CN105044759B - A kind of state estimation of digital nuclear detector is with ensureing maintaining method and system - Google Patents
A kind of state estimation of digital nuclear detector is with ensureing maintaining method and system Download PDFInfo
- Publication number
- CN105044759B CN105044759B CN201510458022.7A CN201510458022A CN105044759B CN 105044759 B CN105044759 B CN 105044759B CN 201510458022 A CN201510458022 A CN 201510458022A CN 105044759 B CN105044759 B CN 105044759B
- Authority
- CN
- China
- Prior art keywords
- detector
- characteristic parameter
- performance
- parameter
- performance characteristic
- 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
Abstract
The invention discloses a kind of state estimation of digital nuclear detector with ensureing maintaining method and system.Can be in the case where not increasing extra hardware circuit using the present invention, give the assessment of nuclear detector and ensure the quantizating index safeguarded, on-line monitoring can be realized and make effective expection, use and the maintenance cost of nuclear detector are reduced, lifts the reliability, stability and availability of nuclear detector.The present invention sets the type attribute parameter of detector first, determines the performance characteristic parameter to be assessed of detector, and the collection data related to performance characteristic parameter to be assessed simultaneously generate list type data；Then the current performance characteristic parameter of nuclear detector is calculated using list type data；Aid decision is safeguarded then in conjunction with historical data, the current performance level and its variation tendency of analysis and evaluation detector, the quantizating index of acquisition current health state, the guarantee that detector is finally provided according to quantizating index state.
Description
Technical field
The present invention relates to radiation detection and signal processing technology field, and in particular to a kind of state of digital nuclear detector
Assess with ensureing maintaining method and system.
Background technology
The either radiation detection such as energy disperse spectroscopy, radiation counter field, or positron life spectrometer, Mossbauer spectrometer
Isobolographic analysis field, and or computer tomography (Computed Tomography, abbreviation CT), single photon emission break
Layer imaging (Single Positron Emission Computed Tomography, abbreviation SPECT) and positron emission
The nucleus medical image fields such as fault imaging (Positron Emission Tomography, abbreviation PET), its detection principle is all
It is that high-energy photon is converted to by measurable pulse electrical signal by detector detecting unit, then recycles its electronics unit to carry
The characteristic point lain in pulse signal is taken, and then obtains the information for meeting different application environmental demand.
Critical component of the nuclear detector as above-mentioned nuclear instrument equipment, is the source for obtaining various radiation informations, its performance
Quality directly influence the measurement accuracy of nuclear instrument equipment.Because nuclear detector output signal-noise ratio is low, background clutter is strong,
Particularly under high temperature, high pressure, intense radiation, strong motion environment, its performance can have greatly changed, so as to cause nuclear instrument to set
Standby measurement result serious distortion, or even instrument and equipment is broken down and be stopped.For example, during logging operation is radiated,
Underground high temperature, high humidity, strong motion environment, nuclear detector coupling is deteriorated, cause output pulse electrical signal amplitude to die down,
The Ping Qu of photoelectron-multiplier-tube narrows and moved forward, and crystal deliquescence causes energy resolution variation etc., and these phenomenons can all make
The degradations such as the uniformity of nuclear instrument equipment, stability, measurement accuracy.Therefore, real-time tracking and monitoring nuclear detector performance
Change, make correct assessment for nuclear detector working condition, effective processing maintenance measure taken in good time, to ensureing core
The stable normal work of instrument and equipment, reduce cost of equipment maintenance, ensure that nuclear safety has a very important role.
At present, in order to ensure the reliability of nuclear instrument equipment, maintenanceability and security, such as nuclear power unit monitoring system
System, generally investigates some obvious physical damages in a manner of manual inspection, but in nuclear detector failure can not then detect
Arrive, and generally there is higher amount of radiation the erecting bed of most nuclear detectors, it is difficult timely and effective hair to depend merely on manual inspection mode
Now with repair these instrument failures, while patrol officer is also highly prone to the danger of production scene intense radiation and toxic and harmful gas
Evil.In addition, correctly screening failure mode, trouble point and the extent of damage of nuclear instrument equipment, it is rich to often require that patrol officer has
Rich experience, be otherwise likely to occur fail to judge, misjudge, maloperation, it is impossible to monitored on-line.
Therefore, Chen Luning et al. proposes trouble-shooter (Chen Luning, Li Zhen, the Qin of a kind of nuclear radiation monitoring system
Triumphant trouble-shooter [p] the Chinese invention patents for waiting a kind of nuclear radiation monitoring systems of of son, Patent publication No
CN103336209.A), the device includes simulation radiation detector and failure diagnostic apparatus, wherein simulation radiation detector is used for pair
Radiation monitoring system signal processing apparatus carries out fault diagnosis, and failure diagnostic apparatus is used to carry out failure point to true radiation detector
Analysis.Although the device can fast positioning failure occur point, on the one hand need to increase extra hardware circuit, as power supply turn
Change circuit, circuit, current occuring circuit, analog quantity sample circuit occur for high-voltage pulse and low voltage pulse signal, on the other hand without
Method realizes on-line monitoring and the assessment of detector working condition.
In addition, face support the army et al. it is also proposed nuclear detector method for diagnosing faults and device (face is supported the army, Zhou Jianliang, king
The method for diagnosing faults and device [P] Chinese invention patents of the nuclear detectors such as celebrating shake, Patent publication No
CN103983881.A), but this method is on the one hand there is still a need for increase new hardware device, such as D/A converter, height piezoelectricity
The additional circuits such as detection module, reference voltage module, variable connector, A/D converter are pressed, undoubtedly add the maintenance of nuclear detector
Cost；Although another aspect this method is connected by trouble-shooter with radiation detector, adopted online by trouble-shooter
Collect nuclear radiation detector shaping circuit, Simulation scale-up circuit and core probe signal data, and by analyzing the system of detector signal
The difference of index, time-frequency domain index and Mishap Database is counted, certain breakdown detector exception can be detected, but due to lacking
In-depth analysis to detector intrinsic characteristics information, effective expection can not be made to the variation tendency of nuclear detector state, and
And fault diagnosis precision is not high.
The content of the invention
In view of this, the invention provides a kind of state estimation of digital nuclear detector with ensureing maintaining method and being
System, in the case where extra hardware circuit need not be increased, it is possible to increase the precision of on-line monitoring simultaneously makes effective expection,
Solve the problems, such as quantitative evaluation and formulate optimal nuclear detector guarantee maintenance measures, reduce the use and maintenance of nuclear instrument equipment
Contribute to reliability, stability and the availability of lifting nuclear instrument equipment while cost.
The state estimation of the digital nuclear detector of the present invention comprises the following steps with ensureing maintaining method：
Step 1: set first according to the application field of digital nuclear detector to be assessed and its detection target component
The type attribute parameter of detector；Then the performance characteristic to be assessed of detector is determined according to the type attribute parameter of detector
Parameter；Then the data related to performance characteristic parameter to be assessed are gathered and generate list type data；Wherein, the type category
Property parameter mainly including the application field of nuclear detector, model, the data type that can be collected, data frame structure and data at
Reason method；The list type data mainly include the time, energy, position, numbering, one or more groups of working environment parameter
Close；
Step 2: the list type data gathered using step 1, the current performance that digital nuclear detector is calculated is special
Levy parameter；
Step 3: with reference to the historical performance characteristic parameter and current performance characteristic parameter that are stored in historical data base, assess
The current performance level of detector, and the variation tendency of assessment performance characteristic parameter；By the current performance level of detector and change
Quantizating index of the change trend as detector current health state；Wherein, historical data base dispatches from the factory comprising digital nuclear detector
When default performance characteristic parameter data set and all previous state estimation with ensureing the property of health status generated in maintenance process
Can characteristic parameter data set；
Step 4: check the quantizating index of the current health state for the detector that step 3 obtains whether in normal, school
Just, failure or in the range of changing, correspondingly make and do not change, correct detector guarantor with calibration, breakdown maintenance or part replacement
Barrier safeguards aid decision.
Further, the step 3 specifically includes following sub-step：
Step 3.1, for historical performance characteristic parameter, with the distributed model of statistical method acquisition performance characteristic parameter
And the relevant parameter of distributed model；
Step 3.2, the current performance characteristic parameter obtained with reference to step 2, and the performance characteristic ginseng that step 3.1 obtains
Number distributed model, assess the current performance level of detector；
Step 3.3, the current performance characteristic parameter obtained with reference to step 2, and historical performance characteristic parameter, assessment obtain
Obtain the variation tendency of performance characteristic parameter；
Step 3.4, the performance characteristic parameter that the current performance level for the detector that step 3.2 obtains and step 3.3 obtain
Variation tendency combination as detector current health state quantizating index.
Further, in the step 3.1, using Gauss model as performance characteristic parameter distribution model, history is utilized
Performance characteristic parameter calculates the covariance matrix Σ and mean vector μ for obtaining Gauss model；
In the step 3.2, using the current performance level of relative error method calculating detector, i.e. current performance state water
Flat E is
Wherein, X is the current performance characteristic parameter that step 2 obtains；
In the step 3.3, the variation tendency value P of performance characteristic parameter is calculated using Support vector regression algorithm, its
In, Selection of kernel function RBF；
In the step 3.4, the quantizating index of current health state is detector performance probability density Σ, μ, current performance
Horizontal E and variation tendency P, i.e. [Σ, μ, E, P].
Further, in the step 4, guarantee is formed using traditional decision-tree and safeguards aid decision, is specifically included as follows
Sub-step：
Step 4.1, detector current performance level is assessed whether in normal performance range, that is, is judgedWhether
More than 0, show that detector current performance exceeds normal performance range if greater than 0, perform step 4.3；Otherwise detector is shown
Current performance belongs in normal performance range, performs step 4.2；Wherein, E is the current performance level that step 3 obtains；A is tune
Parameter is saved, is rule of thumb obtained by the data analysis of reality；Σ is covariance matrix；μ is average；
Step 4.2, the variation tendency of detector characteristic parameter is assessed, that is, is judgedWhether 0 is more than,
Show that detector needs to be corrected if greater than 0, perform step 4.4；Otherwise detector is in health status, by current performance
In characteristic parameter deposit historical data base, assessment terminates；Wherein, d, e are adjustment parameter, rule of thumb by the data analysis of reality
Obtain；P is the variation tendency vector for the performance characteristic parameter that step 3 obtains；
Step 4.3, detector performance deterioration is assessed, that is, is judgedWhether 0 is more than, if greater than 0
Show that detector needs to be repaired, perform step 4.4；Otherwise detector needs to be changed, and performs step 4.4；Wherein, b,
C is adjustment parameter, is rule of thumb obtained by the data analysis of reality；
Step 4.4, the correlation module traced back to by detector performance characteristics parameter in detector electronics unit, to correlation
Module takes corresponding correction, and either breakdown maintenance or part replacement operate with calibration, and assessment terminates.
A kind of state estimation of digital nuclear detector and guarantee maintenance system, including：Type attribute parameter setting module,
Data acquisition, transmission and memory module, performance characteristic parameter calculating module, health status comprehensive assessment module and guarantee maintenance are determined
Plan module；
Wherein, type attribute parameter setting module is used for identified input or manually inputs digital core spy to be assessed
Survey the type attribute parameter of device；
Data acquisition, transmission and memory module according to the type attribute parameter set in type attribute parameter setting module,
Collection data related to the detector performance characteristics parameter, and by the data generation list type data of collection, and by the list
Formula data storage is simultaneously transmitted to performance characteristic parameter calculating module；
Performance characteristic parameter calculating module is according to the type attribute parameter set in type attribute parameter setting module, logarithm
Parsed according to the list type data of collection, transmission and memory module transmission, the digital nuclear detection to be assessed is calculated
The current properties characteristic parameter of device, and current performance characteristic parameter is transmitted to health status comprehensive assessment module；
Historical performance characteristic parameter and performance characteristic ginseng of the health status comprehensive assessment module in historical data base
The current performance characteristic parameter of number computing module transmission, calculates the current performance level and performance change trend for obtaining detector；
And the detector health status quantizating index being made up of current performance level and performance change trend is sent to guarantee maintenance and determined
Plan module；
Ensure that the current health state that maintenance measures module is sent according to detector health status comprehensive assessment module quantifies
Index, digital nuclear detector to be assessed is judged whether in the range of normal, correction, failure or replacing, if in just
It is in normal scope, then without any processing, current performance characteristic parameter is stored in historical data base, if in correction, failure or
In the range of replacing, then the correlation module that is traced back to by performance characteristic parameter in detector electronics unit is taken correlation module
Either breakdown maintenance or part replacement operate with calibration for corresponding correction.
Beneficial effect：
(1) caused every valid data extraction is digital in the course of the work in itself using digital nuclear detector by the present invention
The performance characteristic parameter of formula nuclear detector simultaneously builds historical data base, and the statistics for analyzing nuclear detector properties characteristic parameter is special
The inner link of property and performance characteristic parameter corresponding with historical data base, the health status feelings of Comprehensive Evaluation nuclear detector
Condition, the precision of on-line monitoring is improved, while make it that quantitative basis is more reasonable, confidence level is higher.
(2) extra special failure detector circuit need not be increased, it is only necessary to gather digital nuclear detector in itself in work
Caused every valid data during work, the analysis of Contemporary Digital formula nuclear detector failure can be completed with commenting using the data
Estimate.
(3) the invention provides a kind of guarantee maintenance strategy of life cycle management, user or patrol officer is aided in formulate
Scientific and reasonable calibration correction or maintenance Replacing Scheme.
(4) this method has very strong applicability, by configuring different collections and processing parameter, goes for difference
The digital nuclear detector of function and structure.
Brief description of the drawings
Fig. 1 is the state estimation of the digital nuclear detector of the present invention with ensureing maintaining method flow chart.
Fig. 2 is structured flowchart of the state estimation of the digital nuclear detector of the present invention with ensureing maintenance system.
Performance characteristic parameters of the Fig. 3 by present example using digital pet detector, (a) is digital PET detectors
Temporal resolution (b) be digital pet detector energy resolution (c) be digital pet detector position spectrum.
Fig. 4 uses aid decision decision tree analysis figure for the present invention.
Embodiment
The invention provides a kind of state estimation of digital nuclear detector with ensureing maintaining method and system, its core is thought
Think be：Under conditions of additional hardware circuit is not increased, refine what digital nuclear detector generated by concluding comprehensive and inverting
List type data, deeply excavate digital nuclear detector correlated performance characteristic parameter and join with corresponding performance feature in historical data base
Several difference, so as to obtain the quantizating index of digital nuclear detector current health state and performance of expected change trend, instruct numeral
Formula nuclear detector ensures the formulation of maintenance measures.
The present invention will now be described in detail with reference to the accompanying drawings and examples.
As shown in figure 1, a kind of state estimation of digital nuclear detector provided by the invention is specific with ensureing maintaining method
Comprise the following steps：
Step 1: digital nuclear detection is set according to the application field of digital nuclear detector and detection target component first
The type attribute parameter of device, then determines the performance characteristic parameter of digital nuclear detector according to type attribute parameter, and gathers
The data related to performance characteristic parameter, the data of collection are generated into list type data, then by list type data transfer and deposited
Store up in data analytical Calculation machine.
The digital nuclear detector refers to the pulse electrical signal characteristic information generated to detecting unit in nuclear detector
Extracting method is digital, i.e., carries out digital sample to pulse electrical signal, and pulse characteristic information is carried out with data signal
Characterize, and digital nuclear detector electronics unit (i.e. data acquisition, transmission and memory cell) design follows hardware module mark
Standardization and functional module software implementation principle.Wherein, hardware module standardization refers to that hardware circuit is divided by function, each function mould
Block uses the circuit of unified interface standard, is easy to repair, upgrades；Functional module software implementation refers to make full use of Digital Signal Processing
Technology, script is built to the function of realization with hardware circuit, realized with the mode of software programming, it is same using it is unified, can expand
The interface of exhibition.
The type attribute parameter of the digital nuclear detector is the hexadecimal number of 4 bytes, is the digital core of description
A kind of parameter sets of detector category, application field, model, the data that can be collected mainly including current nuclear detector
The relevant parameters such as type, data frame structure, data processing method, each of which bit have corresponding connotation, by digital core
The application field of detector and detection target component determination, such as：Bit0 represents detector application field, and bit1 represents model,
Bit2 represents data type etc., and this is the most basic expression way of type attribute parameter.The type attribute parameter of nuclear detector
Decide performance characteristic parameter, the extraction side of the data related to performance characteristic parameter for the nuclear detector for needing to monitor and assessing
The appraisal procedure used when method and progress state estimation.The performance characteristic parameter of detector is different according to type attribute parameter
It is and different.Such as：If digital nuclear detector to be assessed is radiation detector, its performance characteristic parameter is exactly energy resolution
Rate, counting rate, operating temperature etc., to being exactly photon energy values, timestamp, operating temperature in requisition for the data of collection；If
Pet detector, then performance characteristic parameter is exactly spatial resolution, temporal resolution, energy resolution, counting rate, in requisition for
The data of collection are exactly lattice element numbering, module numbering, timestamp, photon energy values, position.Its corresponding state estimation side
Method is just different.
The acquisition of type attribute parameter can be by way of automatic identification, i.e., by being communicated with digital nuclear detector
Obtain the type attribute parameter of detector；Also it can manually be configured, i.e., assessment system input is currently needed for locating
The type attribute parameter of the detector of reason.
The list type data of the digital nuclear detector generation include：Time, energy, position, numbering, working environment ginseng
Several one or more combinations, the type attribute parameter of the building form of list type data and digital nuclear detector are close
It is related.Such as：The list type data of energy disperse spectroscopy are the combination of energy, numbering, position, when the list type data of pet detector are
Between, the combination of position, energy, numbering, ambient parameter.
The transmission means of the list type data of the digital nuclear detector generation can determine according to actual application environment, select
Select wired or wireless transmission means.
Step 2: according to digital nuclear detector type attribute parameter to be assessed, and step 1 obtain it is digital
The list type data of nuclear detector, the performance characteristic parameter of digital nuclear detector to be assessed is calculated.
The performance characteristic parameter of the digital nuclear detector includes：Energy resolution, temporal resolution, spatial discrimination
One kind of rate, position spectrum lattice discrimination, detectivity, event count rate, particle resolution, nucleus module operating temperature etc.
Or multiple combinations.
The list type data related to performance characteristic parameter gathered using step 1 can calculate numeral to be assessed
The correlated performance characteristic parameter of formula nuclear detector, such as does that the time meets, Gauss curve fitting can be asked to the time values of list type data
Go out temporal resolution.
Step 3: the performance characteristic parameter for the digital nuclear detector to be assessed being calculated according to step 2, access are gone through
Corresponding performance characteristic parameter in history database, with statistical method and prediction algorithm to digital nuclear detector current health shape
State and performance of expected change trend carry out Comprehensive Evaluation, form quantizating index.
The historical data base is performance characteristic parameter data set default when being dispatched from the factory by digital nuclear detector, Yi Jili
Next state is assessed with ensureing that the effective performance characteristic parameter data set generated in maintenance process forms.
It is described that Comprehensive Evaluation is carried out to digital nuclear detector current health state and performance of expected change trend, refer to utilize and go through
Corresponding performance characteristic parameter in history database, using Bayesian inference, SVMs, artificial neural network, gray theory,
One or more combination creation analysis forecast system models of the multi-intelligence algorithms such as fuzzy reasoning, to digital nuclear detector
Each single item performance characteristic parameter is compared and analyzed one by one respectively, and quantization obtains the comprehensive health of Contemporary Digital formula nuclear detector
It is horizontal.
The quantization method of digital nuclear detector current health state is as follows：
Step 3.1, according to the characteristics of the performance characteristic parameter of digital nuclear detector, performance parameter corresponding to selection is distributed
Model, and estimate using historical data base the parameter of distributed model.
Such as：Selected performance parameter distributed model can be Gauss model：
Using data in historical data base, using the parameter of maximum Likelihood acquisition Gauss model, i.e. covariance
Matrix Σ and mean vector μ.
Step 3.2, the performance characteristic ginseng obtained with reference to the current performance characteristic parameter and step 3.1 of digital nuclear detector
Number distributed model, calculates Current detector performance level, that is, the current performance characteristic parameter for calculating digital nuclear detector deviates
The horizontal situation of nominal or historical performance characteristic parameter.
In this step, the current performance characteristic parameter of digital nuclear detector is estimated using the Gauss model of step 3.1
The situation horizontal from nominal or historical performance characteristic parameter (if the distributed model of selection is other kinds of model, Ye Shili
Selected distributed model is estimated with historical data base, obtains the parameter of the selected distributed model, then uses again
The distributed model estimates the deviation situation of the current performance characteristic parameter of digital nuclear detector).
Typical relative error calculating method can be used by calculating Contemporary Digital formula nuclear detector performance level, be provided currency
Energy characteristic parameter deviates the degree of benchmark.That is, the horizontal E of current performance state of digital nuclear detector is：
Wherein, X is current actual performance characteristic parameter, and mean vector μ is the numeral estimated using Gauss model
Formula nuclear detector average behavior characteristic parameter.
There are multiple performance characteristic parameters for detector, then E, X, μ are correspondingly Vector-Matrix Form, such as numeral
Formula pet detector, E=[e1,e2,e3], X=[X1,X2,X3], μ=[μ1,μ2,μ3], subscript 1,2,3 characterizes time resolution respectively
Rate, energy resolution, the performance level of position spectrum.
Step 3.3, with reference to all previous performance characteristic parameter X in historical data basen-1..., X1, and current actual property
Energy characteristic parameter X, calculates the variation tendency value for obtaining digital nuclear detector performance characteristic parameter.
Support vector regression algorithm can be used to calculate the variation tendency value of performance characteristic parameter, Selection of kernel function is radially
Basic function, obtain variation tendency the vector P, vectorial P=[p of each performance characteristic parameter1,p2,p3], table 1 below, 2,3 characterize respectively
Temporal resolution, energy resolution, position spectrum.
Step 3.4, the P combination outputs that the E and step 3.3 finally calculated step 3.2 is calculated, as detector
Current performance assesses comprehensive quantification index.
The main quantizating index of each performance characteristic parameter of above-mentioned digital P ET detectors is digital nuclear detector
Energy probability density, current performance state level and performance trend, i.e., [Σ, μ, E, P].
Step 4: the digital nuclear detector Current detector performance level quantizating index obtained according to step 3 judges to visit
The current health state of device is surveyed, and then aid decision is safeguarded in the guarantee for forming digital nuclear detector.
It is described to ensure that the aid decision safeguarded is that good, correction, maintenance are provided to each module of digital nuclear detector, is changed
Guarantee decision support.Explanation is how to form maintenance strategy by taking traditional decision-tree as an example.Traditional decision-tree flow such as Fig. 4 institutes
Show, wherein, t0、t1、t2It is as follows for the node of decision tree, idiographic flow：
Step 4.1, assess detector performance level whether in normal performance range (t0), it is as follows to assess formula
Wherein, a is adjustment parameter, is rule of thumb obtained by the data analysis of reality, here, takes a=3, work as t0Be less than or
Show that detector current performance belongs in normal performance range equal to 0, perform step 4.2；t0Show that detector is currency more than 0
Normal performance range can be exceeded, perform step 4.3.
Step 4.2, the change vector of detector characteristic parameter is assessed,
Wherein, d, e are adjustment parameter, are rule of thumb obtained by the data analysis of reality, here, take d=0.5, e=-
0.1.Work as t1Show that detector needs to be corrected more than 0, perform step 4.4；Otherwise detector's status is good, by current performance
In characteristic parameter deposit historical data base, assessment terminates.
Step 4.3, detector performance deterioration is assessed,
Wherein, b, c are adjustment parameter, are rule of thumb obtained by the data analysis of reality, here, take b=6, c=-0.3.
Work as t2Show that detector needs to be repaired more than 0, perform step 4.4, otherwise detector needs to be changed, and performs step
4.4。
Step 4.4, the correlation module traced back to by detector performance characteristics parameter in detector electronics unit, to correlation
Module is corrected or repaired or changes, and assessment terminates.As a result of hardware modularity and functional software strategy so that therefore
Hinder the rapidity of positioning, maintenanceability is substantially improved.
As shown in Fig. 2 the invention provides a kind of state estimation of digital nuclear detector with ensureing maintenance system, specifically
Including：Type attribute parameter setting module, data acquisition, transmission and memory module, performance characteristic parameter calculating module, healthy shape
State comprehensive assessment module and guarantee maintenance measures module.
Wherein, type attribute parameter setting module is used for the type attribute parameter for determining digital nuclear detector, type category
Property parameter setting module in the hexadecimal number comprising 2 bytes, each bit of hexadecimal number has corresponding connotation,
According to the application field of digital nuclear detector to be assessed and detection target, automatic identification inputs or manually inputted the spy
Survey the type attribute parameter of device.
Data acquisition, transmission and memory module according to the type attribute parameter set in type attribute parameter setting module,
The collection data related to the detector performance characteristics parameter, and the data of collection are generated into list type data, by the list type
Data storage is simultaneously transmitted to performance characteristic parameter calculating module.
Performance characteristic parameter calculating module is according to the type attribute parameter set in type attribute parameter setting module, logarithm
Parsed according to the list type data of collection, transmission and memory module transmission, the digital nuclear detector to be assessed is calculated
Current properties characteristic parameter, and current performance characteristic parameter is transmitted to health status comprehensive assessment module.
Historical performance characteristic parameter and performance characteristic ginseng of the health status comprehensive assessment module in historical data base
The current performance characteristic parameter of number computing module transmission, the working condition current to digital nuclear detector and variation tendency are carried out
Synthetic determination, user or patrol officer is set to accurately hold, quickly know the current health status of detector.
Health status comprehensive assessment module includes 4 submodules, is respectively：Performance characteristic parameters history data retrieval submodule
Block, detector performance analysis submodule, detector health forecast submodule and detector health status quantify submodule.The property
Energy characteristic parameter historical data retrieval submodule utilizes all previous performance characteristic parameter stored in historical data base, builds detector
Performance parameter distributed model；The performance of the detector performance analysis submodule binding ability characteristic parameter history sub-module structure
The current performance characteristic parameter of characteristic parameter distributed model and detector, calculate current performance characteristic parameter and deviate nominal or history
The situation of performance characteristic parameter level；All previous performance characteristic in the detector health forecast submodule combination historical data base
Parameter and current performance characteristic parameter, detector performance variation model is built, predict the change of each performance characteristic parameter of detector
Trend；The detector health status quantifies the current performance feature ginseng that submodule combination detector performance analysis submodule obtains
Number deviation situation, and the performance characteristic parameter variation tendency that detector health forecast submodule obtains, provide synthetic determination, shape
Into specific quantizating index, and quantizating index is sent to and ensures maintenance measures module.
Ensure the quantizating index that is obtained according to detector health status comprehensive assessment module of maintenance measures module, be user or
Patrol officer provides scientific and reasonable guarantee maintenance strategy.Ensure that maintenance measures module is divided into 4 submodules, be respectively：Health
State decision-making submodule, correction and calibration submodule, breakdown maintenance decision-making submodule and detector change decision-making submodule.Work as spy
The quantizating index (health status of detector) that survey device health status comprehensive assessment module obtains performs strong in normal range (NR)
Health state decision-making submodule, i.e., without be modified to the state of the art of Contemporary Digital formula nuclear detector, and by the detector
Current performance characteristic parameter is stored into historical data base, participates in follow-up nuclear detector state estimation；When quantizating index is in school
In positive scope, correction and calibration submodule are performed, that is, needs to carry out necessary correction and calibration operation to nuclear detector；Work as quantization
Index performs breakdown maintenance decision-making submodule in fault coverage, that is, needs to carry out necessary trouble hunting behaviour to nuclear detector
Make；When quantizating index is in the range of field-replaceable unit, execution detector changes decision-making submodule, that is, needs to digital detector
Part carries out replacement operation.When performing correction with calibration, trouble hunting, part replacement operation, first by bad performance characteristic
The correlation module that parameter is traced back in detector electronics unit, is then corrected or repairs to correlation module or change.
Pass through state estimation of the specific embodiment data to the digital nuclear detector of the present invention and guarantee again below
Method for maintaining and system do further checking, wherein, it is related to some parameters, these parameters need to be directed to specific processing data and carry out
The parameter of this application instance processes data is listed below to reach good performance in regulation：
The digital nuclear detector used in step 1 belongs to nucleus medical image field, specially positron emission fault into
As equipment, its detector, in γ photons caused by the positron-emitting radionuclides decay on molecular probe, is obtained by snoop tag
Take its energy, position and temporal information.
The list type data format of digital nuclear detector generation is by event time, energy, position, numbering, failure tolerant
Flag bit forms；
In step 2 the performance characteristic parameter of digital nuclear detector be temporal resolution, energy resolution, position spectrum, such as
Shown in Fig. 3.
The horizontal E of the current performance state of digital nuclear detector in step 3, main quantizating index are detector performance
Probability density, detector current performance and detector performance trend, i.e. [Σ, μ, E, P].
Step 4 is analyzed using aid decision using decision tree.Due to the current comprehensive health of digital nuclear detector
State is in good range, therefore without being modified to the state of the art of Contemporary Digital formula nuclear detector.
In summary, presently preferred embodiments of the present invention is these are only, is not intended to limit the scope of the present invention.
Within the spirit and principles of the invention, any modification, equivalent substitution and improvements made etc., it should be included in the present invention's
Within protection domain.
Claims (5)
1. a kind of state estimation of digital nuclear detector is with ensureing maintaining method, it is characterised in that comprises the following steps：
Step 1: detection is set according to the application field of digital nuclear detector to be assessed and its detection target component first
The type attribute parameter of device；Then the performance characteristic to be assessed for determining detector according to the type attribute parameter of detector is joined
Number；Then the data related to performance characteristic parameter to be assessed are gathered and generate list type data；Wherein, the type attribute
Parameter mainly includes application field, model, the data type that can be collected, data frame structure and the data processing of nuclear detector
Method；The list type data mainly include one or more combinations of time, energy, position, numbering, working environment parameter；
Step 2: the list type data gathered using step 1, the current performance feature that digital nuclear detector is calculated is joined
Number；
Step 3: with reference to the historical performance characteristic parameter and current performance characteristic parameter that are stored in historical data base, detection is assessed
The current performance level of device, and the variation tendency of assessment performance characteristic parameter；The current performance level of detector and change are become
Quantizating index of the gesture as detector current health state；Wherein, lacked when historical data base dispatches from the factory comprising digital nuclear detector
The performance characteristic parameter data set and all previous state estimation of province and the performance for ensureing the health status generated in maintenance process are special
Levy parameter data set；
Step 4: check the quantizating index of the current health state for the detector that step 3 obtains whether in normal, correction, event
In the range of barrier or replacing, correspondingly make and do not change, correct and safeguarded with the guarantee of the detector of calibration, breakdown maintenance or part replacement
Aid decision.
2. the state estimation of digital nuclear detector as claimed in claim 1 is with ensureing maintaining method, it is characterised in that described
Step 3 specifically includes following sub-step：
Step 3.1, for historical performance characteristic parameter, obtain the distributed model of performance characteristic parameter with statistical method and divide
The relevant parameter of cloth model；
Step 3.2, the current performance characteristic parameter obtained with reference to step 2, and the performance characteristic parameter point that step 3.1 obtains
Cloth model, assess the current performance level of detector；
Step 3.3, the current performance characteristic parameter obtained with reference to step 2, and historical performance characteristic parameter, assess acquired
The variation tendency of energy characteristic parameter；
Step 3.4, the change for the performance characteristic parameter that the current performance level for the detector that step 3.2 obtains and step 3.3 obtain
Quantizating index of the combination of change trend as the current health state of detector.
3. the state estimation of digital nuclear detector as claimed in claim 2 is with ensureing maintaining method, it is characterised in that described
In step 3.1, using Gauss model as performance characteristic parameter distribution model, obtained using historical performance calculation of characteristic parameters high
The covariance matrix Σ and mean vector μ of this model；
In the step 3.2, using the current performance level of relative error method calculating detector, the i.e. horizontal E of current performance state
For
Wherein, X is the current performance characteristic parameter that step 2 obtains；
In the step 3.3, the variation tendency value P of performance characteristic parameter is calculated using Support vector regression algorithm, wherein, core
Function selects RBF；
In the step 3.4, the quantizating index of current health state is covariance matrix Σ, μ, current performance level E and change
Trend P, i.e. [Σ, μ, E, P].
4. the state estimation of the digital nuclear detector as described in claim 1 or 2 is any, with ensureing maintaining method, its feature exists
In, in the step 4, using traditional decision-tree formed guarantee safeguard aid decision, specifically include following sub-step：
Step 4.1, detector current performance level is assessed whether in normal performance range, that is, is judgedWhether it is more than
0, show that detector current performance exceeds normal performance range if greater than 0, perform step 4.3；Otherwise show that detector is current
Performance belongs in normal performance range, performs step 4.2；Wherein, E is the current performance level that step 3 obtains；A joins for regulation
Number, rule of thumb obtained by the data analysis of reality；Σ is covariance matrix；μ is average；
Step 4.2, the variation tendency of detector characteristic parameter is assessed, that is, is judgedWhether 0 is more than, if greatly
Show that detector needs to be corrected in 0, perform step 4.4；Otherwise detector is in health status, and current performance feature is joined
In number deposit historical data base, assessment terminates；Wherein, d, e are adjustment parameter, are rule of thumb obtained by the data analysis of reality；P
The variation tendency vector of the performance characteristic parameter obtained for step 3；
Step 4.3, detector performance deterioration is assessed, that is, is judgedWhether it is more than 0, shows if greater than 0
Detector needs to be repaired, and performs step 4.4；Otherwise detector needs to be changed, and performs step 4.4；Wherein, b, c are
Adjustment parameter, rule of thumb obtained by the data analysis of reality；
Step 4.4, the correlation module traced back to by detector performance characteristics parameter in detector electronics unit, to correlation module
Taking corresponding correction, either breakdown maintenance or part replacement operate with calibration, and assessment terminates.
5. a kind of state estimation of digital nuclear detector is with ensureing maintenance system, it is characterised in that including：Type attribute parameter
Setting module, data acquisition, transmission and memory module, performance characteristic parameter calculating module, health status comprehensive assessment module and
Ensure maintenance measures module；
Wherein, type attribute parameter setting module is used for identified input or manually inputs digital nuclear detector to be assessed
Type attribute parameter；
Data acquisition, transmission and memory module are according to the type attribute parameter set in type attribute parameter setting module, collection
Data related to the detector performance characteristics parameter, and the data of collection are generated into list type data, and by the list type number
According to storing and transmit to performance characteristic parameter calculating module；
Performance characteristic parameter calculating module is adopted according to the type attribute parameter set in type attribute parameter setting module to data
Collection, the list type data of transmission and memory module transmission are parsed, and the digital nuclear detector to be assessed is calculated
Current properties characteristic parameter, and current performance characteristic parameter is transmitted to health status comprehensive assessment module；
Historical performance characteristic parameter and performance characteristic parameter meter of the health status comprehensive assessment module in historical data base
The current performance characteristic parameter of module transfer is calculated, calculates the current performance level and performance change trend for obtaining detector；And will
The detector health status quantizating index being made up of current performance level and performance change trend is sent to guarantee maintenance measures mould
Block；
The current health state quantizating index that maintenance measures module is sent according to detector health status comprehensive assessment module is ensured,
Digital nuclear detector to be assessed is judged whether in the range of normal, correction, failure or replacing, if being in normal range (NR)
It is interior, then it is without any processing, current performance characteristic parameter is stored in historical data base, if in correction, failure or changing model
In enclosing, then the correlation module traced back to by performance characteristic parameter in detector electronics unit, correlation module is taken corresponding
Either breakdown maintenance or part replacement operate with calibration for correction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510458022.7A CN105044759B (en) | 2015-07-29 | 2015-07-29 | A kind of state estimation of digital nuclear detector is with ensureing maintaining method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510458022.7A CN105044759B (en) | 2015-07-29 | 2015-07-29 | A kind of state estimation of digital nuclear detector is with ensureing maintaining method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105044759A CN105044759A (en) | 2015-11-11 |
| CN105044759B true CN105044759B (en) | 2018-01-23 |
Family
ID=54451432
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510458022.7A Active CN105044759B (en) | 2015-07-29 | 2015-07-29 | A kind of state estimation of digital nuclear detector is with ensureing maintaining method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105044759B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105433964A (en) * | 2014-07-25 | 2016-03-30 | 上海西门子医疗器械有限公司 | X-ray device |
| CN107451133B (en) * | 2016-05-30 | 2023-09-29 | 泰康之家（北京）投资有限公司 | Method and apparatus for evaluating overall feature status of an object |
| CN106024078B (en) * | 2016-06-24 | 2017-08-25 | 西安交通大学 | A kind of method of diagnostic reaction pile neutron detector failure |
| JP2019007825A (en) * | 2017-06-23 | 2019-01-17 | 株式会社島津製作所 | Nuclear medicine diagnostic device and routine maintenance and checkup method of nuclear medicine diagnostic device |
| CN108168736B (en) * | 2017-12-27 | 2020-10-20 | 重庆工商职业学院 | High-precision temperature data correction method |
| CN108536980B (en) * | 2018-04-18 | 2021-11-12 | 中国石油大学(华东) | Gas detector discrete site selection optimization method considering reliability factor |
| CN108630330B (en) * | 2018-05-29 | 2020-03-03 | 岭东核电有限公司 | Pressurized water reactor nuclear power station instrument system detector test processing method, device and system |
| CN112230548B (en) * | 2020-10-27 | 2022-06-17 | 中国核动力研究设计院 | Nuclear power device autonomous control system |
| CN112925007B (en) * | 2021-01-29 | 2022-11-18 | 明峰医疗系统股份有限公司 | Measuring method and system of PET detector and computer readable storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7193216B2 (en) * | 2004-10-22 | 2007-03-20 | Oxford Instruments Analytical Oy | Method and circuit arrangement for compensating for rate dependent change of conversion factor in a drift-type radiation detector and a detector appliance |
| CN102783964A (en) * | 2012-07-02 | 2012-11-21 | 苏州瑞派宁科技有限公司 | Method and system for synchronizing single event list data in positron emission tomography (PET) |
| CN103336209A (en) * | 2013-06-28 | 2013-10-02 | 中国人民解放军海军潜艇学院 | Fault diagnosis device for nuclear radiation monitoring system |
| CN103983881A (en) * | 2014-06-06 | 2014-08-13 | 南华大学 | Fault diagnosis method and device for nuclear detector |
| CN104644204A (en) * | 2015-02-09 | 2015-05-27 | 中国船舶重工集团公司第七一九研究所 | Time correcting method and system for digital PET (positron emission tomography) |
-
2015
- 2015-07-29 CN CN201510458022.7A patent/CN105044759B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7193216B2 (en) * | 2004-10-22 | 2007-03-20 | Oxford Instruments Analytical Oy | Method and circuit arrangement for compensating for rate dependent change of conversion factor in a drift-type radiation detector and a detector appliance |
| CN102783964A (en) * | 2012-07-02 | 2012-11-21 | 苏州瑞派宁科技有限公司 | Method and system for synchronizing single event list data in positron emission tomography (PET) |
| CN103336209A (en) * | 2013-06-28 | 2013-10-02 | 中国人民解放军海军潜艇学院 | Fault diagnosis device for nuclear radiation monitoring system |
| CN103983881A (en) * | 2014-06-06 | 2014-08-13 | 南华大学 | Fault diagnosis method and device for nuclear detector |
| CN104644204A (en) * | 2015-02-09 | 2015-05-27 | 中国船舶重工集团公司第七一九研究所 | Time correcting method and system for digital PET (positron emission tomography) |
Non-Patent Citations (2)
| Title |
|---|
| PET成像中单事件列表式数据的时间同步方法研究;付道俊等;《舰船电子工程》;20150430;第35卷(第4期);第104-106页及124页 * |
| 全数字化小动物PET符合探测系统的设计与研究;刘伟等;《中华核医学与分子影像杂志》;20130630;第33卷(第3期);第223-225页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105044759A (en) | 2015-11-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105044759B (en) | A kind of state estimation of digital nuclear detector is with ensureing maintaining method and system | |
| Zhao et al. | Pattern recognition-based chillers fault detection method using Support Vector Data Description (SVDD) | |
| US20200081054A1 (en) | Power line issue diagnostic methods and apparatus using distributed analytics | |
| CN110598851A (en) | Time series data abnormity detection method fusing LSTM and GAN | |
| CN106529090A (en) | Evaluation method of reliability of aerospace electronic product | |
| US20160369777A1 (en) | System and method for detecting anomaly conditions of sensor attached devices | |
| CN116879662B (en) | Transformer fault detection method based on data analysis | |
| Jin et al. | Kuiper test and autoregressive model-based approach for wireless sensor network fault diagnosis | |
| CN108898311A (en) | A kind of data quality checking method towards intelligent distribution network repairing dispatching platform | |
| CN110221976B (en) | Quantitative evaluation method for quality of metering terminal software based on measurement technology | |
| CN107977301A (en) | Detection method, device, storage medium and the electronic equipment of unit exception | |
| CN109636110A (en) | A kind of method and device obtaining protective relaying device operating status | |
| US20050144537A1 (en) | Method to use a receiver operator characteristics curve for model comparison in machine condition monitoring | |
| CN113283113B (en) | Solar cell array power generation current prediction model training method, abnormality detection method, device and medium | |
| CN111080484A (en) | Method and device for monitoring abnormal data of power distribution network | |
| EP2882139A1 (en) | System and method for IT servers anomaly detection using incident consolidation | |
| CN106195247B (en) | A kind of control system of speed variator based on big data management mode | |
| CN108845285A (en) | Electric energy metering device detection method and system | |
| US20220091601A1 (en) | Method, system and computer program product for evaluation of energy consumption in industrial environments | |
| CN110188040A (en) | A kind of software platform for software systems fault detection and health state evaluation | |
| CN116186976A (en) | Verification method and verification system for accuracy of data collected by equipment platform sensor | |
| CN111382946B (en) | Autonomous evaluation method and system for health state of equipment and industrial internet equipment | |
| CN110455370B (en) | Flood-prevention drought-resisting remote measuring display system | |
| Jiang et al. | Anomaly Diagnosis Method and Condition Assessment of Power Metering Device Based on SSD Algorithm | |
| CN117472898B (en) | Fusion-based power distribution network abnormal data error correction method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |