CN106154814A - Time signal produces system, calibration control device and adjustment controlling means - Google Patents
Time signal produces system, calibration control device and adjustment controlling means Download PDFInfo
- Publication number
- CN106154814A CN106154814A CN201610576571.9A CN201610576571A CN106154814A CN 106154814 A CN106154814 A CN 106154814A CN 201610576571 A CN201610576571 A CN 201610576571A CN 106154814 A CN106154814 A CN 106154814A
- Authority
- CN
- China
- Prior art keywords
- value
- mapping function
- actual value
- historical data
- deviation value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G04—HOROLOGY
- G04F—TIME-INTERVAL MEASURING
- G04F5/00—Apparatus for producing preselected time intervals for use as timing standards
- G04F5/14—Apparatus for producing preselected time intervals for use as timing standards using atomic clocks
Abstract
Present disclose provides a kind of time signal and produce system, calibration control device and method.This calibration control device includes: data acquisition module, is used for obtaining multiple historical data, and each described historical data at least includes a history actual value and this history actual value history deviation value relative to a historical standard value；Deflection forecast module, for calculating the mapping function between described actual value and described deviation value and calculating the following deviation value corresponding with the one of prediction following actual value according to described mapping function according to described historical data；Corrective control module, for calibrating to produce the following actual value after calibrating to current actual value according to described following deviation value.The disclosure can carry out more excellent calibration and control.
Description
Technical field
It relates to Corrective control technical field, in particular to a kind of calibration control device, adjustment controlling means
And apply the time signal of this calibration control device to produce system.
Background technology
In many systems, needing to introduce calibration control program makes the performance of system be improved.Produce with time signal
As a example by raw system:
Within 1967, General Conference of Weights and Measures has redefined second a length of cesium-beam atomic clock Cs atom ground state being positioned on sea level
Two hyperfine energy levels in zero magnetic field transition radiation concussion 9192631770 durations in cycle, so far the mankind enter
Epoch atomic time are entered.Now, Perfect Time metering has become as the basis of national economy, its electric power, communicate, navigate and
The numerous areas such as automatically control and play key player, the most indispensable in national defense construction.International Bureau of Wieghts and Measurements is responsible for generation
And safeguard universal time (also known as UTC, Coordinated Universal Time(UTC)), universal time is by being distributed in more than 70 laboratory in the whole world
Nearly 500 atomic clocks are weighted averagely producing.Each regionality, the temporal frequency measurement unit of endemicity (calling this locality in the following text) and state
Border Weights and Measures Bureau exchanges periodically data, thus the unification of temporal frequency in realizing world wide.
Atomic clock is the frequency source of a kind of precision, produces Perfect Time according to its stable frequency characteristic.It is limited to electricity
The impact of the factor such as device and laboratory environment, the frequency time signal of its output is inevitably present random noise.Therefore,
International Bureau of Wieghts and Measurements minimizes noise by the weighted average of a large amount of atomic clocks, produces accurate, stable universal time.Local
Temporal frequency measurement unit Primary Reference universal time, as standard, maintains local zone time frequency standard.Due to the time this
The particularity of one physical quantity, the time only pass by just can be with accurate measurement, and the time of actual application is continuous signal.In view of
This reason, local time signal produces system to be needed by comparing its inclined with universal time (or other reference standards)
Difference, carries out the calibration of necessity in actual time when frequency signal produces.
But prior art produces the calibration control technology of system or similar system still for above-mentioned local time signal
There are part to be modified.
It should be noted that be only used for strengthening the reason of background of this disclosure in information disclosed in above-mentioned background section
Solve, therefore can include not constituting the information to prior art known to persons of ordinary skill in the art.
Summary of the invention
The purpose of the disclosure is provide a kind of calibration control device, adjustment controlling means and apply this calibration to control dress
The time signal put produces system, causes due to restriction and the defect of correlation technique for overcoming the most to a certain extent
One or more problems.
Other characteristics of the present invention and advantage will be apparent from by detailed description below, or partially by the present invention
Practice and acquistion.
According to an aspect of this disclosure, it is provided that a kind of calibration control device, including:
Data acquisition module, is used for obtaining multiple historical data, and each described historical data at least includes that a history is actual
Value and this history actual value are relative to the history deviation value of a historical standard value；
Deflection forecast module, for calculating the mapping between described actual value and described deviation value according to described historical data
Function also calculates the following deviation value corresponding with the one of prediction following actual value according to described mapping function；
Corrective control module, after calibrating to produce calibration current actual value according to described following deviation value
Following actual value.
In a kind of exemplary embodiment of the disclosure, described actual value is the temporal frequency of a frequency source output, described
Standard value is standard time frequency.
In a kind of exemplary embodiment of the disclosure, described actual value includes time dependent physical quantity, with frequency
Change physical quantity, with dimensional orientation change physical quantity in one or more.
In a kind of exemplary embodiment of the disclosure, described deflection forecast module includes:
Random packet unit, for described historical data is randomly divided into multiple data set, each described data set includes
The described historical data of equal number；
Function Fitting unit, for for each described data set, fitting function is to historical data therein to utilize one to treat
It is fitted obtaining described mapping function；
Weight calculation unit, for assessing the degree of fitting of each described mapping function, and according to the plan of each described mapping function
The weighted value that each described mapping function of right calculating is corresponding；
Deflection forecast unit, for calculating, by each described mapping function, the deviation that described following actual value is corresponding respectively
Value；
Weighted average unit, for the weighted value corresponding according to each described mapping function, to by each described mapping function
Calculated deviation value is weighted averagely obtaining a weighted mean；
Deviation statistics unit, described random packet unit, Function Fitting unit, deflection forecast unit, weight calculation unit
And the multiple described weighted mean of weighted average unit double counting；Described deviation statistics unit is used for counting probability of occurrence
High described weighted mean is as described following deviation value.
In a kind of exemplary embodiment of the disclosure, in described weight calculation unit, for each described mapping function,
Utilize the described historical data in all described data set in addition to the described data set that this mapping function is corresponding to assess this to reflect
Penetrate the degree of fitting of function.
According to an aspect of this disclosure, it is provided that a kind of adjustment controlling means, including:
S1. obtaining multiple historical data, each described historical data at least includes that a history actual value and this history are real
Actual value is relative to the history deviation value of a historical standard value；
S2. the mapping function between described actual value and described deviation value is calculated and according to described according to described historical data
Mapping function calculates the following deviation value corresponding with the one of prediction following actual value；
S3. calibrate to produce the following actual value after calibrating to current actual value according to described following deviation value.
In a kind of exemplary embodiment of the disclosure, described actual value is the temporal frequency of a frequency source output, described
Standard value is standard time frequency.
In a kind of exemplary embodiment of the disclosure, described actual value includes time dependent physical quantity, with frequency
Change physical quantity, with dimensional orientation change physical quantity in one or more.
In a kind of exemplary embodiment of the disclosure, described step S2 includes:
S21., described historical data is randomly divided into multiple data set, and each described data set includes the described of equal number
Historical data；
S22. for each described data set, historical data therein is fitted obtaining institute by fitting function to utilize one to treat
State mapping function；
S23. assess the degree of fitting of each described mapping function, and calculate according to the degree of fitting of each described mapping function each described
The weighted value that mapping function is corresponding；
The deviation value that described following actual value is corresponding is calculated the most respectively by each described mapping function；
S25. according to the weighted value that each described mapping function is corresponding, to by each calculated deviation of described mapping function
Value is weighted averagely obtaining a weighted mean；
S26. repeat the above steps S21～S25, obtains multiple described weighted mean, and it is the highest to count probability of occurrence
Described weighted mean as described following deviation value.
In a kind of exemplary embodiment of the disclosure, in described step S24, for each described mapping function, utilize
The described historical data in all described data set in addition to the described data set that this mapping function is corresponding assesses this mapping letter
The degree of fitting of number.
According to an aspect of this disclosure, it is provided that a kind of time signal produces system, including described in above-mentioned any one
Calibration control device.
It should be appreciated that it is only exemplary and explanatory, not that above general description and details hereinafter describe
The disclosure can be limited.
Accompanying drawing explanation
Accompanying drawing herein is merged in description and constitutes the part of this specification, it is shown that meet the enforcement of the disclosure
Example, and for explaining the principle of the disclosure together with description.It should be evident that the accompanying drawing in describing below is only the disclosure
Some embodiments, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to according to
These accompanying drawings obtain other accompanying drawing.
Fig. 1 schematically shows the block diagram of a kind of calibration control device in disclosure exemplary embodiment.
Fig. 2 schematically shows the schematic diagram of a kind of calibration control device in disclosure exemplary embodiment.
Fig. 3 schematically shows the block diagram of a kind of deflection forecast module in disclosure exemplary embodiment.
The relative weight of each mapping function in table 1 shown in Fig. 4.
The deviation Distribution value of multiple random packet program prediction shown in Fig. 5.
Hydrogen atomic clock shown in Fig. 6 is relative to the frequency of reference time scale.
The relative weight of each mapping function in table 2 shown in Fig. 7.
Following deviation value prediction scheme and the comparison of Kalman filter scheme in disclosure exemplary embodiment shown in Fig. 8
Result.
In disclosure exemplary embodiment shown in Fig. 9 and Figure 10, following deviation value prediction scheme is at different predicted times
The root-mean-square of the deviation value of the prediction under the conditions of length.
Figure 11 schematically shows a kind of adjustment controlling means flow chart in disclosure exemplary embodiment.
Figure 12 schematically shows the flow chart of step S2 in Figure 11.
Detailed description of the invention
It is described more fully with example embodiment referring now to accompanying drawing.But, example embodiment can be with multiple shape
Formula is implemented, and is not understood as limited to example set forth herein；On the contrary, it is provided that these embodiments make the disclosure will more
Fully and completely, and by the design of example embodiment those skilled in the art is conveyed to all sidedly.Described feature, knot
Structure or characteristic can be combined in one or more embodiment in any suitable manner.In the following description, it is provided that permitted
Many details thus provide fully understanding of embodiment of this disclosure.It will be appreciated, however, by one skilled in the art that can
That omits in described specific detail to put into practice the technical scheme of the disclosure is one or more, or can use other side
Method, constituent element, device, step etc..In other cases, be not shown in detail or describe known solution a presumptuous guest usurps the role of the host avoiding and
The each side making the disclosure thickens.
Additionally, accompanying drawing is only the schematic illustrations of the disclosure, it is not necessarily drawn to scale.Accompanying drawing mark identical in figure
Note represents same or similar part, thus will omit repetition thereof.Some block diagrams shown in accompanying drawing are merits
Energy entity, it is not necessary to must be corresponding with the most independent entity.Software form can be used to realize these merits
Energy entity, or in one or more hardware modules or integrated circuit, realize these functional entitys, or at heterogeneous networks and/or place
Reason device device and/or microcontroller device realize these functional entitys.
This example embodiment provide firstly a kind of calibration control device, can apply to time signal and produce system
Or other need the automatic control system that the following deviation of prediction is controlled, i.e. for described school in this example embodiment
The accurate range of application controlling device does not do particular determination.With reference to shown in Fig. 1, the calibration control device in this example embodiment
Data acquisition module, deflection forecast module and Corrective control module can be included.Wherein, data acquisition module may be used for obtaining
Taking multiple historical data, each described historical data at least includes that a history actual value and this history actual value are gone through relative to one
The history deviation value of history standard value.Deflection forecast module may be used for calculating described actual value with described according to described historical data
Mapping function between deviation value also calculates corresponding with the one of prediction following actual value following inclined according to described mapping function
Difference.After Corrective control module may be used for calibrating current actual value to produce calibration according to described following deviation value
Following actual value.As a example by controlling by the calibration for frequency time signal below, this example embodiment alignment is controlled
The each several part of device is described in detail.
Data acquisition module may be used for obtaining multiple historical data, and each described historical data at least includes that a history is real
Actual value and this history actual value are relative to the history deviation value of a historical standard value.For example:
With reference to shown in Fig. 2, as a example by the calibration for frequency time signal controls, described actual value can be a this locality
The temporal frequency data of frequency source output, for example, it is possible to be local atomic clock or the stable output conduct of other precison frequency source
Basis signal produces temporal frequency data.Described standard value can be standard time frequency data, such as, can be world standard
Time, Cesium atomic fountain clock or other can be as the benchmark of temporal frequency reference.Obtaining the frequency source output of history this locality
Temporal frequency and and historical standard temporal frequency after, both history deviation values can be calculated.But other examples in the disclosure
Property embodiment in, it is also possible to be obtaining described actual value and directly obtain both deviation values during standard value value, and then permissible
As above-mentioned history deviation value, i.e. this example embodiment is not entered for directly or indirectly obtaining history deviation value
Row particular determination；Depending on the time interval that deviation value obtains can be according to the target of application and practical situation, such as, it is chosen as 1 point
Clock, 1 hour, 1 day, 5 days or interval etc. At All Other Times；Additionally, this example embodiment is carry out as a example by the deviation of frequency
Illustrate, but in other exemplary embodiments of the disclosure, it is also possible to being deviation or other deviations of time, this also belongs to this
Disclosed protection domain.
Deflection forecast module may be used for calculating between described actual value and described deviation value according to described historical data
Mapping function also calculates the following deviation value corresponding with the one of prediction following actual value according to described mapping function.Citing and
Speech:
As a example by the calibration for frequency time signal controls, local zone time produce the target that system pursues be stable and
Accurately, therefore Corrective control scheme need to consider stability and the accuracy of the final temporal frequency exported.Rationally correct control
Time processed produces the accurate prediction that the basic condition of system is future time exemplary frequency deviation values, it was predicted that the essence of scheme is that estimation is former
Secondary clock is in following most probable variation tendency.Inevitable property due to noise, it was predicted that value be the theoretical estimated value of atomic clock (and
Non-measured value).The wide variety of prediction scheme of one that following deviation value is predicted be fitting of a polynomial prediction (as
Linear prediction), the program has that model is simple, amount of calculation is little, execution efficiency is high, can realize the advantages such as medium-and long-term forecasting, but its
Bad adaptability；When atomic clock is by noise jamming, it was predicted that error is relatively big, it is delayed to feed back.Another kind of wide variety of prediction side
Case is Kalman filter prediction, compared to former scheme, has in directions such as adaptabilities and promotes significantly, can be the most anti-
Answer the variation tendency of atomic clock, it was predicted that performance is better than former scheme；But its major defect is to realize medium-and long-term forecasting.This
Outward, all there is phase place or frequency hopping point, strong noise in data over-fitting problem, i.e. atomic clock historical data in above two scheme
Point is bigger on the impact of future anticipation performance.
Based on foregoing, this example embodiment also provide for a kind of new following deviation value prediction scheme, wherein adopt
By the distribution space of the following deviation value drop point of stochastic sampling construction of strategy prediction, the probability distribution of i.e. following atomic clock theoretical value
Space.Its sampling approach defines according to type and the historical data characteristic of atomic clock.For hydrogen atomic clock frequency difference data and caesium
Atomic clock time difference data generally can use linear model.Assume with the prediction of the historical data of hydrogen atomic clock a period of time following certain
The deviation Value Data of one time point, along with the increase of predicted time length, the prediction space of the following deviation value drop point of its prediction
Also will become big therewith, the uncertainty i.e. predicted gradually expands.For example, with reference to shown in Fig. 3, deflection forecast module is permissible
Including random packet unit, Function Fitting unit, weight calculation unit, deflection forecast unit, weighted average unit and deviation
Statistic unit etc..Wherein:
Random packet unit may be used for being randomly divided into described historical data multiple data set, each described data set bag
Include the described historical data of equal number.
For example, following deviation value is entered by N × m the historical data that can utilize atomic clock in this example embodiment
Row prediction, historical data can be randomly divided into N group, often m historical data of group by each stochastic sampling scheme.
Function Fitting unit is used against each described data set, and fitting function is to history number therein to utilize one to treat
Obtain described mapping function according to being fitted.
For example, can be according to atomic clock or other precison frequency source and standard time frequency in this example embodiment
Historical data characteristic, choose and treat fitting function (characteristic function), for matching historical data；Treat that fitting function can be to have
The function of different mathematic(al) representations, or there is family's function of same mathematical expression-form, or not there is mathematical expression
The vector of formula or matrix.In this example embodiment, treat that fitting function can be such asOr y
=a+bx, wherein x is independent variable, such as, can represent above-mentioned historical time frequency；Y is dependent variable, such as, can represent above-mentioned
History deviation value；A, b, c represent each term coefficient treating fitting function respectively.It will be readily appreciated, however, that show at other of the disclosure
In example embodiment, it is also possible to choose other functions as treating this is not done by fitting function, this exemplary embodiment special limit
Fixed.Selecting after fitting function, for each described data set, method of least square can used, to m history number therein
According to being fitted, determine each term coefficient treating in fitting function, obtain described mapping function；For all data sets, Zong Gongke
To obtain N number of described mapping function.
Weight calculation unit may be used for assessing the degree of fitting of each described mapping function, and according to each described mapping function
Degree of fitting calculates the weighted value that each described mapping function is corresponding.
For example, for each described mapping function, it is possible to use except described data set corresponding to this mapping function it
The outer described historical data in all described data set assesses the degree of fitting of this mapping function.For example, it is possible to by the 1st to the n-th-
Historical data in data set described in 1 data set and the (n+1)th to n-th substitutes into the n-th mapping function, to assess n-th
The degree of fitting of mapping function；Certainly, in other exemplary embodiments of the disclosure, it would however also be possible to employ other historical datas.Intend
Right can by distance, quantity of information, similarity and or the combination in any of above-mentioned three define, such as can use and a little arrive line
Distance, the root-mean-square of forecast error, the method digital simulation degree such as the statistical probability of deviation value of prediction.The degree of fitting of mapping function
The highest, then the weight that this mapping function is corresponding is the biggest, otherwise the least.Such as, the weight of the n-th mapping function can be designated as wn,
Then wnCan pass throughCalculate, wherein, giRepresent the degree of fitting of i-th mapping function bar.But art technology
Personnel are it is easily understood that the degree of fitting numerical procedure of other modes and weighted value calculation also belong to the guarantor of the disclosure
Protect in scope, this exemplary embodiment and be not limited thereto.
Deflection forecast unit may be used for calculating corresponding inclined of described following actual value by each described mapping function respectively
Difference.
For example, this example embodiment can being predicted, following actual value, i.e. future time frequency are xf, by xfGeneration
Enter each mapping function, the functional value y of correspondence can be calculatedp；The deviation value of the prediction that wherein i-th data set is corresponding can
To be designated as yip。
Weighted average unit may be used for the weighted value corresponding according to each described mapping function, to by each described mapping letter
The calculated deviation value of number is weighted averagely obtaining a weighted mean.
For example, in this example embodiment, N number of described mapping function calculated deviation value can be used to carry out
Weighted average, the weighted mean obtainedypIt is corresponding the predicting of this stochastic sampling scheme
Deviation value.
Deviation statistics unit, described random packet unit, Function Fitting unit, deflection forecast unit, weight calculation unit
And the multiple described weighted mean of weighted average unit double counting.Described deviation statistics unit is used for counting probability of occurrence
The highest described weighted mean is as described following deviation value.
For example, in this example embodiment, described random packet unit, Function Fitting unit, deflection forecast unit,
Weight calculation unit and weighted average unit can obtain more than 100, with double counting more than 100 times, the deviation value predicted
yp.Every time during random packet, due to the difference of packet, it is predicted that deviation value ypBy slightly difference；Therefore can collect every
Deviation value y predicted in secondary random packet schemepForm prediction space, the deviation value drop point predicted is carried out statistical
Analysis.Owing to N × m historical data comes from the output the most in the same time of same atomic clock, the intrinsic propesties that they characterize is close,
Therefore the probability distribution general approximate Gaussian distribution model in space is predicted.It is the highest that statistical separates out probability of occurrence in prediction space
The y of (occurrence number is most)p, it is designated as Yp, YpCan be as following xfThe deviation value in moment, the i.e. output of local zone time frequency source
Temporal frequency and the following deviation value of standard time frequency.
Corrective control module may be used for calibrating to produce calibration to current actual value according to described following deviation value
After following actual value.
For example, in this example embodiment, the short-term stability not affecting local zone time frequency can ensured
Under premise, input the calibrator quantity that is worth to of following deviation according to prediction to frequency adaptation device (jumping device as phase place is micro-), to tie up
Hold local zone time frequency constantly near standard time frequency.
This example embodiment have employed a kind of random pursive strategy and predict time dependent physical quantity, become with frequency
The physical quantity changed, with the physical quantity of dimensional orientation change, or the combination in any of above-mentioned three, by suitable deviation compensation
Realize the calibration following time, frequency, dimensional orientation etc..
In above-mentioned exemplary embodiment, it is to illustrate as a example by temporal frequency, but exemplary at other of the disclosure
In embodiment, described actual value includes time dependent physical quantity, with the physical quantity of frequency change, with dimensional orientation change
One or more in physical quantity, i.e. scheme in this example embodiment can be equally used for the calibration of other kinds of system
Control.Additionally, the calibration control device in above-mentioned exemplary embodiment can also include information display module, power supply module
Deng other, this is not done by this exemplary embodiment particular determination.By the technical scheme in this example embodiment, a side
Face, can promote degree of stability and the accuracy of echo signal, on the other hand, can improve the prediction energy for following deviation value
Power, another further aspect, abnormity point or the impact on following deviation value prediction of the strong noise point in historical data can be reduced.
Although additionally, it should be noted that being referred to some modules of the equipment for action executing in above-detailed
Or unit, but this division is the most enforceable.It practice, according to embodiment of the present disclosure, above-described two
Or more multimode or the feature of unit and function can embody in a module or unit.Otherwise, above-described
One module or the feature of unit and function can embody with Further Division for by multiple modules or unit.
Further, the technique effect of the following deviation value prediction scheme in above-mentioned example embodiment is also entered by inventor
Go experimental analysis.Such as, inventor uses the punctual hydrogen atomic clock data of about 1 year (UTC (NIM) is as reference time scale), enters
Go the experimental verification of three aspects, including the conceptual checking of basic skills, historical data medium frequency trip point to following deviation
The impact of value prediction and the potential advantages of medium-and long-term forecasting ability.Experiment results shows, in this example embodiment
Following deviation value prediction scheme in calibration control program can obtain the support of random chance theory of statistics；Compare Kalman filter
Ripple scheme, the following deviation value prediction scheme in this example embodiment has potential advantages at short-term forecast aspect of performance；Phase
Ratio deviation from linearity value prediction scheme, the following deviation value prediction scheme in this example embodiment has substantially in medium-and long-term forecasting
Advantage；Additionally, the following deviation value prediction scheme in this example embodiment is jumped in integrated forecasting ability, detection phase place or frequency
Height aspect has outstanding advantage.Specific experiment checking is as follows:
1) Forecasting Methodology example
A. description is tested: use 49 points of hydrogen atomic clock frequency as historical data (as shown in Figure 2), be randomly divided into 7
Group often 7 data points of group, in a random packet scheme, the coefficient of 7 mapping functions of matching is as shown in table 1.In table 1 often
The weight of individual mapping function is as shown in Figure 4.By historical data being re-started 100 random packet, form 100 different
Scheme (each scheme is similar with table 1, often group 7 data points, totally 7 mapping functions), deviation value y to each program predictionp
Being analyzed, result is as shown in Figure 5.
A table 1 stochastic sampling scheme under the conditions of relative smooth historical data
Note: the deviation value of historical data 49 days (MJD 57197-57245) prediction target (MJD=57246) is 1.3611
×10-12。
B. experiment conclusion: under the conditions of the historical data of relative smooth, the following deviation value in this example embodiment is pre-
Survey scheme is capable of predicting target.In each sampling scheme, the weight of 7 mapping functions is more or less the same.Repeat 100 times at random
Packet, it was predicted that deviation value be Gauss distribution, meet prediction theory model, demonstrate the following deviation in this example embodiment
The spatial distribution probabilistic model (Gauss distribution) of the following deviation value drop point of prediction in value prediction scheme.
2) the frequency hopping point impact on estimated performance
A. test description: as shown in Figure 6, when historical data exists time frequency hopping point, use experiment 1) in phase
The experiment parameter answered repeats experiment, it was predicted that the frequency values as MJD=57350, in the result of a random packet scheme such as table 2
Shown in, in table 2, the relative weighting of each mapping function is as shown in Figure 7.Use the following deviation value in this example embodiment pre-
The hydrogen atomic clock data of 1 year by a definite date are predicted by survey scheme and Kalman filter scheme, and its comparative result is as shown in Figure 8.
A table 2 stochastic prediction scheme under the conditions of historical data exists frequency agility
Note: historical data 49 days (MJD 57301-57349), wherein there occurs frequency agility on the 46th day, it was predicted that target
Deviation value is 1.9132 × 10-12。
B. experiment conclusion: when there is frequency hopping point in historical data, comprises frequency hopping point in random packet scheme
The error of mapping function relatively big, weight is less.Therefore, the following deviation value prediction scheme in this example embodiment has and subtracts
Little forecast error, the significant advantage of raising predictive ability.Future compared with Kalman filter scheme, in this example embodiment
Deviation value prediction scheme medium frequency trip point only has minor impact to future prediction error.
3) long-range predictive ability in promoting
A. description is tested: use the following deviation value prediction scheme in linear prediction method and this example embodiment to hydrogen
Atom clock frequency time rate data are predicted, two kinds of situations of high spot reviews:, there is not frequency hopping point in historical data,
Two, historical data exists frequency hopping point (as shown in Figure 6).The deviation value randomly choosing three random packet schemes is pre-
Surveying result, 40 data points closing on of extraction give a forecast the root mean square statistics of resultant error, and statistical result such as Fig. 9 (does not has frequency
Trip point) and Figure 10 (having a frequency hopping point) shown in.Wherein a length of 1～45 day of predicted time, each data point is spaced
It it is 1 day.
B. experiment conclusion: along with the increase of predicted time length, it was predicted that error is gradually increased, this point meets prediction reason
Opinion.The estimated performance of three random packet schemes is superior to linear prediction scheme.When historical data exists time frequency hopping
During point, the following deviation value prediction scheme compared in linear prediction method, this example embodiment has in middle long-range estimated performance
There is significant advantage.
Further, this example embodiment additionally provides a kind of adjustment controlling means.With reference to shown in Figure 11, described
Adjustment controlling means can be with following steps:
Step S1. obtain multiple historical datas, each described historical data at least can include a history actual value and
This history actual value is relative to the history deviation value of a historical standard value.
Step S2. calculates the mapping function between described actual value and described deviation value basis according to described historical data
Described mapping function calculates the following deviation value corresponding with the one of prediction following actual value.
Current actual value is calibrated to produce the following reality after calibrating by step S3. according to described following deviation value
Value.
In this example embodiment, described actual value is the temporal frequency of a frequency source output, and described standard value is mark
Quasi-temporal frequency.
In this example embodiment, described actual value can include time dependent physical quantity, with frequency change
Physical quantity, with dimensional orientation change physical quantity in one or more.
With reference to shown in Figure 12, in this example embodiment, described step S2 may include that
Described historical data is randomly divided into multiple data set by step S21., and each described data set can include identical number
The described historical data of amount.
Step S22. is for each described data set, and historical data therein is fitted by fitting function to utilize one to treat
To described mapping function.
Step S23. assesses the degree of fitting of each described mapping function, and calculates each according to the degree of fitting of each described mapping function
The weighted value that described mapping function is corresponding.
Step S24. calculates, by each described mapping function, the deviation value that described following actual value is corresponding respectively.
Step S25. is according to weighted value corresponding to each described mapping function, to calculated by each described mapping function
Deviation value is weighted averagely obtaining a weighted mean.
Step S26. repeat the above steps S21～S25, obtain multiple described weighted mean, and count probability of occurrence
The highest described weighted mean is as described following deviation value.
In this example embodiment, in described step S24, for each described mapping function, utilize except this mapping letter
The described historical data in all described data set outside the described data set that number is corresponding assesses the degree of fitting of this mapping function.
In above-mentioned adjustment controlling means, the detail of each step has been carried out in detail in corresponding calibration control device
Thin description, therefore here is omitted.
Further, it should be noted that above-mentioned accompanying drawing is only according to included by the method for exemplary embodiment of the invention
Process schematically illustrates rather than limits purpose.It can be readily appreciated that above-mentioned process step shown in the drawings is not intended that or limits
These time sequencings processed.It addition, be also easy to understand, these process can be such as either synchronously or asynchronously to hold in multiple modules
Row.
Those skilled in the art, after considering description and putting into practice invention disclosed herein, will readily occur to its of the disclosure
Its embodiment.The application is intended to any modification, purposes or the adaptations of the disclosure, these modification, purposes or
Person's adaptations is followed the general principle of the disclosure and includes the undocumented common knowledge in the art of the disclosure
Or conventional techniques means.Description and embodiments is considered only as exemplary, and the true scope of the disclosure and spirit are by appended
Claim is pointed out.
Claims (10)
1. a calibration control device, it is characterised in that including:
Data acquisition module, is used for obtaining multiple historical data, each described historical data at least include a history actual value with
And this history actual value is relative to the history deviation value of a historical standard value；
Deflection forecast module, for calculating the mapping function between described actual value and described deviation value according to described historical data
And calculate the following deviation value corresponding with the one of prediction following actual value according to described mapping function；
Corrective control module, for calibrating to produce the future after calibrating to current actual value according to described following deviation value
Actual value.
Calibration control device the most according to claim 1, it is characterised in that described actual value be one frequency source output time
Between frequency, described standard value is standard time frequency.
Calibration control device the most according to claim 1, it is characterised in that described actual value includes time dependent thing
Reason amount, with frequency change physical quantity, with dimensional orientation change physical quantity in one or more.
4. according to the calibration control device described in claims 1 to 3 any one, it is characterised in that described deflection forecast module
Including:
Random packet unit, for described historical data is randomly divided into multiple data set, each described data set includes identical
The described historical data of quantity；
Function Fitting unit, for for each described data set, historical data therein is carried out by fitting function to utilize one to treat
Matching obtains described mapping function；
Weight calculation unit, for assessing the degree of fitting of each described mapping function, and according to the degree of fitting of each described mapping function
Calculate the weighted value that each described mapping function is corresponding；
Deflection forecast unit, for calculating, by each described mapping function, the deviation value that described following actual value is corresponding respectively；
Weighted average unit, for the weighted value corresponding according to each described mapping function, calculates by each described mapping function
The deviation value obtained is weighted averagely obtaining a weighted mean；
Deviation statistics unit, described random packet unit, Function Fitting unit, deflection forecast unit, weight calculation unit and add
The multiple described weighted mean of weight average unit double counting；Described deviation statistics unit is the highest for counting probability of occurrence
Described weighted mean is as described following deviation value.
Calibration control device the most according to claim 4, it is characterised in that in described weight calculation unit, for each
Described mapping function, utilizes the described history in all described data set in addition to the described data set that this mapping function is corresponding
The degree of fitting of this mapping function of data assessment.
6. an adjustment controlling means, it is characterised in that including:
S1. obtaining multiple historical data, each described historical data at least includes a history actual value and this history actual value
History deviation value relative to a historical standard value；
S2. the mapping function between described actual value and described deviation value is calculated and according to described mapping according to described historical data
Function calculates the following deviation value corresponding with the one of prediction following actual value；
S3. calibrate to produce the following actual value after calibrating to current actual value according to described following deviation value.
Adjustment controlling means the most according to claim 6, it is characterised in that described actual value be one frequency source output time
Between frequency, described standard value is standard time frequency.
Adjustment controlling means the most according to claim 6, it is characterised in that described actual value includes time dependent thing
Reason amount, with frequency change physical quantity, with dimensional orientation change physical quantity in one or more.
9. according to the adjustment controlling means described in claim 6～8 any one, it is characterised in that described step S2 includes:
S21., described historical data is randomly divided into multiple data set, and each described data set includes the described history of equal number
Data；
S22. for each described data set, fitting function reflects described in being fitted historical data therein obtaining to utilize one to treat
Penetrate function；
S23. assess the degree of fitting of each described mapping function, and calculate each described mapping according to the degree of fitting of each described mapping function
The weighted value that function is corresponding；
The deviation value that described following actual value is corresponding is calculated the most respectively by each described mapping function；
S25. according to the weighted value that each described mapping function is corresponding, enter by each described mapping function calculated deviation value
Row weighted average obtains a weighted mean；
S26. repeat the above steps S21～S25, obtains multiple described weighted mean, and counts the institute that probability of occurrence is the highest
State weighted mean as described following deviation value.
10. a time signal produces system, it is characterised in that include according to the calibration described in Claims 1 to 5 any one
Control device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610576571.9A CN106154814B (en) | 2016-07-20 | 2016-07-20 | Time signal generation system, calibration control device and adjustment controlling means |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610576571.9A CN106154814B (en) | 2016-07-20 | 2016-07-20 | Time signal generation system, calibration control device and adjustment controlling means |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106154814A true CN106154814A (en) | 2016-11-23 |
| CN106154814B CN106154814B (en) | 2018-12-07 |
Family
ID=58059351
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610576571.9A Active CN106154814B (en) | 2016-07-20 | 2016-07-20 | Time signal generation system, calibration control device and adjustment controlling means |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106154814B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107193207A (en) * | 2017-07-28 | 2017-09-22 | 中国电子科技集团公司第五十四研究所 | A kind of multi-mode rubidium clock calibrating installation for scatter communication |
| CN108229747A (en) * | 2018-01-12 | 2018-06-29 | 中国计量科学研究院 | Adjustment controlling means, device and time signal generation system |
| CN108255047A (en) * | 2016-12-28 | 2018-07-06 | 德克萨斯仪器股份有限公司 | For the method and apparatus of the chip-scale atomic clock of magnetic compensation |
| WO2018166418A1 (en) * | 2017-03-17 | 2018-09-20 | 杭州海康威视数字技术股份有限公司 | Network timing method and device |
| CN110617916A (en) * | 2019-09-29 | 2019-12-27 | 歌尔股份有限公司 | Calibration method and device of air pressure sensor |
| CN110989326A (en) * | 2019-12-26 | 2020-04-10 | 中国计量科学研究院 | Local high-precision time frequency real-time comprehensive device |
| CN112363383A (en) * | 2020-10-26 | 2021-02-12 | 上海感探号信息科技有限公司 | Time axis unifying system and method |
| CN112737762A (en) * | 2020-12-25 | 2021-04-30 | 深圳深宝电器仪表有限公司 | Time calibration method and device, equipment to be calibrated and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101692163A (en) * | 2009-09-24 | 2010-04-07 | 中国计量科学研究院 | Method and system for remotely calibrating frequency standards |
| CN102004258A (en) * | 2010-09-09 | 2011-04-06 | 中国计量科学研究院 | Time frequency transfer method and receiver based on multiple global navigation satellite system (GNSS) integration |
| CN103926549A (en) * | 2014-04-25 | 2014-07-16 | 中国计量科学研究院 | Method and device for phase calibration |
| CN104199274A (en) * | 2014-09-24 | 2014-12-10 | 北京市计量检测科学研究院 | Pre-estimating method for frequency modification value of rubidium clock |
| CN105676627A (en) * | 2015-12-25 | 2016-06-15 | 中国科学院国家授时中心 | Time keeping system primary and standby main clock seamless switching system and method |
-
2016
- 2016-07-20 CN CN201610576571.9A patent/CN106154814B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101692163A (en) * | 2009-09-24 | 2010-04-07 | 中国计量科学研究院 | Method and system for remotely calibrating frequency standards |
| CN102004258A (en) * | 2010-09-09 | 2011-04-06 | 中国计量科学研究院 | Time frequency transfer method and receiver based on multiple global navigation satellite system (GNSS) integration |
| CN103926549A (en) * | 2014-04-25 | 2014-07-16 | 中国计量科学研究院 | Method and device for phase calibration |
| CN104199274A (en) * | 2014-09-24 | 2014-12-10 | 北京市计量检测科学研究院 | Pre-estimating method for frequency modification value of rubidium clock |
| CN105676627A (en) * | 2015-12-25 | 2016-06-15 | 中国科学院国家授时中心 | Time keeping system primary and standby main clock seamless switching system and method |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108255047A (en) * | 2016-12-28 | 2018-07-06 | 德克萨斯仪器股份有限公司 | For the method and apparatus of the chip-scale atomic clock of magnetic compensation |
| CN108255047B (en) * | 2016-12-28 | 2021-09-28 | 德克萨斯仪器股份有限公司 | Method and apparatus for magnetically compensated chip-scale atomic clocks |
| US10708033B2 (en) | 2017-03-17 | 2020-07-07 | Hangzhou Hikvision Digital Technology Co., Ltd. | Network time correction method and apparatus |
| WO2018166418A1 (en) * | 2017-03-17 | 2018-09-20 | 杭州海康威视数字技术股份有限公司 | Network timing method and device |
| CN107193207A (en) * | 2017-07-28 | 2017-09-22 | 中国电子科技集团公司第五十四研究所 | A kind of multi-mode rubidium clock calibrating installation for scatter communication |
| CN108229747B (en) * | 2018-01-12 | 2020-12-15 | 中国计量科学研究院 | Calibration control method, device and time signal generation system |
| CN108229747A (en) * | 2018-01-12 | 2018-06-29 | 中国计量科学研究院 | Adjustment controlling means, device and time signal generation system |
| CN110617916A (en) * | 2019-09-29 | 2019-12-27 | 歌尔股份有限公司 | Calibration method and device of air pressure sensor |
| CN110617916B (en) * | 2019-09-29 | 2021-08-31 | 潍坊歌尔微电子有限公司 | Calibration method and device of air pressure sensor |
| CN110989326A (en) * | 2019-12-26 | 2020-04-10 | 中国计量科学研究院 | Local high-precision time frequency real-time comprehensive device |
| CN112363383A (en) * | 2020-10-26 | 2021-02-12 | 上海感探号信息科技有限公司 | Time axis unifying system and method |
| CN112363383B (en) * | 2020-10-26 | 2022-04-05 | 上海感探号信息科技有限公司 | Time axis unifying system and method |
| CN112737762A (en) * | 2020-12-25 | 2021-04-30 | 深圳深宝电器仪表有限公司 | Time calibration method and device, equipment to be calibrated and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106154814B (en) | 2018-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106154814A (en) | Time signal produces system, calibration control device and adjustment controlling means | |
| CN105740203B (en) | The passive collaboration DF and location method of multisensor | |
| CN110837219B (en) | Virtual atomic clock system for monitoring entity atomic clock and working method | |
| Cao et al. | Switching Markov Gaussian models for dynamic power system inertia estimation | |
| Pinson et al. | From probabilistic forecasts to statistical scenarios of short‐term wind power production | |
| Fuss et al. | Investment under market and climate policy uncertainty | |
| De Giorgi et al. | Error analysis of short term wind power prediction models | |
| Pegoraro et al. | PMU and smart metering deployment for state estimation in active distribution grids | |
| CN108288231B (en) | method for evaluating influence of distributed photovoltaic access on load characteristics of power distribution station | |
| Davis et al. | A Kalman filter UTC (k) prediction and steering algorithm | |
| CN105868853A (en) | Method for predicting short-term wind power combination probability | |
| Bastola et al. | Sensitivity of the performance of a conceptual rainfall–runoff model to the temporal sampling of calibration data | |
| Sigrin et al. | Representation of the solar capacity value in the ReEDS capacity expansion model | |
| CN103970008A (en) | Timekeeping method based on crystal oscillator error compensation | |
| Tang et al. | Enhancement of distribution load modeling using statistical hybrid regression | |
| Calver et al. | Comparative analysis of statistical and catchment modelling approaches to river flood frequency estimation | |
| Wu et al. | A hybrid BAG-SA optimal approach to estimate energy demand of China | |
| CN105631556A (en) | Non-constant condition non-linear optimal parameter disturbance calculating method | |
| Xue et al. | Dynamic transverse correction method of middle and long term energy forecasting based on statistic of forecasting errors | |
| CN111610710B (en) | Local atomic time generation method based on cesium fountain clock and hydrogen atomic clock group | |
| Datta et al. | Electricity market price-spike classification in the smart grid | |
| CN104821854A (en) | Multidimensional spectrum sensing method for multiple main users based on random sets | |
| CN104537448A (en) | Method for improving state division of wind power Markov chain model based on longitudinal moments | |
| CN101483432A (en) | High precision GPS clock used in lightning time difference detection station | |
| CN114492923A (en) | Long-time scale power prediction method |
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 |