * | 2009-03-03 | 2009-09-02 | 华中科技大学 | Method for evaluating service reliability of numerical control equipment |
| CN101718634A (en) * | 2009-11-20 | 2010-06-02 | 西安交通大学 | Equipment state comprehensive dynamic alarming method based on multivariate probability model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103150417A (en) | 2013-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106529090B (en) | A kind of aerospace electron class Reliability Assessment method | |
| CN106291381B (en) | A kind of method of Combined estimator electrokinetic cell system state-of-charge and health status | |
| CN105572499B (en) | A kind of eye pattern generation method based on vector network analyzer | |
| CN108445759B (en) | Random fault detection method for networked system under sensor saturation constraint | |
| CN105353226A (en) | EMI noise source impedance equivalent parameter extraction method based on scattering parameter and intelligent algorithm | |
| Huang et al. | Measurement and characterization of channel delays for broadband power line communications | |
| CN110296833B (en) | Soft measurement method and system for hydraulic cylinder comprehensive test board | |
| CN110907763A (en) | Power cable fault detection method based on time-frequency domain reflection method | |
| CN113438061B (en) | Measuring and calculating method for statistical eye pattern, eye pattern analysis device and storage medium | |
| CN109978259A (en) | Method and system for predicting residual life of product based on q-type Weibull distribution | |
| CN114218778A (en) | Method and device for analyzing sonic boom test data | |
| CN113792438B (en) | Method and device for evaluating electromagnetic anti-interference performance of whole vehicle | |
| CN110889207A (en) | System combination model credibility intelligent evaluation method based on deep learning | |
| CN105891701B (en) | A kind of score filtering wave by prolonging time device delay performance test method and its test device | |
| CN104297711A (en) | Uncertainty analysis method for vector network analyzer | |
| CN101689836A (en) | The method and apparatus that is used for the linearizing non-linear power amplifier | |
| CN116520095B (en) | Fault location method, system and computer readable storage medium | |
| CN103150417B (en) | Weapon equipment complete machine measurement technique index determination method | |
| Angrisani et al. | New proposal for uncertainty evaluation in indirect measurements | |
| CN108107392A (en) | Multi-thread TRL calibration methods and terminal device | |
| Galvão et al. | Identification of fractional-order transfer functions using exponentially modulated signals with arbitrary excitation waveforms | |
| US9673862B1 (en) | System and method of analyzing crosstalk without measuring aggressor signal | |
| CN105656453B (en) | A kind of optical fiber current mutual inductor random noise Real-Time Filtering method based on time series | |
| CN111695501B (en) | Equipment soft fault detection method based on operating system kernel calling data | |
| CN108376123A (en) | A kind of satellite gravity observing data outliers extracting method, device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |